Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Denaturing nature : philosophical and historical reflections on the artificial-natural distinction in… Inkpen, S. Andrew 2014

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2014_september_inkpen_stephen.pdf [ 3.69MB ]
Metadata
JSON: 24-1.0165947.json
JSON-LD: 24-1.0165947-ld.json
RDF/XML (Pretty): 24-1.0165947-rdf.xml
RDF/JSON: 24-1.0165947-rdf.json
Turtle: 24-1.0165947-turtle.txt
N-Triples: 24-1.0165947-rdf-ntriples.txt
Original Record: 24-1.0165947-source.json
Full Text
24-1.0165947-fulltext.txt
Citation
24-1.0165947.ris

Full Text

?DENATURING NATURE ?Philosophical and Historical Reflections on the Artificial-Natural Distinction in the Life Sciences ??by  ?S. Andrew Inkpen ?B.Sc. Saint Mary’s University 2008 ?????A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF ?Doctor of Philosophy ?in ?The Faculty of Graduate and Postdoctoral Studies ?(Philosophy) ?????THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) ?August 2014 ?© S. Andrew Inkpen, 2014 ??ABSTRACT ?? The philosopher Georges Canguilhem observed that the “physician’s thought and activity are incomprehensible without the concepts of the normal and the pathological.” I argue similarly regarding the biologist, only it is “the artificial” and “the natural” that are indispensable. Whether it is their objects of study, the methods used to investigate those objects, or even fellow researchers, biologists have habitually classified aspects of their discipline in a way that reflects the artificial-natural distinction. Why this way of classifying? What purpose does it serve? What principles guide its application? With what repercussions?  Tracing the transformation of these concepts through a series of historical episodes, I explore the reasons why biologists use this distinction and how it has influenced the practices and directions of certain biological fields—specifically evolutionary biology and ecology. The argument of this dissertation is that in biology decisions concerning the choice and evaluation of experimental and evidential practices, objects of study, and even assessments of scientific personas betray the artificial-natural distinction. Invocations of this distinction, like the normal-pathological, code normative contentions about proper biological practice. “The natural,” I argue, often functions as an epistemic authority.   The methodology I employ in this dissertation is conceptual and historical. The arguments marshalled are supported by conceptual-philosophical analysis, close readings of primary texts, and archival work. In the end I aim to problematize a set of widely invoked, but heterogeneously used, biological concepts. My arguments undermine a commonplace view according to which the collapse of the artificial-natural distinction is a prerequisite for contemporary science. This distinction is not, I argue, an outdated, pernicious relic; it has continued to exert a significant influence on scientific practice, and should not be ignored. 
?ii?PREFACE ?? This dissertation is original, unpublished, independent work by the author, S. Andrew Inkpen. A shorter version of Chapter 4 has been accepted for publication in the December 2014 issue of Endeavour as “‘The Art Itself is Nature’: Darwin, Domestic Varieties, and the Scientific Revolution.” ?????iii?TABLE OF CONTENTS LIST OF TABLES VI ............................................................................................................LIST OF FIGURES VII ..........................................................................................................ACKNOWLEDGEMENTS  VIII ...............................................................................................DEDICATION X ...................................................................................................................CHAPTER 1 1 .....................................................................................................................The Artificial-Natural Distinction An Introduction CHAPTER 2 32 ...................................................................................................................Art, Nature, and the Scientific Revolution Making Modern Experimental Science CHAPTER 3 57 ...................................................................................................................Denaturing Nature Disturbing Conditions and Classifications CHAPTER 4 84 ...................................................................................................................The Art Itself is Nature Darwin, Domestic Varieties and the Mechanical Philosophy CHAPTER 5 112 .................................................................................................................Selection in the State of Nature Darwin and Wallace on Domestication CHAPTER 6  159 ................................................................................................................Searching for What Nature has Wrought Dobzhansky and the “Natural” Experimental Fruit Fly ?ivABSTRACT .....................................................................................................................iiPREFACE .......................................................................................................................iiiTABLE OF CONTENTS ...................................................................................................ivviviiviiix??????CHAPTER 7  219 ................................................................................................................Communities, Natural Experiments, and “Soft” Science Jared Diamond’s Community Ecology CHAPTER 8  267 ................................................................................................................Conclusion REFERENCES  275 .............................................................................................................APPENDIX 1 305 ...............................................................................................................A Note on the Artificial-Natural Classifications of Persons APPENDIX 2 311 ...............................................................................................................What is a Natural Experiment? APPENDIX 3 316 ...............................................................................................................Diamond on Experimental Tradeoffs?v?LIST OF TABLES ???Table 7.1 Abridged table from Diamond 1986a comparing the advantages and disadvantages of different types of experiments in ecology......................251Table A3.1 Full table from Diamond 1986a comparing the advantages and  disadvantages of different types of experiments in ecology......................318?vi?LIST OF FIGURES ??Figure 2.1 The Mirror of Nature and the Image of Art..........................................43Figure 2.2 An Air Pump Used by Robert Boyle....................................................45Figure 4.1 An Elaborate Astronomical Clock in Strasbourg (c. 1875)..................89Figure 4.2 Clockmaker About to Fix a Clock........................................................92Figure 4.3 A Mechanical Toy.................................................................................94Figure 4.4 Victorian’s in Domesticated Nature, “Botanising”...............................95Figure 4.5 The Polish Pigeon or Polish Cock......................................................102Figure 5.1 Wallace’s Variation in the State of Nature..........................................128Figure 5.2 Darwin’s Variation in the State of Domesticity..................................129Figure 5.3 An Argus Pheasant Sitting on a Branch of a Tree..............................144Figure 6.1 The Mutant Type: “eyeless”...............................................................171Figure 6.2 The First Linear Linkage Map............................................................172Figure 6.3 A Postcard from Dobzhansky to L. C. Dunn......................................204Figure 7.1 Diamond’s Geometric Principles for Nature Reserve Design............227Figure 7.2 Distributions of Two Macropygia Cuckoo-Dove Species..................229Figure 7.3 Experimental Types............................................................................313?viiACKNOWLEDGEMENTS  ?? I owe a tremendous debt to two people in particular. First, my supervisor, John Beatty. Some time ago John took a risk on an eager biology undergraduate with little training in philosophy or history and who was, probably unbeknownst to John, predisposed to intellectual wandering. John has seen me through every step of graduate education and has been a good friend as well. The best parts of this project are a result of his insight. The worst parts are of my own doing. I owe him a debt I could not possibly repay. Without him, this project would never have begun.  Second, my partner in crime, Dani Hallet. She has unscrambled more drafts and heeded more complaining than anyone should be asked to bear, and through this has remained her marvelous and clever self. Her mark is indelibly left on the pages that follow. Without her, this project would never have concluded.  I have accrued less tremendous, but still significant, debts to a number of other people. My transition to philosophy can be blamed on Lisa Gannett and John MacKinnon at Saint Mary’s University. The former has become a close friend and has provided needed feedback on this project. The other members of my dissertation committee, Alan Richardson and Chris Stephens, were helpful throughout course work and with aspects of this project as well. Other members of the UBC philosophy faculty have been influential on the direction I have taken, in particular, Margaret Schabas and Sylvia Berryman. Finally, I would like to thank those members of my dissertation examination committee not mentioned above, Bob Brain (university examiner) and Garland Allen (external examiner).   Thanks to all my friends at UBC, particularly Chris French, Taylor Davis, Tyler DesRoches, Alirio Rosales, and Jiwon Byun. You have made graduate school both stimulating and enjoyable. Also to all members of the UBC philosophy of biology group.   Nissa Bell and Rhonda Janzen have an organizational wherewithal I have no hope of comprehending or approaching, but of which I have made great use and am sincerely thankful.  At Harvard University, I’d like to thank, in particular, Janet Browne, Sam Schweber Everett Mendelsohn, and the Modern Sciences Reading Group, the Losos Biological Laboratory, and the Incubator Series for reflecting on the material that formed the foundation of these chapters. Janet and Sam provided feedback on aspects of this dissertation dealing with the 19th century and Charles Darwin, and Janet provided me with a much-needed job. ?viii Thanks to support from the University of British Columbia, the Social Sciences Research Council of Canada, the Darwin Correspondence Project, and the American Philosophical Society. 
?ixDEDICATION ????????For Dani and Rook, Steve and Kathy 
?x??CHAPTER 1 ?The Artificial-Natural Distinction An Introduction ???????it is the duty of the naturalist [...] to separate artifice from Nature; and never to confound the animal with the slave, the beast of burden with the creature of God   Georges-Louis Leclerc, Comte de Buffon, 1749-88, Histoire Naturelle ?Domestic animals are abnormal, irregular, artificial; they are subject to varieties which never occur and never can occur in a state of nature: their very existence depends altogether on human care Alfred Russel Wallace, 1858, On the Tendency of Varieties ?we do not have any basis for seeing in the process of origin of mutations the result of the artificial influence of man [...] this process goes on just as regularly under natural conditions  Sergei Chetverikov, 1926, On Certain Aspects of the Evolutionary Process ?studies dealing with natural populations did far more to convince the “Darwinians” of the Mendelian nature of selectively important “natural” variation than either the “artificial” (as they called them) Drosophila mutations or mathematical calculations Ernst Mayr, 1973, The Recent Historiography of Genetics  1? The philosopher of medicine Georges Canguilhem observed that a “physician’s thought and activity are incomprehensible without the concepts of the normal and the pathological.”  I argue similarly regarding the biologist, only it is “the artificial” and “the 2?1 All emphases mine.1 Canguilhem 2008, 121.2natural” that are indispensable. Whether it is their objects of study, the methods they use to investigate those objects, or even fellow researchers, biologists have habitually classified aspects of their discipline in a way that reflects the artificial-natural distinction.  Why this 3way of classifying? What purpose does it serve? What principles guide its application? With what repercussions?  Tracing the transformation of these concepts through a series of historical episodes, I will explore the reasons why biologists use the artificial-natural distinction to think through and classify aspects of their science, as well as how these classifications have influenced the practices and directions of certain biological fields—in particular evolutionary biology and ecology. The argument of this dissertation is that in biology decisions concerning the choice and evaluation of objects, evidential and experimental practices, and even assessments of scientific personas betray the artificial-natural distinction. I will argue that invocations of the artificial-natural distinction, like the normal-pathological, code normative contentions about proper biological practice. “The natural,” I will argue, often functions as an epistemic authority. To the extent that we understand science to be knowledge-making through the investigation of phenomena according to specific methods and practices done by individuals, these scientific decisions, and thus the artificial-natural distinction framing them, shape ?2 The third category I highlight here, “researchers themselves,” may cause some head-scratching. I have two 3things in mind, as explained below. First is that biologists often categorize other biologists according to a dichotomy of naturalist versus experimentalist. This was especially common in the early decades of the 20th century and still happens today. Second, and relatedly, whether biologists draw the artificial-natural distinction or deny it, they will be viewed by the other camp as being naive, uninformed, cognitively or methodologically limited etc. Naturalists are more likely to draw this distinction than experimentalists. Chapter 3 and 6 consider these issues in more detail.   As an illustration of the extent to which this language commonly but is diversely used, a 2010 JSTOR search of the journals Ecology and Evolution showed that the artificial-natural distinction was used to categorize the following things (this from only the first 50 results): selection, environment, organisms, life, populations, ponds, calls, hybridization, survival, manipulation, species, habitats, settings, ecosystems, variation, community, strains, conditions, soil, forest, enemies, patterns, plants, nests, fertilization. biological science.  These arguments are made ostensively, by providing examples which 4show these points to be true of biologists and their science.  The argument of this dissertation undermines a commonplace view according to which the collapse of the artificial-natural distinction was essential for the rise of modern science, and furthermore, according to which the distinction currently lacks significance and relevance. I argue, instead, that this distinction is not merely an outdated, pernicious relic left-over from antiquated anxieties, but has continued to exert a significant influence on scientific practice, thus meriting philosophical and historical treatment.   The origins of the artificial-natural distinction are ancient, pre-dating the origins of biology. Its antiquity and embeddedness makes the invocation of these categories obvious in many cases, yet nonetheless consequential. Motivating this dissertation is the belief that sometimes the most fascinating aspects of scientific discourse and practice are concealed by their everyday familiarity. As 17th century scholar Edward Tayler observed, “ages, like men, rarely demand that the language they allow to think for them be brought to the surface of consciousness.”  My aim is to bring these concepts to the surface so as to better understand 5their significance.  The methodology I employ in this dissertation is an amalgamation of philosophical-conceptual analysis and intellectual history; I analyze manifestations of these concepts as words in historical sites.  The concepts with which I deal are “artificial” and “natural”; the 6?3 A note about my use of “shape” here. These are the sorts of decisions that we look to in order to distinguish the 4science of biology from other disciplines or other systematic knowledge-making activities. In this sense they shape what we mean by “biology.”  Tayler 1966, 25.5 See Hacking 2001, Chapter 3, for a discussion of the methodology draw on.6sites are case studies in the history of life sciences up to the present day. The evidence marshaled is supported by conceptual-philosophical analysis, close readings of primary texts, and archival work.  What follows are reflections on a set of commonly but heterogeneously used concepts. It is perhaps helpful to begin at the end, by summarizing seven general conclusions that will emerge throughout this dissertation: ?1. Invoking the artificial-natural distinction is common in biology and has been central to recalcitrant debates. Biologists, of the stripe considered here, have a habit of approaching their discipline and its puzzles with these two alternatives in mind. To steal a helpful metaphor from historian William Cronon, it serves as their “conceptual map” for understanding the world.  72. There is a standard historical narrative, which serves as a touchstone for discussions of the artificial-natural distinction in science, that relates the dissolution of the distinction to the rise of modern science. This narrative is highly problematic and cannot accommodate the life sciences. In the life sciences this distinction has remained central to its modern history. 3. The distinction is commonly invoked in biology because it piggybacks on a more fundamental concern about how human actions relate to nature’s “normal” development. A common sentiment encountered in biology is that investigations involving significant manipulations of nature are unwarranted—human manipulation “denatures nature” and misleads investigators with regard to their object of study. I argue that in many cases when this distinction is drawn, humans are treated as, what philosophers call, “disturbing conditions.”  4. The concepts “Nature” and “the natural” do not refer to a physical place, as one might prima facie expect, but to an ideal state free from human disturbance. 5. Artificial things are often treated as lacking the reality of natural things, and this is because the former are, in some way, human disturbed. “The natural” is often, for this ?4 Cronon 1995c.7reason, given a privileged epistemic position in the biological sciences: it is the “ultimate” object of knowledge. 6. As Canguilhem showed for “normal” and “pathological,” underlying the artificial-natural distinction are normative concerns about, for example, what objects and methods should be used to gain valid biological knowledge. The dichotomy influences scientific practice because “artificial” and “natural” are analogous to what moral philosophers call “thick concepts.” They are, as Bernard Williams said, both world-guided and action-guiding.  It follows that classifications made according to 8the artificial-natural dichotomy are not innocuous: how we classify things changes how we act towards them. 7. These concepts, as well as their relation to one another, have been in a process of continual reconceptualization. Phenomena, practices, etc., once deemed artificial may no longer be so called, which again changes how biologists act towards them.   ? This chapter has two further divisions. The first introduces recurrent themes, and summarizes my methodology and the theoretical resources I draw on and develop. The second provides a layout of the dissertation. ?A GENERAL PRIMER INTRODUCING RECURRENT THEMES AND THEORETICAL RESOURCES  The artificial-natural distinction is discussed in a variety of ways and in diverse literatures. As these have collectively influenced my thoughts in this dissertation, I will highlight those most pertinent and explain how the present study builds on and departs from them. This will help to situate what follows as well as highlight its novelty.  ?DEFINITIONAL ATTEMPTS AS IDEAL TYPES ?5 See Williams 1985, Chapter 8.8 There have been many attempts to provide necessary and sufficient conditions for distinguishing between artificial and natural things. All of these draw attention to human influence. The Oxford English Dictionary shows that going back to the 15th century, “artificial” has commonly referred to that which is made or constructed by human skill or involving human intervention. The natural, when opposed to the artificial, is that which is independent of human influence or contrivance.  Immediately problems arise. In what way 9does human influence stamp an object as artificial? What degree of human influence is required for something to count as artificial?   Aristotle answered these questions in terms of “natures”: natural objects have natures, that is, internal impulses to change in a goal-directed manner; artificial objects, say, beds or clothing, on the other hand, may be said to have goals only insofar as humans have thwarted nature’s goals to meet their own.  Artifacts do not have natures; there is nothing that has as 10its nature to develop into a coat. Whether Aristotle consistently adheres to this ontological categorization is a problem for his historical commentators. What is certain is that this straightforward classification of natural and artificial things did not endure. By 1740 the philosopher David Hume warned of the ambiguity of “nature,” it is the “repository of anything and everything.”  Likewise, the 19th century philosopher John Stuart Mill’s 11reflections make it clear that he not only concurred with Hume, but considered the artificial-natural relationship to be particularly problematic.  12?6 See Sagoff 2001 plus the literature below for further discussions.9 By “change” Aristotle has in mind change by place, growth, or other alteration; not just, that is, locational 10change. Schabas 2008, 72.11 Mill 1874. In 1965 Arthur Lovejoy documented 66 different meanings of nature! See Lovejoy 1965.12 Although Aristotle’s demarcation criteria did not endure, the idea that artificial things are made by humans and so lack ontological identity has been a common and important one, particularly in biology. To many 18th and 19th century naturalists, for example, domestic animals were ignored or even treated with scorn because they were created by man and thus not a part of the natural, that is, God’s order. Buffon expresses this sentiment in the epigraph.  In terms of more recent definitional attempts, the polymath Herbert Simon is at least prima facie more successful than Aristotle.  In The Sciences of the Artificial (1981) he 13discusses the relation between “natural worlds” and “artificial worlds,” demarcating between the objects of the natural and the design sciences (engineering, architecture, business, etc.). While many today would question his distinction between epistemic cultures, his definitional criteria provide ideal types (or an ideal space) useful for thinking about the artificial-natural distinction.  14?1. Artificial things are synthesized (though not always or usually with full forethought) by man. 2. Artificial things may imitate appearances in natural things while lacking, in one or many respects, the reality of the latter. 3. Artificial things can be characterized in terms of functions, goals, adaptation.  15?7 For another entertaining attempt, see Jacques Monod Chance and Necessity (1971), Chapter 1. The Nobel 13prize-winning biologist Monod introduces his famous discussion of chance and necessity in biology with a failed “imaginary experiment” to distinguish artificial from natural objects. Because these terms cannot be distinguished “objectively” he is compelled to consider the unique characteristics of living systems, in particular, their (apparent) purposiveness (Monod 1971, 13).  See Pauly 1987 and Chapter 7 of Daston and Galison 2007. But see Dear 2003 on differences between these 14epistemic cultures tracing back to differences between “natural philosophy” and “instrumentality” in ancient and early modern thought. For difficulties maintaining the distinction between natural objects and artifacts in terms of their functions, 15see Sperber 2007.4. Artificial things are often discussed, particularly while they are being designed, in terms of imperatives as well as descriptives (how they should be as well as how they are).  16? We will see these demarcation criteria invoked by biologists throughout the discussions that follow. The connection between the first and second should again bring to mind Aristotle and Buffon—things made lack the reality of things found. However, my aim in this dissertation is not to offer, much less evaluate, definitions of these concepts. Nor is it to survey uses, and extract minimal definitional criteria that are collectively and timelessly held amongst biologists. I am concerned with a different enterprise, to understand how and why biologists invoke this distinction as well as how it influences scientific practice. Instances of dispute or inconsistency often illuminate best what is at stake.  Aristotle’s and 17Simon’s criteria are, in other words, helpful only insofar as they help illuminate our understanding of biologists’ own reasoning. ?THE EPISTEMIC AUTHORITY OF NATURE?  The artificial-natural distinction is often discussed in a different way in environmental history and policy.  For example, when Jared Diamond, an influential ecologist and 18conservation biologist, was asked to reflect on the ends of restoration ecology, he drew attention to a widely held goal “to restore communities to their natural state.”  “The 19?8 Simon 1981, 8.16 Controversies “often involve disagreements over the reality of entities or propriety of practices whose 17existence or value are subsequently taken as unproblematic or settled” (Shapin and Shaffer 1985, 7). See also, Hacking 2001, 189, and Chapter 12; Cronon 1991, xix. A very good collection is Cronon 1995a; see especially Cronon 1995b.18 Diamond 1987b, 334. For examples of this sort of thinking throughout ecology see the provocative essay 19“Natural Landscapes, Natural Communities, and Natural Ecosystems” by Shrader-Frechette and McCoy (1995).problem,” he said, “is with the very idea expressed in the word ‘natural’. What do we mean by this word?”  The goal is arbitrary and yet potentially dangerous in its consequences. He 20continued, ?There are simply too many problems and ambiguities about such a goal. Over most of the surface of the earth there are no really ‘natural’, undisturbed communities left. If we arbitrarily define natural to mean ‘as first seen by Europeans’, we face major ambiguities. Even if we did define some nineteenth century condition as the ‘natural’ condition, there are many communities whose species composition we couldn’t restore even with an infinite amount of time and money.  21?The artificial-natural distinction is of consequence in this case because it dictates what we see as worthy of protection or as the goal of restoration, and can distract us from what is really feasible.  In other words, wider considerations about what counts as natural or what counts 22as artificial can influence practical decisions made by restoration ecologists. Is the rainforest more natural than the surrounding farmlands? Are indigenous song birds more natural than introduced starlings?  Human disturbed environments or organisms may be neither the goal 23at which to aim nor worthy of protection in their own right, but such decisions ultimately depend on, or are entangled with, how the artificial-natural distinction is drawn. ?9 Ibid., 33120 Ibid., 33421 It is also of course possible that the distinction is dictated by what we see as worthy of protection—we call 22something artificial, that is, because we don’t value it. Rather than consider this Euthyphro-ian dilemma a problem, let me just allow for an obvious feedback loop: we protect things because they are natural, and likewise they are natural because we want to protect them. This sort of problem is considered by Daston and Galison (2007) with regards to the virtue of objectivity.  See a recent issue of The Chronicle Review: The Chronicle of Higher Education for an interesting article 23about the “Battle for the Soul of Conservation” related to the first question (Nov. 15th, 2013). For the second, see Davis et al. 2011. Seen as both an important environmental category, and yet one that is conceptually problematic and liable to mislead, or worse, put to unwanted political use, many environmentalists and environmental historians try to salvage the category of “the natural” by endorsing a distinction made famous by Karl Marx between first nature (“original, prehuman nature”) and second nature (“the artificial nature that people erect atop first nature”).  These 24two concepts are subsumed under the larger blanket concept “Nature,” meaning, “all the powers [or laws of nature] existing in either the outer or the inner world and everything which takes place by means of those powers [or laws of nature].”  I flag this distinction for a 25particular purpose: when biologists draw the artificial-natural distinction, they are not denying that humans are a part of Nature in the broadest sense—that human actions are supernatural. They are drawing a distinction between first nature and second nature. The French political philosopher Georges Sorel meant to draw this same distinction but did so in an atemporal way.  He called this distinction “artificial nature” versus “natural nature.” 26When I speak of the distinction, I have his categories in mind.  Considerations like Diamond’s bring to light the normative aspects inherent in the concept of nature. Historians and philosophers in recent decades have drawn attention to the moral authority of nature: the ways in which “nature” and “the natural” have been, and still ?10 See Cronon 1991, xix. See also Marx 2005, 61 for the context.24 Mill 1874 [1969]. Or Darwin’s, “I mean by Nature, only the aggregate action and product of many natural 25laws, and by laws the sequence of events as ascertained by us” (Darwin 1872 edition of the Origin). This way is preferable because first nature and second nature make it sound as though the artificial-natural 26distinction has a temporal dimension, which I think it easier done without. After all, first nature is meant to evoke the feeling of the “untouched” New England wilderness, or at least for William Cronon’s influential usage.are, used as a cultural value or social norm.  Such is the basis for the famous naturalistic 27fallacy—the is-ought gap—in moral theory, which holds that it is a fallacy to assume that what is natural is for that reason good and dictates what we should do, how we should act. A cursory look at history or at contemporary political debates suggests that we have a particular knack for committing this fallacy.  28 While it is commonly acknowledged by historians and philosophers that “natural” carries normative weight, these accounts rarely observe, for example, that the phrase “in nature” is a common expression in both ecology and evolutionary biology which has descriptive and normative dimensions interlinked specifically in the context of knowledge-making. As do its antonyms (e.g., “artificial,” “laboratory”). I will argue in this dissertation that “the natural” is also an epistemic authority: just as decisions about which types of organisms or communities we should protect have been made in the broader context of the artificial-natural dichotomy in environmental thinking, so to have decisions about what is to count as a valid object or method in biology.  This argument is founded on the combination 29of two literatures: one devoted to understanding the moral authority of nature and the ?11 See Daston and Vidal 2004 and Bensaude-Vincent and Newman 2007 for excellent historical-philosophical 27discussions and examples. For similar issues in environmental science from a variety of perspectives, see Shrader-Frechette and McCoy 1995, Cronon 1996a, Ereshefsky 2007, Davis et al. 2011. See also Sagoff 2001 for these issues in the context of genetic engineering. See Haraway 1991 for (challenging) discussions about narratives of “nature” and their political-scientific intentions and consequences. For examples, see Sagoff 2001.28 I have found three studies particularly helpful in this regard. Two are about the 18th century (“Human 29experimentation in the eighteenth century: natural boundaries and valid testing,” by Londa Schiebinger, and “Attention and the Values of Nature in the Enlightenment” by Lorraine Daston). While these studies touch on the evaluation of objects and methods of study, they are primarily concerned with uncovering the sources of the moral authority of nature, and so are more useful as formal guides rather than helpful with regard to content. The final paper, by historian Graeme Gooday, entitled “‘Nature’ in the Laboratory,” considers mid-19th century rhetorical and pedagogical uses of “nature” in arguments over the legitimacy of the microscope. Thomas Henry Huxley and others argued that the microscope had privileged access to nature by drawing on discourse traditionally found in natural history: one only truly experiences nature by looking through the microscope. With its focus on valid methods and instruments of study this study comes the closest to the topic of this dissertation.complex mechanisms which imbue nature with value; the other directed at more straightforwardly philosophy of science problems about the evaluation of scientific objects, methods, and evidential, experimental, and explanatory practices.    Let us consider a few examples. When geneticist Theodosius Dobzhansky referred to another biologist’s work derisively as being based on merely “laboratory flies!,” he was both describing the fact that these flies were bred in captivity rather than taken directly from the field (a description that depends as we will see on how the artificial-natural distinction is drawn), and exploiting the negative connotations that accompanied the term “laboratory,” a place of artificiality and artifice, in the eyes of some naturalists.  To the consumate 30naturalist, what happens in the lab should stay in the lab—laboratories are not for studying natural things. As we will see in Chapter 6, particular views about what was “natural” and what was “artificial” played a role in structuring Dobzhansky’s research practice and his evaluations of others and their work.   The artificial-natural distinction also influences the choice and classification of research site and organism. When ecologist and co-founder of the Natural Capital Project, Gretchen Daily, began ornithological research as a graduate student she “only had eyes for the rain forest.” It alone seemed to offer a pristine and untouched natural world, perfect for a budding conservationist. But following a failed attempt at studying bird species in the Costa Rican rain forest, and unwilling to waste a summer of research, she turned to the farmland and surrounding countryside. And what she found, to her surprise, was that the “natural ?12 Dobzhansky to Ernst Mayr, March 29, 1975. William Provine has argued that Dobzhansky was wrong about 30the flies because Alfred Sturtevant, the biologist he is referring to, used flies Dobzhansky had sent him directly from the mountains east of Pasadena (Provine 1981). I believe that Dobzhansky is more likely referring to the fact that Sturtevant used free-living populations of D. melanogaster, which he thought involved problems of human interaction, and were unsuitable for this reason. See Chapter 6 and Dobzhansky 1939, 358.world had never left this man-made system.” So began her life-long project to change the way we understand conservation by restoring an anthropocentric perspective (to the great frustration of her 1980s forebears).  While she is now happy with the outcome, her initial 31disappointment in not being able to study undisturbed nature, was, and still is, a common reaction amongst conservation biologists—something she is attempting to change.  This is 32to say, her initial choice of research site was influenced by what she at the time believed to be artificial and natural and her present work aims to undermine these widespread beliefs.  The biologist, and now philosopher, Massimo Pigliucci recalls a similar story from his graduate student days. When he decided to switch research organism from the Connecticut native perennial, Lobelia, to the domesticated “weed,” and model organism, Arabidopsis thaliana, his dissertation committee was appalled. The latter, they insisted, “wasn’t a real plant.” Here, again, we see the interplay of Simon’s first and second criteria: this plant’s unnaturalness makes it less real. The resistance to Pigliucci’s decision is put forward against a backdrop of the artificial-natural dichotomy, in which “the natural” is understood to be preferable.  In both of these cases—that of Daily and of Pigliucci—it is not 33just material or scientific features of the organisms or sites in question that determine biologists’ evaluation of them, but also wider considerations about what counts as artificial or natural and what this entails. Some objects and places are thought to have scientific or epistemic value precisely because they are thought not to be human disturbed. ?13 We should not divorce the choices of research objects and sites made by conservation biologists and 31ecologists in the 1980s from the conviction, held among many college students in the 1960s and 70s, that everything artificial is bad, and everything natural is good (see Yankelovich 1972, 167-85. See also Rome 2013, 46.). The story and quotes come from Voosen 2013.32 The story and quotes from Pennisi 2000. See also Leonelli 2007, 212.33 Classifications according to the artificial-natural distinction extend beyond the objects, sites, methods, and other aspects of the biological “toolbox,” to styles of researchers themselves (or their typical personas). This case is much less straightforward (see Appendix 1). In the late-19th and early-20th centuries, for example, the terms “naturalist” and “experimentalist” were used to classify two sets of biologists.  The former studied nature 34“undisturbed,” the latter made use of experimental methods. The former tended to draw a distinction between the artificial and the natural and approached experimental methods with skepticism; many of the latter did not draw the distinction, and few saw it as impinging on the validity of experimental methods. Such classifications similarly conceal normative dimensions behind seemingly descriptive claims.  Each camp viewed the other as being 35naive, uninformed, or cognitively or methodologically limited; to classify one’s opponent as a member of the other category did not simply indicate different methodological commitments, but called into question her ability to produce valid research.   36 The part played by concepts of the artificial and natural in biology is analogous to that played by what Bernard Williams called “thick concepts.” Williams coined this term in ?14 See Allen 1979. 34 The case becomes even more complicated when these are used as historical categories as opposed to actors 35categories. For example, evolutionary ornithologist Ernst Mayr’s uses of the terms “naturalist” and “experimentalist” to classify sets of biologists’ contributions to the evolutionary synthesis movement of the 1930-40s, conceals a similarly normative dimension, as one acute commentator—the historian William Coleman—pointed out: according to Mayr, the naturalist had the right scientific character to properly recognize what the experimentalist could not. In all of these ways, concepts of the artificial and the natural are used to express epistemic value. Such uses 36are not a new development in science. We might return to Aristotle. On a common reading, medieval Aristotelianism respected an ontological distinction between artificial and natural objects, as I said above. Natural objects possessed natures that marked them as one type of thing rather than another. Artificial objects, lacking such natures, also lacked ontological identity. Since natural philosophy—our “science”—aimed to describe “natures,” it focused on natural objects and the sorts of things they ordinarily do. Artificial objects, lacking natures, were not the object of study, nor were they helpful for learning about the objects of study. In this case, the artificial-natural distinction corresponds to one between invalid and valid objects of study. Given that our intellectual landscape has changed considerably since this time, to what extent does a similar sort of thing happen in current science and why? See Dear 1995, 2005.the context of moral theorizing to describe concepts such as treacherous or courageous.  37These are both “world-guided” in the sense that they refer to features of the world to guide their application, but they are also “action-guiding” insofar as they give us reason to, or not to, act. This is what I mean when I say that the artificial-natural distinction conceals underlying descriptive and normative dimensions. Classifying something as “natural” is to evaluate its scientific validity, not just to describe its properties. When embedded in larger norms of scientific behavior, as we saw in the case of Daily and Pigliucci’s choice of research organism and site, the artificial-natural distinction can have a direct impact on scientific practices. In other words, as a thick concept that goes beyond superficial description, natural, in opposition to artificial, acts as an epistemic authority that makes itself known through real decisions and practices. ?REVERSED ROLES FOR NATURAL AND ARTIFICIAL  When “natural” is invoked in the context of environmental thinking it typically corresponds to “good.” This is not the case for either term in the artificial-natural distinction in science. The early-20th century experimentalist Jacques Loeb, for example, lauded the artificial and promoted a “trivialization of nature” as part of his engineering ideal for biology.  To Loeb, “the natural”—the nonhuman object of biological study—was just one 38amongst a set of possible states for investigation, and not one he considered particularly useful or interesting. What humans did in the laboratory was in no way “natural”—Loeb was ?15 Thick concepts are contrasted with “thin concepts,” such as “good” or “right.” These concepts are action-37guiding but not world-guided. See Williams 1985, Chapter 8. See also Jenkins 2006, Chapter 6, for a general discussion. As it is put by Philip Pauly (1987).38creating animals with bums on each end of their bodies—but this only added to the potential value of such work, as he saw it. For Loeb, “natural” was value-neutral at best.  In this dissertation, however, the cases I consider are biased towards those in which, when opposed to artificial, “natural” corresponds to good, valid, or advantageous, and “artificial,” bad, invalid, or disadvantageous. There are three deliberate reasons for this bias. First, I am primarily interested in the field practices of ecology and evolutionary biology, and in these disciplines, “natural” typically carries positive connotations. The easiest way to define the subfield of biology of which I’m interested is to follow an arbitrary division common to major biology departments, between (i) organismal and evolutionary biology and (ii) molecular and cellular biology. While this leaves some fields unspecified it is fair to say that I focus predominantly on biologists in the former category. Second, I want to complicate a trend in historical-philosophical research that focuses on the rise of laboratory science in the 20th century and emphasizes the accompanying unnaturalness of much of modern science.  It’s not that this history is misrepresentative, but that the problems it seeks to 39grapple with are not always exportable to the sciences of which I’m interested. To a laboratory experimentalist, many of the issues I consider may appear unintelligible, but that is the point. Third, and finally, I simply have insufficient space, time, and mental capacity to do justice to all subfields of biology. ??16 Perhaps the earliest example is Garland Allen’s Life Sciences in the Twentieth Century. We will consider this 39further in Chapter 6. The historical philosopher Hans-Jorg Rheinberger’s work is a good example of this focus on laboratory science. See Rheinberger 2010.THE COMMONPLACE VIEW: THE DISSOLUTION OF THE ARTIFICIAL-NATURAL DISTINCTION   Within historical literature about the origins of modern science, the artificial-natural distinction is often discussed in a different way still. A common narrative portrays the collapse of this distinction during the early modern period as central to the rise of modern experimental science. On this view, Aristotle’s distinction was nothing but a hindrance. It took the rise of the “mechanical” philosophy in the 17th century to put Aristotle’s distinction finally to rest and to found experimental science on its collapse. The recurrence of the distinction throughout the modern history of the life sciences should then be somewhat unexpected. Are the life sciences a “modern” science?  Furthermore, it is often suggested by early modern scholars that these concepts structured thought and practice in the ancient and early modern worlds more fundamentally and less vaguely than they do today.  In an intellectual world that was dominated by order, 40hierarchies, correspondences and analogies, understanding the relation between “the vast and all-comprehending Dominions of Art and Nature” was part and parcel of understanding other similar categories: particular-universal, conventional-natural, passive-active, primitive-progressive, pastoral-courtly, etc.  Discussions of the artificial-natural distinction were 41commonplace and radiated into many areas of social-political life. When Shakespeare toys with these concepts in The Winter’s Tale to produce both wit and social commentary he is assuming an audience which has been well-versed in how these concepts purportedly provide the world with a particular order. Historian Lorraine Daston writes that, compared to the ?17 See Tayler 1961. See Daston and Park 1998.40 Tayler 1961, 22ff. See also Foucault 2008, Chapter 5. See many examples throughout Daston and Vidal 2004 41of the importance of these categories during the enlightenment. early modern period, “in the current metaphysical vernacular, the artificial has been swallowed up by the natural.”  In other words, in the modern world, these categories have 42become less distinct and are thus, it often follows, less significant.  43 These historical perspectives combine with one that focuses on the rise of laboratory experimentalism to support a view of contemporary science as fundamentally artificial, in which the artificial-natural distinction lacks relevance.  This view grew out of a healthy 44reaction to earlier historical and philosophical literature that had a predilection for discovery in science. Although this is still a controversial topic in some quarters, many would now agree that much of current science is about creating, manufacturing or constructing phenomena. It is hard to deny this change in perspective when one considers the amount of human work necessary for designing and creating scientific apparatuses, producing and interpreting results, stabilizing results, as well as the rarity or even absence of the phenomena produced in the universe outside the laboratory.  Science no longer works on Aristotle’s 45undisturbed nature, it is said, but on “artificial phenomena.”  The epistemological 46advantages of the laboratory arise precisely from their not having to study natural objects as ?18 Daston 1998, 153.42 One way to push against this narrative would be to point to the work of Leo Marx (see Marx 2005). Marx 43shows that the idea of “nature,” along with its opposite “artificial,” were a powerful categories structuring early american social and political life (from the 18th century through to 1920 at least). The early puritans, for example, saw America as a “state of nature,” and took it as their religious duty to domesticate and civilize the savage land.  One edited volume which attempts to move past this view is Bensaude-Vincent and Newman’s The Artificial 44and the Natural: An Evolving Polarity (2007). Most of the essays, however, are about ancient and early modern periods, and there is nothing about the life sciences. None of the essays consider the concern that humans denature nature or, as I explain below, act as disturbing conditions, and thus it is only so helpful with regard to the following study. See Hacking 1983 for early summary. See Golinski 2005, and Sismondo 2010, Chapter 14, “The 45Unnaturalness of Science and Technology,” for recent summaries. See Hacking 1991, Kohler 1992, Knorr Cetina 1999; see Sismondo 2010, Chapter 14 for summary of 46literature.they are, where they are, or when they are.  As philosopher of physics Nancy Cartwright 47writes, “nature in the mundane world seems obstinately unruly. Outside the supervision of the laboratory [...] what happens in one instance is rarely a guide to what will happen in others.”  When it comes to the rise of model organisms as standing for wild ones, 48philosopher Sabina Leonelli argues that it is precisely their ability to transcend the artificial-natural boundary that is central to their scientific success.  According to this recently 49popular view of science, the artificial-natural distinction is eroding away as modern science moves further from the unruly nature it once studied.  50 Such perspectives might be contrasted with one arising from a smaller literature that often focuses on the field sciences and treats the artificial-natural distinction as both prevalent and important. Historian Robert Kohler concludes a study on the laboratory-field border—one of the modern manifestations of the artificial-natural distinction—with the following remark, “The line between nature and artifice can be blurred but not erased. Natural places cannot be made so lablike that they become unnatural; laboratories cannot be made so natural that they lose the artifice that gives them their power.”  Kohler’s remark 51should remind us that just because scholars of science understand the sharpness of the ?19 See Knorr Cetina 1992, 117, as just one example. See references in Sismondo 2010, Chapter 14.47 Cartwright 1999, 86.48 Leonelli 2007. Donna Haraway’s work on cyborgs also draws attention to the multiple things which call this 49distinction into question. See Haraway 1991. The question raised by this literature is whether science still studies “natural” objects and whether we are still 50primarily concerned with learning about the “natural” world rather than the engineered world. The microbiologist Carl Woese describes many of his molecular biologist colleagues as having lost the vision of understanding nature, instead caring only about technological advance (Woese 2004). Many science studies scholars agree with Woese’s assessment and see the phenomena studied as essentially artifactual, and thus also see current science as primarily focused on the artificial, not the natural—or, maybe more accurately, they understand the secret of modern science to have been the dissolution of these categories altogether. Kohler 2002, 308.51distinction to be somewhat dulled, and have rightly called attention to its permeability, we need not conclude that the distinction is unimportant, especially to field biologists themselves. Instead we should ask why the distinction remains important to these biologists, given that we see many problems with its conceptual coherence. For many working biologists, the artificial-natural distinction is real because it determines the conditions in which careers and decisions are made. What is real, as the Thomas Theorem conjectures, is real in its consequences. Moreover, the artificial-natural distinction might matter more to some sciences (and scientists) than to others, having to do with a complicated mix of history, ontology, and objectives. To many of the biologists we will consider, some parts of nature count as more natural and thus also as epistemically more valuable. ?METHODOLOGICAL ORIENTATION: HISTORICAL CONCEPTUAL ANALYSIS  As a philosopher, my analysis is fundamentally conceptual. This is not to suggest that I am interested in seeking timeless definitions for “natural” and “artificial.” Rather, I am interested in how changes in their relationship have influenced the practice of biology from the 19th century to the present. As I have said above, I am most interested in those places in which “natural” is endowed with epistemic authority, and in which “the artificial” is approached uneasily, as if it possessed an air of unreality, or even deceitfulness. The present study aims to uncover some of the sources of this characterization.   I am indebted to three methodological approaches that combine the insights of history and philosophy. The first is the most straightforward. Philosophers often take it as their primary objective to question the hidden assumptions that inform arguments. I see myself as ?20applying this methodological orientation to historical examples. In the following chapters I seek the hidden assumptions informing uses of the artificial-natural distinction.   52 The second approach is best summarized by philosopher Alan Richardson:    only by attending to very specific themes and argumentative moves can one tease out quite different connotations in philosophical terms that one might be tempted to think were used identically across great stretches of time. [...] One virtue of the historiography of philosophy that I offer is that it makes the past history of philosophy more interesting, more nuanced, and more subtle; it is not the hack work of hundreds of years of people trying to work out one or two mistaken ideas. A more important virtue is that it can make the future history of philosophy more interesting. Just now it is hard to say that epistemology and philosophy of science have quite enunciated the problem(s) of knowledge appropriate to our time and place. If the historical record can reveal five or fifteen rather than one or two accounts of the problem of knowledge, we have that many more epistemological resources with which to think. I can think of nothing for which we contemporary philosophers could be more grateful.  53?Richardson, and others like him, hold that detailed studies of history provide us with resources for thinking about our present state of affairs. The ways in which the artificial-natural distinction has been conceptualized in the past, for instance, might provide us with resources for thinking about the distinction at present. Of course, Richardson is discussing a more traditional philosophical problem—the problem of knowledge and the place of experience—but I see no reason why the artificial-natural distinction is not equally amenable ?21 The difference between what I’m doing and history of ideas is a matter of degree. I am both more concerned 52with context than traditional history of ideas (in this case, my methodology is more similar to intellectual history), and more concerned with hidden assumptions informing the use of concepts than their broad transformations over time. Harvard history professor Peter Gordon has a nice essay on his faculty page, which I have found useful, entitled, “what is intellectual history.” This is not published and I’m not allowed to cite it without permission. Richardson 2003, 66.53to this manner of philosophical interrogation. We might ask, how does our current distinction differ from how Aristotle drew that distinction? This may give rise to a new way of thinking about our current distinction, or simply prompt us consider the assumptions behind our distinction, regardless of its similarity to Aristotle’s. Surely even this second scenario is enough to warrant the attempt.  The final methodology informing this work is that used by philosophers such as Georges Canguilhem, Michel Foucault, and Ian Hacking, which often goes under the label “history of the present.” This method pushes one to seek the historical conditions that made possible our present organization of concepts.  One might think that in applying this method 54to problematic concepts like “child abuse” or “madness,” by telling their history and unearthing their preconditions, we would make them less problematic. My presentism, like Hacking and Foucault, has a different aim. This is to problematize a distinction that many would see as mundane, inconsequential, or possibly even dissolved. For example, consider the phrase “in nature.” It is a common clause appended to the end of sentences within, and titles of, biological papers. I argue that the phrase “in nature” does not refer to a physical place, but to an ideal state involving no human disturbance. This should cause a moment’s pause and a question, why is this phrase so common? With what sort of epistemic weight is evidence valid “in nature” endowed? Why is it important to flag something that seems self-evident? Why not also add “is true”? We can only make sense of uses of the artificial-natural distinction by paying attention to how these concepts are and have been used and by thinking about how the history of their usage obliges us, as its heir, to think in a certain way. ?22 Canguilhem 2008; Foucault 2007, Chapter 5; Hacking 2001, Chapter 3; See Rheinberger 2010, on 54Canguilhem.?THEORETICAL RESOURCE 1: HUMANS AS “DISTURBING CONDITIONS”  I introduce and develop two theoretical resources that help facilitate the discussions that follow. The first treats human as “disturbing conditions.” In general terms I see contestations over the artificial-natural distinction in modern biology as negotiations about how and why human actions count as interferences or disturbances in the systems of biological study under discussion. The Oxford English Dictionary defines “disturbance” as “interference with the regular or due course or continuance of any action or process.” Philosophers of science also have a convenient and general way of characterizing such interfering factors. They call them “disturbing conditions.” Throughout the recent history of biology I argue (subject to necessary refinement in Chapter 3) that humans themselves have been thought of as disturbing conditions, in a more narrow philosophical sense, and that this is reflected in the artificial-natural language commonly used. It must be kept in mind, however, that what counts as “human disturbance,” “artificial,” or “natural” is constantly being redefined. Throughout this dissertation we will look at some of the ways in which humans have been thought of as disturbing conditions. ?THEORETICAL RESOURCE 2: META-LEVEL CLASSIFICATION  The second resource I develop treats the artificial-natural distinction as a classification system. In the past, philosophers have often been concerned with “classification” in a narrow sense. For example, classification within the sciences, as when individuals are classified as tokens of a species type. Classification of this type is often ?23discussed in relation to natural kinds: whether there are more objective or natural ways of carving up the world and whether we can know what these ways are.  I am concerned with a 55much broader notion of “classification,” as discussed in science and technology studies literature, which is concerned about the kind of work that classification practices can do.  56This way of looking at classification focuses on how classifying things changes the way we act towards them and how particular ways of classifying become standardized, rather than whether there are “objective” classifications. We will look at how objects, broadly construed, in the life sciences are classified according to the artificial-natural distinction. Classifications, it will be argued, are consequential. They have the power to define disciplines.  ?RELATION TO RECENT PHILOSOPHY OF BIOLOGY  Biology differs from other sciences in a number of ways. In comparison to physicists,  biologists tend to be wary of grand theories and explanations, are not driven by the search for universal laws, use a set of heterogeneous methods and evaluation criteria, and are more concerned about the actual than the possible.  A friend once wrote to me, “who wants to be a 57philosopher of biology if biology is just physics applied to organisms?” Indeed. I think that the artificial-natural distinction has an epistemic relevance in biology that it does not have in some other sciences, particularly the so-called “harder” sciences. And because philosophers of biology have been particularly invested in arguments about what does and does not make biology unique, this dissertation is properly in the scope of philosophy of biology.  ?24 See David Hull’s paper in Douglas and Hull 1992.55 See Hacking 2001, Chapter 6; Bowker and Star 1999; Epstein 2007.56 See for example Beatty 1995, Keller 2002, Waters 2007, Woodward 2010.57 Yet, my goal, and this I think is different from a currently popular approach—one Michael Ruse characterized as “handmaiden to the science”—is not to aid the sciences or scientists so much as understand them and their historical development.  The concepts I 58focus on are not concepts used by biologists, such as “natural selection” or “fitness.” “Natural” and “artificial” are meta-level concepts that I argue are central to biology whether biologists recognize this or not (although some certainly do).   This study also impinges on a number of general themes in the philosophy of science. My concern to understand how the artificial-natural distinction influences practical decisions about choice of object, site, or method, is driven by a concern to understand how biologists evaluate evidence. Controversies within biology are often about “relative significance,” for example the relative significance of natural selection versus random genetic drift. What is at issue in such cases is “the extent of applicability of a theory or mechanism within a domain [...] not whether the theory or mechanism is the correct account of the domain.”  Rather than 59performing crucial experiments, biologists often therefore end up amalgamating evidence from a number of various sources, like mathematics, computer simulations, laboratory experiments, field experiments, natural experiments, etc. to support a theory of their choice—biologist Jonathan Losos compares this approach to detective work.  The question of how 60they evaluate different types and sources of evidence is a pressing one, and one that I think the artificial-natural distinction has some bearing on. ??25 Ruse 2008.58 See Beatty 1995, 229.59 See Losos 2007.60LAYOUT OF THE DISSERTATION  Rather than following a single culminating narrative, the chapters of this dissertation offer reflections on the artificial-natural distinction from different philosophical and historical perspectives. The second chapter, Art, Nature and the Scientific Revolution, begins with the touchstone for discussions of the artificial-natural distinction in science: the “scientific revolution.”  I ask, what is it about this time period that compels historians and 61philosophers to consider how the artificial-natural distinction was drawn? The answer will be that the ideological origins of modern science—in particular the origin of modern experimentalism—is regarded as intimately tied to the collapse of an ancient ontological distinction between nature and artifice. I will develop this point through a brief intellectual history of this time period. This history will (i) introduce different ways of thinking about the artificial-natural relationship—many of which are still with us today—and help conceptually clarify these relationships for future discussions, and (ii) provide a reason why the artificial-natural distinction is often ignored or regarded as a relic of ancient debates no longer relevant.  At the end of Chapter 2 I draw out three philosophical tropes that will reemerge throughout the thesis. These are phrased as questions and are as follows: (i) how does the breakdown of the distinction relate to the rise of manipulative approaches to the study of nature? (ii) how do different ways of thinking about art and nature relate to one another and continue to exist in tandem? (iii) how did the ontological collapse of this distinction make room for the distinction to be drawn on epistemic grounds? ?26 One exception is Bensaude-Vincent and Newman (2007).61 I begin Chapter 3, Denaturing Nature, with another intellectual history. I argue that, contrary to the narrative developed in Chapter 1, there has been a continuous invocation of the artificial-natural distinction in the biological sciences since the time of the scientific revolution when it was supposedly dissolved in all of science. This also shows that the artificial-natural distinction we draw on today is old, and that the problems surrounding it are long-standing. I provide a number of examples throughout the history of the life sciences where the artificial-natural distinction arises, saving the details of these cases for later chapters.  In the second half of Chapter 3 I will ask whether there is a common, general thread guiding the historical cases considered in the first half. I will suggest two. First, I explain what the artificial-natural distinction consists in, and argue that the distinction is a manifestation of the extent to which humans count as, what philosophers call, “disturbing conditions.” To draw the artificial-natural distinction is to acknowledge that humans are—for reasons which are context-dependent—disturbing conditions. Second, I explain that the artificial-natural distinction is used as a classification system; biologists use it widely to classify various aspects of their discipline. Drawing on literature in philosophy and science and technology studies, I explain that such classifications are not innocuous—they have practical consequences for the way in which we act towards the objects we classify. The chapters that follow aim to show some of these consequences, as they play out in the constituting of biological thought and practice.  Common to both the scientific revolution and the Darwinian revolution of the 19th century were discussions impinging on the relationship between art and nature. Was art a part ?27of nature? Could art be used as a model for nature? In Chapter 4, The Art Itself is Nature, I consider this intellectual congruence and suggest that it is more than just nominal. Charles Darwin and Asa Gray, for example, were well-aware of the 17th century debates which preceded them about the relation between art and nature through the works of such revered English writers as William Shakespeare and Thomas Browne. Furthermore, they used their understandings of these debates to inform and express their own thinking about the relation between artificial and natural selection. Darwin’s reliance on the art of breeding as a central element in his theory of natural selection cohered well with 17th century texts that argued for a breakdown of the art-nature distinction by claiming that Nature was simply the art of God. This chapter demonstrates one of the ways in which the artificial-natural distinction made its way from the literature we considered in Chapter 2 to the 19th century: the concepts are transferred, even though the exact concerns they are used to express change.  Building on the previous chapter, Chapter 5, Selection in the State of Nature, focuses more closely on how the artificial-natural distinction was understood and used by naturalists in the 19th century. I investigate a disagreement between Darwin and Alfred Russell Wallace with regard to Darwin’s famous analogy between artificial and natural selection. Though a proponent of Darwin’s views throughout his life, Wallace remained uneasy about Darwin’s principal analogy. I first show that the reason commonly provided by scholars for the disagreement is historically unfounded. I then approach their disagreement in a different light, by looking at how each of them reflected more broadly on the artificiality of domestication and on how these reflections interacted with the conclusions they drew about whether the process of domestication was representative of natural selection in the “state of ?28nature.” Wallace—drawing on a long naturalist tradition—upholds the artificial-natural distinction and thus opposes Darwin’s analogy, whereas Darwin’s discussions act to collapse the distinction itself. In effect, Darwin’s discussions render unanswerable questions such as, “is this a product of art or nature?” From these considerations, as well as those discussed in the last chapter, I conclude that different ways of conceiving of the artificial-natural distinction were used to think through and express deep seated differences of opinion about whether domesticated organisms could be used to shed light on natural species change.  Jumping ahead to the 1930s, Chapter 6, Searching for what Nature has Wrought, considers a period known as the “evolutionary synthesis.” Following a 1974 conference on the synthesis, the historian William Provine wrote to conference attendee Theodosius Dobzhansky asking him how it felt to be singled out as “the perpetrator of the great synthesis of the 1930’s and 40’s.” Dobzhansky’s work was seen to have fruitfully blurred the line between the outdoor, “natural” world of naturalists and the indoor, “artificial” world of experimentalists—a line that had impeded the synthesis of evolution and genetics. The naturalist Ernst Mayr wrote to him, “I still remember how delighted I was when I read your first papers [...] I exclaimed at the time ‘Finally a geneticist who talks sense!’” In what way did Dobzhansky talk sense, while his experimental colleagues did not? To Mayr, his work on natural populations proved that he had an acquaintance with “nature” which was generally lacking among geneticists. Although Dobzhansky was perhaps concerned with “the natural” in a way that his immediate predecessors were not, his experimental practice was fundamentally similar to theirs and manipulative laboratory work continued to play a dominant and necessary role throughout his career.  ?29 Questions abound: Why was “the natural” meaningful and useful for Dobzhansky? Why did Dobzhansky believe that a study of evolution based on natural populations would provide different and better results than one on artificial populations? Given that his practice remained fundamentally laboratory-based, in what way did he think his experimental practice met this “natural” goal? I address these questions by considering a crucial experimental move Dobzhansky made in the 1930s: a switch of research organism, from the—in his words—“domestic,” “artificial,” or “cosmopolitan” fruit fly, D. melanogaster, to its “wild,” “natural,” or “noble” cousin D. pseudoobscura. I argue for two theses: (i) that assumptions about what was artificial and what natural played a role in structuring Dobzhansky’s research practice and his evaluations of others and their work; (ii) Dobzhansky was skeptical of studies that were too artificial, and this skepticism was reinforced throughout his career, because he believed they promoted both misrepresentative and undesirable views of the nature and evolutionary consequences of genetic variation.  In the final episode, Chapter 7, Communities, Natural Experiments, and “Soft” Science, I consider 1980s community ecology and the use of natural experiments therein. In all its oxymoronic guise, the idea of the “natural” experiment has been a valuable one throughout the history of biology. It simultaneously possesses both the authority to speak about nature as well as the rigor associated with the rise of laboratory experimentalism. Thus, bringing the artificial-natural distinction back into the heart of the territory in which it was apparently collapsed. But, why exactly is it advantageous that an experiment be “performed” by nature rather than by human experimenters? And why has the artificial-natural distinction been a useful and commonplace way for biologists to classify the types of experimental work ?30done in biology? Community ecologists confronted these questions in the 1980s during a significant internal dispute about what it meant to do “good” science. This chapter will address these questions by focusing on ecologist Jared Diamond’s use of natural experiments and his widely cited and disputed reflections on proper experimental methodology. I will argue that, contra any narratives about the collapse of the distinction due to the rise of experimentalism, concepts of “the natural” and “the artificial” have been central to experimentation in ecology, particularly as a way to clarify and maintain methodological distance between ecology and the so-called “harder” sciences. This will also give us reason to think that the artificial-natural distinction in biology is here to stay.  The final chapter, Chapter 8, Conclusion, is a summary which showcases limitations and possible future research directions. 
?31??CHAPTER 2 ?Art, Nature, and the Scientific Revolution Making Modern Experimental Science ????????For even as in the business of life, a man’s disposition and secret workings of his mind and affections are better discovered when he is in trouble than at other times, so likewise the secrets of nature reveal themselves more readily under the vexations of art than when they go their own way. Francis Bacon, 1620, Novum Organum, Book One ? The touchstone for discussions of the artificial-natural distinction in science continues to be the scientific revolution.  In addressing the literature on this topic I seek to understand 62what it is about this time period that compels us to consider the artificial-natural distinction? I will (i) introduce different ways of thinking about the artificial-natural relationship and help clarify these relationships for future discussions, and (ii) provide a reason why in current discussions of science the artificial-natural distinction is ignored or seen as a relic of ?32 One discussion which goes beyond the scientific revolution is Bensaude-Vincent and Newman’s 2007 edited 62volume The Artificial and the Natural: An Evolving Polarity. This volume is the first historical treatment of the distinction and while it does consider more than the scientific revolution, such discussions are rare. Most of the volume is about ancient science and the scientific revolution period. antiquated, irrelevant debates. The business of the next chapter will be to complicate and undermine this idea.  63?THE MYTHIC ORIGINS OF MODERN SCIENCE  What was the scientific revolution? Historian Steven Shapin opens his recent introduction to the topic with the following provocative assertion: “There was no such thing as the Scientific Revolution, and this is a book about it.”  This is as good a characterization 64of the current state of affairs as any. It reflects the paradoxical place of revolutions in the history of science.   On the one hand, there is general agreement that between roughly the 16th and 18th centuries large-scale transformations occurred in discourses, practices, institutions, and attitudes that aimed at explaining, understanding and controlling nature.  It is believed that 65through many of these transformations aspects of our current scientific methodology become visible, if only in the broadest outlines. Three famous examples will make this point: (i) the term “law” was applied to regularities found in the natural world and nature itself thus came to be seen as lawful or law-governed; (ii) mathematics was applied liberally to describe phenomena throughout the natural world. The Italian astronomer Galileo Galilei, for ?33 One may well ask, why bother to tell the historical story at all? One reason is that the stories we tell about the 63past significantly influence how we view the present. This is especially so for deeply engrained stories, such as those about the “scientific revolution” and the origins of “modern science.” By reevaluating a theme central to these stories, I think we will come to see the present in a new light and in a way that raises a host of philosophical issues. This is not to say that the following is a Whig history—one that assesses the accomplishments of the past in terms of current standards. I will suggest rather that we alter our current standards through a study of their prehistory.  Shapin 1996, 1.64 See Cohen 1994 and the references therein. For classic, but not so old, discussions see, Westfall 1971; Kuhn 651976; Wilson 1995 as well as footnotes below.example, is famously paraphrased as saying that the book of nature was written in the language of mathematics; (iii) the merging of the mechanical arts with natural philosophy led to an increased use of experiments and instruments.   66 Yet, the once popular grand narratives about the scientific revolution, citing such canonical names as Nicolaus Copernicus, Johannes Kepler, Galileo, René Descartes, Francis Bacon, Robert Boyle, and Isaac Newton, no longer convince.  Early on, many studies of the 67scientific revolution sought to explain how today’s science came to be. The emphasis has shifted towards understanding what these transformations meant for those living in the 17th century, rather than how they do or do not contribute to the progress towards modern science. Hardly anyone today would disagree that the activities of early moderns were too various to be penned into a single, coherent story about the genesis and rise of modern science.   Furthermore, today’s science does not map onto early modern science in any simple intelligible fashion. Our “science” was spread out over natural philosophy, natural history, alchemy, mixed mathematics, and the mechanical arts, each with its own styles of reasoning, ?34 For these reasons, historians spent a good chunk of the 20th century sorting out both what was so “new”—as 66the early modern thinkers themselves put it—about these ways of studying nature, and the ever-increasingly complicated mishmash of social-cultural, scientific, technological and political causes that gave rise to such novelty. For a nice overview see Wilson 1995, Chapter 1. For laws see Ruby 1986; Milton 1998; for mathematics see Westfall 1971, Dear 1995; for experimentation see Shapin & Schaffer 1985. See Kuhn for 1976 for a classic discussion of the differences between the experimental and mathematical traditions and the changes occurring in each. Although it existed before, the phrase “Scientific Revolution” was given wider readership by Alexandre Koyré in 1939. See Cohen 1994 and Shapin 1996 for excellent discussions of these causes and their historiographies. See Lindberg and Westman 1990 and Osler 2000 for reappraisals of the themes that dominated early historical research surrounding the Scientific Revolution. See Park and Daston 2008 for the early modern use of “new.” See Dijksterhuis 1961; Koyré 1957; Burtt 1925; Butterfield 1950; Cohen 1994.67methods, and goals.  Our idea of the “renaissance man” is likely more a manifestation of our 68ignorance of how early modern areas of knowledge were sub-divided, than it is of the abilities of early moderns to transcend disciplines.  As with many other grand narratives, such as those of the Enlightenment or Modernism, the scientific revolution has taken on a mythic quality.  In spite of, or perhaps 69because of, its paradoxical nature, the scientific revolution still enters into academic and popular discussions of the rise of modern science and in doing so reflects what we take to be important about science and influences how we think about current science.  This “myth,” as 70historians Katherine Park and Lorraine Daston put it, “expresses in condensed and sometimes emblematic form themes too deep to be unsettled by mere facts, however plentiful and persuasive,” and “holds both proponents and opponents in its thrall.”  As an origin myth, it 71acts to reinforce our current conceptions of the essence of science and, if only implicitly, what marks science from other intellectual endeavors and cultural practices.  The 72philosopher Evelyn Fox Keller wrote that “Unexamined myths, wherever they survive, have a subterranean potency; they affect our thinking in ways we are not aware of, and to the ?35 See Park and Daston 2008, 3-5, for a discussion of the different knowledge-producing bodies/practices/68disciplines in the early modern period. See Willson 1995, 11ff for a discussion of how these different disciplines entered into university curriculum. See Hacking 2001 for a discussion of different “styles of reasoning.” Each discipline typically dealt with the following: natural philosophy with organic and physical change, natural history described nature’s particulars often for medical purposes, alchemy with distillation and the transmutation of metals, mixed mathematics with geometry, astronomy, optics, and harmonics, and the mechanical arts with architecture, engineering, clock-making, agriculture, breeding, horticulture. See Park and Daston 2008. See also Wilson 1995, who also refers to the scientific revolution as mythic.69 Peter Dear’s recent Revolutionizing the Sciences (2009) is another example. See for a reflection on this issue 70Cunningham and Williams 1993. Park and Daston 2008, 15. See also a great discussion by Cunningham and Williams (1993) about the ‘big 71picture’ version of the scientific revolution, and why this is inadequate in light of our current historiographical aims and more recent history of science. For some thoughts on the function of myths see Kirk 1970.72extent that we lack awareness, our capacity to resist their influence is undermined.”  While 73the scientific revolution is hardly unexamined, there are aspects of it that continue to influence our thinking in mythic ways.   Here I characterize one particular aspect of the mythology. By rehearsing the myth my point is not to reinforce a just-so story about the early modern period, but to tell an intentionally selective story that illuminates what many have found to be a particularly compelling feature.  Doing so will put us in a better position to reflect philosophically on the 74extent to which this aspect of the making of modern science is indeed true of science and to assess the influence it has had on our thinking about science.  The aspect of the scientific revolution I want to analyze is the “breakdown” of the artificial-natural distinction. This breakdown enters into discussions of early modern science in several ways. Lorraine Daston writes, “The insight that art was merely a part of nature, governed by the same rules was of cardinal importance to the early stages of the Scientific Revolution [...] The central precept of the mechanical philosophy, that nature was actually composed of microscopic machines, presupposed the destruction of the boundary between works of nature and handiwork.” Steven Shapin writes, “Unless it was accepted that there was a basic similarity between the products of nature and those of human artifice, experimental manipulations with machines could not stand for how things were in nature [...] there could be no secure inference from what experimental apparatus made manifest to the natural order of things.” Ian Hacking says that Robert Boyle’s air-pump “inaugurated ?36 Keller 1985, 76.73 For a much better introduction to the origins of modern science, that is pluralistic both in terms of episodes, 74time periods, sciences, and larger historiographic themes see Bowler and Morus 2005.laboratory science. Before the air pump, one aimed at solving the phenomena given in nature [...] Afterwards a new kind of science answered to [...] phenomena that fleetingly exist by artifice.” Today, Hacking continues, “the category of matters of fact, answering to artificially produced phenomena, is sacrosanct.” Richard Westfall writes, “It is difficult to speak of natural phenomena in relation to [Newton's experiments ...] the design of the experiment determined that nature had no choice but to answer 'yes' or 'no.' [...] By the end of the 17th century, the scientific revolution had forged an instrument of investigation that it has wielded ever since.” And finally, H. Floris Cohen writes that “Every interested thinker in the 17th century had to grapple with the distinction between ‘natural’ and ‘artificially produced’ nature and the root issue of how one was related to the other.”   75 We can see that the collapse of the artificial-natural distinction is an oft-cited idea in the secondary literature. Let me move to the history of the breakdown of this distinction and its relevance for science in the early modern period. ?THE ONTOLOGICAL BREAKDOWN OF THE ARTIFICIAL-NATURAL DISTINCTION  The early moderns responsible for the scientific revolution inherited the distinction between the artificial and the natural from a scholastic understanding of natural philosophy. The text widely cited, in this regard, throughout the 17th century was Aristotle’s treatise on Physics.  Consider the following passage, 76??37 Daston 1988, 464; Shapin 1996, 97-8; Hacking 1991, 235 and 240; Westfall 1971, 116; Cohen 1994, 188 75(respectively) Daston and Park 1998, 263-4. See Close 1969. 76Of the things that exist, some exist by nature, some from other causes. By nature the animals and their parts exist, and the plants and simple bodies (earth, fire, air, water) - for we say that these and the like exist by nature. All the things mentioned plainly differ from things which are not constituted by nature. For each of them has within itself a principle of motion and of stationariness (in respect of place, or of growth and decrease, or by way of alteration). On the other hand, a bed and a coat and anything else of that sort, qua receiving those designations - i.e. in so far as they are products of art - have no innate impulse to change.  77?For Aristotle, distinguishing between art and nature meant distinguishing between external sources of change and internal sources of change.  The natural object contained within itself 78a principle of motion to change or develop in accordance with its nature. The nature of an acorn is to develop into an oak tree. The acorn, in other words, does this regardless of any external influence. The “natures” of natural objects were to be determined, according to this tradition, by their ends.  79 The artificial object, in contrast, does not have an internal nature or essence. Its source of change is imposed externally by us—what Aristotle calls accidental, rather than substantial. There is nothing that has as its nature to develop into a coat or a bed. We make these things by using, combining and in some cases thwarting, the natures of natural objects. Artificial objects are thus “necessarily posterior to and parasitic upon natural objects.”  80Artificial objects, since they lacked natures, also lacked a clear ontological identity. What kind of thing an artifact was could not be specified by its nature, since it did not have one. ?38 Aristotle Physics in Works, vol. 1, 329; quoted in Daston and Park 1998, 264.77 See discussions of this passage as compared to Aristotle’s work in other areas in Schiefsky 2007.78 See Dear 1995, 153-5.79 Daston and Park 1998, 264.80The artificial-natural distinction was an ontological distinction; artificial and natural objects differed in kind, not in degree.  81 The understanding of “natural philosophy” that dominated in medieval universities prior to the scientific revolution, was Aristotelian in the above sense. It was, according to historian Peter Dear, a contemplative endeavor, aimed at understanding the natural world in terms of how things happened most of the time or as per their usual course.  This 82contemplative endeavor can be juxtaposed with how “Art” was understood in classical antiquity and preserved through the renaissance: art was “any rationally organized activity which has a practical rather than a speculative end (e.g. rhetoric, carpentry, politics, painting, drama), and as the system of theoretical knowledge or the intellectual expertise or the technical proficiency which such activities presuppose.”  Art was practical and instrumental.  83 In contrast to the instrumentality of art, natural philosophy promoted an understanding that was intricately tied to the Aristotelian idea of “natures.” It aimed at providing an account of the natural world in terms of the natures of objects. This difference lay at the heart of distinctions in natural philosophy between, for instance, violent motion and natural motion. Smoke rises, it was supposed, because it is composed primarily of the element air whose natural place is above the earth, which in turn falls to its natural place at the centre of the universe. Since the Aristotelians were concerned with understanding the natural motion of physical bodies, the study of human artifacts—which operated according to ?39 Because of the entrenchment of this view, historians call the opposition of art and nature during the 81renaissance a “habit of the understanding” or “conceptual reflex.” See Park and Daston 2006, and also Tayler 1964 and Close 1969. See Dear 1990, 680ff; Dear 2005, 393-4; Hansen 1986, 129-30; Daston 1991, 340-1; Shapin 1996, 30-46; 82Newman 2004; Reill 2008, 26; Merchant 1980. Close 1969, 467.83violent motion—was of less value.  The launched projectile, which disobeyed its natural 84inclination to fall, told of what humans could do rather than what happened by nature. One 17th century thinker, reflecting on this tradition and its particular distinction between types of motion, wrote, this “seems much to favour the Opinion of the Naturists, since ‘tis grounded upon a Supposition, that what is violent, is, as such, contrary to Nature.”  Similarly 85machines and other artificial constructions functioned because humans had constrained their natural motion.  In order for these types of motion to fall under the purview of natural 86philosophy, its epistemological and ontological foundations had to be reconceptualized.  87 The ontological opposition between art and nature began to break down as the goal of natural philosophical knowledge changed from one of contemplation to one of instrumentalism.  This gradual placing of natural knowledge in the service of the 88improvement of arts and trades was brought on, as historians have pointed out, by a plethora of factors present by the early modern period, including the increasing role that machines—such as large clocks erected in town centers—played in day-to-day human affairs, the wide-spread beliefs that we can securely know only what we construct by hand and that the authority of ancient texts was secondary to the “authority of nature” known by personal experience, influential observations of various heavenly phenomena through the telescope and other discoveries using mechanical instruments, and political instability that called into ?40 Violent motion is also called “constrained” or “artificial” motion.84 Boyle [1686] 1999, 507.85 DesChene 2007.86 Also required was a new understanding of “experience” itself. See Dear 2005. Add a note here about this.87 Daston and Park (1998) examine how this change was preceded by challenges to the art-nature distinction 88through the Wunderkammern. See Daston 1995, 40ff and especially Dear 2005, 2006. As shown by Sylvia Berryman (2009; see especially the Appendix), the mechanical philosophers were not reacting to or rejecting ancient philosophy directly—in fact, many of the ways in which “mechanical” was used in the early modern period have their roots in antiquity.question the established uses of natural philosophical knowledge, and created a climate in which political authority could be symbolized by the sovereignty’s dominion over nature.  89An early example of this instrumentalism combined with a reexamination of the art-nature distinction can be found in the work of philosopher Francis Bacon.   Bacon opposed the glorification of Aristotelian natural philosophy at the expense of the mechanical arts. He accused his philosophical predecessors of illegitimately employing the art-nature divide as an excuse to focus on nature in its unrestrained state (natura in cursu), at the expense of nature “constrained, moulded, translated, and made as it were new by art and the hand of man” (natura vexata).  In doing so, Bacon believed they had failed to 90grasp the benefits of the crafts for improving the human condition, and even worse, had founded philosophy on an “impoverished and inadequately evaluated, stock of experience,” comparable to founding a kingdom upon the mere “gossip of the streets.”  “The secrets of 91nature,” he wrote in opposition to the Aristotelian view, “reveal themselves more readily under the vexation of art than when they go their own way.”   92 This was more than just an opposition, it was a complete reversal of the aim of traditional natural philosophy. Bacon was clear that the ontological distinction between art and nature, that which had given rise to a reverence for “nature” over art, was no longer tenable.  He wrote, 93?41 See broad social histories of the period: Shapin 1996, 123ff and his bibliographic essay on this topic. See also 89Shapin and Shaffer 1985. Bacon 1861-1879, 506. See also Newman 2004, 25890 Bacon 1620 [1999], Book 1, §6391 Ibid.92 The reverence for nature Bacon attributed to natural philosophers is more a reaction to that dominant in his 93own time than to what the ancients actually did or said. See Schiefsky 2007 for a discussion of this point.?it is the fashion to talk as if art were something different from nature, so that things artificial should be separated from things natural, as differing totally in kind; whence it comes that most writers of natural history think it enough to make a history of animals or plants or minerals, without mentioning the experiments of mechanical arts (which are far the most important for philosophy); and not only that, but another and more subtle error finds its way into men’s minds; that of looking upon art merely as a kind of supplement to nature; which has power enough to finish what nature has begun or correct her when going aside, but no power to make radical changes, and shake her in the foundations; an opinion which has brought a great deal of despair to human concerns. [...] things artificial differ from things natural only in the efficient [cause]  94?This discussion is unmistakably directed at Aristotelian understandings of natural philosophy based on a distinction between art and nature. Bacon, redraws the distinction between art and nature not in terms of natures or essences, but in terms of their efficient causes, that is, the immediate processes which bring them about.  Nature, personified, is now an artisan; just as 95an artisan makes an artificial object, so nature makes a natural object (Figure 2.1).  It is in 96this sense which the natural world is, for Bacon, highly wrought—the world is contrived in the same way that humans contrive mechanical engines. This lead Bacon to claim that his history of the arts, an important part of his redefined philosophy, was a “mechanical” history.  The word “mechanical” came to greater prominence in the later 17th century, but in 97?42 Bacon 1861-1879, 506.94 In this way he is taping into a distinct ancient tradition of thinking about the creative forces in the cosmos and 95their relation to one another. Commonly these were: art, nature, and chance. See Close 1969.  See Daston 1995.96 Ibid., 506.97a way subtly yet importantly different from what Bacon intended, and corresponding to a different understanding of the relation between art and nature.  ??  Figure 2.1: The Mirror of Nature and the Image of Art. During the renaissance it was common to portray Nature, or Natura, as an artisan giving the task of looking over the earthly realm. This image comes from Robert Fludd’s 1617 Utriusque cosmi maioris scilicet et minoris metaphysica, physica atque technica historia. The hand of God, emerging from the cloud in the upper middle, holds a chain that symbolizes His mastery of Natura. Natura similarly holds a chain attached to simian, Ars, holding a globe. This symbolizes the attempts of art to ape the creations of Nature and ultimately God. (From: Tayler, Nature and Art in Renaissance Literature.) (© Wellcome Library, London) ??43 Bacon’s philosophy was promoted most ardently in the later 17th century by members of the Royal Society of London (founded 1660) through the so-called “experimental” or “mechanical” philosophy. These philosophers advocated similarly instrumental goals. One prominent member, Joseph Glanvill, writes that the objective of natural philosophy is to “enlarge knowledge by observation and experiment [...] so that nature being known it may be mastered, managed, and used in the services of human life.”   98 The mechanical philosophy also took their means—the experiment—from Baconian philosophy. As the quote above suggests, Bacon significantly elevated the role and status of experiments in natural philosophy.  He saw experiments as not just “thought experiments” 99and not just for demonstrating a conclusion known in advance or deduced from theory, but as a means to see how nature would behave under previously unobserved circumstances that could not have been brought about without human intervention.  This method he called, 100“twisting the lion’s tail.”  When later members of the Royal Society experimented by 101putting rodents into an air pump and evacuated the air inside, they followed in the wake of Bacon’s experimental methodology (Figure 2.2). ?44 Glanvill 1668 [1958]. See Merchant 1980 for a broader discussion. 98 See Kuhn 1976, 41ff for a discussion of these themes.99 As an example of simply demonstrating prior knowledge, Kuhn writes, “Roger Bacon writes that, though 100one can in principle deduce the ability of flame to burn flesh, it is more conclusive, given the mind’s propensity for error, to place one’s hand in the fire” (Kuhn 1976, 43).  Kuhn 1976, 44. 101Figure 2.2: An Air Pump Used by Robert Boyle. If you look closely you’ll notice that there is an unhappy rodent in the glass chamber. From: Boyle’s 1669 A Continuation of New Experiments Physico-Mechanical, Touching the Spring and Weight of Air. (© Wellcome Library, London) ? However, the mechanical philosophy advocated by the Royal Society was more than just an elaboration of Baconian methods. It involved a new understanding of the relation between art and nature that grounded arguments for the primacy of experimentalism as a method of gaining natural knowledge. This new understanding of nature was influenced by French thinkers like René Descartes, and while it was nominally similar to Bacon’s—it was ?45also called “mechanical”—it proposed something very different.  Robert Boyle was the 102foremost proponent of this mechanical-experimental philosophy and he is for this reason often praised as the father of experimental science. Rather than an artisan, Boyle likened nature to an intricate, but passive, work of art: an engine, such as a clock. He offered the following understanding of nature: ?‘tis like a rare Clock, such as may be that at Strasbourg, where all things are so skillfully contriv’d, that the Engine being once set a Moving, all things proceed according to the Artificers first design, and the Motions of the little Statues [...] do not require [...] the peculiar interposing of the Artificer, or any intelligent agent imployed by him, but perform their functions on particular occasions, by vertue of the General and Primitive Contrivance of the whole Engine.  103?For Boyle, nature was artificial in the sense that it was God’s work of art. This collapsed the gap between art and nature more fundamentally than had Bacon: the natural was essentially subsumed under the artificial.  Like Bacon, nature was highly wrought; but unlike Bacon, 104this was not because Nature, or Natura, brings about its objects in the same contrived way as humans bring about artificial objects.  It was because nature was nothing other than a very 105complicated mechanical engine. Against the distinction between natural and violent motion, ?46 Regarding “Mechanical”: it has four separate senses throughout the scientific revolution that can be 102confusing. (i) Mechanical as in the mechanical arts; both highly wrought and made with hands (e.g., Francis Bacon). (ii) Mechanical as in like a machine; being characterized by uniformity and regularity (e.g., Robert Boyle); (iii) Mechanical as in made of parts or corpuscles (e.g., René Descartes); (iv) Mechanical as in can be described by the language of mathematics (e.g., Isaac Newton). Regarding the influence of René Descartes; he wrote for example that “it is no less natural [...] for a clock constructed with this or that set of wheels to tell the time than it is for a tree which grew from this or that seed to produce the appropriate fruit” (Descartes 1998[1644], 209[326]).  Boyle 1686 [1999], 448.103 This is the starting place for natural theological arguments from design. Nature’s contrivance is indicative of 104its creator, in the same way that a watches contrivance is indicative of a watchmaker. See Merchant 1980, 6-7, on the related distinction between natura naturata (the natural creation) and natura 105naturans (nature as a creative force). Boyle wrote, “It may be justly Question’d [...] Whether there be any Motion, among Inanimate Bodies, that deserves to be call’d Violent, in Contradistinction to Natural”, since “among such, all Motions [...] are made according to Catholick, and almost, if not more than almost, Mechanical Laws.”  106 This account, as Boyle laid out in his Vulgarly Received Notion of Nature (1686), was prompted by fears that ancient understandings of nature, along with those of Bacon, were anthropomorphic and stole praise away from God by portraying Nature as an artisan or as having some degree of autonomy or vitality. To Boyle, God needed no intermediary; nature was simply a machine and God the engineer that put the clockwork system in motion. (We will return to Boyle’s view in Chapter 4.)		 The mechanical philosophers argued that traditional natural philosophical explanations of nature were—to borrow their favorite terms of abuse—unintelligible, animistic and anthropomorphic. The ancient saws “Nature does nothing in vain” and “Nature abhors a vacuum” lacked explanatory power, they argued. To the explanation that bodies descended because they were heavy, philosopher Thomas Hobbes responded, “But if you ask what they mean by heaviness, they will define it to be an endeavour to go to the centre of the earth [...] which is as much to say that bodies descend, or ascend, because they do.”  The 107French natural philosopher Pierre Gassendi wrote of Aristotle’s definition of motion, “the act of being in potentiality insofar as it is in potentiality”: “Great God! Is there any stomach strong enough to digest that?”  In contrast, many mechanical philosophers offered bottom-108?47 Boyle 1686 [1999], 507.106 Hobbes 1651 [2002], 504107 Cited in Cartwright 1999, 78.108up explanations involving the interaction of variously shaped, invisible, micro-level corpuscles. The interactions proposed were analogous to the ways in which everyday, macro-level bodies interacted and hence, they argued, intelligible.  The view of nature as a passive work of artifice cohered well with the rise of experimentalism and with the belief that nature, highly regular and uniform, can be described according to laws. When nature became machine-like, it was easy to see why experiments—and in general methods involving significant manipulations of nature’s mechanisms—were deemed useful.  Recall the oft-cited analogy to a clockmaker. Just as a clock is the product 109of a designer, so nature was seen as a great engineering feat set in motion by the hand of God. If nature was like a clock, one should approach it in the same way as a clockmaker approaches a foreign watch: determine the laws which govern the regular and uniform movement of its variously sized and shaped parts by taking it apart and studying its pieces.   Joseph Glanvill writes, for instance, that anatomy was useful because it tended “mightily to the eviscerating of nature, and disclosure of the springs of its motion,” and the microscope important because “the secrets of nature are not in the greater masses, but in those little threads and springs which are too subtle for the grossness of our unhelped senses.”  The metaphor runs deep since it could be employed to justify experimental or 110manipulative practices and the uses of instruments which helped one get beyond the unmediated senses or created unnatural environments. As the distinction between the artificial and the natural broke down, the methods appropriate for the study of machines and ?48 There is a difference to be pointed out here. See the beginning of Wilson 1995. Nature as a machine versus 109nature is to a machine as a clock-face is to clock mechanism.  Glanvill 1668 [1958], 1668.110the mechanical arts, and the rigor these philosophers saw accompanying these methods, could be applied to the study of nature. And so, it is said, begins modern experimentalism.  The narrative I have summarized concerns the ideological origins of the experimental methods we see as partly constitutive of modern science. The scientific revolution is characterized as a period when the study of nature changes from one of passive observation to one of experimental manipulation and control; from the study of “natural phenomena” in their due course to the study of “artificial phenomena” hidden from the immediate senses but brought out through human intervention under artificial circumstances.  This was the 111scientific revolution because we seen in it the dim origins of our own scientific attitudes, discourses, and practices.  Accompanying this change is a reconceptualization of the art-112nature relationship, from the ontological distinction drawn by Aristotle, through Bacon’s likening of nature to a human artisan, to nature as artificial, in the sense of being highly regular and uniform, describable in terms of laws of nature, and modular or reducible to independent parts.  Before analyzing this narrative, it is useful to take stock of another important and somewhat paradoxical idea which is often said to emerge during the scientific revolution: the idea that humans are separate from, or outside of, nature.  This theme is evident in the 113political sphere, for instance, when Thomas Hobbes theorized that the formation of human society was a flight from the “state of nature.” The idea is also instantiated in Descartes' res ?49 See amendments to this standard history from especially William Newman: Newman 2004, 2006; Bensaude-111Vincent and Newman 2007; see also Moran 2005. See Wilson 1995.112 Paradoxical because this is seen to arise through the same historical forces that resulted in the breakdown of 113the art-nature distinction. See Merchant 1980, Williams 1980; Daston 1995; Cronon 1995a, b.cogitans and mind-body dualism. As a theme of the literature on method in the scientific revolution, however, it is captured by the discourse of “experimentation on nature,” “manipulation of nature,” and “intervention in nature” which became prevalent during this time. Humans were intelligent and could choose to act in certain ways, but the nature that they they studied and experimented on was “‘brute, passive, stupid matter’ in seventeenth century parlance.”  This view of nature was in a sense necessary for the experimental 114methodology of the scientific revolution. The methodology required that humans view nature as controllable, lacking a will of its own, and thus also to a large extent as separate from themselves.  115?ANALYZING THIS MAKING OF MODERN SCIENCE  In analyzing this narrative, the historian would point out the ways in which it misrepresents the diversity of projects of the early moderns or their predecessors.  I am not 116qualified to analyze the narrative in this way, and it is not my intention to do so. I am interested in what it advocates as the central aspects of modern science, here represented in their proto-form. In this regard, the narrative prompts three philosophical questions to which we will return throughout the dissertation. The questions are, (i) how does the collapse of the distinction relate to the justification of manipulative approaches to the study of nature? (Or, its converse: how does drawing the distinction relate to arguments against the use of manipulative approaches?) (ii) how do these different ways of thinking about art and nature ?50 Daston 1995, 39.114 Carolyn Merchant’s work has focused most specifically on the alienation of nature during this time period. 115See Merchant 1980. The clearest example of this project is Newman 1997. 116relate to one another? (iii) how did this ontological breakdown make room for the distinction to be drawn on other grounds?  I mean to pose the first question in a philosophical manner, rather than historical; the question is about justification in general, rather than what Boyle or Glanvill thought counted as justification.  According to the medieval Aristotelian position considered above, for 117example, there is an ontological difference between artificial and natural objects and this is paired with a particular epistemic goal to understand what happens ordinarily by nature. This pairing counted against the usefulness—in natural philosophy—of experiments in the Baconian sense of “twisting the lion’s tail.” Such experiments demonstrated human control over nature, useful for practical purposes only.  Humans are in this case seen as 118disturbances or accidents in nature’s ordinary course. Consider 16th century natural philosopher Thomas Erastus’ objections to alchemy, and 17th century alchemist Daniel Sennert’s answers. Erastus denied that “the ‘chymical resolutions’ [...] can ever reveal the true principles of things, because they are ‘not natural, but artificial.’”  Sennert responded, 119in a way similar to Boyle, “But if you consider the proximal agent of mixture, it must be denied that chymical resolutions are not natural, even if the artisan participates in his own fashion. They are brought out by fire and heat, by means of a natural cause.”  For 120opponents of the alchemical proto-experimentalism, such as Erastus, Sennert’s response ?51 Historically, the connection between experimentalism and the breakdown of the artificial-natural distinction 117is not necessary (although the connection was exploited by the Royal Society). There were certainly experiments being done long before the mechanical philosophy. No one has argued this point more forcefully than William Newman. See Newman 1997, 2006. As one historian writes of pre-Baconian natural philosophy, “Their attitude might best be described as a-118practical, since practical purposes are utterly irrelevant to the science of nature as such. [...] there is a certain disdain, almost contempt, for utility in the more practical sense” (Reif 1962, 56). Newman 1997, 313. My emphasis.119 Newman 1997, 313.120would be meaningless: nature’s ordinary course is the object under investigation and human contrived situations are thus fundamentally misleading. They are considered “violent,” hence opposed to nature, and do not count, as Erastus argued, as revealing the true principles of things, that is, their natures.  121 This was not true for humans alone, but all processes which impeded nature’s ordinary course—all “violent” motions or accidents. How do we learn that the nature of an acorn is to develop into an oak tree? We watch its development under ordinary conditions where it actualizes its potential, in other words, we look for what it does most of the time. Of course, most acorns do not end as oak trees. Whatever impedes their normal development counts as “violent” motion and thus as a disturbance. As one 17th century scholastic textbook writer summarizes, ?In the individual kinds of changes two changes are found: the one natural, the other beyond nature or violent. There is natural generation, as when roses blossom in the spring; violent, when they blossom in winter by means of human endeavor. [...] Natural deterioration [occurs] when a man becomes old in his sixtieth year; violent, when he ages more quickly because of sickness or mental anxiety. Alteration is natural, when water cools; violent, when it becomes hot. Locomotion is natural, as when earth descends; violent, when it ascends.  122??52 Newman 1997, 314. 121 Translated in Reif 1962, 224.122Experiments that twist the lion’s tail involve violent motion and thus teach us about disturbances, not about natures.  123 The later scientific revolution involved a significantly different pairing: the collapse of the artificial-natural distinction, by likening nature to artifice, was paired with an instrumental goal to investigate what humans could do, rather than what happened in nature’s due course. This opened the door for manipulative projects. Experimentalism involves the assumption, as historian Garland Allen wrote, that “intervention into the workings of the organism [or nature], although artificial, still reveals something about natural or normal states of nature.”  The assumption that intervention or disturbance can reveal what is natural 124conflicts with the Aristotelian position just discussed.  It involves a fundamental shift in 125what counts as a “natural” state of nature. The Aristotelian would deny the assumption. The connection between justifying experimentation and collapsing the artificial-natural distinction will be considered throughout what follows.  The second question is prompted by the progressive nature of this narrative. It gives the impression that each new understanding of the art-nature relationship supplants the one that came before; that Aristotle’s distinction was supplanted by Boyle’s collapse. Through ?53 See Reif (1962) for a discussion of the common type of Aristotelian Natural Philosophy taught through 123scholastic textbooks in the 17th century (1962 238ff). As she writes, “due to their [the textbook writers] conception of natural philosophy as a wholly speculative science whose subject is natural, not artificial, bodies, they relegated [...] practical endeavors to the various arts.” Peter Dear also points out that, more often than not, we do not learn about the acorn by actually watching anything. Nature’s are revealed through universal statements of commonly held experience—statements like, “the sun sets in the evening.” A lot more could be said here. There is a further complication as well which I have left out of this discussion and that is the role of experience itself, and its relation to understanding cause and effect as a necessary connection. See the discussion in Dear 1995, 153-5. Allen 1994, 90.124 Nancy Cartwright (see Cartwright 1999) argues that our current scientific experimentalism is actually much 125more similar to Aristotle’s than we give it credit for. The difference is that for Aristotle we learned about natures through unmediated, ordinary circumstances, whereas the natures of modern science reveal themselves only under very contrived circumstances. See below.this dissertation we will see that this is not the case. Each of these different ways of thinking about the art-nature relationship continue to exist in tandem in biology, sometimes creating tension. The arguments between early-20th century naturalists and experimentalists will call to mind those highlighted above between the Aristotelians and Bacon, or Erastus and Sennert: naturalists were concerned that experimental methods did not teach one about “undisturbed nature”; experimentalists chided them for their “reverence” of what they considered “natural or ‘normal’ phenomena.”  Throughout the dissertation we will consider 126how these understandings of the relationship between natural and artificial relate to one another.  127 Finally, the third question, to the extent that the ontological distinction between art and nature was undermined during this period, it was not for this reason simply ousted from scientific discourse, as the historical narrative suggests. The collapse of the ontological distinction only opened the door for the distinction to be drawn on other grounds. Historians have been particularly interested in its being drawn on aesthetic, moral and theological grounds.  But it was also drawn on epistemic grounds: how can we learn about, as Allen put 128?54 MacDougal 1909, 121; Kohler 2002, 88; see Kingsland 1991. Livingston 1917; quoted in Kohler 2002, 93-4.126 Darwin’s Origin of Species, presents examples of all three types of relationship between the artificial and the 127natural considered above. At times Darwin speaks of nature in the Baconian sense: nature is like a breeder that creates forms fitted to different external circumstances in the same way that humans create pigeons fitted to their (the breeders’) aesthetic sensibilities. He utilizes this notion of nature to dodge the problem that natural species are different from man’s domestic pigeons. Isn’t it clear, he asks rhetorically, that nature creates works of art much greater than man ever could? Darwin also uses nature in Boyle’s sense. He says we can look at each creature as like “any great mechanical invention”, as “the summing up of many contrivances” (Darwin 1859, 486). Maybe most surprisingly he also speaks of nature in an Aristotelian sense, if only to discredit it. This comes out when he speaks of “reversion.” The position he argues against is one that takes it that “no deductions can be drawn from domestic races to species in a state of nature” because domestic races are simply the result of disturbances of nature, and once the disturbance is alleviated, they revert back to their “natural” condition. In other words, this so-called “principle of reversion” posits an internal source of change; an internal potential to return to a natural state. Human interference is simply accidental and cannot teach one about species in the “state of nature.” Daston and Park 1998, 280, 300. For example, in the case of 18th century naturalists separated “artificialia” 128from “naturalia”; similarly, “natural” is commonly used as a moral authority, and contrasted with the artificial. it, natural states of nature through the study of objects under artificial circumstances? The collapse of the artificial-natural distinction does not solve this stumbling block; it simply introduces another one. Artificial circumstances become only one possible state of nature—what makes us think they will stand for the others? What makes us think they are representative? This is a problem because what we study under controlled circumstances are rarely the exact phenomena of which we’re interested. What we hope is that they accurately represent the type of phenomena of which we’re interested.  We will consider these issues 129in the following chapters.  130?CONCLUSION  Myths about the origins of modern science reflect what we take to be the salient features of, and influence how we think about, contemporary science. These myths often involve the collapse of the artificial-natural distinction and the ideas that nature became passive, manipulable, and distinct from humans, while human artifice rose to the level of an ?55 This obviously depends on what circumstances we’re interested in understanding—what modern 129commentators call the target system (to be contrasted with the experimental system). One recent attempt to explicate what science learns about natural circumstances from experimental 130circumstances involves re-invoking Aristotle. Philosopher Nancy Cartwright argues that the best way to understand what physicists do when they perform experiments is to bring back Aristotle’s idea of natures. Physicists perform experiments that reveal the stable natures of the phenomena of which they’re interested. Unlike Aristotle those natures can no longer be revealed in the everyday, common experience of the natural world however. They are only revealed under very specific, particular and idealized circumstances that physicists create in the laboratory. Like Aristotle, however, the knowledge that is gained is of natures: the typical or ordinary behavior of the phenomena under investigation. What physicists do, according to Cartwright, is less the creation of circumstances that are representative of natural states of affairs in the laboratory, and more the understanding of natures of physical phenomena under very unnatural circumstances. These natures can be used to understand more complicated, non-experimental, “natural” circumstances because their typical behavior or nature will remain stable across large changes in the natural systems in which they occur. (See Cartwright 1999a. See Morgan 2003 for a discussion of how this approach fails when applied to biological phenomena.) Beginning with the next chapter we will start to see many examples of biologists who are worried about human disturbance, and thus continue to invoke the artificial-natural distinction. The way in which this might be considered Aristotelian is very different from how Cartwright understands physics to be Aristotelian—many biologists, as we will see, do not think that the nature of the biological phenomena can only be revealed under very contrived circumstances—quite the contrary.epistemically privileged method. The following chapters will complicate and undermine the collapse narrative. This chapter has also introduced different ways of thinking about the artificial-natural distinction and provided a reason why their discussion is not widely represented in historical and philosophical scholarship. In the end I proposed three questions which arose from this narrative and which we will consider in further detail in what follows. 
?56??CHAPTER 3 ?Denaturing Nature Disturbing Conditions and Classifications ????????Despite Darwin, we are not, in our hearts, part of the natural process. Lynn White Jr, 1967, The Historical Roots of Our Ecological Crisis ? In the previous chapter I considered a historical narrative that has done more than any other to direct our intuitions about how the artificial-natural distinction impinges on scientific discourse and practice. Because this narrative emphasizes the connection between the breakdown of the artificial-natural distinction and the origin of modern experimental science, it creates the impression that, firstly, the distinction matters little to contemporary science, and secondly, that any science invoking the distinction must be old-fashioned or premodern.  My goal throughout the rest of this dissertation is to move beyond this 131narrative.   In the first half of this chapter I provide a general overview of how the artificial-natural distinction is used in the life sciences through a brief, albeit longue durée, intellectual ?57 Or more like a design science. By modern science here I really mean something more like “modern natural 131science.” That “artificial” enters into design sciences—architecture, engineering, business administration, artificial intelligence, etc.—is well known. Simon (1981) argues at great length that such design sciences are “sciences.” The scientific revolution narrative is about modern natural sciences, and largely ignores design sciences.history spanning from the late-18th century onwards. My argument contrary to the above narrative, is that the artificial-natural distinction has been repeatedly invoked since the time of the scientific revolution when it was supposedly dissolved in all of science. This section will also act as a prelude by introducing the episodes that will occupy later chapters.  In the 132second half I ask whether there are any common, general threads to be found among the historical cases. I think there are two: ?(i) We should treat the history of the artificial-natural distinction as a negotiation about the place of human activity within the world of biological study. (ii) The artificial-natural distinction is used to demarcate the contours of the discipline according to multiple axes, in terms of objects, results, methods, and even personas.  ?ARTIFICIAL AND NATURAL THROUGHOUT THE HISTORY OF THE LIFE SCIENCES  The breakdown of the artificial-natural distinction that accompanied the rise of Boyle’s “mechanical” philosophy did not happen as cleanly as the standard myth suggests. This is especially true for the life sciences.  A common historiographical sentiment is that 133the mechanization was “a 'failed' aspiration.”  Of course, this is not true for today’s biology 134tout court. Especially in those areas “closest” to the physical sciences, like molecular biology, the boundary between the artificial and the natural doesn't have the same influence ?58 This history will be selectively focused on a set of cases where the artificial-natural distinction arises and is 132contested—in other words, it should not be taken as representative of the history of the life sciences in general. This was not, however, because of some general lack of initiative to collapse the distinction: many of the 133early mechanists and physiologists—as well as many later figures in biology and philosophy—thought that all of life could and should be understood as like a machine. Moreover, René Descartes, whom we encountered in the last chapter as having a strong influence of the Royal Society’s thinking, was directly influenced by physiologist William Harvey's experiments on the circulation of blood. Ernst Mayr, the evolutionary biologist and philosopher, wrote, “When evolutionary biology was examined 134for its “scientificness” according to the criteria of mechanics, it flunked the test” (Mayr 1997, 28). Shapin 1996, 185; see Wilson 1995, chapter 1 regarding the historiography of this sentiment.on practice nor does it play as large a role in theory. Nonetheless, the relationship between these two antithetical terms looms behind many of the major and most stubborn issues in the history of the life sciences. Here I will introduce examples from the recent history of biology to bring out some of the ways in which the distinction has been and still is drawn and contested.  135 Late-18th century debates among pre-Darwinian physiologists over the importance of experimentation are especially germane. The Laplacian-style determinists of the 18th and 19th centuries—most famously, Antoine Lavoisier and Claude Bernard—argued that artificial experimentation should be the major methodological approach to physiology because only through experimental means could life-phenomena be isolated, stabilized and calculated; life, for them, was law-governed and could be represented by predictable regularities determined through extensive experimentation.  Physiologists in the tradition of 136the Montpellier school of medicine—notably Xavier Bichat—favoring a less manipulative observational physiology, could not disagree more; to isolate and stabilize biological phenomena—to obtain constant outcomes from experiments—was to fail to comprehend living phenomena altogether.  They believed that there was something spontaneous and 137indeterminate about life such that one could obtain predictable outcomes only at the expense of “denaturing” the objects under investigation.  Thus, the human-manipulated, artificial 138experiment, they argued, was of limited value for the study of nature. ?59 As a clarification, in saying that the distinction is contested in biology, I mean to claim that the extent to 135which the distinction is collapsed is an issue prevalent in biology. Coleman 1971; Westfall 1971, 82ff; Gigerenzer et al. 1989, 124-7; Canguilhem 2008. See Bichat 1801.136 Williams 2003, 157-8; Reill 2008. We will see how these concerns arise for Darwin in the next chapter.137 Albury 1977, 61; Gigerenzer et al. 1989, 124-7.138 Concerns about the artificial-natural boundary in the pre-Darwinian era did not arise in physiology alone. In natural history, for instance, such concerns can be found throughout the works of both Georges-Louis Leclerc, Comte de Buffon and Georges Cuvier.  Buffon's 139natural history, laid out in his multi-volume Natural History (1749–1788), was a direct response to the methods of the mechanical philosophy and its definition of matter and therefore dealt directly with the artificial-natural distinction.  “The springs that [nature] 140uses,” he writes, exploiting a favorite analogy of the mechanists, “are living forces,” wholly different from the dead matter of machines.   141 While Buffon maintained that the mechanical philosophy impressively dealt with dead matter, living matter, to be comprehended correctly, required a different methodology entirely. In volume five of Natural History, for instance, he warns naturalists about complications posed by domestic animals, which carry the “stigma of slavery.”  Domestic 142animals have been shaped by humans under “unnatural” conditions and are misleading with regard to “savage,” “wild” or “natural” animals.  Their study is therefore of limited use for 143natural history. Thus it is the duty of the naturalist to “examine them with care [...] to separate artifice from Nature; and never to confound the animal with the slave, the beast of burden with the creature of God.”   144 In the introduction to his famous systematic study of the structure of animals, The Animal Kingdom (1817-1830), Georges Cuvier contrasts natural history with general physics ?60 See Sloan 1995 for natural history more generally.139 Sloan 1976.140 Quoted in Reill 2005, 47.141 Buffon 1749-1788, 50.142 Ibid., 301143 Ibid., 301-2; my emphasis.144(i.e., at that time: mechanics, dynamics, chemistry, etc.), and relies on a distinction between the artificial and the natural to do so. Although natural history should attempt to apply the same methods as general physics—calculation and experimentation—Cuvier suggested that this would seldom be possible, since life-phenomena must be taken in their entirety and cannot be reduced to their elements.  When we attempt to isolate and suppress “the 145numerous phenomena which compose the life of an animal,” life itself is “wholly annihilated.”  The artificial methodology of general physics which relies on human 146manipulation is useless for the study of living nature—a concern similar to that expressed by the Montpellier tradition above. Within natural philosophy, that is, both natural history and general physics, each investigative strategy has its particular place: “Calculation, so to speak, commands Nature; it determines phenomena more exactly than observation can make them known: experiment forces her to unveil; while observation watches her when deviating from her normal course, and seeks to surprise her.”  Adopting an observational, comparative 147methodology may be the only way to study the natural history of life without annihilating its very naturalness—that is, without annihilating the very object of study.  Later in the 19th century Charles Darwin would find himself involved in debates surrounding the boundary between the artificial and the natural. Traditionally construed, the argument of the Origin of Species (1859) is built on an analogy between natural and artificial selection: Darwin exploits the fact that breeders have had particularly good success in creating new varieties of domesticated plants and animals, through the process of artificial ?61 Cuvier 1817-1830, 14.145 Ibid.146 Ibid.147selection, to argue for the efficacy of natural selection.  While the former depends on an 148anthropogenic power, the latter depends only on a natural process of competition, that is, “the struggle for existence.” Darwin even chose the phrase “natural selection,” “in order to mark its relation to man's power of selection” over domesticated animals and plants.  Alfred 149Russel Wallace, co-founder of the theory of evolution by natural selection, disagreed strongly with Darwin's use of artificial selection (see Chapter 5). From his earliest papers Wallace makes it clear that these two categories of selection are disanalogous, arguing that natural competition between species in the wild is a completely separate process from domestic breeding.  For instance, 150?A wild animal has to search, and often to labour, for every mouthful of food—to exercise sight, hearing, and smell in seeking it, and in avoiding dangers [...] There is no muscle of its body that is not called into daily and hourly activity [...] The domestic animal, on the other hand, has food provided for it, is sheltered, and often [...] carefully secured from the attacks of its natural enemies, and seldom even rears its young without human assistance.  151?Domestication, practiced by even the best breeders, is unstable and impermanent—domesticated animals revert to their wild type when human selection is alleviated. For these reasons, Wallace saw the artificial process of domestication as unable to tell us anything ?62 See chapters 4 and 5. See also Evans 1984; Provine 1986; Waters 1986; Richards 1998; Alter 2007; Burnett 1482009. Darwin 1859, 61.149 Darwin and Wallace 1858, 54.150 Ibid., 59-60.151about natural processes like natural selection. The debate between Darwin and Wallace about the importance of artificial selection lives on in contemporary evolutionary theory.  152 The artificial-natural distinction remains prominent in post-Darwin biology as well and the concerns expressed are quite similar. In Chapter 1 I alluded to the engineering ideal of early-20th century biologist Jacques Loeb. Loeb thought that biology should be reformulated as an engineering science, rather than a natural science; one that aimed, harkening back to Bacon and Glanvill, to produce practical knowledge first-and-foremost. If the aim was to understand the nature of life, for example, the surest way to meet that aim was to try to synthesize life in the laboratory: “we must either succeed in producing living matter artificially, or we must find the reasons why this is impossible.”  His stance on the proper 153aims and practices of biology was meant to be in direct opposition to naturalists of the 19th century, such as Darwin and Wallace.  154 A similar perspective can be found among many experimentalists of the early decades of the 20th century. The physiologist and ecologist Burton Livingston, for example, tactfully wrote that by 1917 the “older reverence for natural or ‘normal’ phenomena [had] largely disappeared.” And echoing Loeb’s engineering program, ?We have learned that the range of conditions offered by nature does not generally happen to be great enough to allow adequate experimental interpretation of plant processes, [...] if a student has not a liking and talent for creating physical and ?63 See reviews by: Hill and Caballero 1992; Falconer 1992; Rice and Hostert 1993; Harshman and Hoffman 1522000; Gregory 2008; Garland and Rose 2009; Kawecki et al. 2012. Loeb 1912, 5-6. See also Keller 2002, 18.153 This perspective was influential on a number of later biologists including the famous geneticist Hermann J. 154Muller as well as scientists in related disciplines such as the behaviorist B. F. Skinner.chemical conditions such as never have occurred in nature, he should not cast his lot with plant physiologists, for the next generation.  155?Livingston’s argument should be seen more as rhetoric than accurate representation—who is this “we” that has “learned,” the experimentalist? The “older reverence” for natural phenomena still existed in many fields and physiological ecologists continued to argue about such issues for many years.   156 These engineering and experimentalist conceptions of biology met with antagonism from biologists of other traditions, especially evolutionary naturalists, who considered them to be misguided. Members of Thomas Hunt Morgan’s laboratory—the famous “fly group”—recalled the “violent opposition” they encountered from those who saw the mutations they discovered in the fruit fly Drosophila melanogaster as mere pathological changes, brought about by the laboratory, and having no relevance to what goes on in nature.  (The mutations 157referred to included flies with black bodies, white eyes, or no eyes at all.) Henry Fairfield Osborn, curator at the American Museum of Natural History in New York, for example, wrote that “Speciation is a normal and continuous process,” he argued, “it governs the greater part of the origin of species [...] Mutation [what the fly group investigated] is an abnormal and irregular mode of origin, which while not infrequently occurring in nature is not essentially an adaptive process; it is, rather, a disturbance of the regular course of ?64 Livingston 1917; quoted in Kohler 2002, 93-4.155 For a history of ecology which brings out many of these issues see Kingsland 2005.156 Quote from an interview between historian Garland Allen and the geneticist Theodosius Dobzhansky, 1966, 157at Rockefeller University. Dobzhansky worked in Morgan’s laboratory. Emphasis in first quote mine. Osborn 1927, 40-1. The second quote has emphasis throughout in the original 158which I have removed.speciation.”  The idea that mutations were disturbances, and their importance an artifact of 158laboratory experimentalism, was a commonly held belief among naturalists.  159 The distinction played a role in many debates associated with the “modern evolutionary synthesis” of the 1930s and 40s—an influential movement which aimed at unifying a set of seemingly distant sub-disciplines of biology under the general rubric of evolution (see Chapter 6).  Theodosius Dobzhansky, one of the primary “architects” of the 160modern synthesis, relied heavily on the “artificial setting” of the laboratory to derive evolutionary conclusions about populations of fruit flies (Drosophila) in the “wild.”  He 161often and openly expressed his reservations however about the artificiality of such settings and continually sought to make the laboratory more natural by bringing in flies, for example, directly from the field rather than using strains which have spent long periods of time under artificial conditions.  In a 1939 paper Dobzhansky expressed his uneasiness: 162?The fact that most animal and plant species which have served as material for genetic investigations might be classed as domestic or semi-domestic has repeatedly been used to cast aspersions on the validity of the resulting data for an understanding of the evolutionary process. To some writers the word “domestication” has become a kind of scarecrow. Since there is no evidence that domestication per se either induces or prevents the appearance of any class of genetic changes, this attitude is untenable. In the last analysis domestication is merely a special case of “natural” conditions, this ?65 Evolutionary biologist Ernst Mayr recalled that this comment “well expressed the feelings of the naturalists.” 159See his keynote address for Evolutionary Synthesis Conference, 13, 33, APS.  Provine 1971, 1986; Smocovitis 1996; Cain 2003, 2009; Cain and Ruse 2009.160 Provine 1981.161 Kohler 1994, 288. Ironically, this practice is now frowned upon because the evolution of flies not yet 162adapted to laboratory conditions is seen as too artificial!—most laboratory strains have thus been domesticated for a decade or more. See Harshman and Hoffman 2000, 34-5.latter term being in reality a name subsuming a great variety of diverse conditions. On the other hand, domestication does modify in some ways the balance of forces acting upon the genetic composition of a population, and hence cannot be entirely disregarded in studies concerning population dynamics. In the ecology of an animal, even so little “domesticated” as D. melanogaster, one can perceive certain special features: temporary relaxations of natural selection caused by overabundance of food and lack of enemies, extreme shrinkages and excessive increases of the population size, introduction or removal from a given locality of masses of individuals by man.  163?As I mentioned in Chapter 1, Dobzhansky disdainfully referred to a paper by another prominent geneticist and evolutionary biologist, Alfred Sturtevant, as being based upon “laboratory strains!.”  Dobzhansky's concerns that model organisms, such D. melanogaster, 164might be purely human artifacts, non-representative of their natural or “wild” counterparts, continues to be hotly debated today.  165 The application of the artificial-natural distinction extends to biologists’ personas and their evaluations of one another. The idea that biologists who spend too much time in the laboratory don’t have enough of an appreciation for “nature” to pronounce on evolution persisted throughout Dobzhansky’s lifetime. Dobzhansky himself, for example, wrote to Ernst Mayr in 1974, “I also agree with you that [geneticist Hermann J. Muller] was ‘rather naive’ in many of his pronouncements on evolution,” but this was to be expected, “from a man of great laboratory achievements but no familiarity with organisms as they live ?66 Dobzhansky 1939, 345-6.163 Provine 1981, 52.164 See for instance Wolff 2003, Creager et al. 2007, Ankeny and Leonelli 2011.165outside.”  In other words, to say anything substantial about the natural process of evolution, 166one needs some acquaintance with nature.  More recently, in ecology debates surrounding the artificial-natural distinction arise not only with regard to model organisms, but with regard to the experimental set-up itself. Laboratory experiments in ecology are often done with microorganisms or small arthropods kept in bottles because this set-up is cheap, easy to maintain, and easy to control genetically and environmentally. Critics contend that these conditions go little way towards answering ecological questions.  Ecologist Steven Carpenter puts the worry in the following way: “a 167molecular biologist who isolates ribosomes is working on ribosomes; an ecologist who isolates organisms in bottles may not be working on communities and ecosystems in any relevant sense.”  According to Carpenter, the objects of molecular study can be and should 168be isolated. It makes little difference to the molecular study of ribosomes—their structure and function, for instance—whether they are considered in vitro or in vivo. On the other hand, to isolate the objects of ecological study from their natural context is, in a sense, to sacrifice those objects altogether; they are altered so fundamentally that little worthy of ecological study remains.   Ecologists like Jared Diamond argue, therefore, that for ecology the most fruitful types of experiments are natural experiments, in which biologists study the effects of large natural disturbances (e.g., storm, drought, fire, etc.), rather than the human-induced ?67 Theodosius Dobzhansky to Ernst Mayr, 1974, Mather, Sierra, California; APS. It is also worth flagging that 166Muller was influenced by Loeb’s conception of biology in which the main aim was to control life processes and to create or engineer new biological possibilities (as we saw from the quote above). This aim was inconsistent with many evolutionary biologists who wanted to understand how nature worked undisturbed by human influence (see Pauly 1987, 177-83). Diamond 1983, 1986; Carpenter 1996, 1999.167 Carpenter 1996, 678.168disturbances of laboratory and field studies (see Chapter 7).  Natural experiments should 169not be confused with the study of natural history; that is, natural experiments do not belong to the observational or descriptive side of science. Rather they are often utilized to test specific hypotheses, involve significant manipulations— albeit “natural,” rather than human-induced, manipulations—and aim at determining the causal factors at play in a particular ecological situation. Most natural experiments involve a comparison of two ecological “places” that ideally differ with regard to only one important factor, e.g., two lakes that differ with regard to mineral composition or two islands that differ with regard to a particular predatory species. A claim is then made about the causal influence—or lack thereof—of this factor. Diamond and others argue that natural experiments far outweigh other types in terms of realism (i.e., whether results can be extrapolated to natural communities), generality (i.e., the number of natural communities to which the study applies) and scope (i.e., the experimental manipulations possible—natural experiments, for instance, do not face as many ethical or environmental obstacles).  Parallel concerns arise in ecology in a number of different venues. Biologist Earl McCoy and philosopher Kristin Shrader-Frechette have highlighted and problematized the ways in which the natural-artificial distinction arises throughout the entire discourse of ecology. “Natural places,” “natural divisions,” “natural systems,” “natural experiments,” “natural laboratories,” “natural communities,” are all commonly employed phrases. Furthermore, natural history and ecology both, they argue, presuppose concepts of “natural ?68 Diamond 1983, 1986, 2001; see also Dunning 2008 and Diamond and Robertson 2010 for the use of natural 169experiments outside of ecology; and see Kohler 2002 for a history of natural experiments in ecology.place.”  Should a distinction that is for a number of reasons conceptually incoherent play 170such a fundamental role in ecology, they provokingly ask?  In a similar fashion, the artificial-natural distinction underlies recent arguments surrounding the distinction between “native” and “non-native” (or “alien”) species. This distinction has acted as a guiding principle in conservation and restoration management, where alien species are seen as disruptive or unwanted and large-scale environmental policies are administered to control or remove them.  It is argued, for instance by conservation 171ecologist Daniel Simberloff, that alien species threaten the natural environment.  Critics of 172the native-alien distinction, however, point out that the very distinction between native and alien species is itself inconsistent, and that such claims are often not based on any evidence of the disruptive influence of alien species at all. Instead, claims are usually based on the idea that the artificial act of human-dispersal accomplishes a “denaturing” of the dispersed species—above we saw this phrase used in a similar fashion by Xavier Bichat and the Montpellier school of medicine.  As it is put in a recent article, Simberloff's claim “makes sense only if 173a human act of dispersal renders nature unnatural.”  174 The issue of native versus non-native species is part of a broad category of what we might call applied issues in which the artificial-natural divide plays a role.For example, the issues emerge clearly from a set of influential articles by Thomas Henry Huxley, the evolutionist and vocal proponent of Darwin’s theory.  When Huxley was writing these 175?69 See Shrader-Frechette and McCoy 1995.170 Davis et al. 2011.171 Simberloff 2005. 172 Chew and Hamilton 2011, 36.173 Ibid., 36.174 Huxley 1893, 1894.175articles, questions were emerging about whether evolution justified a particular social-ethical order—a question which resonates still. Huxley argued that civilized human society was not in the “state of nature,” where the struggle for existence reigned supreme, but instead in the “state of art.” He likened human civilization to a domestic garden: when we build a garden, he said, we put a wall around it, literally or figuratively, so that natural forces —what he called the “cosmic process”—are kept out; we might control, for instance, the influence that sunlight, temperature, or pests have on our garden. Civilization is currently in a similar state, he argued, and those concerned with ethical progress within society would do well to keep this in mind. “The thief and murderer follow from nature,” he argued, “just as much as the philanthropist.”  A functioning society therefore depends on its members acting contrary to 176nature, so that the struggle for existence is alleviated within the artificial conditions of human evolution. For these reasons he professed, “Let us understand, once for all, that the ethical progress of society depends, not on imitating the cosmic (i.e., natural) process, still less in running away from it, but in combating it.”  For Huxley, and the many that have followed 177him, human evolution is a process altogether different from natural evolution.  178?DISTURBING CONDITIONS AND CLASSIFICATIONS  What should we make of these myriad examples? One thing at least should be clear. In many areas in the history of the life sciences, significant and recurrent issues arise from the different ways in which the artificial-natural distinction is drawn and contested. This ?70 Huxley 1893, 80.176 Ibid., 83177 See Williams 1992.178distinction has been used to frame the debates even when the stakes are very different. Bichat and Bernard’s concerns about life’s supposed spontaneity are very different from Darwin and Wallace’s dispute over the status of domestic organisms. In other words, the collapse of the artificial-natural distinction has not occurred in many areas of biology.  Without ignoring their historical specificity, is there anything more we can say about these cases? What threads run through and link them? I think there are two: ?(i) The history of the artificial-natural distinction is a negotiation about the place of human activity within the world of biological study. (ii) The artificial-natural distinction is used to demarcate the contours of the discipline according to multiple axes, in terms of objects, results, methods, and even personas. ?I will explain these below but illustrate them in greater depth throughout the succeeding chapters. ?(I) HUMANS AS DISTURBING CONDITIONS  The reason why the artificial-natural distinction arises frequently in biology may be, at one skeptical extreme, attributed simply to convention. In the distant past something significant was conveyed through its employment, but unlike then, its current prevalence can be attributed to tradition. This dissertation is meant to prove otherwise. The wide-ranging set of examples considered so far are enough to call this into question: the frequency with which the distinction is actively debated would be somewhat mysterious if it were merely ?71conventional. I believe there are deeper reasons why this language permeates biologists’ characterizations of manifold aspects of their discipline.   It is helpful to reflect on this by thinking about the rhetorical use of the artificial-natural dichotomy. People, Loeb is one, exploit the artificial-natural language to further their own goals against those of their opponents and to evaluate each other’s work or even scientific identity. “Your fruit flies are artificial constructions!,” is a good or bad thing depending on the context. Such rhetorical uses of this distinction, however, are effective—indeed rhetoric in general is effective—only in so far as they take advantage of a tacit but potent common framework of assumptions for understanding and communicating. If there was not already an underlying framework, such uses would not be rhetorically effective. This should push us to ask, what are the features of this underlying framework? How did this framework come into being?  In Chapter 1 I hypothesized that the artificial-natural distinction is useful because it succinctly conveys a more fundamental concern about how human actions relate to nature’s normal development. This concern is a feature of the underlying framework which makes rhetorical uses of the distinction effective. The concern is that “the artificial”—that which is human-mediated—is misleading with regard to “the natural”—that which exists by (or in) nature. This is expressed many times above—think of the complaint that humans “denature” nature. Even those who do not themselves express this concern employ tactics that feed off its prevalence. Loeb exploits it to argue not that the distinction doesn’t exist, but that the artificial is more useful. The artificial-natural distinction is a useful way of sorting out what ?72is created by humans from what exists by nature; of sorting that which is misleading from that which is useful and valid for biological study.   Of course, all natural science requires sorting artifacts from natural phenomena, where the latter are most commonly the ultimate object of study.  In this sense, artifacts are 179things that are observed in a scientific investigation but that are a result of preparation, rather than naturally occurring. Examples from microscopy are perhaps the easiest to grasp. Many artifacts are introduced as a result of the preparative procedure that goes into creating a microscopic slide. These can be as simple as distortion due to the size or color of the specimen, but also include much more misleading and problematic cases, for instance, when an artifact is mistaken for a natural object. What T. H. Huxley thought were organic structures in the late 1850s, possibly even the most primitive and fundamental form of life, turned out to be nothing but preparative artifacts created by immersing the specimen in a particular solution of alcohol and sea water.  Although the cases vary in complexity, this is 180a common practical scientific problem.  A distinction thus arises for all sciences between preparative artifacts and natural phenomena, so the interesting question is why do the concerns appear to be more exaggerated, fundamental, and derisive in biology. There are at least two reasons. First, in those areas of biology where this distinction arises most frequently—ecology, organismal and evolutionary biology—the systems studied are extremely complex. Since complete control over all experimental variables is in these cases impossible, and often even undesirable, the ?73 But see Herbert Simon on the design or engineering sciences versus the natural sciences (Simon 1981). See 179Weber 2005, Chapter 9, for a general discussion of artifacts and a study case of the mesosome. See Sperber 2007 on artifacts more generally. Rehbock 1975.180introduction of artifacts is more common and the sorting of artifacts from natural objects is arduous, uncertain and can remain incomplete.   Second, and more provocatively, in these areas of biology the question is often not just about which objects are artifacts, so they can be ignored or removed and relegated to the category of error. The question is whether the artifacts are actually natural phenomena. One reason why this question arises is because humans themselves are often active participants in the biological systems studied. They are arguably parts of ecosystems and causes of natural selection. In such cases, debates over whether something counts as an artifact cannot be resolved simply by appealing to technical or practical factors, such as Huxley’s sea water-alcohol solution. Instead, one is forced to appeal to wider concerns about drawing the artificial-natural distinction. The way the distinction arises in biology thus betrays a more fundamental set of questions about the relation between human activities and the objects of biological study: are humans a part of natural ecosystems or are they a part of the natural evolutionary process? If they are, how are their activities similar to or different from the rest of the system studied?  Consider an example introduced above. One of the debates between Darwin and Wallace—the focus of Chapter 5—was about whether domesticated varieties are artifacts or natural creations. At the time the answer to this question had a direct bearing on whether they were considered valid objects of study in natural history. For example, Asa Gray, a Harvard botanist and friend of Darwin’s, regretted the fact that domesticated varieties had been neglected by naturalists merely “because these races are not in a state of nature.”  Were 181?74 Gray [1860] 1963, 21.181domesticated varieties to be ignored in the study of species evolution or were they examples of species evolution? This disagreement between Darwin and Wallace differs from the microscopy case in that it is not just a sorting of artifacts from natural phenomena, but a negotiation about what is to count as an artifact in a more fundamental sense: how does human breeding relate to natural evolution? Are humans creating artifacts or natural phenomena? In what ways do these things differ? And what does this entail for the ways we might know about them? These are more than just technical or practical questions, but are substantive questions in biological theory.  In very general terms what is negotiated in such cases is whether and in what ways human actions should be seen as interferences in systems of biological study. Philosophers of science have a convenient and general way of characterizing such interfering factors. They call them “disturbing factors” or “disturbing conditions.”  I think that the ways in which 182human-mediation is treated in the above cases suggests that it is analogous to the factors philosophers call disturbing conditions. Consider one definition from philosopher of ecology Chris Eliot. Disturbing conditions, Eliot writes, are ?factors which interfere with the applicability of a model (or model system) to the members of the designated set of instances to which a model can be, is, or is intended to be, applied.  183?75 The language of disturbing conditions comes from economics, where it still occurs regularly. John Stuart 182Mill argued, for instance, that economic laws were true only in the abstract: they describe what would happen in the absence of disturbing causes (Persky 1990, 187-90; Hausman 1992, 123-51). He reasoned that since economic laws depend on a definition of “mankind as solely occupied in acquiring and consuming wealth,” anything that causes mankind to deviate from this definition would count as a disturbing cause (Reutlinger et al. 2011). Ecologists often use similar language, as when they designate a particular ecosystem “disturbed.” The Oxford English Dictionary defines “disturbance” in a similar way as “Interference with the regular or due course or continuance of any action or process.” Eliot 2004, 3.183?To provide an illustration, during the 1920s the mathematicians Alfred Lotka and Vito Volterra independently proposed a model that accounts for how populations of predators and prey—mice and owls, say—interact. By building into their model sizes and growth rates of each population as well as the efficiency of both predators and prey, they could provide predictions of oscillations in the size of each population relative to the other.   Although this model does outline a general trend, it will never accurately predict what happens in natural populations of owls and mice. This is because, for example, the size of mice populations depends on more than just owls. It depends on the abundance of their own food sources, the availability of suitable habitat, the presence of other predators, disease, and so on. Relative to this model, all of these factors are disturbing conditions. The application of the model is valid—accurately describes the dynamics of the natural system—only when these factors are excluded (that is, when all disturbing conditions are excluded).  While Eliot’s definition is appropriate for this particular case, to capture the general philosophical use of this phrase we need a definition that is applicable beyond the context of scientific models.  In addition to characterizing those “outside” factors in mathematical 184model-building, “disturbing conditions” are used in two other contexts. In the context of laws of nature or scientific generalizations they describe those factors that must be excluded in order for a law or generalization to apply. In the context of experimentation they describe those factors that must be excluded or controlled for in order for a scientific test to be ?76 Unless of course “model” is understood very broadly to capture any representation—see below.184reliable.  What these different uses of disturbing conditions have in common is related to 185scientific representation. In these instances, a model, generalization, or experimental system is used to represent a part of the natural world. Disturbing conditions more generally are thus factors not included in a representation that, when present, impede its accuracy or suitability. They impede its ability represent.  The above history suggests that human-mediation has been considered a disturbing condition in certain areas and at certain times in biology. Although I mean the phrase “disturbing condition” in a general way, I think there is something analogous between how, say, humans are considered in the above examples and how disease is considered in the predator-prey example.  Human disturbance is the “outside” factor which hinders the 186validity of an object, method, etc., to represent a portion of nature. The history of the artificial-natural distinction is thus a negotiation over whether and in what ways humans count as disturbing conditions; in what ways human activity is outside of the system of study (as disease was outside of the Lotka-Volterra model) and in what ways this matters.  It should go without saying that it is not my contention that what is significant about biology is that biologists as a collective consider humans as disturbing conditions. My contention is that throughout the history of biology, concerns about whether humans are ?77 In terms of debates over laws of nature more generally, such as whether laws require implicit ceteris paribus 185clauses that exclude disturbing conditions and whether such laws express empirically testable content, see Earman, Roberts and Smith 2002. In terms of experiments where disturbing conditions are factors that must be excluded or controlled for in order for a scientific experiment to be considered reliable, see Reutlinger et al. 2001. I realize that this characterization is more general than some philosophers would allow, but I believe it is 186consistent with the diverse way in which this term is used in ecology, economics, and philosophy. See Reutlinger et al. (2011) for many examples that corroborate my claim about this diversity. Most commonly “disturbing conditions” are discussed in terms of laws of nature. The language of “disturbing conditions” (“factors” or “causes”), and the related language of “ceteris paribus clauses,” came into prominence in the nineteenth century, predominantly in economics. See Persky 1990; Hausman 1992, 124; Kaufer 1997.disturbing conditions have arisen frequently and have been at the heart of a number of notable debates. The discourse of biology is thus permeated by the artificial-natural dichotomy because this language is a useful way to express such concerns.  By way of summary, what I have been pursuing here is the common framework that gives rise to the usefulness of the artificial-natural distinction. We might say that throughout the history of biology humans themselves have been thought of as disturbing conditions and this is reflected in the commonly used language of artificial and natural. We must keep in mind, however, that what counts as “human disturbance,” “artificial,” or “natural” is constantly being redefined. As a result of this discussion we should amend our first thread to say: ?(i) The history of the artificial-natural distinction is a negotiation about whether and in what ways humans count as disturbing conditions. ?(II) THE ARTIFICIAL-NATURAL DISTINCTION AS A META-CLASSIFICATION SYSTEM  In what ways is the artificial-natural distinction deployed? To what objects is it applied? The answer to this question is quite simple, but has interesting consequences. When the distinction is drawn, biologists use the artificial-natural distinction as a classification system; in other words, they classify aspects of their work and discipline according to the artificial-natural distinction. Consider these common classifications made by biologists: ????78? Notice first that this classification is wide in scope. It is not on the same level as classification within science, such as classifying a specimen as a particular species. This is a meta-level classification system, applied to a variety of objects of study, processes, experimental perturbations, experiments, and even persons or types of researcher (although the latter becomes slightly more complicated). The concept of human disturbance enters in at all of these levels (for instance, the naturalist, according to common opinion, is one whose object of study is not disturbed nature).  Classification practices have been a central object of study in recent science and technology studies.  What I borrow from this research is its general definition of 187classification and two general points about classification systems.  188Natural Artificialnatural ecosystem artificial ecosystemnatural community artificial or synthetic community wild organism domestic organismnatural variation artificial variationnatural mutations artificial mutationsnatural perturbation artificial perturbationnatural selection artificial selectionnatural experiment laboratory experimentnaturalist experimentalist?79 This research builds principally on the work of Michel Foucault, Mary Douglas, and Nelson Goodman. See 187Foucault 2008 [1966]; Douglas 2002 [1966]; Goodman 1978. This literature focuses on many other issues which are irrelevant for our current discussion. Many of these 188have to do with classification systems reflecting hierarchies of power, shaping moral and social order, and natural kinds. See Douglas and Hull 1992, Hacking 2001, chapter 6, and Epstein 2007. In one important analysis Geof Bowker and Susan Leigh Star suggest that a classification is a “set of boxes (metaphorical or literal) into which things can be put to then do some kind of work—bureaucratic or knowledge production.”  They argue that a 189classification system exhibits the following abstracted or ideal properties: ?1. There are consistent, unique classificatory principles in operation. 2. The categories are mutually exclusive. 3. The system is complete. (Provides total coverage of the world it describes.)  190? Of course very few actual classification systems meet these goals—the classificatory principles are often not consistent or unique, and classification systems gradually change over time—but this account provides a regulative ideal nonetheless. The artificial-natural distinction meets each of these properties as well as one can expect. The classificatory principle in action relates to treating humans as disturbing conditions—in the sense that human-disturbed things are artificial. The categories are mutually exclusive and complete in any given case, albeit quite variable between cases. In other words, the artificial-natural dichotomy is being used as a classification system.  The much more interesting points have to do with common properties of all classification systems. First, classification systems are not innocuous. They have consequences. This is so for everyday things as well as scientific things, because how we classify things changes how we act towards them—this is the point of many classification systems. If something is classified as A we treat it as a member of A, if it is classified as B we ?80 Bowker and Star 1999, 10.189 Ibid.190treat it as a member of B. In Chapter 1, I provided an example of this in terms of the goal of restoration ecology, commonly, to restore communities to their “natural” state. Obviously in this case deciding what is to count as natural has practical consequences.  While this case is perhaps more extreme—and, in this regard, useful for making the present point—I do not think we should overlook that a similar thing happens with regard to the artificial-natural distinction as it is used in biology more widely. For the rest of this dissertation it will be important to keep in mind that classifications have consequences. How biologists use this classification system changes how they act towards those aspects of the discipline classified.  The second point is that, “As classification systems get ever more deeply embedded,” write Bowker and Star, “they risk getting black boxed and thence made both potent and invisible.”  With embedded, Bowker and Star mean to draw attention to the degree of their 191commonality in working infrastructures. By infrastructures they have in mind more than just physical structures, and mean to highlight any organizing structure, including beliefs and practices, that support a system (whether bureaucratic, scientific, or belief). The artificial-natural distinction is deeply embedded in the working infrastructures of modern biologists because they use these categories as part of their daily practice and these categories provide important organizational support. They become invisible as their use becomes unconscious: they are invoked without recourse to the principles of their application. In other words, they become invisible as their use becomes obvious. A goal of what follows is to bring the ?81 Bowker and Star 1999, 325.191classifications biologists make out in the open, as a way of understanding what consequences such classifications might have.  By invoking the artificial-natural distinction, biologists do more than categorize aspects of their discipline, they are also defining their very discipline. By determining what objects to study, how to do so, and who shall do so, biologists shape the content and scope of their discipline. As a classification system applied to the whole of biology, the artificial-natural distinction is constitutive of biological practice. Especially in biology, where “the natural” can act as an epistemic authority, classification according to the artificial-natural distinction has a strong influence over what is and what is not considered a valid object, site, method, etc., for study. Recall again Daily’s and Pigliucci’s choices of site and research organism, or Dobzhansky’s assessment that biologists who spend too much time in the lab cannot speak authoritatively about evolution. By classifying, biologists are demarcating the contours of their discipline: they are defining what does and does not count as properly scientific. ?CONCLUSION  This chapter was split into two parts. The first was historical and aimed to move beyond the implications of the myth narrated in the previous chapter: the artificial-natural distinction is of relevance to modern science—it has been central to the modern history of biology. In the second half of this chapter I asked whether there were any general threads to be extricated from the historical cases and have argued for two threads which will be central to succeeding chapters, again they were: ?82?(i) The history of the artificial-natural distinction is a negotiation about whether and in what ways humans count as disturbing conditions. (ii) The artificial-natural distinction is used to demarcate the contours of the discipline according to multiple axes, in terms of objects, results, methods, and even personas. ?While being central to negotiations about valid objects or methods of study, as the above examples attest, the artificial-natural distinction both reflects and informs our understanding of the relation between human activity and that part of the natural world biologists attempt to understand. As the distinction is defined and redefined so too is what counts as valid for biological study, as are the ways in which humans count as disturbing conditions. 
?83??CHAPTER 4 ?The Art Itself is Nature Darwin, Domestic Varieties and the Mechanical Philosophy ?????????there is no reason art should gain the point of honor of our great and puissant mother Nature. We have so much by our inventions surcharged the beauties and riches of her works that we have altogether overchoked her; yet wherever her purity shineth, she makes our vain and frivolous enterprises wonderfully ashamed. Michel de Montaigne, 1580, Of the Cannibals ? Common to both the scientific and Darwinian revolutions were discussions impinging on the relationship between art and nature. Was art a part of nature? Could art be used as a model for nature? This intellectual congruence, however, is more than just nominal. Charles Darwin and Asa Gray, for example, were well-aware of the 17th century debates which preceded them about the relation between art and nature through the works of such revered English writers as William Shakespeare and Thomas Browne. Furthermore, they used their understandings of these debates to inform and express their own thinking about the relation between artificial and natural selection. ???84DOMESTIC VARIETIES AND HUMAN INTERFERENCE  In a well-known 1860 review of Darwin’s Origin of Species (1859), the Harvard botanist Asa Gray wrote the following about the art of breeding (what Darwin had termed “artificial selection”):  ?“the art itself is Nature,” since the whole art consists in allowing the most universal of all natural tendencies in organic things (inheritance) to operate uncontrolled by other and obviously incidental tendencies. No new power, no artificial force, is brought into play.  192?Perhaps a little enigmatic, this quotation succinctly captures and defends a major theme in Darwin’s work. In the first four chapters of the Origin Darwin developed an extended analogy between artificial and natural selection. Likening nature to a breeder, he compared the creation of domestic varieties through artificial selection to the creation of natural races in the “state of nature.” Just as human selective breeding produces new domestic varieties with desirable characteristics, he argued, selection in nature, a result of the “struggle for existence,” produces new varieties and even species, filling up cracks in the economy of nature. In a letter written prior to the publication of the Origin, he explained to Gray that Nature is an “unerring” breeder, “which selects exclusively for the good of each organic being” (Darwin to Gray, Sept 5, 1857).  193?85 Gray [1860] 1963, 27; Darwin 1859.192 There is a voluminous literature about Darwin’s analogy, much of which is tangential. See Young 1971; Ruse 1931975; Evans 1984; Waters 1986; Bartley 1992; Richards 1998; Gayon 1998; Theunissen 2012. Unless otherwise noted, Darwin letters were retrieved from the Darwin Correspondence Database: https://www.darwinproject.ac.uk. In the eyes of his contemporaries, however, the value of Darwin’s comparison hinged on whether the creation of domestic varieties was merely a result of the artificial conditions in which they were raised and bred. A common sentiment in circulation prior to the publication of the Origin held that breeding was too unnatural to represent a natural process. This sentiment was often used to buttress arguments against drawing an analogy between domestic varieties and wild species. Charles Lyell wrote in 1832 that domesticated varieties were “extreme cases brought about by human interference, and not [...] phenomena which indicate a capability of indefinite modification in the natural world.”  The mathematician-194geologist William Hopkins asserted in a review of the Origin, “we have no right whatever to assume that nature will necessarily produce such effects at all when left to her own unobstructed operations, as those which she produces under man's interference”; he maintained that we “commit an error” when we assert that under ordinary conditions nature is capable of producing modifications similar to those observed in domestic varieties.  And 195Gray’s colleague at Harvard, Louis Agassiz, wrote in his copy of the Origin that Darwin’s mistake “has been to study the origin of species among domesticated animals” rather than wild ones.  Darwin disagreed with these assessments, responding to one commentator, “it is 196an error to speak of man ‘tampering with nature’.”  In the opening quotation, Gray defends 197?86 Lyell 1832.194 Hopkins 1860, 75.195 Agassiz wrote in his copy—sent to him to review—of the first edition of Darwin’s Origin of Species, “The 196mistake of Darwin has been to study the origin of species among domesticated animals exclusively instead of wild ones; his results concerning species are founded not on an investigation of species but on an investigation of breeds” (Agassiz papers, Ernst Mayr Library, Harvard University). Agassiz repeated his opinion in a review: “this process of raising breeds by the selection of favorable subjects, is in no way similar to that which regulates specific differences. Nothing is more remote from the truth than the attempted parallelism between the breeds of domesticated animals and the species of wild ones.” Agassiz 1860, 147. These comments were made by the French anatomist Georges Pouchet. See Darwin 1868, 2.197Darwin: artificial selection should not be seen as an unnatural interference in nature’s otherwise normal development. Instead, he responds, “the art itself [that is, of breeding] is Nature,” “no artificial force [...] is brought into play.”  Understood in this light, Gray’s comment should be recognized as part of a broader theme running through the Darwinian revolution—a theme referred to by historian John Cornell as, “Darwin’s reinterpretation of the meaning of art and nature.”  Darwin said, or 198implied, something significant—albeit highly contested—about the art-nature relationship which was not lost on his peers. This theme, we might recall, is central to another very well-known revolutionary period: the reconfiguration of 17th century natural philosophy known as the scientific revolution. Floris Cohen, author of the most comprehensive historiographical inquiry of the scientific revolution, writes, “Every interested thinker in the 17th century had to grapple with the distinction between ‘natural’ and ‘artificially produced’ nature and the root issue of how one was related to the other.”  Lorraine Daston provides my favorite 199articulation of this theme: “It is a platitude among historians of the Scientific Revolution that the seminal thinkers of the seventeenth century, most notably Bacon and Descartes, abolished the nature/art distinction by subsuming the artificial under the natural. However, it would be just as, if not more, accurate to claim that the distinction was collapsed [...] by subsuming the natural under the artificial.”  Regardless of the direction of subsumption, scholars of the 200early modern period generally agree that the relationship between these two categories was transformed in the 17th century, as we have seen above. ?87 Cornell 1984, 308. See also, Burnett 2009.198 Cohen 1994, 188.199 Daston 1995, 41-2.200 If discussions impinging on the relationship between nature and art played a pivotal role in both the scientific and Darwinian revolutions, what similarities do they share and what can we learn from them? It is my contention that this intellectual congruence is more than just nominal. I will argue that Darwin and Gray were aware of 17th century debates about art and nature, and furthermore, used their understandings of these debates to inform and express their own thinking about the relation between artificial and natural selection. Let me begin with a typical 17th century discussion of art and nature. After which I shall return to the Darwinian revolution. ?NATURE AND ARTIFICE IN THE SEVENTEENTH CENTURY  Early modern discussions of the relationship between art and nature found expression in a variety of contexts.  Here I will again take as an exemplar the mechanical philosophy 201of Robert Boyle, but I will elaborate beyond the discussion in Chapter 2, in a way that introduces the terms of 17th century debates about art and nature that are relevant for the present context.  Robert Boyle likened nature to an intricate work of art, or as he characteristically put it, an “engine.” In his view, clocks provided a commonplace and appropriate model for the natural world (Figure 4.1).  ??88 See, Daston 1998. and Bensaude-Vincent and Newman 2007.201Figure 4.1: An Elaborate Astronomical Clock in Strasbourg (c. 1875). An example of the type of artifice likened to nature by Robert Boyle. (© Wellcome Library, London) ? In likening nature to an clock, Boyle sought to distance himself from what he called the “vulgarly received notion of nature,” a confused and confusing notion, fabricated by the ancients, and passed down through the Medieval period and Renaissance.  The vulgar 202notion of nature was problematic for a number of reasons, notably, it encouraged what Boyle deemed unintelligible anthropomorphisms that appropriated praise rightfully belonging to God. A typical example was the oft-cited phrase, “nature abhors a vacuum.” Nature, Boyle ?89 Boyle [1686] 1999.202thought, could not abhor anything, nor could it serve as an active arbitrator in philosophical debates.   Although his primary motivation was theological, likening nature to artifice was consistent with Boyle’s natural philosophy as it helped justify experimentation as a legitimate form of knowledge production.  According to the “vulgar,” Aristotelian-inspired philosophy 203Boyle denounced, the proper object of study was nature’s due course. Natural processes, it was purported, were goal-directed and the aim of natural philosophy was to understand these goals. The “nature” of an acorn, for example, could be found by watching an acorn develop into an oak tree under typical, everyday conditions.  204 As a consequence, and as we saw in Chapter 2, this “vulgar” philosophy advocated a distinction between artifice and nature—a point of view so entrenched throughout the renaissance that historians have referred to it as a “habit of the understanding” or “conceptual reflex.”  Art was understood as an intervention upon or frustrating of nature’s ends. It was 205able to imitate or overcome nature, but it could not reveal nature, since through art, nature’s ends were subverted by human ends.  The object of Aristotelian natural philosophy could 206not be apprehended through the contrived circumstances characteristic of experimentation.  In contrast, for Boyle nature was artificial because it was God’s work of art. This view of nature as a passive work of artifice cohered well with the rise of experimentalism and ?90 Boyle thought that the vulgarly received notion was blasphemous as it often portrayed nature as an artisan 203who oversaw the natural world. This imbued nature with an autonomy that stole praise away from God—God needed no intermediary (see Boyle [1686] 1999; see also Daston 1995). To Boyle, nature was simply a machine and God the engineer that set the clockwork system in motion. See, Dear 1995.204 Park and Daston 2006, 265 and 276. See also, Tayler 1966 and Close 1969.205 See, Dear 1990.206with the rising belief that nature is regular, uniform, and can be described according to laws. By likening nature to artifice in this way, Boyle provided a foundation for experimentation.  Once nature was understood as machine-like, the usefulness of 207experiments—and in general methods that involved manipulating nature’s mechanisms—became obvious. Just as a clock is the product of a designer, so nature was seen as a great engineering feat created and set in motion by the hand of God (Figure 4.2). If nature was like a clock, one should approach it in the same way as a clockmaker approaches a foreign watch: determine the laws that govern the regular and uniform movement of its variously sized and shaped parts by taking it apart and fiddling with its pieces. Lambasting the vulgar practice of passive observation, Boyle wrote, “He must be a very dull Enquirer, who, demanding an Account of the Phaenomena of a Watch, shall rest satisfied with being told, that ‘tis an Engine made by a Watch-Maker; though nothing be thereby declar’d of the Structure and Co-aptation of the Spring, Wheels, Ballance, and other Parts of the Engine; and the manner, how they act on one another.”   208??91 Shapin 1996, 97-8.207 Boyle [1686] 1999, 558208Figure 4.2: Clockmaker About to Fix a Clock. Published in The universal magazine (1748).             (© Wellcome Library, London) ? The clock metaphor runs deep since it could be used to justify experimental-manipulative practices and the uses of instruments which aided the unmediated senses or created unnatural conditions. Joseph Glanvill, recall, wrote that anatomy was useful because it tended “mightily to the eviscerating of nature, and disclosure of the springs of its motion.”  Given these colorful metaphors, it is unsurprising that Boyle wrote, “Proper 209comparisons do the imagination almost as much service as microscopes do the eye.”  As 210?92 My emphasis. Glanvill 1668 [1958].209 Quoted in Mayr 1986, 82.210the distinction between the artificial and the natural broke down, the methods appropriate for the study of machines and the mechanical arts, and the rigor these philosophers saw accompanying these methods, could be applied to the study of nature.  Likenings of nature to artifice throughout the early modern period were certainly not unique to Boyle and the Royal Society.  An influx of artifacts, like clocks and automata, 211and artifactual techniques, like plant grafting, in practical, philosophical and leisurely affairs, sparked debates about how artifice related to nature which permeated the wider culture, and found expression in the works of playwrights and scholars, such as William Shakespeare and Thomas Browne, both of whom we will encounter below (Figure 4.3).  The conceptual 212coherence of the art-nature distinction was widely challenged on both practical and theoretical grounds in the 17th century, and was discussed in a variety of contexts. As we saw in Chapter 2, modern commentators often present the collapse of the art-nature distinction during this period as a prerequisite or ingredient for “modern science” because it helped underwrite experimentation as a legitimate source of natural knowledge.  As we’ve seen, 213the art-nature relationship was far from resolved and returned for further disputation in the context of early evolutionary biology. ??93 See for the early modern Italian context: Grafton 2007. See also Daston and Park 1998.211 See Mayr 1986 and Riskin 2007.212 See Shapin 1996. For a dissenting view see Newman 2004 and 2006.213Figure 4.3: A Mechanical Toy. Hydraulic power drives a musical box and works a skeleton puppet. From Mechanica hydraulico-pneumatica, by Gaspar Schott (1657). (© Wellcome Library, London) ?THE ART ITSELF IS NATURE  Just as Boyle’s clock metaphor was a product of the technological innovation and fascinations of its time—a fact Boyle exploited—so too was Darwin’s use of breeding.  As 214urban populations increased dramatically throughout the Victorian era, agriculture, horticulture, animal husbandry, and so on, developed as never before (Figure 4.4). ?94 See Young 1985.214Periodicals such as the Gardener’s Chronicle disseminated information about these practices widely.  Darwin’s father-in-law (and uncle) Josiah Wedgwood was himself a leading sheep 215breeder as well as a successful businessman.  It is thus not surprising that Darwin noted in 216the margin of William Youatt’s famous volume on cattle, “As this simple principle [of selection] only lately discovered even in most reliable practice, no wonder not discovered as theory of species.” Darwin was perhaps aware that technological or practical innovation—in this case the art of breeding—often predates the discovery of a theory about the natural world.  217Figure 4.4: Victorian’s in Domesticated Nature, “Botanising.” From New illustration of the sexual system of Carolus von Linnaeus, by Robert John Thornton (1807) (© Wellcome Library, London) ?95 Ruse, M. 1979. Secord 1985. See also Browne 1995.215 Ruse1979, 178.216 March, 1840. Quoted in Cornell 1984, 325.217? Although Darwin’s likening of Nature to a breeder may have been timely, there were notable obstacles to overcome. First there was a social obstacle: although breeding occupied a critical place in the wider cultural context, naturalists tended to distrust breeders and their art (and vice versa), creating—if not full-on animosity—then at least mutual ignorance through a divergence of interests.  My goal in this chapter, however, is to address obstacles 218of a more intellectual nature, as exemplified by Lyell’s and Hopkins’ remarks about the Origin: should the art of breeding be seen as an unnatural interference? how did this impinge on debates about the proper object of study in natural history?   In asking these questions, my discussion parts ways with previous histories of the relation between artificial and natural selection. Such histories have focused on, for example, the use of analogies in science, the use of scientific rhetoric, the genesis of Darwin’s own thinking, Darwin’s debt to the philosophy of science of his day, and whether such an analogy was or is justified.  I’m interested in a different issue that Darwin had to worry about, 219namely, how to justify the study of domesticated varieties as a way to learn about natural species, when the former (and the practices giving rise to them) were traditionally looked at as being outside the “state of nature” and thus improper objects of study. In their efforts to deal with this issue, Darwin and Gray gained insight from the 17th century literature they were reading. This worry was especially germane because Darwin’s first-hand experience ?96 Secord 1985, 522. But see, Ritvo, H. 1987, on “aristocratic” breeding as an exception to the trend Secord 218outlines. See the citations above.219with the natural species he wrote about diminished significantly following his return from the Beagle voyage in 1837.   Allow me then to return to Asa Gray’s review of the Origin. His comment, “the art itself is Nature,” is taken from Shakespeare’s 1623 play, The Winter’s Tale. It should not be surprising that Gray quotes Shakespeare, since early modern writers like Shakespeare, Michel de Montaigne, John Milton and Thomas Browne were widely read and beloved in the Victorian era, often even more so than in their own day.  In The Winter’s Tale, Shakespeare 220contrasts two perspectives on the relation between art and nature through a dialogue between the King Polixenes and the shepherdess Perdita. The dialogue represents early modern debates and gives a sense of the broader intellectual atmosphere in which men like Boyle were working. Perdita says she has no “streaked gillyvors [carnations]” in her garden, referring to them disdainfully as “Nature’s bastards,” since they are a product of both art (techniques of grafting and breeding) and of nature, rather than “great creating Nature” alone.  King Polixenes objects that manipulated nature should not be treated with scorn and 221points out that the art of producing hybrid flowers is itself a part of nature and therefore worthy of appreciation.   222 Shakespeare’s intention in this section was to signal to his audience that the already well-worn topic of the relations between art and nature could speak to the broader social issues discussed in the play.  Similarly, Gray’s intention in quoting Shakespeare is to signal 223?97 Beer 1983 and 1985.220 Shakespeare [1623] 1998, 68.221 For a 17th century discussion of the horticultural arts see Austen 1653. See pp. 28-9 for grafting. For a broad 222discussion of how art and nature were compared in 17th century literature on gardening, see Tayler 1966, 16ff. See the Introduction by Frank Kermode, in Shakespeare The Winter’s Tale. For an extended discussion, see 223Tayler 1966, 121ff.to his audience that the playwright’s well-known dialogue is relevant to his scientific discussion of the relation between domestication and natural history. Gray sides Darwin with King Polixenes: artificial selection is itself a part of nature. Domesticated organisms are no less worthy of study because they are the products of a human art; in fact, they are particularly valuable for this very reason. Gray laments that domesticated races have “been generally neglected by naturalists, because these races are not in a state of nature; whereas they deserve particular attention on this account, as experiments [...] ready to our hand.”  It 224is worth noting that, as with Boyle, Gray’s endeavor to dissolve the art-nature distinction also ends up underwriting an experimentalist agenda.  Seventeenth century literature was in the immediate background of Darwin’s own thinking when he first likened the art of breeding to nature. Many Darwin scholars locate his earliest use of the analogy between artificial and natural selection to November 1838.  It is 225during this period that Darwin starts to employ this familiar analogy in his notebooks; in early December he records, “if nature had had the picking she would make such a variety far more easily than man” and “It is a beautiful part of my theory, that domesticated races of organics [sic] are made by precisely the same means as species.”  It is suggestive that a few 226months earlier, in late August, Darwin had been rereading the 1643 Religio Medici, by Thomas Browne.  Browne was a physician and natural philosopher, influenced by Bacon 227?98 Gray [1860] 1963, 21.224 See Hodge 2009.225 Notebook E, 63 [after Dec 4th] and Notebook E, 71 [after December 16th]. Unless otherwise noted, 226Darwin’s notebooks were retrieved from Darwin Notebooks Online: http://darwin-online.org.uk/reproductions.html. For the Darwin enthusiasts: this reading occurred after Darwin’s reading of the Sebright and Wilkinson 227breeding pamphlets in early 1838, just following his reading of Brewster’s review of Comte in early August, just before he starts to refer to domesticated organisms as adapted—rather than ill-adapted—, and before his reading of Malthus in late September.and admired by Boyle. Darwin’s notes on Browne’s text—crammed in a transmutation notebook between thoughts on varieties versus species, natural arrangements of animal groups, and domestication—make it clear what sections interested him most.  In these 228sections Browne dealt directly with the relation between art and nature in a characteristically 17th century way.  He writes, 229?now nature is not at variance with art, nor art with nature, they being both the servants of his providence: art is the perfection of nature [...] In brief, all things are artificial; for nature is the art of God.  230?Like King Polixenes, Browne suggests that the categories of art and nature should not be opposed in the classic, Aristotelian way. He describes nature, similarly to Boyle, as being composed of “narrow engines” and “contrived parts” with God as its “excellent Artist.” His conclusion is unequivocal: “all things are artificial.”  Browne sought to subsume the natural 231under the artificial, undermining a clear distinction between the two by rendering Nature God’s work of art. Darwin records the last line in his notebook, “p. 23 for Nature is the art of God.”  232 Browne’s discussion is instructive for other reasons as well. The passages Darwin read consider also the beauty of God’s design. Browne wrote, “I hold there is a general ?99 Kohn 1987, 350.228 Daston 1995, 40. Daston and Park 2006, 300; Preston 2005, 34.229 Browne, T. [1643] 1862, 35.230 Browne [1643] 1862, 30-5.231 Notebook D, 54e; Kohn “Notebook D,” p. 350. There is a discrepancy between the online and print 232transcription of Darwin’s notebooks. Online the Browne passage is transcribed as “for Nature is the act of God,” which is incorrect and fails to capture the meaning of Browne’s discussion and why Darwin would care about it. The print version of Darwin’s notebooks is transcribed correctly and is thus preferable.beauty in the works of God, and therefore no deformity in any kind or species of creature whatsoever.”  This included creatures such as toads, bears, or elephants, but also those 233labeled monstrosities, “wherein notwithstanding there is a kind of beauty, nature so ingeniously contriving the irregular parts, that they become sometimes more remarkable than the principal fabric.”  In the writings of Darwin’s contemporaries domesticated varieties 234were derogatorily referred to as monstrosities, and often neglected by naturalists for that reason.  Lyell, for instance, objected that species hybrids could not be used as examples to 235corroborate theories of transmutation in nature since a significant “number of such monsters” could only be “obtained by art.”   236 Darwin’s notes demonstrate that through the first half of 1838 he would have largely agreed with Lyell’s assessment. Until at least July 1838 he frequently contrasted the formation of wild “natural” species with the making of domesticated varieties or “monstrosities” by humans.  His famous analogy—a comparison of wild, natural species 237formation and the creation of domestic varieties—was thus a complete reversal of his earlier thoughts on the matter.  Darwin’s reading of Browne in August 1838 thus fell at a time 238when his thinking about the relation between the art of selective breeding and wild species ?100 Browne [1643] 1862, 34-5.233 Ibid., 35.234 See Secord 1985. Again, see Ritvo 1987 on aristocratic breeding. As Ritvo shows, not all domesticated 235species were judged in such a way. My interest is not in how domesticated varieties were viewed tout court, but how they were viewed by the naturalists with which Darwin would have engaged and held in high esteem. In this regard, Darwin and Gray both lament the lack of “scientific” treatment of domesticated varieties. This argument is consistent with Ritvo’s claim that animal-human relations changed throughout the early-18th and late-19th centuries such that animals became significant primarily as objects of human manipulation. See Ritvo 1987, 2-3. Lyell 1932, 51.236 See Hodge 2009, 66. For further discussions of Darwin’s changing views during this time, See Hodge and 237Kohn 1985, 197-200; Ospovat 1981. I’d like to thank Jon Hodge for pointing this out to me. 238formation was fruitfully transforming from contrast to comparison. It should thus not be overlooked that Browne’s discussion put considerable tension on the coherence of the label “monstrosity” itself, and again like King Polixenes, recommended that manipulated nature not be treated with scorn. In terms of Darwin’s changing point of view, such a discussion was fuel for the fire. After August 1838, we should note, Darwin’s entries on artificial selection become considerably more frequent.   239 Perhaps we should also keep this literature in mind when considering Darwin’s later reflections on monstrosity and abnormality, as when he warns that “We should [...] be cautious in deciding what deviations ought to be called monstrous.”  After all, “the great 240plume of feathers on the head of the Polish cock has been thus designated, though plumes are common with many kinds of birds” (Figure 4.5).   241??101 See Schweber 1977, 257.239 Darwin, 1868: 2, 413.240 Ibid.241Figure 4.5: The Polish Pigeon or Polish Cock. From Darwin’s 1868 The Variation of Animals and Plants under Domestication. Often the Polish Pigeon was referred to as a monstrosity because of the great plume of feathers on its head. Darwin was wary of such attributions. (© Wellcome Library, London) ? 17th century discussions of art and nature permeated Darwin’s intellectual milieu when he was thinking about the relation between domestication and nature. When Darwin described the history of domestication as “an experiment on a gigantic scale” and protested that “It is an error to speak of man ‘tampering with nature’,” it is hard to imagine that he ?102wasn’t aware of the similarities between his own reliance on the art of breeding and well-known 17th century debates about art and nature.  Similarly, when Darwin wrote in 242conclusion to his 1842 Essay, “We must look at every complicated mechanism and instinct, as the summary of a long history, of useful contrivances, much like a work of art,” and then in 1844 changed “work of art” to “great mechanical invention,” we likewise see his continuity with the 17th century as we find him modifying an early modern turn-of-phrase. His use of manipulated nature to produce knowledge applicable to unmanipulated nature brought art under the purview of natural history, a bold move which could gain legitimacy from comparisons with the works of Shakespeare, Browne and Boyle that undermined the distinction between art and nature.  243 Darwin’s later, more systematic writings also undermined that distinction. Brief but profound examples can be found throughout his discussions of unconscious selection and sexual selection, other types of selection that have traditionally received less scholarly attention. In the next chapter we will consider these cases in a different context. ??103 Darwin 1868:1, 2-3.242 There are a number of complexities which, although slightly tangential, are worth mentioning. First, there is 243a large literature about what Darwin was using studies of domesticated varieties for prior to the publication of the origin. Was he looking at them as examples of natural selection? Was he looking more specifically for laws of inheritance? See Evans 1984 and Bartley 1992. For the philosophical context of experiments when Darwin was performing them see, Schweber 1977. In the text I have left this unspecified on purpose, as my concerns are more general than this. Second, one of the problems with the artificial-natural selection analogy was that natural species were infertile with one another, whereas domestic varieties, no matter how extreme their trait differences, continued to interbreed. There is another literature about the extent to which Darwin saw this as a problem, and following from this, whether he employed the analogy solely didactically. See, e.g., Kottler 1985. I believe, as hinted at in the below text, there are good independent reasons to think Darwin saw artificial selection as a type of natural selection, especially later in his life. I believe he possibly held this belief as early as November 1838, when he started to see domestic varieties as adapted to their particular circumstances (See Hodge and Kohn, 1985, 198). Even in the Origin, he didn’t refer to the category of “artificial selection” until page 109—prior to that referring to “selection by man” which was a category including a range of types of selection from unconscious to methodical.??NATURE’S EXPERIMENTS  Needless to say, Darwin’s and Gray’s primary interlocutors were not 17th century natural philosophers, but their own contemporaries. Darwin’s reliance on breeding located him amidst 19th century debates over the proper place of experiment in generating knowledge, and sets him at the crossroads between two strongly supported traditions. The first tradition was more closely aligned with natural history and dismissed the value of human manipulation in the study of nature. This tradition had a rich lineage, itself tracing back to the 17th century. Georges Cuvier was a notable endorser; his work was tremendously influential in early-19th century England and pervaded Darwin’s intellectual environment.  244Late in life Darwin referred to Cuvier as one of his “two gods” (the other being Linnaeus)—an expression of his general admiration for the French naturalist.  245 Cuvier’s authoritative discussion of domestication in the Discours Preliminaire (1812; translated 1813) argued that variation under domestication is superficial, a result of humans tampering with nature’s order and nothing more.  “Nature is [...] careful to prevent 246the alteration of species that could result from their mixture,” he writes, “It requires all kinds of ruse and human constraint to achieve these unions.”  Through domestication, “man 247develops all the variations of which the type of each species is susceptible, and draws out ?104 Ospovat 1981.244 Darwin to William Ogle, 12 Feb 1882.245 The English translation of the Discours Preliminaire was commissioned by none other than Robert Jameson, 246Darwin’s dull-mannered geology professor at Edinburgh; Rudwick 1997, xi. Outram 1984, 141ff. Cuvier 1812, translated in Rudwick 1997, 227.247forms that the species, left to themselves, would never have produced.”  The degree of 248variation under domestication is unnatural and proportional to its cause, which he referred to as “slavery.”  Cuvier’s view influenced Charles Lyell’s thought, as should be evident in the 249above quotation from the Principles of Geology, where Lyell dismissed the epistemological value of domesticated varieties.   Cuvier’s view of domestication was consistent with the suspicious outlook towards experimentation and manipulated nature he inherited from French physiologists like Xavier Bichat.  In Le Règne Animal ([1817] 1840), Cuvier explained that there were three general 250scientific methods: ?Calculation, so to speak, commands Nature; it determines phenomena more exactly than observation can make them known: experiment forces her to unveil; while observation watches her when deviating from her normal course, and seeks to surprise her.  251?He concluded, however, that because of the holistic, complex nature of living phenomena, “Natural History will long remain [...] one of pure observation.”  The most effectual mode 252of observation in natural history, he argued, was non-manipulative and comparative. These represent “kinds of experiments ready prepared by Nature, who adds to or subtracts from ?105 Ibid.248 Ibid. The 18th-century French naturalist and polymath, Georges-Louis Leclerc Comte de Buffon, similarly 249referred to domesticated animals as carrying the “stigma of slavery” in his Histoire naturelle. It is the duty of the naturalist, he wrote, “to separate artifice from Nature; and never confound the animal with the slave, the beast of burden with the creature of God.” See Buffon 1780, 301-2. Also see, Roger 1997, 303. Albury 1977; Outram 1986. Bichat himself was in this regard influenced by earlier physiologists of the 250Montpellier school who were reacting to the type of mechanical philosophy advanced by natural philosophers like Boyle and Browne. See, Williams 2003. Cuvier, G. [1817] 1840, 14.251 Ibid.252each of them different parts, just as we might wish to do in our laboratories.”  By 253successively studying the same bodies in “the different positions in which Nature places them” we are able to recognize constant relations that indicate underlying causal relationships.  For example, Cuvier found that the relative size of the liver was inversely 254proportional to the relative size of the respiratory apparatus across many animal species. Insects lacked a liver and underwent cutaneous respiration over the entire outer surface of their bodies. Mammals, on the other hand, had a liver and breathed only through a single bodily system, the lungs. He concluded from this comparison, or “natural” experiment, that in mammals the liver assists the action of the lungs.  255 The Cuvier outlook permeated Darwin’s early life, especially the time he spent studying medicine in Edinburgh (1825-1827). Robert Grant, a teacher at Robert Knox’s famous extramural school of medicine and later Professor of Comparative Anatomy at University College London, allowed the young Darwin to assist him collecting specimens near Edinburgh.  Grant, like Knox, was familiar with Cuvier’s work and even spent time in 256France studying under Cuvier himself. Given that Darwin and Grant were close and conversed over many topics while collecting, it would be strange if Cuvier’s work did not arise for discussion.  At this time in Edinburgh “vitalistic” theories were also a commonplace, and while the content of these theories is perhaps less relevant to Darwin, the methodological ?106 Ibid.253 Ibid., 15.254 Example from Outram 1984, 331.255 Browne 1995.256prescriptions they espoused often had direct ties to Cuvier’s work.  Darwin’s readings in 257Edinburgh, for instance, included works by John Fleming and John Abernathy, both of whom sympathized with Cuvier and espoused a form of vitalism.  In his Philosophy of Zoology 258(1822), for instance, Fleming cites Cuvier specifically, referring to Le Règne Animal as “the most valuable of modern synthetic arrangements.”  Methodologically speaking, Fleming’s 259work paralleled Cuvier’s in many ways. Particularly relevant for the types of allegations lobbed against Darwin’s work, Fleming referred to domestic breeds as “monstrosities” and discussed the “deranging influence of domestication” on animals species.  Darwin was thus 260presented with Cuvier’s views on domesticated varieties through works such as these.   Standing at the crossroads and looking in the opposite direction, Darwin would have encountered another tradition, more closely aligned with the philosophy of the physical sciences and one that had less scruples with experimentation. For example, in early August 1838, shortly before his reading of Thomas Browne, Darwin closely studied a review of Auguste Comte’s Cours de philosophie positive—so closely he developed an intense headache.  The review was an overview of Comte’s philosophy in general, but particularly 261relevant was an outline of the role of experiment which stressed the usefulness of studying phenomena under artificial or unnatural conditions. Through this powerful instrument of investigation, the reviewer commented, “we observe bodies out of their natural state; by ?107 For a summary of vitalism in Edinburgh in the early 19th-century, see Jacyna 1983.257 Browne 1995, 85. Darwin re-read Flemings book on December 15th, 1840, when he wrote “—well read—258references at end.—,” and potentially also between June 1st, 1838 and December 1839. Fleming 1822, xiv.259 Ibid., 36.260 See Schweber 1977, 241. Brewster 1838. The review is published anonymously. On the implications of this 261reading, see: Schweber 1977 and Schweber 1978.placing them in artificial aspects and conditions contrived for the purpose of exhibiting to us, under the most favourable circumstances, their phenomena and their properties.”  As shown 262by historian Silvan Schweber, such quotations bear a close resemblance to the text of Darwin’s earliest sketches of his theory.  Darwin’s reading of 17th century literature which 263muddled the distinction between “the artificial” and “the natural” was carried out alongside this literature which exalted the study of unnatural nature. The 17th century debates would have thus appeared quite germane, even timely, given the philosophical context in which Darwin was situated.   While certainly opposed to dismissals of the experimental value of domesticated varieties, Darwin was nonetheless in a position to appreciate the concerns of Cuvier’s following. Despite his reliance on domestication as a gigantic experiment, Darwin did at times express his own reservations towards experimentation. For instance, in the early 1860s he was preoccupied with experiments on the fertilization of orchids by insects.  One 264experiment had him artificially remove the nectaries of flowers and observe whether insects still visited the flowers. After performing these experiments he wrote to the English Botanist George Bentham: ?I have for years been attending to insect fertilisation of Orchids, & I shd. very much from several curious reasons like to see what effect no nectary will have produced on ?108 Brewster 1838.262 He writes, likely in 1842, “The most favourable <conditions> circumstances for variation seems to be 263propagation for many generations in domesticity [...] the influence of domesticity seems to resolve itself into conditions different from those under which nature placed the organisms.” See Schweber, 1977, 256. Vorzimmer 1975. This early “Pencil Sketch” of Darwin’s theory was likely written in 1842, not 1839. The “<>” indicate corrections Darwin made at the time of writing. See Darwin 1862.264the visits of insects. I once tried cutting off the nectaries; but nature's cutting off would be much better.  265?Bentham had recently discovered a variety of Orchis pyramidalis that was naturally lacking a nectary. In this letter Darwin was asking for a few samples so that he may repeat his experiment with them. But, we might inquire, why would Darwin consider “nature’s cutting off” to be preferable to his own experimental intervention? Here I think we can see his attempting to preempt objections to the use of experiments in natural history. The flowers Bentham discovered were a perfect opportunity for a “natural” experiment in accordance with Cuvier’s prescriptive reflections on method, indicating that Darwin took such prescriptions to heart. ?CONCLUSION ?Up to the age of thirty, or beyond it, poetry of many kinds, such as the works of [John] Milton, [Thomas] Gray, [Lord] Byron, [William] Wordsworth, [Samuel Taylor] Coleridge, and [Percy Bysshe or Mary] Shelley, gave me great pleasure, and even as a schoolboy I took intense delight in Shakespeare, especially in the historical plays.  266Charles Darwin, 1876, Autobiography ?As these autobiographical reflections reveal, Darwin, like many 19th century intellectuals, was well-versed in 17th and 18th century literature and its 19th century heirs—his reading notebooks are full of references to this body of work. Much of this literature was ripe with discussions of art and nature, as was, it’s worth mentioning, a 19th century disagreement ?109 Darwin to Bentham, June 17th, [1861].265 Barlow 1958, 138.266between Coleridge and Wordsworth over the proper interpretation of art and nature in Shakespeare’s work.  It is unsurprising, then, that Darwin, Gray, and their intellectual peers, 267would have been particularly well prepared to understand and debate the relation between art and nature. Darwin’s use of this dichotomy to categorize different types of selection should be approached with this literature in mind.   My aim in this chapter was to assess an intellectual congruence between the scientific and Darwinian revolutions: the role played by a likening of nature to artifice. The brief history I have provided indicates that this congruence is more than just nominal. Darwin and Gray were well-aware of important 17th century debates about the relation between art and nature through the works of such revered and widely-read English writers as William Shakespeare and Thomas Browne. These 17th century debates played a pivotal role in the ideology of the scientific revolution, as demonstrated through the philosophical reflections of Robert Boyle. I have argued that Darwin and Gray used these 17th century debates to inform their own discussions, particularly those concerning the relation between artificial and natural selection. Darwin’s reliance on the art of breeding as a central element in his theory of natural selection cohered well with 17th century texts that argued for a dissolution of the art-nature distinction by claiming that Nature was simply the art of God. Darwin was aware of incompatible understandings of natural history that themselves trace back to 17th century reactions to a mechanical worldview. That he did not fully accept these methodological ?110 On Coleridge and Wordsworth, see Abrams 1958, 122. Their disagreement, maybe unsurprisingly, is 267expressed through Shakespeare’s dialogue between King Polixenes and Perdita, the same dialogue used by Gray to defend Darwin above.prescriptions against human-manipulated nature show him to be taking a stand on a 19th century debate that could be illuminated by one that occurred throughout the 17th century. 
?111??CHAPTER 5 ?Selection in the State of Nature Darwin and Wallace on Domestication ????????I find myself having wondered several times at what time—how late you had to be born or at least professionalized in evolutionary and populations terms before the word selection lost all vestiges of its everyday meaning. It was many years before it occurred to me that that word was the same as the word selection [Lewontin pauses to spell it out: s-e-l-e-c-t-i-o-n] in English. When I realized that, then I reunderstood the whole plan of writing of Darwin’s Origin of Species. Richard Lewontin, 1974, Amidst a Conference on the Evolutionary Synthesis ? Although there are many similarities between their theories of evolution by natural selection, Alfred Russel Wallace and Charles Darwin continued throughout their lives to disagree about the significance of domestication. Was it the differences between artificial and natural selection which mattered or the similarities? What were the purported differences? Building on the analysis in the preceding chapter, I will demonstrate how different ways of conceiving of the artificial-natural distinction were used to think through and express deep seated differences of opinion about whether domesticated organisms could be used to shed light on natural species change. ???112DOMESTICATION, REVERSION, AND NATURAL SELECTION  One of the now famous differences between Darwin’s and Wallace’s theories of evolution is the role played by domestication, or what Darwin called “artificial selection.”  268This difference persisted throughout their careers, onwards from their joint papers delivered to the Linnaean Society in 1858—the first time the learned public heard of evolution by natural selection.   In his 1858 contribution, Darwin drew repeatedly on an analogy between domestic varieties and natural races to argue for natural selection. He writes, “Let this work of selection on the one hand, and death on the other, go on for a thousand generations, who will pretend to affirm that it would produce no effect, when we remember what, in a few years, Bakewell effected in cattle, and Western in sheep, by this identical principle of selection?”  269He repeats, “I can see no more reason to doubt that these causes in a thousand generations would produce a marked effect, and adapt the form of the fox or dog to the catching of hares instead of rabbits, than that greyhounds can be improved by selection and careful breeding.”  At which point, he introduces natural selection, 270?Man, by this power of accumulating variations, adapts living beings to his wants [...] Now suppose there were a being who did not judge by mere external appearances, but who could study the whole internal organization, who was never capricious, and should go on selecting for one object during millions of generations; who will say ?113 See Young 1971; Bowler 1984; Cornell 1984 Rheinberger & McLaughlin 1984; Beddall 1988; Bartley 1992; 268Gayon 1998, etc. Darwin and Wallace 1858, 49269 Ibid., 49.270what he might not effect? [...] I think it can be shown that there is such an unerring power at work in Natural Selection  271? Darwin had accepted the importance of this comparison many years before, writing to his friend J. D. Hooker in 1844, “I believe all these absurd views [older views of immutability and mutability of species alike], arise, from no one having, as far as I know, approached the subject on the side of variation under domestication, & having studied all that is known about domestication.”  When the Origin of Species was published in 1859, the 272analogy between domestication and natural selection only seemed to gain in significance, extending throughout the first four chapters of the book. His 1868 book, The Variation of Animals and Plants under Domestication, lay in the same vein, as the name implies.   In his 1858 paper, Wallace, on the other hand, argued for the same conclusion—species mutability and common descent—by drawing on the differences between domesticated varieties and “wild animals in a state of nature.” Unaware of Darwin’s thinking but aware that an analogy with domestication had recently been drawn to support the opposite conclusion, species immutability, Wallace opted to argue against the analogy. He concludes, “no inferences as to varieties in a state of nature can be deduced from the observation of those occurring among domestic animals. The two are so much opposed to each other in every circumstance of their existence, that what applies to the one is almost sure not to apply to the other.”  273?114 Ibid., 51.271 Darwin to J D Hooker, 10 Nov 1844, No789.272 Darwin and Wallace 1858, 61.273 While we might expect this difference to simply reflect their ignorance of each other’s theories in 1858, this was not the case. As time went on Darwin remained committed to the analogy between artificial and natural selection, and Wallace continued to approach the topic with extreme caution and noted reservation. When Wallace wrote Darwinism (1889) after Darwin’s death—a textbook of evolution by natural selection laid out in a structure intentionally similar to the Origin—he tellingly professed in the Preface: “It has always been considered a weakness in Darwin’s work that he based his theory, primarily, on the evidence of variation in domesticated animals and cultivated plants,” and thus, he, “endeavoured to secure a firm foundation for the theory in the variations of organisms in a state of nature.”  274 This longstanding difference in their views is often explained by drawing attention to how each of them dealt with the phenomenon of “reversion”: whether varieties—natural or domesticated—have a tendency to physically revert over time back to their ancestral type.  275This common view creates the impression that Wallace thought reversion undermined any attempt at drawing an analogy between domestication and natural selection, whereas Darwin ?115 Wallace [1889] 1897, vi.274 Rheinberger and McLaughlin 1984; Beddall 1988; Richards 1998, 2005; Gayon 1998. Two other ways touch 275on differences between Darwin and Wallace with regard to (i) the importance of selection on permanent varieties, rather than individuals (see Bowler 1976, Gayon 1998; and see Kottler 1985 and Fagan 2008 for dissenting views) and (ii) relatedly, the fact that domesticated varieties could diverge from one another a great deal physically, but still interbreed (T. H. Huxley didn’t like this; see Huxley 1906, 256). The first is tied somewhat to reversion and I will return to it below. This view can explain why Darwin and Wallace differed in 1858, but not by itself why the difference maintained itself despite Wallace later accepting individual variation as important. The second is much more complicated. Although Darwin and Wallace had a fairly extensive discussion of hybrid sterility and whether species sterility could be brought about through natural selection (Wallace thought it could, Darwin thought it couldn’t), Wallace never raises the objection that domestic sterility makes them different from wild species. In Darwinism (1889, 184-5) he says that the sterility of domestic varieties can be explained away in terms of our preserving their sterility (because it is important to us) through domestication. Unlike Huxley, he did not think this was a problem for natural selection because he didn’t draw the artificial selection analogy in the first place. Darwin and Wallace’s discussions are further complicated by an underlying ambiguity over the meaning of “species” (Beatty 1982, 1986; Rheinberger & McLaughlin 1984).saw things differently. It creates the further impression that the connection between accepting reversion and denying an analogy with domestication was unproblematic.   This explanation is supported by what Wallace says in his 1858 paper and how Darwin responds in Chapter 1 of the Origin.  Although somewhat scattered, Darwin’s 276discussion is worth quoting in full. ?Having alluded to the subject of reversion, I may here refer to a statement often made by naturalists—namely, that our domestic varieties, when run wild, gradually but certainly revert in character to their aboriginal stocks. Hence it has been argued that no deductions can be drawn from domestic races to species in a state of nature. I have in vain endeavoured to discover on what decisive facts the above statement has so often and so boldly been made. There would be great difficulty in proving its truth: we may safely conclude that very many of the most strongly-marked domestic varieties could not possibly live in a wild state. In many cases we do not know what the aboriginal stock was, and so could not tell whether or not nearly perfect reversion had ensued. It would be quite necessary, in order to prevent the effects of intercrossing, that only a single variety should be turned loose in its new home. Nevertheless, as our varieties certainly do occasionally revert in some of their characters to ancestral forms, it seems to me not improbable, that if we could succeed in naturalising, or were to cultivate, during many generations, the several races, for instance, of the cabbage, in very poor soil [...], that they would to a large extent, or even wholly, revert to the wild aboriginal stock. Whether or not the experiment would succeed, is not of great importance for our line of argument; for by the experiment itself the conditions of life are changed. If it could be shown that our domestic varieties manifested a strong tendency to reversion,—that is, to lose their acquired characters, whilst kept under unchanged conditions, and whilst kept in a ?116 It is assumed that Darwin is responding to a Wallacean-type argument when he discusses reversion at the 276beginning of the Origin. See Rheinberger & McLaughlin 1984 and Gayon 1998. This puzzle is also further complicated by confusing things Darwin sometimes says in letters, for example, he writes to Wallace in one of their earliest exchanges, “I have acted already in accordance with your advice of keeping domestic varieties & those appearing in a state of nature, distinct; but I have sometimes doubted of the wisdom of this, & therefore I am glad to be backed by your opinion” (May 1, 1857).considerable body, so that free intercrossing might check, by blending together, any slight deviations of structure, in such case, I grant that we could deduce nothing from domestic varieties in regard to species. But there is not a shadow of evidence in favour of this view: to assert that we could not breed our cart and race-horses, long and short-horned cattle, and poultry of various breeds, and esculent vegetables, for an almost infinite number of generations, would be opposed to all experience.  277? This quotation suggests that Darwin sees nothing compelling in an argument from “reversion” and also that he disagrees with how Wallace employed reversion in his 1858 paper. Here I want to complicate this account and, in doing so, clear up a number of common confusions surrounding the phenomenon of “reversion” as it relates to Darwin and Wallace.   Part 1 of this chapter is largely deflationary. I argue, against this common view, that whatever disagreements Darwin and Wallace might have had over the nature of “reversion,” these cannot by themselves (or in any straight-forward way) explain their differing views towards the importance of domestication. It will be shown that Darwin and Wallace largely agreed with regard to reversion and (paradoxically, given the above quotation) Darwin was—or at least became—more permissible when it came to tolerating reversion. Yet they continued to disagree about domestication.   In Part 2 I want to approach their disagreement in a different light, by looking at how each of them reflected more broadly on the artificiality of domestication and on how these reflections interacted with the conclusions they drew about whether the process of domestication was representative of natural selection in the “state of nature.” At stake was the ability of domestic varieties to serve as objects for the study of nature. ?117 Darwin 1859, 14-5.277?PART 1. A MISCELLANY OF REVERSIONS  Charles Lyell’s Principles of Geology (1832) presents numerous examples of reversion as it was characteristically portrayed in the early-19th century: ?The black cattle that have run wild in America, where there were many particularities in the climate not to be found, perhaps, in any part of the old world, and where scarcely a single plant on which they fed was of precisely of the same species, instead of altering their form and habits, have actually reverted to the exact likeness of the aboriginal wild cattle of Europe.  278?It seems now admitted by horticulturists, that none of our garden varieties of fruit are entitled to be considered strictly permanent, but that they wear out after a time; and we are thus compelled to resort again to seeds; in which case, there is so decided a tendency in the seedlings to revert to the original type, that our utmost skill is sometimes baffled in attempting to recover the desired variety.  279? Lyell’s Principles was widely read, it was influential on the views of both Darwin and Wallace. It is also presumably the work Wallace is responding to in his 1858 paper.  280Published in 1832, Lyell’s text represents commonly held views about the relation between organisms under domestication and those that exist in the “state of nature.”  In these passages, Lyell—largely occupied with discrediting Lamarck’s progressive theory of evolution—draws an analogy between domestic varieties and natural species in order to argue for the immutability of species. As he says, “The best authenticated examples ?118 Lyell 1832:2, 30. My Emphasis.278 Lyell 1832:2, 33. Second emphasis mine.279 See Ruse 1979 for Darwin; see Beddall 1968 and McKinney 1972 for Wallace. 280of the extent to which species can be made to vary, may be looked for in the history of domesticated animals and cultivated plants.”  But alas, when human interference is 281alleviated domesticated varieties, and thus by analogy also natural varieties, have a tendency to revert to the “likeness” of their original type.  All variation is, according to this view, 282ephemeral. Domesticated varieties “may be regarded as extreme cases brought about by human interference, and not as phenomena which indicate a capability of indefinite modification in the natural world.”  They could “never be perpetuated in a wild state for 283many generations, under any imaginable combination of accidents.”  284 Two things should be said about early descriptions of reversion. First, commonly cited examples are vague and are for the most part hearsay; this is not a phenomenon that had undergone systematic study (leaving aside the fact that it had been documented by frustrated breeders for a long time).   285 Second, they are used to argue for species immutability by analogy. The reversion of domesticated species to their original type only reinforces the idea of species stability and the permanence of species types. The “best authenticated examples of the extent to which species can be made to vary” still indicate an inherent principle towards type. Though Lyell’s discussion of animal breeding and horticulture is sporadic, it is understandable why Wallace—who wrote his 1858 paper while in the Malay Archipelago taking Lyell’s Principles as his authoritative text—felt the need to argue against drawing an analogy between domesticated ?119 Lyell 1832:2, 53.281 As the examples above attest, this is true for both animals and plants.282 Lyell 1832:2, 32.283 Lyell 1832:2, 32.284 See Müller-Wille & Rheinberger 2012, chapters 3 and 4.285varieties and natural species: it was commonly held that such an analogy (if anything) only proved the immutability of species.  As one historian summarizes, “the belief had grown 286[from the 17th century onwards] that domesticated varieties were quite unlike wild species, being much more variable as a result of better nutrition and care and liable to revert in its absence.”  287 Lyell’s discussion highlights one type of 19th century “reversion,” (i) Reversion by an inherent principle towards an ideal type. According to this view, when species reversion occurs, it is the result of an inherent tendency to return to an ancestral species type. Variations are accordingly monstrosities; luckily by way of reversion Nature has a means for removing them.  288 By 1858 Darwin and Wallace both deny the credibility of this type of reversion and they both have means for explaining it away. In the Origin, as we saw above, Darwin refers to this type or reversion as not having a “shadow of evidence” in its favor. His argument is a thought experiment: in order to prove the existence of this type of reversion it would have to be shown that varieties demonstrated reversion under unchanged external conditions. Otherwise it would be ambivalent whether reversion was caused by the change of conditions or, as purported, an inherent tendency towards type. But, Darwin argues, this experiment has never been performed in reality, and furthermore, seems to be in contradiction to all evidence from the literature on breeding.  ?120 In the last chapter we saw many other examples of this from people like William Hopkins and Louis Agassiz. 286Darwin himself held this position in the late-1930s. Evans 1984, 133.287 In his Sketch of 1842, Darwin called attention to this type of reversion as an analogue of vis medicatrix, a 288principle of Hippocratic medicine referring to the “healing powers of nature” (Darwin 1909, p. 3). This is an apt characterization because it emphasizes an inherent tendency towards a fixed and ideal type. In full it refers to vis medicatrix naturae. Darwin had a copy of a book containing a discussion of this principle in his library.  In later work, Darwin allows for the other type of reversion mentioned in his thought experiment, (ii) Reversion by changed external conditions. For instance, in the Variation, when discussing the “best known case of reversion, and that on which the widely-spread belief in its universality apparently rests,” Darwin writes “with feral pigs, exposure to the weather will probably favour the growth of the bristles, as is known to be the case with the hair of other domesticated animals, and through correlation the tusks will tend to be redeveloped.”  Domesticated varieties may be caused to revert back to the likeness of the 289wild individuals because of similar external conditions. This is a far cry from the first type of reversion.   In 1858, Wallace provided a different argument against the first type of reversion. He writes, “Domestic varieties, when turned wild, must return to something near the type of the original wild stock, or become altogether extinct.”  Since he was not given the chance to 290emend the 1858 paper before it was published, Wallace later clarified the passage: “That is, they will vary, and the variations which tend to adapt them to the wild state, and therefore approximate them to wild animals, will be preserved. Those individuals which do not vary sufficiently will perish.”  By “will be preserved” he is making direct reference to natural 291selection. He continues to argue throughout his career that reversion is to be expected given the theory of natural selection. When domesticated varieties are released back into the state of nature they will either (i) perish altogether and be replaced by their wild counterparts or (ii) gradually evolve by natural selection to the likeness of the wild species, since the wild ?121 Darwin 1868:2, 33 and 47289 Darwin and Wallace 1858, 60.290 Wallace 1895, 31.291species exemplify adaptation to local conditions.  The wild species “must be that in which 292the various powers and faculties are so proportioned to each other as to be best adapted to procure food and secure safety.”  In either case, an observant onlooker would notice that 293the domesticated variety had ceased to exist (and here Wallace means ceased to exist in the same form as it had under domestication). This influential argument against the first type of reversion gives rise to another, (iii) Reversion by natural selection. Wallace’s argument was taken up by Darwin and even Lyell in later work.  294 The most straight-forward explanation for reversion was through blending inheritance. When domesticated varieties are released, they breed with their wild counterparts, producing offspring that have inherited some of the “wild” characteristics. As Darwin puts it, when domesticated varieties are released into the state of nature “crossing alone would tend to obliterate their proper character.”  Over time the differences between 295domesticated and wild varieties will tend to dissolve, resulting in a type of reversion to the wild form (especially effective in combination with the previous type of reversion), (iv) Reversion by crossing with wild varieties. ?122 Keep in mind here that Wallace was a proto-adaptationist.292 Darwin and Wallace 1858, 60.293 Darwin writes, “It has often been argued that no light is thrown, from the admitted changes of domestic 294races, on the changes which natural species are believed to undergo, as the former are said to be mere temporary productions, always reverting, as soon as they become feral, to their pristine form,” however, he continues, “This argument has been well combated by Mr. Wallace” (Darwin 1868:2, 425). Similarly, in the Antiquity of Man, Lyell writes, “It had usually been supposed by the advocates of the immutability of species that domesticated races, if allowed to run wild, always revert to their parent type. Mr. Wallace had said in reply that a domesticated species, if it loses the protection of Man, can only stand its ground in a wild state by resuming those habits and recovering those attributes which it may have lost under domestication” (Lyell 1863, 327). I thus strongly disagree with Gayon (1998, 40) that Darwin is making “Wallace say the exact opposite of what he did in fact say!” here. Darwin and Wallace saw eye-to-eye that this type of reversion is explained by natural selection. I do agree that Darwin and Wallace differed with regard to the implications of this reversion, as explained in the next section. Darwin 1868: 2, 32.295 By the time he published the Variation in 1868, Darwin’s fascination with “reversion” had clearly intensified. He now referred to the phenomena as a “great principle of heredity” and recognized that any theory of inheritance would have to explain it.  His theory of 296pangenesis did just this. And in his discussion of it, he also highlights another type of reversion, (v) Reversion through expression of latent characters. This type of reversion occurs regardless of changing an animal’s external conditions. As opposed to its closest relative, reversion type (i), reversion by latent characters could be explained without reference to an inherent tendency towards an ideal type.   Darwin repeatedly drew on two examples to demonstrate this type of reversion: first, “that in every horse of every generation there should be a latent capacity and tendency to produce stripes, though these may not appear once in a thousand generations” and second, “that in every white, black, or other coloured pigeon, which may have transmitted its proper colour during centuries, there should be a latent capacity in the plumage to become blue and to be marked with certain characteristic bars.”  297 The latent capacity of pigeon plumage to become marked with certain characteristic bars through breeding may be encouraged by changes in external living conditions or outbreeding, but neither is necessary. There is something of an inherent tendency towards latent characters and Darwin used pangenesis to explain it.  When introducing his 298“provisional hypothesis” he thus mentions reversion as one of its explananda. Wallace also ?123 See Bartley 1992, pp. 323-7.296 Darwin 1868:2, 56.297 Darwin writes, “For instance, there is a latent tendency in all pigeons to become blue, and, when a blue 298pigeon is crossed with one of any other colour, the blue tint is generally prepotent. When we consider latent characters, the explanation of this form of prepotency will be obvious.” Darwin 1868:2, 386.accepted this type of reversion. Referring directly to Darwin’s example, he writes, “This is considered to be a case of reversion to the ancestral type, just as the slaty colours and banded wings of the wild rock-pigeon sometimes reappear in our fancy breeds of domestic pigeons.”  299 What this section has demonstrated is, firstly, that reversion was not a simple, unambiguous phenomenon, and secondly, contrary to the common view, differences with regard to reversion cannot by themselves explain Darwin’s and Wallace’s views towards the importance of domestication. Far from being at odds, they largely agreed about the facts of reversion and were united in their rejection of type (i).  And yet still they disagreed with 300regard to domestication.  301????124 Wallace [1889] 1897, 107.299 Although Wallace without doubt saw (iii) as the most important explanation for reversion and played down 300the importance of (ii). Not only did Darwin allow for reversion, he also put the phenomenon to work. In the Variation Darwin used 301reversion to deal with one of the stickiest problems for his theory: the fact that “transitional varieties” are seldom found and thus could not provide evidence of the slow process of continuous modification predicted by the theory of natural selection. As Darwin states the problem: “why, if species have descended from other species by insensibly fine gradations, do we not everywhere see innumerable transitional forms? Why is not all nature in confusion instead of the species being, as we see them, well defined?” (Darwin 1859, 171).   Reversion-type (v) was exploited on many occasions to fill in the gaps. Darwin used the fact that varieties often expressed clearly defined latent characters through breeding to infer their ancestral form. He writes, “From these facts we see that the crossing of the several equine species tends in a marked manner to cause stripes to appear on various parts of the body, especially on the legs. As we do not know whether the primordial parent of the genus was striped, the appearance of the stripes can only hypothetically be attributed to reversion. But most persons, after considering the many undoubted cases of variously coloured marks reappearing by reversion in crossed pigeons, fowls, ducks, &c., will come to the same conclusion with respect to the horse-genus; and in this case we must admit that the progenitor of the group was striped on the legs, shoulders, face, and probably over the whole body, like a zebra” (Darwin 1868:2, 43; emphasis mine). Rather than rejecting “reversion” to support an analogy between domestication and natural selection, Darwin actually puts the concept to work within an evolutionary framework.  Using reversion in this way continued into the 20th century. A very interesting discussion of this can be found in Wang 2012 about the political motivations for attempting to bring back the Aurochs, the supposed wild ancestor of domesticated cattle, in 1920-40s Germany.PART 2. ARE DOMESTICATED VARIETIES ARTIFICIAL AND WHAT DOES THIS ENTAIL?  WALLACE’S STATE OF NATURE AND STATE OF DOMESTICITY  While differing with Darwin as to its significance, and continuing to express a level of uneasiness, Wallace did not discount all knowledge gained from the study of domestication, especially later in his career. This is best demonstrated by how Wallace approaches domestication in his book Darwinism (1889). In the Preface, Wallace (in)famously states, ?It has always been considered a weakness in Darwin's work that he based his theory, primarily, on the evidence of variation in domesticated animals and cultivated plants. I have endeavoured to secure a firm foundation for the theory in the variations of organisms in a state of nature; and as the exact amount and precise character of these variations is of paramount importance in the numerous problems that arise when we apply the theory to explain the facts of nature, I have endeavoured, by means of a series of diagrams, to exhibit to the eye the actual variations as they are found to exist in a sufficient number of species. By doing this, not only does the reader obtain a better and more precise idea of variation than can be given by any number of tabular statements or cases of extreme individual variation, but we obtain a basis of fact by which to test the statements and objections usually put forth on the subject of specific variability; and it will be found that, throughout the work, I have frequently to appeal to these diagrams and the facts they illustrate, just as Darwin was accustomed to appeal to the facts of variation among dogs and pigeons.  302? This quotation is often taken as an illustration of his aversion to domestication. It says much more than this however. Firstly, Wallace is primarily interested in explaining evolution among organisms in the state of nature, not in the “state of domesticity.”  As we saw in the 303?125 Wallace [1889] 1897, vi.302 So-called in his 1858 paper.303last chapter, this was a common distinction drawn by naturalists in the 19th century. Darwin’s work was founded on domesticated varieties and will thus always involve a further inference from principles that hold within the state of domesticity to principles that hold within the state of nature. A common objection was that the variation of domesticated varieties was artificial, a product of human interference, and thus not representative of the natural world.  304Since many people saw this inference as problematic, Wallace attempted to remove it by going straight to the source: variation in the state of nature.   Secondly, Wallace, paying lip service to Darwin’s Variation, does not argue that domestication is epistemically useless. He even includes a chapter in Darwinism about variation under domestication. This chapter demonstrates that domestication provides us with information about the origin of variation and about heredity. It demonstrates various general hereditary “facts” necessary for the action of natural selection: “the occurrence of incessant slight variations in all parts of an organism, with the transmission to the offspring of the special characteristics of the parents; and also, that all such slight variations are capable of being accumulated by selection till they present very large and important divergencies from the ancestral stock.”  These are the “facts” of variation upon which is founded Darwin’s 305theory.   306?126 See citations in last chapter.304 Wallace [1889] 1897, 100. Many scholars have pointed out that Darwin saw domestication as an “experiment 305on a gigantic scale” (1868:1, 3; e.g., Schweber 1978; Rheinberger & McLaughlin 1984; Bartley 1992; Richards 1998). As we see here, this cannot straightforwardly explain differences between Darwin and Wallace with regard to their valuations of domestication. It might be argued that Wallace sees domestication as an artificial experiment, whereas Darwin as a natural experiment. This would be in accordance with what I claim below, but would require much more argumentation. He also draws an analogy with domestic varieties and natural species to defend a point: add this here.306 But, thirdly, Wallace supplements (and supplants) this discussion with one about “the variations of organisms in a state of nature.” This chapter precedes the one about domestication and prepares the reader to appreciate that what is demonstrated through variation and heredity in domesticated organisms is not altogether inconsistent with what is found in the state of nature. This discussion makes the analysis of domestication largely superfluous, and acts to preempt objections that were leveled at Darwin’s theory. Wallace uses diagrams demonstrating variation in the state of nature in the same way that he understood Darwin to be appealing to variation in domesticated varieties: to demonstrate to the reader that variation is adequately supplied for natural selection to be effective (Figure 5.1 and 5.2). Wallace’s diagrams were based on measurements taken in the wild and thus directly represented wild variation. ?127      ?  Figure 5.1: Wallace’s Variation in the State of Nature. The left image represents measurements of the length of various traits for a number of species of the lizard genus Lacerta. The right image represents measurements of the length of various traits for twenty males of Icterus baltimore, Baltimore Oriole. (Wallace 1889, Darwinism) (© John van Wyhe, Darwin Online) ??128? ?  Figure 5.2: Darwin’s Variation in the State of Domesticity. The left image is of a Pouter pigeon. The right image is of a number of domestic pigeon skulls (from top to bottom: Wild Rock-pigeon, Columba livia; Short-faced Tumbler; English Carrier; Bagadotten Carrier). (from: Darwin 1868. Variation of Animals and Plants under Domestication) (© John van Wyhe, Darwin Online) ? Illustrating variation in the state of nature for the reader was certainly important. Most natural history texts at this time focused on the stable, characteristic traits of species in the state of nature, playing down any variation within that might upset systematic description. The influential 17th century botanist John Ray, for example, famously argued that characters that could be externally manipulated (e.g., by humans) or that were for other reasons too variable, were insufficient to distinguish species and thus should be ousted from taxonomy.  307Such views advocated what historian and biologist Ernst Mayr famously called the ?129 See Müller-Wille & Rheinberger 2012, p. 60, and also chapt. 2-3.307“typological” view of species: that members of a species are all instances of the specific type, and that the perpetuation of the species consists in the production of ever more instances of the type.  According to this view, all variation was error; variation was the result of 308disturbances that interfered with the natural tendency of organisms of a particular type to produce offspring of that type. This was obviously a view hostile to evolutionary change. What Wallace and Darwin wanted to argue, using among other things the above diagrams, was that variation was natural. (Although as we will see below, Wallace never could shake the idea that artificial selection, rather than variation, was unnatural and a disturbance.)  Without having spent time in the state of nature and observed the variations first-hand—as Wallace had done—the common 19th century reader would have no way of adjudicating whether natural selection could work at all. Wallace thus says of the Duke of Argyll’s reading of the Origin, “The noble author represents the feelings and expresses the ideas of that large class of persons who take a keen interest in the progress of science in general, and especially that of Natural History, but have never themselves studied nature in detail, or acquired that personal knowledge of the structure of closely allied forms [...] which is absolutely necessary for a full appreciation of the facts and reasonings.”  By “nature” Wallace means “state of 309nature,” as opposed to the domesticated nature. His diagrams provided a solution to this problem.   Although domestic varieties were not epistemically useless, Wallace continued from 1858 onwards to avoid invoking Darwin’s analogy between domestication and natural selection. He held his ground in this regard for two reasons, both having to do with ?130 Mayr 1982.308 Wallace 1868 [1895], 141.309differences between the state of nature and the civilized state—and both expressed views that were deep-seated among Victorians.   First, Wallace worried that the analogy with domestication suggested that natural selection required an intentional selector and thus imbued the natural world with a problematic level of consciousness.  It is easy to forget today that Darwin brought the term 310“selection” into common biological parlance, and changed its meaning considerably (as Lewontin points out in the epigraph). For instance, in his 1860 review of Darwin’s Origin, the Harvard professor and natural historian Louis Agassiz wrote, “The fallacy of Mr. Darwin's theory of the origin of species by means of natural selection, may be traced in the first few pages of his book, where he overlooks the difference between the voluntary and deliberate acts of selection applied methodically by man [...] and the chance influence which may effect animals and plants in the state of nature. To call these influences ‘natural selection,’ is a misnomer.”  Wallace addressed such issues in a paper section entitled, “Mr. 311Darwin’s Metaphors Liable to Misconception,” where he argued that Darwin was merely speaking metaphorically and did not mean to anthropomorphize “nature.” But even before this in-print defense, Wallace worried about Darwin’s analogy. He wrote to Darwin in 1866 that he had been struck by the “utter inability of numbers of intelligent persons to see clearly or at all” that natural selection was “self-acting” and did not involve an intelligent chooser behind the scenes.  Such misunderstandings, Wallace thought, could be avoided by 312removing the terminology of selection itself. He suggested that “survival of the fittest” was ?131 Wallace 1868 [1895], 144-5310 Agassiz 1860, 147.311 Wallace to Darwin, July 2, 1866, Wallace to Darwin.312more apt. “Selection” was, after all, a civilized human activity, preformed by a selector with a high degree of consciousness and agency. Selective breeding occurred, he said, in the “state of domesticity,” not the state of nature. To say that nature selects individuals best fitted to their environments, just as humans might select the swiftest greyhounds, is simply misleading.  Eventually Wallace gave in and accepted “natural selection,” but he never 313fully accepted the analogy with domestication.  314 Wallace also held his ground for a different reason. Although the origin of variation and the heredity principles were essentially the same for domestic varieties and natural species, he believed that the accumulation of that variation was entirely different, in effect, it was diametrically opposed. In the state of nature, natural selection resulted in organisms that become progressively well fitted to their environments (in the mid-19th century sense of fulfilling the requirements of a particular environment or way of life). Selection in the state of nature is for fitness. In the state of domesticity, however, artificial selection led to organisms that were maladapted. Selection under domestication was for useless, fanciful or superficial characters and thus often indirectly at odds with fitness. Consider a few of Wallace’s comparisons between domestication and natural selection: ?132 Although Darwin recognized the possibility that the phrase would be misunderstood he remained reluctant to 313replace it. He added a section to the 3rd edition of the Origin (1861, 84-5) explaining how the phrase should be understood. Even by the 2nd edition (1860) Darwin had begun to worry about his use of “selection,” adding the word “metaphorically” to the following sentence on page 84: “It may metaphorically be said that natural selection is daily and hourly scrutinising, throughout the world, every variation, even the slightest” (my emphasis). See also Thomas Henry Huxley's paper “Criticisms on 'The Origin of Species'” (1864) which addresses M. Flourens' concerns about Darwin's use of “selection.” For an historical treatment of the Darwin-Wallace disagreement about Darwin's use of “selection” see Rheinberger & McLaughlin (1984) and Gayon (1998, chapter 1). See Young (1971, 1993) for excellent discussions of the reception more generally of natural selection. And furthermore he continued to speak of natural selection as a passive “weeding out” process, rather than 314an active “selection” process. For more detail and the broader context see discussion in Young 1985, Chapter 4. For Wallace’s use of “weeding out”, see for example, Wallace 1890, 1892.?The essential difference in the condition of wild and domestic animals is this,—that among the former, their well-being and very existence depend upon the full exercise and healthy condition of all their senses and physical powers, whereas, among the latter, these are only partially exercised, and in some cases are absolutely unused. A wild animal has to search, and often to labour, for every mouthful of food [...] There is no muscle of its body that is not called into daily and hourly activity [...] The domestic animal, on the other hand, has food provided for it, is sheltered, [etc.] [...] Half of its senses and faculties are quite useless [...] in the domesticated animal all variations have an equal chance of continuance; and those which would decidedly render a wild animal unable to compete with its fellows and continue its existence are no disadvantage whatever in a state of domesticity. Our quickly fattening pigs, short-legged sheep, pouter pigeons, and poodle dogs could never have come into existence in a state of nature [...] The great speed but slight endurance of the race horse, the unwieldy strength of the ploughman's team, would both be useless in a state of nature [...] Domestic animals are abnormal, irregular, artificial; they are subject to varieties which never occur and never can occur in a state of nature: their very existence depends altogether on human care; so far are many of them removed from that just proportion of faculties, that true balance of organization, by means of which alone an animal left to its own resources can preserve its existence and continue its race.  315?As man has considered only utility to himself, or the satisfaction of his love of beauty, of novelty, or merely of something strange or amusing, the variations he has thus produced have something of the character of monstrosities. Not only are they often of no use to the animals or plants themselves, but they are not unfrequently injurious to them. In the Tumbler pigeons, for instance, the habit of tumbling is sometimes so excessive as to injure or kill the bird [...] This peculiar character of domestic productions distinguishes them broadly from wild species and varieties, which, as will be seen by and by, are necessarily adapted in every part of their organisation to the conditions under which they have to live.  316??133 Darwin and Wallace 1858, 59-61.315 Wallace [1889] 1897, 100.316 Wallace’s use of “artificial” to disparage domesticated varieties has a multidimensional meaning. It means, firstly, created by human interference and existing in a state of domesticity. Secondly, it means ephemeral, capricious, or useless and associated largely with changing fashion. The artificiality of domestic varieties is contrasted with natural varieties which exhibit a “true balance of organization” and are “necessarily adapted in every part of their organization to the conditions under which they live.” And third, “artificiality” has an air of “unreality” about it. Artificial, domesticated varieties are not as real as natural varieties; they are fleeting examples of human interference, not the products of natural law.  317 The fact that the accumulation of variation in the state of nature and under domestication were at odds indicated to Wallace that the analogy was a poor one. His argument involves the amalgamation of two ideas constitutive of his thinking. The first is that Wallace, like Darwin, was strongly influenced by Lyell’s “uniformitarianism,” which held that explanations of former changes in the history of life must only avail themselves of natural causes now in action.  If a cause cannot be demonstrated to be acting at present, 318then we cannot simply invent it to explain events that happened in the past.   319?134 This was a commonly held view among naturalists; in the last chapter we considered Asa Gray’s lament that 317the study of domestication “has been generally neglected by naturalists, because these races are not in a state of nature” (Gray [1860] 1963, 20). McKinney 1972, 32-33. Wallace brought Lyell’s book with him on his first trip to South America. See also 318Beddall 1968, Wilson 1970 and Berry 2002, p. 156. Lyell’s famous Principles of Geology was aptly subtitled “Being an Attempt to Explain the Former Changes of the Earth’s Surface by Reference to Causes Now in Operation.” To explain how species had evolved in the state of nature, Wallace had to demonstrate how causes now in 319action in the state of nature could by analogy have led to species evolution in the past. As Wallace wrote in an early notebook, “While the inorganic world has been strictly shown to be the result of a series of changes from the earliest periods produced by causes still acting, it would be most unphilosophical to conclude without the strongest evidence that the organic world so intimately connected with it had been subject to other laws which have now ceased to act.” Wallace 1855, quoted in McKinney 1972, 23. Second, Wallace’s presentation of evolution often gives the impression of a two-level process.  The first is the creation of “permanent varieties” which differ from the natural 320species because of (something like) selection on individual differences through individual competition.  The second, and the more important level in Wallace’s early writings, is the 321struggle between “permanent varieties.” Wallace writes, if “any species should produce a variety having slightly increased powers of preserving existence, that variety must inevitably in time acquire a superiority in numbers,” relative to the parent species. When conditions become unfavorable the parent species may be replaced by the variety. “The superior variety would then alone remain, and on a return to favourable circumstances would rapidly increase in numbers and occupy the place of the extinct species and variety.” In such a case, “The variety would now have replaced the species.”  322 A species’ departure from its original type thus occurs as a permanent variety supplants the natural species.  The permanent variety is, in return, supplanted by a better 323?135 See Bowler 1976 and Gayon 1998, chapter 1. Wallace’s use of “variation” and “variety” in his 1858 paper is 320quite perplexing and while much literature has abounded which addresses this problem, there is no obvious answer as to what Wallace meant by these terms in 1858 (see the early disagreement between H. F. Osborn and E. B. Poulton in 1896 quoted in Kottler 1985; also see for a summary of recent literature). While I have here presented Wallace’s theory of evolution as a two-step process (like Bowler 1976, 2013) because I believe this to be the most sensible, my discussion does not hinge on this view and can equally be interpreted in terms of individual selection only: in this case, Wallace treats domesticated varieties as still unfit and unable to outcompete or eventually replace the parent species, and the second step is instead understood in terms of individual selection. “[A]s the individual existence of each animal depends upon itself, those that die must be the weakest,” writes 321Wallace, “while those that prolong their existence can only be the most perfect in health and vigour—those who are best able to obtain food regularly, and avoid their numerous enemies. It is, as we commenced by remarking, ‘a struggle for existence,’ in which the weakest and least perfectly organized must always succumb” (Darwin and Wallace 1858, 56-7). See Bowler 2013, 63ff. Some historians argue that Wallace did not mean to present evolution as a two-level process. I agree with Peter Bowler, Janet Browne, and Michael Ruse that at least in 1858 he did.  Ibid., 1858, 58.322 The latter really just another permanent variety, since Wallace admits there is “generally no means of 323determining” which is the natural species and which the permanent variety (Darwin and Wallace 1858, 53).adapted permanent variety and so on. This leads to the “tendency of varieties to depart indefinitely from the original type,” the title of Wallace’s 1858 paper.   Thus, being a uniformitarian, Wallace was looking for a theory of evolution that could be demonstrated to be currently in action and that could by analogy explain species evolution in the past. Domestication did not provide a model of this process since the varieties it created could never outcompete the parent species in the state of nature. Why was this? Because domestication gave rise to varieties that were unfit. The human disturbance which created domestic varieties was not the same as the process which caused species evolution in the past—the latter consisting of better fitted varieties replacing those that are less well fitted. Domestication was ephemeral, capricious, and useless and associated largely with changing fashion. It created monsters with lopsided characters, not organisms perfectly adapted to their surrounding environment and displaying a “true balance of organization.”  Wallace’s strategy when it came to explaining natural selection was thus to deduce natural selection from a number of laws or principles or facts which held true in the state of nature itself, without recourse to domestication.  This strategy is lucidly expressed in his 3241868 paper “Creation by Law” where Wallace presents a “series of facts or laws” that “are mere statements of what is the condition of nature.”  These laws include: (1) law of 325multiplication in geometrical progression; (2) law of limited population; (3) law of likeness of offspring to their parents; (4) law of variation; (5) law of unceasing change of physical conditions upon the surface of the earth; (6) equilibrium or harmony of nature (i.e., well-?136 This was more consistent with strategies employed in the physical sciences. See for example William 324Hopkins’ review of Darwin’s Origin. Hopkins 1860. Wallace 1868 [1895], 143.325adapted species flourish). Wallace saw evolution by natural selection to be straight-forwardly entailed by these facts and needing no support by way of analogy to domestication.  326 As radical as Wallace was in other respects, his views reflect the teachings of traditional natural history. The artificiality of domestic varieties had always entailed a kind of unreality. They were products of humans interfering with natural laws, and not stand ins for natural processes themselves. As products of interference, domestic varieties were not outside of nature per se, but they were definitely outside of “natural nature,” to use Sorel’s terms.  Wallace’s discussions were therefore compatible with mainstream accounts of 327domestic varieties—as not themselves the objects of speculation or study. The objects of study were well-adapted species in the state of nature. What happened when organisms left the state of nature? To Wallace, and many others, they became unfit.  Wallace’s belief that the accumulation of variation through domestication was at odds with fitness extended even to domesticated humans. In an uncharacteristic moment for an otherwise self-effacing man, Wallace summarized his intellectual contributions, emphasizing two papers on “human selection,” ?137 More accurately Wallace used the phrase “survival of the fittest,” rather than “natural selection” in this paper. 326Natural selection, he thought, reminded readers of human selection, and he wanted to distance himself from that view. Darwin also at times appears to be emphasizing that natural selection is a consequence of certain laws, e.g., “These laws, taken in the largest sense, being Growth with Reproduction; Inheritance which is almost implied by reproduction; Variability from the indirect and direct action of the external conditions of life, and from use and disuse; a Ratio of Increase so high as to lead to a Struggle for Life, and as a consequence to Natural Selection, entailing Divergence of Character and the Extinction of less-improved forms” (1859, 489-90). I quoted George Sorel’s distinction between “natural nature” and “artificial nature” in Chapter 1. Domestic 327varieties were either seen as having themselves triumphantly emerged from the state of nature to serve humans (as civilized domestication practices replaced those of savage domestication practices) or having been enslaved and degenerated to appease human fancy, depending on the author’s point of view. (For former see Ritvo 1987 and Crawford 1863; for latter see quotes and footnotes in last chapter. The latter seems to have been predominantly the view of the naturalists to which Darwin and Wallace were indebted. See Secord 1985.) Either way, since they were not part of the state of nature, it was argued that domesticated animals and the processes involved in domesticating them could not be used to shed light on the state of nature—and often times such an inference was strongly advised against. There are many examples of this which were cited in the last chapter. See especially Louis Agassiz’s notes on Darwin’s Origin as well as subsequent publications.?These deal with different aspects of the same great problem—the gradual improvement of the race by natural process; and they were also written partly for the purpose of opposing the various artificial processes of selection advocated by several English and American writers. I showed that the only method of advance for us, as for the lower animals, is in some form of natural selection, and that the only mode of natural selection that can act alike on physical, mental, and moral qualities will come into play under a social system which gives equal opportunities of culture, training, leisure, and happiness to every individual. This extension of the principle of natural selection as it acts in the animal world generally is, I believe, quite new, and is by far the most important of the new ideas I have given to the world.  328?As with animals, artificial selection on humans is ineffective if the goal is the general “improvement” of the race. Only natural selection can accumulate variation in a direction that increases fitness. The civilized human will is too susceptible to fancy, caprice, and most importantly, wealth.  Wallace’s views also reflect a much discussed popular understanding of the relation between the state of nature and the state of domesticity as a recovery narrative, from “savagery” to “civilization.”  The language of the “state of nature” comes from 17th 329century political texts that endorsed the human flight from nature by way of social contracts. This progressive narrative permeated the imaginative consciousness of 19th century intellectuals. “With every step of this progress in civilization,” T. H. Huxley famously writes ?138 Wallace 1905:2, 389. By “natural selection,” Wallace has in mind sexual selection. While he argued against 328Darwin’s theory of sexual selection as applied to animals, he promoted it as a form of selection in the state of domesticity. For a brief history of the recovery narrative see Merchant 1995. See Ruse 1996 for a discussion of 329“progress” more broadly speaking, and its relation to evolutionary theories.near the end of the century,  “the colonists would become more and more independent of the state of nature; more and more, their lives would be conditioned by a state of art.”   330 At times, Wallace promoted these views himself. Man is, “a being apart,” Wallace agrees, “since he is not influenced by the great laws which irresistibly modify all other organic beings,” including most importantly natural selection. Rather than existing in a state of nature, man has become “fitted for the social state.” Not only had man escaped natural selection, but “he is actually able to take away some of that power from nature.” We can anticipate the time, for better or worse, “when the earth will produce only cultivated plants and domestic animals; when man's selection shall have supplanted ‘natural selection.’”  331And to Darwin’s great chagrin, Wallace in 1869 went one step further, claiming that natural selection could not account for the mental qualities of modern man—some higher force must be invoked.   332 In summary, according to Wallace, domestication was superfluous once “facts” about the origin of variation and heredity could be proven to exist in the state of nature. In addition, the analogy between domestication and natural selection was misleading because the accumulation of variation that resulted from each process was at odds—domestication created maladapted monsters, whereas natural selection created organisms fitted for survival in nature. Domestication and domestic varieties were, for Wallace, unnatural. That is, Wallace invoked the artificial-natural distinction to argue against the domestication analogy. ??139 Huxley 1894, 19.330 Wallace [1864] 1870, 325-8. 331 Wallace 1869. He was an outspoken proponent of spiritualism for the rest of his life.332CHARLES DARWIN ON ART AND NATURE  Darwin’s discussions, in contrast, collapse the artificial-natural distinction itself.  333Many of the properties characteristic of Wallace’s “state of domesticity”—most uniquely tied to civilization—are shown by Darwin to have counterparts in the state of nature.  I will 334focus on three cases that demonstrate how strongly Darwin’s discussions encouraged a merging of the artificial world of domestication and civilization with the wild and savage world of the state of nature (all of the emphasized phrases being loaded with cultural significance that Darwin exploited repetitiously). The first case focuses on the artificiality of the state of nature itself, the second on the naturalness of domestication, and the third undermines attributions of “monstrous.” Darwin’s strategy is to render unanswerable questions such as, “is this a product of art or nature?”  335?THE ART OF SEXUAL SELECTION  Initially, Darwin’s theory of sexual selection had the appearance of an afterthought.  336It was called on to explain what he believed natural selection could not: the interesting and ?140 I am indebted in this section to papers by Cornell 1984, Rheinberger & McLaughlin 1984, Alter 2007, 333Burnett 2009, and Müller-Wille 2009. Unsurprisingly many of these ideas were also strongly contested.334 Since Darwin’s reasoning with regard to the analogy from artificial to natural selection has attracted much 335attention (and was briefly dealt with in the last chapter), I will spend less time explaining his views and instead just point out a few interesting, yet often overlooked, cases. The discussions I indicate here are in contrast to a different way in which Darwin exploits the artificial-natural distinction. For instance, “We have seen that man by selection can certainly produce great results, and can adapt organic beings to his own uses, through the accumulation of slight but useful variations, given to him by the hand of Nature. But Natural Selection, as we shall hereafter see, is a power incessantly ready for action, and is as immeasurably superior to man’s feeble efforts, as the works of Nature are to those of Art.” (Darwin 1859, 61) Here he is being, what we might call, strategically inconsistent. By emphasizing the differences between Art and Nature he is trying to get around the problem that domestic varieties freely interbreed. For a discussion see Burnett 2009. In the Origin Darwin introduces sexual selection, but it doesn’t take centre stage until Darwin's (1871) The 336Descent of Man, and Selection in Relation to Sex.unusual sexual differences in ornamentation and behavior found within many groups of animals, for example, the male peacock’s large and conspicuous tail. Its significance, however, increased over the course of his life and it came to sit interestingly between the processes of domestication and natural selection: a kind of choice-based selection occurring in the state of nature.  Sexual selection differed from natural selection in that the selective force was not a result of the “struggle for existence,” but was instead a result of the struggle between individuals of one sex to find and secure mates. It was a weaker type of selection, which resulted not in “death to the unsuccessful competitor, but few or no offspring.”  Even still, 337sexual selection was invoked to explain the most anomalous and beautiful aspects of the state of nature—for example, the argus pheasant’s plumage and the bower of the bower bird.   This struggle for mates took one of two forms. It either took the form of a contest between the males, where the victor secured a mate, or was more peaceful and involved elaborate displays or songs. In its more peaceful guise, sexual selection depended on choice, not unlike in domestication.  Consequently, in many of the examples illustrating sexual 338selection Darwin relies on an analogy between these two types of selection to substantiate his case. The result is surprising and fascinating. Just as Wallace observed above, choices can be quite whimsical, often based on a love of beauty or just novelty and tied to current fashion ?141 Darwin 1859, 88.337 Here I am considering how Darwin compares sexual selection and artificial selection. The parallels between 338sexual selection and natural selection, in terms of the degree of conscious planning behind the “choice” made, can be brought out by considering the other form of sexual selection, what might be called the less peaceful “contest” form. In this form, sexual selection does not involve choice at all; it is, rather, the result of the contest that determines who is able to find a mate. In this sense, the “struggle between the males for possession of the females” seems to involve no more consciousness than the “struggle for existence” between individuals in nature. The “contest” form of sexual selection, therefore, would seem to fit in at the low end of the spectrum, next to natural selection.rather than fitness.  For Wallace, this entailed that domesticated varieties were 339unrepresentative of their counterparts in the state of nature. For Darwin, the theory of sexual selection highlights the capriciousness of the state of nature itself.  340 In the Origin, Darwin describes sexual selection as follows, ?successive males display their gorgeous plumage and perform strange antics before the females, which standing by as spectators, at last choose the most attractive partner. Those who have closely attended to birds in confinement well know that they often take individual preferences and dislikes [...] It may appear childish to attribute any effect to such apparently weak means: I cannot here enter on the details necessary to support this view; but if man can in a short time give elegant carriage and beauty to his bantams, according to his standard of beauty, I can see no good reason to doubt that female birds, by selecting, during thousands of generations, the most melodious or beautiful males, according to their standard of beauty, might produce a marked effect.  341?As can be seen, he is not at all shy in allowing for the female bird to “choose” a mate based on her “preferences and dislikes,” drawing the reader's attention to the similarity between the female bird (i.e., an instance of sexual selection) and the human breeder (i.e., an instance of domestication) to argue for the efficacy of sexual selection. Furthermore, it is by a standard of beauty—and not of utility—that the female bird makes consistent choices. “In the same manner as man can give beauty, according to his standard of taste, to his male poultry,” he writes, “so it appears that in a state of nature female birds, by having long selected the more ?142 Again, “fitness” here is meant in the mid-19th century sense of fulfilling the requirements of a particular 339environment or way of life.  See Cronin 1991 and Milam 2010b for reflections on this theory and its history. See also Müller-Wille 2009.340 Darwin 1859, 87.341attractive males, have added to their beauty.”  Darwin thus suggests that it is not just in the 342state of domesticity that beauty, rather than utility, may affect the production of varieties.  But, Wallace had asked, what of changing fashion? What of capriciousness? Are these not characteristics of domestication alone? Darwin responded, just as “any fleeting fashion in dress comes to be admired by man, so with birds a change of almost any kind in the structure or coloring of the feathers in the male appears to have been admired by the female.”  343Darwin was even prepared to go “a short distance” with the Duke of Argyll—not someone he would normally go any distance with—when the he wrote, “that variety, mere variety, must be admitted to be an object and an aim in Nature.”  “It would even appear that mere 344novelty, or change for the sake of change,” Darwin agreed, “has sometimes acted like a charm on female birds, in the same manner as changes of fashion with us.”  And against 345one writer, who “oddly fixed on Caprice ‘as on of the most remarkable and typical differences between savages and brutes’,” Darwin writes,  “lower animals are, as we shall hereafter see, capricious in their affections, aversions, and sense of beauty.”  The male 346Argus pheasant, Darwin concluded, is “more like a work of art than of nature” (Figure 5.3).  347?143 Darwin 1871:1, 259. My emphasis.342 Darwin 1871:2, 74.343 Darwin 1871:2. 230.344 Darwin 1871:2, 230. Darwin is so confident that birds have a similar, although less developed, sense of 345aesthetic appreciation governing their selection of mates that he says in the Origin, “the tuft of hair on the breast of the turkey-cock [...] can hardly be either useful or ornamental to this bird;—indeed, had the tuft appeared under domestication, it would have been called a monstrosity” (Darwin 1859, 90). In case we might misunderstand, in later additions of the Origin Darwin replaced “ornamental to this bird” with “ornamental in the eyes of the female bird” (Darwin 1872, 70; my emphasis). In other words, we should be leery of attributing all sexual differences to sexual selection because we find traits, like this tuft of hair, which cannot be favoured by the female bird because it is ugly in the eyes of human breeders! Darwin 1871:1, 64-5.346 Darwin 1871:2, 92.347?Figure 5.3: An Argus Pheasant Sitting on a Branch of a Tree. Etching by John Le Keux, 1783-1846. (© Wellcome Library, London) ? To make such claims seem plausible—that birds choose at all would strike many Victorians as problematically anthropomorphic—Darwin draws comparisons between the abilities of the female bird and “the savage,” what many in the 19th century saw as a middle-ground between the human-animal dichotomy and the object much of speculation. When it comes to selection based on aesthetic discrimination, he says, female birds may even out-rank savages. “The taste for the beautiful, at least as far as female beauty is concerned, is not of a special nature in the human mind,” he writes. And judging “from the hideous ornaments ?144and the equally hideous music admired by most savages, it might be urged that their aesthetic faculty was not so highly developed as in certain animals, for instance, in birds.” Of course, “no animal would be capable of admiring such scenes as the heavens at night, a beautiful landscape, or refined music,” but then again, such “high tastes” are “not enjoyed by barbarians or by uneducated persons.”  348 By selecting certain types of individuals over others—pigeons with shorter beaks, feathers that run backwards or carunculated skin—human breeders create long-term changes in their stocks. These choices are often based on an assessment of what is most beautiful, an assessment that is whimsical, certainly liable to caprice, and based on current fashion. Breeders are guided in their selection by external “circumstances” and “fashions,” or even idiosyncrasies associated with their individual “frame of mind” or “character.”  What 349Darwin draws our attention to when discussing sexual selection is that the taste for the beautiful and liability to caprice or fashion can also be found in nature. Darwin is not here denying the artificiality of domestication, but is instead playing up the artificiality of the state of nature.  ?UNCONSCIOUS SELECTION IN THE STATE OF NATURE  Darwin undermines the artificial-natural distinction in a different way by exploiting the concept of “unconscious selection.” The scholarly focus on Darwin’s analogy between artificial and natural selection has tended to hide the fact that “artificial selection” is graded. Darwin spills a lot of ink—especially in the Variation—explaining and exploiting types of ?145 Darwin 1871:1, 64.348 Darwin 1868:2; 212, 208-210, 214, 214.349artificial selection. Most notably methodical and unconscious selection, though Darwin is aware that “Unconscious selection so blends into methodical selection that it is scarcely possible to separate them.”   350 Methodical selection is the most artificial. It requires that the selector be consciously aware of their favourite animals, be consciously aware of the modifying power of selecting them over many generations, and work towards the goal of producing new varieties. This type of selection “is the magician's wand, by means of which [the breeder] may summon into life whatever form and mould he pleases.”  It is this type of selection which, in 351characteristic portrayals, figures in Darwin’s argument by analogy.  But Darwin’s other type of selection, unconscious selection, should not be overlooked. This type of selection, as the name implies, occurs when breeders unintentionally alter their breeds by unconsciously selecting the best or most fancied individuals to mate.  352As Darwin writes in the Origin, “a man who intends keeping pointers naturally tries to get as good dogs as he can, and afterwards breeds from his own best dogs, but he has no wish or expectation of permanently altering the breed.” He continues, “I cannot doubt that this process, continued during centuries, would improve and modify any breed, in the same way as Bakewell, Collins, &c., by this very same process, only carried on more methodically, did greatly modify, even during their own lifetimes, the forms and qualities of their cattle.”  353Darwin believed that this form of selection led to far more significant results than ?146 Darwin 1868:2, 211.350 Darwin 1859, 31.351 See Darwin 1859, 34-6.352 Ibid, 34-5.353methodological selection, and moreover, was theoretically interesting because it closely resembled natural selection.   354 This resemblance is two-fold. Unconscious selection resembles natural selection because of its “unconsciousness”—Darwin demonstrated that breeds can be significantly altered despite the “selectors” having no intention to do so. This was one way in which he evaded the criticism that natural selection required intention to be efficacious. Unconscious selection also resembles natural selection in terms of its physical locality; unconscious selection occurs not just on the estates of those who keep pointers, but also within the state of nature itself.  To make the second point, Darwin again exploited a 19th century category 355that captured the imagination of Victorians and was the topic of much discussion: the “savage” or “barbarian.” Living within the state of nature, indigenous peoples, many Victorians suspected, were those that the progressive narrative failed to capture; they were uncivilized, Godless and lawless. To some they were a link between humans and the state of nature, to others they were a different species entirely, more akin to the “brutes” with which they cohabited.  Darwin drew repeatedly on the “primitive” domestication practices he 356observed in the Tierra del Fuego region while on the Beagle voyage. This type of domestication was essentially occurring in the state of nature.   These dimensions of unconscious selection are brought out in the following story that Darwin returned to on many occasions and in different works: ?147 Darwin writes in the Variation, “This form of selection has probably led to far more important results than 354methodical selection, and is likewise more important under a theoretical point of view from closely resembling natural selection.” Darwin 1868:2, 424; my emphasis. See Burnett 2009.355 See Browne 1995, Chapter 10. See also Marx 2005 for a later 19th century and American perspective.356?If there exist savages so barbarous as never to think of the inherited character of the offspring of their domestic animals, yet any one animal particularly useful to them, for any special purpose, would be carefully preserved during famines and other accidents, to which savages are so liable, and such choice animals would thus generally leave more offspring than the inferior ones; so that in this case there would be a kind of unconscious selection going on. We see the value set on animals even by the barbarians of Tierra del Fuego, by their killing and devouring their old women, in times of dearth, as of less value than their dogs.  357?The barbarians of Tierra del Fuego do not consciously attempt to create new strains from their domesticated animals, nor do they even consciously breed the best individuals. They may not even be conscious of their own preference for certain animals or make a distinction between species, as the purported fate of their old women suggests.  In times of famine, 358these savages simply kill off and devour those least preferred by them (including their own elders!) leaving the rest to reproduce, in a “kind of unconscious selection.” Darwin is drawing attention to the fact that selective breeding activities do not require much consciousness at all, which is significant if we remember that natural selection is supposed to be an unconscious analogue of human selection.  This “unconscious selection” is also occurring in the state of nature. As I said, many of Darwin’s readers would not have considered the Fuegians to be of the same species as ?148 Darwin 1859, 36.357 Darwin elaborates further on this example in the Variation, “When the Fuegians are hard pressed by want, 358they kill their old women for food rather than their dogs; for, as we were assured, 'old women no use—dogs catch otters.' The same sound sense would surely lead them to preserve their more useful dogs when still harder pressed by famine.” (1858:1, 214-5) Darwin also adds that the same process of unconscious selection occurs when “aborigines of Australia” are given European kangaroo dogs. Burnett 2009.359themselves, let alone civilized.  With examples like these, Darwin sought to bring his 359readers closer to the border with natural selection.   Darwin took advantage of the resemblances between unconscious selection and natural selection to underwrite the naturalness of domestication, subverting the claim that breeding is merely unnatural interference. For example he writes, “But the gardeners of the classical period, who cultivated the best pear they could procure never thought what splendid fruit we should eat; though we owe our excellent fruit, in some small degree, to their having naturally chosen and preserved the best varieties they could anywhere find.”  These 360“gardeners” did not consciously breed the best fruit, they just naturally preserved the ones that tasted best, in the same way as bees might be said to naturally choose to pollinate those flowers most accessible by flight.  The art of human breeding, in this light, appears quite 361natural.  Or, as Gray observed in the last chapter, “the Art itself is Nature.” 362 In light of this discussion, I find Darwin’s remark about domestication at the beginning of the Descent of Man striking. He writes, ?It has been asserted that man alone is capable of progressive improvement; that he alone makes use of tools or fire, domesticates other animals, possesses property, or employs language; that no other animal is self-conscious, comprehends itself, has the power of abstraction, or possesses general ideas; that man alone has a sense of ?149 Darwin 1859, 37; my emphasis.360 As two historians put it, “Man’s natural historical past shows him to be a part of nature himself. Considered 361as a species, man has even in a state of nature always been an instance of selection for the nature which surrounds him; he has always functioned as an element of natural selection” (Rheinberger and McLaughlin 1984, 355). See also Burnett 2009 and Rheinberger and McLaughlin 1984.362beauty, is liable to caprice, has the feeling of gratitude, mystery, &c.; believes in God, or is endowed with a conscience.  363?When considering a set of commonly ascribed differences between humans and nature—ones he will ultimately attempt to debunk—he includes among this list domestication. As we have seen, the way Darwin puts the concept of unconscious selection to work acts to undermine a distinction, founded on conscious intent and physical locality, between selection as it occurs in the state of nature and domestication as it is done by humans.  Darwin at times entertained an even more extreme view, the logical conclusion of the last quotation. In his unpublished manuscript Natural Selection, Darwin muses over an interesting relationship between a species of ant and aphid.  Though certainly unconscious, 364these ants have modified a species of aphid for their own benefit to such an extent that, Darwin says, we might call the aphids “their domesticated cattle!”  Darwin ends the 365passage with the following striking remark: “Hence we see that man is not the only animal which keeps domesticated animals.”  Such examples, we should note, are not the same as 366his classic examples of natural selection, which involve Nature selecting, but hint at particular animals doing the selecting just as humans select. These provide an indication that Darwin toyed with applying “domestication” outside of even the most unconscious of human ?150 Darwin 1871:1, 49; my emphasis.363 Natural Selection was the book he was writing when he received Wallace’s letter in 1858 that prompted their 364joint publication and the subsequent publication of the Origin of Species. Stauffer 1975, 511-2.365 Ibid. Furthermore, in an early paper published in The Gardeners' Chronicle and Agricultural Gazette Darwin 366considers the relationship between “insect-agents” and the orchids they pollinate over many generations (especially pertinent given the above “gardeners” quotation) (Darwin 1860, 528). See also the Origin where Darwin draws the readers attention to the relationship between natural selection and insect selection: “When our plant, by this process of the continued preservation or natural selection of more and more attractive flowers, had been rendered highly attractive to insects, they would, unintentionally on their part, regularly carry pollen from flower to flower; and that they can most effectually do this, I could easily show by many striking instances.” (93)breeders.  Domestication crossed the divide between art and nature, making plausible 367arguments by natural selection. ?ARTIFICIAL MONSTERS  Domestic varieties were at this time often described by naturalists as “monstrosities,” unfit because they were created and maintained by man, not by nature.  Purportedly human 368interference created awkward organisms with lopsided characters, not organisms perfectly adapted to their surrounding conditions and displaying a “true balance of organization.” For many this was enough to warrant—or could be rhetorically used to defend—their choice of a different object of study, natural species. Darwin was critical of such a perspective.  Darwin argued at times that what is considered a “monstrosity” is largely arbitrary and does not reflect a tangible difference between the state of nature and the state of domesticity. We should be “cautious in deciding what deviations ought to be called monstrous,” he writes. Although “there can hardly be a doubt that, if the brush of horse-like hair on the breast of the turkey-cock had first appeared on the domesticated bird, it would have been considered a monstrosity.” And while, “we might call the wattle or corrugated skin round the base of the beak of the English carrier-pigeon a monstrosity, [...] we do not thus ?151 Another example of this can be gleaned from outside Darwin's work. Dr. Hermann Müller and Darwin 367corresponded from about 1867 till Darwin's death in 1882. Müller wrote a series of articles published in Kosmos in 1878 entitled, “Die Insekten als unbewusste Blumenzüchter” or “Insects as unconscious selectors of flowers.” In these articles he—according to a 1879 issue of American Naturalist—splits the evolution of flowers into peculiarities due to natural selection and peculiarities due to insect-selection. Müller seems to have run to the logical conclusion with Darwin's concept of unconscious selection, and sees no problem attributing this to insects. See Trelease 1879. Darwin also corresponded (more extensively) with Hermann Müller's brother Johann Friedrich Theodor (Fritz) Müller. See also Secord 1986, 523.368speak of the globular fleshy excrescence at the base of the beak of the male Carpophaga oceanica.”  369 In other words, if we are consistent with our attributions of “monstrous,” it would have to apply to many things that exist throughout the natural world as well, he argues. Lopsidedness is as much a property of creatures in the state of nature, as it is domestic productions. Such attributions are therefore not a good way to divide domestic varieties from natural species: many of our domestic varieties “cannot be said to have an unnatural appearance; and no distinct line, as it seems to me, can be drawn between natural and artificial races.”  Although in “extreme cases the distinction is plain, in many other cases an 370arbitrary line has to be drawn.”  371 Moreover, Darwin attacked the two most common reasons why domestic varieties were labeled monstrosities: they lacked a true balance of organization and they were maladapted. To argue that species in nature, such as the giraffe, could have gained their harmoniously coordinated characters through the natural selection of small, continuous variations, Darwin takes the opportunity to draw attention to the harmoniously coordinated characters of domesticated varieties. “If the short-faced tumbler-pigeon, with its small conical beak, globular head, rounded body, short wings, and small feet—characters which appear all in harmony—had been a natural species,” he writes, “its whole structure would have been viewed as well fitted for its life.”  Not the way many naturalists would have seen 372things. He continues, “Look at the greyhound, that perfect image of grace, symmetry, and ?152 Darwin 1868:2, 413.369 Ibid., 245.370 Ibid., 413.371 Ibid., 221.372vigour; no natural species can boast of a more admirably co-ordinated structure, with its tapering head, slim body.”   373 Darwin also repeatedly drew attention to the adaptedness of domestic varieties. This, however, required seeing “adaptation” in a more context-specific or relative light, and in this way differently from Wallace and many others. The reason Wallace considered domestic varieties to be maladapted was because they were not adapted to conditions in the state of nature; they could never outcompete a wild species. Adaptation to man’s purposes is, for Wallace, to be maladapted. Darwin on the contrary seems not to restrict the application of the term to conditions in the state of nature. Of the 18th century naturalist Johann Blumenbach’s remark that “many dogs, such as the badger-dog, have a build so marked and so appropriate for particular purposes, that I should find it very difficult to persuade myself that this astonishing figure was an accidental consequence of degeneration,” Darwin retorts, “had Blumenbach reflected on the great principle of selection, he would not have used the term degeneration, and he would not have been astonished that dogs and other animals should become excellently adapted for the service of man.”  374 The idea that domestic varieties were adapted, rather than maladapted, appears also to be a fairly early discovery for Darwin. In March 1839, he notes, “Get instances of adaptation in varieties — greyhound to hare. — waterdog, hair to water — bulldog to bulls. — primrose to banks — cowslip to fields.” These being “adaptations just as much as Woodpecker. — ?153 Ibid.373 Darwin 1868:2, 220. By the “great principle of selection” is Darwin here referring just to artificial selection 374or a more broad category? Quoted in Hodge and Kohn 1985, 200.375only here we see means.”  What is fascinating about this quote is how freely he passes from 375adaptation under domestication to adaptation in nature.  Darwin’s multi-angled criticisms of the concept of a “monstrosity” put considerable stress on the distinction between domestication and natural selection. Again, as the boundary between this distinction is dissolved, domestication starts to appear less like a disturbance in nature’s “normal development,” and more like a part of nature. ?CONCLUSION: HUMAN DISTURBANCE AND ARTIFICIALITY FOR DARWIN AND WALLACE  In the first part of this chapter I argued against the common view that disagreements between Darwin and Wallace over the nature of reversion can explain their differing views about the significance of artificial selection. In the second part I argued that a better way to make sense of their differing views is by taking into account the broader context in which their discussions were embedded. Their disagreement over the appropriateness of the analogy between artificial and natural selection took place in the broader context of Victorian speculations about the state of nature (“the natural”) and the state of domesticity (“the artificial”). Each supported their view by drawing on a common stock of normatively loaded, tropes and perceptions (consciousness, monstrosity, caprice, etc.). Wallace argued against the analogy between artificial and natural selection and in the process reinforced a distinction between the state of domesticity and the state of nature. This fit well with Wallace’s beliefs regarding the unique status of humans: they alone were endowed with a mental and moral capacity that required a supernatural explanation. Darwin argued for the analogy and ?154collapsed the distinction.  This fit well with Darwin’s larger project to reduce the gap 376between human beings and the rest of nature (which often consisted in displaying standard human abilities, activities, creations, and so on, and showing that they differ only in degree from those found in nature). To Wallace, Darwin’s analogy was too anthropomorphic: human domestication was unlike, in fact it was in direct opposition to, what happened in wild nature. To Darwin, Wallace’s anthropomorphic injunctions were simply anthropocentric.  This disagreement between Darwin and Wallace is, among other things, about the status of domestic varieties and has a bearing on my discussion of artifacts in Chapter 3. Are domestic varieties artifacts caused by human disturbance or are they natural phenomena in their own right? Wallace, as was typical of 19th century naturalists, treats domestication as human disturbance that results in artifacts, not natural phenomena, that is, well-adapted organisms in the state of nature. Human actions are treated as factors that must be excluded from our analysis as they are outside the proper object of study. This is reflected in the judgment that domestic varieties are maladapted—to be properly adapted is to be adapted to conditions characterizing the state of nature. That is the type of adaptation that natural selection is supposed to explain.  Wallace pays lip service to Darwin and uses domestication to illustrate the origin of variation and the principles of heredity, directing the reader to Darwin’s “remarkable” Variation.  But he intends this discussion only to supplement one about the origin of 377?155 I have not looked into the more historical question of what might have caused them to hold different views 376about the artificiality of the state of nature (or the naturalness of domestication). Certainly it had something to do with where they developed their theories and what sort of organisms there abounded: Darwin crossing the boundary between the world of breeding and the world of natural history, Wallace working to classify species in the field (see Secord 1981 and 1986 for Darwin; see Fagan 2008 for Wallace). Morgan and Morrison 1999.377variation and the principles of heredity in the state of nature. And, importantly, human selection is not a snap-shot of natural selection in action for Wallace. Domestication is like the production of seedless watermelons; not supernatural creation, that is, within the bounds of physical law, but also not something that could stand for or explain how the present constitution of species has been created from past causes that are still now in action. The existence of domesticated varieties is ephemeral and tied too closely and contingently to human disturbance. Wallace would therefore rather just show how natural selection follows from a set of descriptive laws that he then demonstrates to hold in the state of nature. Domestication is useful, but it is not natural selection.   Darwin’s view is less straightforward; he is often frustratingly, but strategically, inconsistent. It helps to think in terms of Darwin’s uses, instead of his use, of domestication. Here I have called attention to cases where it appears that Darwin’s strategy is to make questions such as, “is this a product of art or nature?” unanswerable, maybe even to collapse the distinction altogether. Of course fancy pigeons are artificial, but what about sheep modified by completely unconscious selection, dogs modified by the Fuegians, male birds with their uselessly fashionable colors, or ants and their domesticated cattle? In these cases a distinction between the artificial and the natural is put under considerable strain. And with it the distinction between the state of domesticity and the state of nature. We should not forget that these distinctions themselves were central to common disparagings of domestic varieties as proper objects of study.  Darwin’s strategy, in the face of opposition such as that expressed by Wallace, could have been to just deny the “artificiality” of domestication altogether—to play down the ?156capriciousness, ephemerality, consciousness, or whimsical nature of domestication. In places he does do this, for instance, he tries to show that the conscious intent of the selector is not a prerequisite for the selection to be efficacious. But he also does the inverse: he plays up the artificiality of the state of nature. He concentrates on concepts classically associated with civilized society and demonstrates that they can be not only found in the state of nature but are constitutive of the selection therein.   For Darwin, domesticated varieties are not created through the human disturbance of a natural process, they are not unnatural, but are instead the result of a natural process. They are adapted to a particular set of circumstances outside the state of nature—and we shouldn’t forget that the domestic dog was certainly doing better than the wolf in England. Darwin’s Variation, in effect, is as much a natural history of domesticated organisms as it is about the specific facts or laws of variation.  Through this work Darwin actively made domestication 378a part of natural history by writing the authoritative text, thus further corroborating his prior uses of domestication in the Origin.   379 These differences between Darwin and Wallace are not simply about finding artifacts and removing them—as is common in all scientific practice—but about whether a purported artifact is actually a natural object. They could not simply be resolved by appealing to technical or practical factors having to do with the production of domestic varieties. There is a more fundamental difference in their outlooks that compels us to appeal to wider concerns about the artificial-natural distinction itself. Are domestic species artifacts (Wallace’s view) or are they natural phenomena (Darwin’s view)? Are human actions outside of the state of ?157 Gayon 1998, 47.378 See Secord 1986, 536, who originally makes this point.379nature? These questions can only be answered through attempts to navigate the boundary between artificial and natural.  As we have seen in this chapter, this boundary is navigated through classification. Depending on which way domestic varieties are classified makes a difference to whether they are considered valid objects for the study of nature, including whether or not they can tell us about natural selection. As the artificial-natural distinction is defined and redefined so too is what counts as valid for biological study and whether humans count as disturbances.  As one final consideration, in Chapter 1 I hypothesized that the concepts of “nature” and “natural” were not about physical place. In the discussions of the last two chapters I have demonstrated that what is meant by “nature” is “state of nature”—the term inherited from 17th century political philosophy. This term is used 39 times in Darwin’s Origin of Species alone, and was a standard category that stood for the proper object of study in natural history, that is, things as they exist in their uncivilized or undomesticated state. The “state of nature,” however, does not refer to a specific physical place—it is an ideal involving no civilized human disturbance. Darwin exploits this fact: there were people there, the Fuegians, and they practiced a “natural” domestication.  ???158??CHAPTER 6  ?Searching for What Nature has Wrought Dobzhansky and the “Natural” Experimental Fruit Fly  380???the work on natural selection is of necessity confined mainly to experiments in which the environment of the organism is modified artificially [...] Skeptics may contend that if the change in the environment is wrought directly or indirectly by man, the resulting selection is no longer “natural.”  Theodosius Dobzhansky, 1937, Genetics and the Origin of Species ?It is advisable to avoid collecting in orchards or too close to human habitations because in such places the Drosophila populations consist chiefly of cosmopolitan introduced species [...] Such “garbage” species are relatively rare in woods or other “natural” habitats. Theodosius Dobzhansky to J. George Harrar, 1950 ?The standpoint which I for one find extremely difficult to understand is that any such thing is contrary to nature. What is contrary to nature and what isn’t, and who is there to tell? [...] Well, again, it is perfectly reasonable to ask, is Muller’s “Bravest New World” project contrary to Nature? Theodosius Dobzhansky, 1962, Reminiscences ?quite authoritative biologists in the 1920s and 1930s [contended that ...] Drosophila mutants described by the Morgan school were monstrosities, and they were all found in laboratory bottles, not in natural populations, products of some kind of disruption of hereditary materials in highly artificial environments. Theodosius Dobzhansky, 1980, The Birth of the Genetic Theory of Evolution ? Of the many biologists involved in the evolutionary synthesis of the 1930s and 40s, Theodosius Dobzhansky is often singled out as its instigator. The research program endorsed in his Genetics and the Origin of Species (1937) fruitfully blurred the line between the ?159 A significant portion of the research for this chapter was completed at the American Philosophical Society. 380Throughout this chapter, I will use the abbreviation APS.outdoor, “natural” world of naturalists and the indoor, “artificial” world of experimentalists—a line that had impeded the synthesis of evolutionary theory and genetics. Although Dobzhansky was perhaps concerned with “the natural” in a way that his immediate laboratory predecessors were not, his experimental practice was fundamentally similar to theirs and manipulative laboratory work continued to play a dominant and necessary role throughout his career. What then was so “natural” about Dobzhansky’s research and why might this have mattered? ?THE ARTIFICIAL AND THE NATURAL IN THE EVOLUTIONARY SYNTHESIS PERIOD  As the founding of the Journal of the History of Biology in 1968 testifies, the 1960s and 1970s saw an unprecedented interest in the history of the life sciences. Many members of this younger generation had trained, worked or were influenced by the renowned evolutionary biologist turned historian Ernst Mayr. Mayr was never afraid to tell these younger historians when he thought they had overlooked details that were of crucial importance for the historical actors themselves, and when it came to the 20th-century, Mayr often exercised the authority of being an historical actor as well as an historian.   In 1973, Mayr sent the geneticist Michael Lerner—also an historian and historical actor—a draft of an essay review of recent books on the historical origins of genetics. “It is obviously a rather personal and subjective interpretation but I am trying to point out things that are always ignored,” Mayr wrote. “I am trying to make future students of the subject more critical.”  He was “greatly relieved” when Lerner did not object.  What “things” did 381 382?160 Ernst Mayr to I. Michael Lerner, 19 Oct 1972, APS.381 Mayr to Lerner, 6 Nov 1972, APS. 382he think this younger generation had missed? Mayr’s published article shows how closely related they are to the central topic of this dissertation: ?One major development of the 1915-1930 period is entirely ignored [...] I am referring to the work on the genetics of geographic races. [...] studies dealing with natural populations did far more to convince the “Darwinians” of the Mendelian nature of selectively important “natural” variation than either the “artificial” (as they called them) Drosophila mutations or mathematical calculations [the traditional historical foci]. [...] This work was a crucial factor in making the “new [evolutionary] synthesis” of the 1930s and 1940s possible.  383? Mayr’s allusion to “natural” populations, “natural” variation, and “artificial” mutations illustrates a way of classifying the objects of biological study that became increasingly prevalent during the 20th century. The natural-artificial dichotomy was used to classify races, hybrids, populations, species, mutations, variations, processes, environments, methods of study, you name it. These classifications had complex descriptive-normative dimensions; they often concealed underlying concerns about which sorts of objects and methods one should employ to produce valid knowledge—which were the right ones for testing evolutionary theory and which were not. As I mentioned in Chapter 1, although historians and philosophers widely recognize that “natural” (as well as its antitheses) carries normative weight, it is surprising that such accounts rarely extend this analysis to the above uses in biology, or to epistemic rather than moral concerns in general.   384?161 Mayr 1973, 152-3. My emphasis.383 But see Gooday 1991 and Schiebinger 2004. For “nature” as a moral authority in science, see: Daston and 384Vidal 2003; for similar issues regarding non-native species, see: Chew and Laubichler 2003, Chew and Hamilton 2011; for similar issues in environmental philosophy, see Ereshefsky 2007 In this chapter I will demonstrate that concerns about the relation between “the artificial” and “the natural” had an important part to play in the historical period referred to by Mayr as the evolutionary synthesis. It is necessary to say a brief word about the synthesis itself. The phrase “modern synthesis” (later “evolutionary synthesis”) was introduced by biologist Julian Huxley in 1942 to mark a recent general consensus amongst biologists that Mendelian genetics (the study of the principles of inheritance and mutation) was compatible with the theory of Darwinian selection (that evolution is gradual and the result of natural selection acting on continuously varying characters).  Since this time the “synthesis” has 385taken on a much more complicated meaning; in fact, depending on who and what you read (of both historians and historical actors), you will get very different impressions of what intellectual or material “pieces” were synthesized and what made the synthesis possible or even significant.  This chapter is not about the synthesis; it is about the time period 386associated with the synthesis (roughly 1930-60) and more specifically about the thought and practice of someone who is often referred to as a synthesis “architect,” Theodosius Dobzhansky.   Dobzhansky makes for a particularly interesting case study. Not only did concepts of “the natural” and “the artificial” play important roles throughout his career, as is suggested by the epigraph quotations, but his uses of the distinction were quite influential. First of all, the synthesis of which Dobzhansky is often said to have instigated is associated with the breakdown of a previous divide between naturalists and experimentalists, a divide that itself ?162 Huxley 1942.385 See especially Cain 2009 for a discussion of uses of “evolutionary synthesis.” Cain argues that we abandon 386the unit concept of “evolutionary synthesis” altogether as this assumed, unitary narrative tends to hide, rather than illuminate, many of the interesting changes occurring in the life sciences throughout this time period.has a lot to do with drawing a distinction between the artificial and the natural (see Appendix 1).  His scientific work of the late-1930s is heralded as productively blurring the line 387between the outdoor, “natural” world of natural historians and the indoor, “artificial” world of experimental biologists. His friend, the geneticist L. C. Dunn, prefaced Dobzhansky’s seminal Genetics and the Origin of Species  (1937) with the comment that Dobzhansky epitomizes a new “Back to Nature” movement in biology: “The methods learned in the laboratory are good enough now to be put to the test in the open and applied in that ultimate laboratory of biology, free nature itself.”  The book is described by one geneticist and 388editor of Dobzhansky’s work as “the first book that had consequences in the biological world of the twentieth century comparable to the effects of [Charles Darwin’s] The Origin of Species by means of Natural Selection in the nineteenth century.”  389 Dobzhansky’s movement called for a new object of study, the “natural population,” which was distinguished from the “artificial” or laboratory population. The natural population was to be found in the field, rather than created in the laboratory.  This 390movement also required a switch of research organism: from the—in Dobzhansky’s words—“domestic,” “artifi