Read an Excerpt
CHAPTER 1
SITUATING
In nature and in human societies, the course of every action and event depends on the context in which it happens: that is the premise of this book. And its thesis is that scientific investigations of these activities are likewise best carried out in context. How could anyone think otherwise, one might ask, when the force of situations is a fact of daily experience? Yet in the culture of modern science — at least at its high end — true knowledge is taken to be what is true everywhere, in any context, or in no context. So the scientific procedures considered exemplary — experiment, hypothetico-deduction, randomized trials, statistical proofs of two-variable cause and effect — are those that eliminate or define away contexts as extraneous complications. A sharp social and epistemic boundary is thus drawn between the methods of science and the customary ways of everyday life. Desituating has become a signature characteristic of modern science, while science that does not control for context but accepts it is regarded as less than best practice.
These are, of course, ideals. In reality, much science does not evade or suppress situation in a quest for laws that transcend all particulars of place and time. Situating practices are common in the human and social sciences; and in field and environmental sciences, investigations in situ are no less valued than those performed ex situ, in labs and offices. The one creates conceptual space and investigative opportunities for the other, and mixed modes are not uncommon. The more one looks for science in situ, the more one finds. Situating, arguably, is no less characteristic of modern science than its opposite: just less visible and less valued. It is seen as the preliminary groundwork for real science, assembling useful facts but yielding no proofs: that is the modern default view. Part of my purpose in writing this book is to make science in situ a bit more visible, and to expose the disconnect between the powerful but restrictive ideals of desituating science and the reality of creative achievement in sciences carried out in place. I write from the context side, and for it: as a humanist historian whose business it is to recover the forgotten or misremembered contexts of historical actions and events, I have a brief to make for context.
My particular subject here is an intensive kind of situated science that I call resident science. The paradigmatic and most familiar resident practice is the "participant observation" of social anthropologists, who reside in a community for months or years, observing its customs and culture and taking part in a limited way in its activities. Participation is not a necessary element of resident science, however. In studies of animal communities, for example, observers cannot, or should not, take part. A virtue of resident science as a historical category is that it treats a range of human and natural sciences together that are customarily treated separately. And though it is a small subset of the more capacious and diverse category of situated science, resident science epitomizes the defining character of that larger group. I chose it for my subject because it is a well-defined and bounded topic and a manageable one for a shortish book on a very large subject.
What, then, are the defining features of resident science? One is its reliance on intensive personal observing in situ, by observers who in effect operate inside their objects of study. Resident science is strongly observational, and often open-ended and exploratory. It seeks generalities in patterns of observed particulars more than in deductions from abstract "laws" and theories. Resident observers treat the contexts and situations in which they and their subjects act, not as stage settings for actions, but as essential elements of phenomena. Resident science is coresident. It is situated, in that observers are themselves present in the situations and actions they observe. And it is situating, in that subjects are observed in the natural or social contexts in which they normally act. This double situatedness of cohabiting observers and observed is the essence of resident science as I understand it.
My current venture into resident science builds on the extensive inquiries into place and practice in science that have in the last thirty years become a defining issue in science studies. The principle themes of this body of scholarship are now widely familiar: that places are not neutral stages for scientific activities but directly affect how they are carried out and received; that all scientific knowledge is initially the product of some particular material and social locale; that some places — "truth spots" — can give credibility to knowledge claims; and that in becoming what everyone knows — the universal view from nowhere — knowledge must travel and transcend locality of place. These issues of place are important and fruitful. They are not my subject here, however. I am concerned solely with the relations of scientists and their subjects in their common contexts of coresidence: how these working relations are created and how they operate (or don't). Whereas concepts of place have typically emphasized geographical or spatiocultural features, residence is less a spatial than a relational and ecological instrument of thought. So I have kept the more familiar physical issues of place in the background and focused on the activities and experience of investigation — typically of human or animal communities — from the inside, and how coresidence leads not to the erasure of contexts but to their persistence in the facts and concepts of the science produced. Resident science stays in context and keeps context in. Its best stories are inside stories.
SITUATING SCIENCE
To see how situating is achieved, it is useful to look first at its more familiar opposite. Desituating has been accomplished by various means. Most simply (and superficially) it is achieved rhetorically, by stripping written accounts of situating particulars of who and where, in order to create the impression of timeless universality — of Nature speaking, not I or we. This is the process famously described by Bruno Latour and Steven Woolgar in their landmark book on making facts in laboratory science. More substantively, desituating is achieved conceptually, by embedding things or phenomena in an abstract framework of theory or modeling. Subjects may also be materially desituated, by physically relocating them to sequestered environments designed for the sole purpose of scientific analysis, in which disturbing particularities of context can be controlled or (more or less) eliminated. Laboratories, museums, gardens, aquaria, clinics, field stations, workshops, and offices are familiar types. This actual desituating is the favored method of modern science.
The iconic images of desituated nature are familiar: rats in mazes, small animals in cages, rows of maize in weeded and watered plots, fruit flies in jars, pure-strain bacteria in petri dishes; monkeys in vivaria or backyard caged colonies. In the human sciences there are native informants answering questionnaires on an ethnographer's veranda, kids in classrooms watched through one-way glass, or college undergraduates in psychology experiments — perhaps the closest thing there is to a "standard" human organism. And there are the objects extracted from their natural or social contexts: animal skins in cabinet drawers, dried plants pressed in folios, archaeological and ethnographic artifacts in cabinets, and excised tissues and organs in vitro that stand in for intact organisms in vivo. And let us not forget the models of population cycles that may or may not match the real things, or the reassuringly smooth curves of neoclassical economics — stand-ins for economies that in reality are given to "irrational" (yet perfectly natural) swerves, swoops, and dives.
The practical and epistemic advantages of desituating are clear enough. Things and phenomena are more precisely categorized and measured without the complications of real-world contexts, and more securely treated statistically and experimentally. Hypotheses of cause and effect are more easily generated and tested when single variables are isolated and analyzed without the uncertainties of particular situations. Observations and experiments that are reliably reproducible are more easily credited as true than are those that vary with circumstance. Although the noise of context is inherently part of the scientific signal, noisy signals are harder to sell as true knowledge. So science that is presented as clear signal without the context enjoys an epistemic free ride, while situated science must be continually justified and defended.
The social advantages of desituating are no less vital, given a society where professional reputation and advancement in a career depend on regular publications produced by communally sanctioned procedures. Desituating practices facilitate the steady production of results, enabling practitioners to publish and not perish; and because they stand up to critique and peer review, they are more effective in making and defending reputations. Situating practices are in contrast less reliably productive, and because they are carried out in person and in particular contexts, they will likely be harder to represent as objective. Results that are not presentable as definitive answers to specific hypotheses will be harder to get funded and published. It is little wonder, then, that desituating sciences are high on the totem pole of scientific prestige, while those that embrace situation are lower down.
All this is not to say that situating science has been abandoned in modernity: it obviously has not. Natural history has lost standing in the age of experiment, but it may well be pursued by more people and more productively now than ever before. In ecology and environmental sciences — the modern descendants of natural history — in situ observing is common practice. In some sciences it defines the mainstream, as in social anthropology and wildlife ecology; in others it animates significant subfields, as in animal behavior and small-group sociology and human relations. In many sciences recurring efforts (not always sustained) are made to resituate abstracted objects of study, as in evolutionary biology (see below) or social psychology, and even a science as devotedly abstract as economics.
There are good reasons why, despite their practical and epistemic disabilities, situating practices have retained a significant place in modern science. They restore contextual meanings that are too easily sacrificed to single-minded pursuit of analytic rigor. They keep scientists mindful of the fact that contingency and uncertainty are essential aspects of phenomena, and that analytic categories and cause-and-effect connections may not be so securely real as they seem to be when identified in unnatural isolation. Open-ended exploration revitalizes domains of theory that with time and use have become merely self-confirming, and enlarges conceptions of "data" that have become narrowed to just what can be used in rigorous "proofs" of cause and effect. Intensive study of particular cases in context challenges a methodology of deductive hypothesis testing that becomes unhealthily hegemonic, as the political scientist John Gerring has observed. Situating is a useful reminder that "just because an idea is true doesn't mean it can be proved. And just because an idea can be proved doesn't mean it's true." Observing phenomena in context is, in sum, not an imperfect substitute for proper science: in many instances it is proper science. Abstraction and universalism are not in themselves the problem: only the assumption that they are the best method for every subject and situation. There seems to be a growing recognition in science and in history of science that universal "laws" may not be the best way, and are certainly not the only way, to deal with complex phenomena. One size does not fit all.
Those who have explored the social and epistemic advantages of decontextualizing have generally not symmetrically inquired what is lost when phenomena are stripped of context. In his pathbreaking essay on Louis Pasteur's anthrax field trials, for example, Bruno Latour demonstrates how Pasteur captured anthrax for laboratory medicine by relocating the phenomenon from a pasture, where infection occurred contingently and mysteriously (it seemed), and farmers and veterinaries were the authorities; to a laboratory from which contingency was banished, and where infection was made reliably repeatable and explicable as the effect of a certain species of microbe. That indisputable cause and effect gave bacteriologists effective ownership of the anthrax disease: the one road to truth passed through their labs. It remained nonetheless true, however, that in the real world of herds and meadows, infection depends crucially on the particulars of environment and circumstance. And these cannot be properly studied by Pastorians in labs but only in pastures by those who know them and know how nature and human husbandry work. The question here is, How are objects of science kept in their pastures, so to speak, or returned to them from Paris labs for scientific study in situ?
There are degrees and varieties of situating practices. Rhetorical situating — I was there, and here's what I saw and did, believe me — was once the usual form of natural histories and is still occasionally used to create a sense of virtual witnessing. Situating is also accomplished by expanding the conceptual frames in which phenomena are set: genes reset in the whole machinery of gene expression, cells in nets of cell communication, microbial species in microbiomes of vast diversity, plant and animal species (including the human species) in whole ecosystems. A striking recent case is the resituating of the mathematical genetics of altruism and eusociality (think ants and termites, or us) into the larger frame of ecology and evolution by Edward O. Wilson and others. It is not just any species that develops eusocial traits, it turns out, but those few that have chanced to evolve ecologically to a point where eusocial tendencies can be amplified by natural selection. "[T]he origin of altruism cannot be deduced by aprioristic reasoning based on general models," Wilson writes. "It can, however, be revealed by reconstructing actual histories with empirical data."
Abstract models may similarly be given richer conceptual surrounds. Theoretical economics affords a good example in the logical game of prisoner's dilemma, in which two idealized "rational actors" kept incommunicado weigh the benefits and costs of cooperation or betrayal. This universalized model, admitted by all to be totally unrealistic, is made (modestly) more true to life by adding little stories with more actors and interactions, and particulars of social context. A reductive logic of decision making is thus resituated in imagined but plausible scenes of choice. In principle a story could be invented for any actual situation, though that would in the end transform a reductio ad absurdum into an elaboratio ad absurdum. One is reminded of Jorges Luis Borges's fable of an ancient college of cartographers, who created a map of their empire as big as the empire itself — a perfect but perfectly useless object that was finally abandoned, along with the college and all cartographers.
More concretely, the situations of experimental subjects can be made materially more like those of nature or society. Mimetic situating, one might call such practices. Minimally, "amenities" (exercise wheels, natural nesting materials) may be added to animal cages in the hope of making captives' behavior more like what it would be in free life. Vivaria and aquaria are made more naturalistic by adding species, or environments that afford cover from predators. Entire laboratories may be located in places whose climates and environments cannot be simulated: alpine labs for thin-air physiology, marine stations for littoral and intertidal environments, arctic stations for cold-climate science, or gardens in various climate zones for study of plant variation. Or actual situations in nature may be discovered or arranged that do in context what experiments do in labs — nature's experiments and what I've called "practices of place."
(Continues…)
Excerpted from "Inside Science"
by .
Copyright © 2019 The University of Chicago.
Excerpted by permission of The University of Chicago Press.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.
