The Evolution of Phylogenetic Systematics aims to make sense of the rise of phylogenetic systematics—its methods, its objects of study, and its theoretical foundations—with contributions from historians, philosophers, and biologists. This volume articulates an intellectual agenda for the study of systematics and taxonomy in a way that connects classification with larger historical themes in the biological sciences, including morphology, experimental and observational approaches, evolution, biogeography, debates over form and function, character transformation, development, and biodiversity. It aims to provide frameworks for answering the question: how did systematics become phylogenetic?
The Evolution of Phylogenetic Systematics
Reflections on the History of Systematics
Robert E. Kohler
Of all the life sciences, systematics is probably the one whose history is least studied. Its celebrity founders have been well historified: Linnaeus, whose universal system of binomial nomenclature still endures; Darwin, who gave classification a biological foundation; and a few others. But of the activities of the hundreds of collectors, curators, and classifiers who have found, preserved, named, and ordered the million-plus species whose world we share-of these our knowledge remains scattered and fragmentary. This is paradoxical, because of all the sciences systematics has the deepest living memory, thanks to rules of nomenclature that oblige those who would name a new species to actively engage the literature back to the Linnaean big bang.
This situation is, happily, changing; substantial histories have been quietly accumulating, some by historians with a sustained devotion to the subject. These include Jim Endersby (2005, 2008), Paul Farber (1976, 1985), Jürgen Haffer (1992), Joel Hagen (1984, 1999), David Hull (1998), Gordon McOuat (1996, 2003), Ernst Mayr (1982, chaps. 4-6), Bruce Patterson (2000, 2001), Harriet Ritvo (1997), Peter Stevens (1986, 1994), Keith Vernon (1993), and Mary P. Winsor (1976).
There are also circumstances external to biological systematics that may stimulate greater interest in taxonomy. One is the decided uptick of interest among historians of non-life sciences in classifying, both as practice and as a way of knowing. This is especially marked in the history of chemistry (Ursula Klein, Michael Gordin) and mineralogy (Matthew Eddy) and also in the history of ecology and generally in sciences that deal with collections of material objects, like archaeology or ethnology (Klein 2003, 2005; Klein and Lefèvre 2007; Gordin 2002, 2004; Eddy 2003; Bensaude-Vincent, García-Belmar, and Bertomeu-Sánchez2003; Müller-Wille 2003; and on ecology and thing-rich sciences, see Kohler 2007, 2008). It may be, too, that growing public concern about loss of biodiversity and anthropogenic mass extinction could make systematics and its history matters of broad interest. Attitudes are changing, both in the world of scholarship and in political culture, that have for over a century relegated systematics to the low end of the totem pole of prestige in science.
Meanwhile, we can imagine what we would ideally like to know about the subject. Basic facts, for a start: who first gave that great multitude of creatures their scientific names and identities; and when, where, and how. There is the human encounter with nature in the field-my own particular interest: how fieldwork has been organized, who paid for it, how it was experienced; and the encounter in museums and herbaria with the ever-rising flood of specimens and species to be put in order. How have systematists dealt with tens of millions of fragile objects and a million-plus natural kinds? It is a problem of data handling that few other sciences have had to confront.
Systematists themselves are no less interesting subjects of historical inquiry. What sort of people are these energetic finders and sorters, and what are their distinctive folkways-their rules of categorizing and naming; their customs of rewarding discovery and pruning errors and redundancies? How do they maintain (as all communities must) a sense of common identity, and by what social arrangements do they renew that community across generations? Recruitment is a particular problem for systematics, because the science is based in museums rather than in university departments with their ready access to young talent; so training is more like an informal personal apprenticeship than in most sciences.
This comprehensive history of systematics is the larger undertaking to which my own recent book on American systematists and their global activities between 1870 and 1940 is a modest contribution (Kohler 2006).1 I want to use this chapter, first, to summarize some of the main ideas of my book, which deals mainly with fieldwork; then, to think more broadly how my approach might apply as well to systematics in other times and places.
It turns out that the history of discovery of new species-at least in vertebrate groups-is surprisingly lumpy and episodic. I did not know that at first. Only when I had nearly finished my book did it occur to me that it would not be hard to chart the historical pace of discovery. The data were ready to hand-thanks to systematists' custom, when compiling species lists, of giving the date that each species received its currently accepted name.
I'm not sure what I expected to find when I began to tabulate these data: perhaps a more or less steady pace of discovery, with spurts and pauses reflecting the randomness of individual initiative and serendipity. So I was surprised-riveted, in fact-when the data on vertebrate groups began to show distinct waves. Following the initial Linnaean burst of naming before 1800 there was a strong second wave of activity peaking in the 1830s and 1840s; then a third wave (for most vertebrate groups though not for birds) that topped out in the 1890s and 1900s (my period); and a fourth, much smaller peak (more a bump) of discovery in the 1960s and 1970s (Kohler 2006, 6-7, fig. 1.1).2 (See fig. 1.1.)
I was not surprised that a heightened pace of discovery occurred in my period of interest. I had found much anecdotal evidence that it did, and knew just how zealously American museums in this period were collecting around the world. What I had not anticipated was that the entire history of species discovery, before and after my period, would be episodic, with decades of intensified activity separated by relative lulls. It was obvious from these data that individual opportunities and initiative were not idiosyncratic and random but structured-by some thing or things, and on a global scale. But by what things? The same ones that I had found operating in the United States circa 1880-1930? Or by others distinctive of their own times and places?
A grand Noachian narrative of species inventory is there to be written-that became clear-but too big for me, a novice in the history of taxonomy, and perhaps too big for any scholar working alone. But seeing a pattern so concretely silhouetted in the data, I had to wonder how the interpretation I had devised for my chapter of the larger narrative might inform the whole.
We aren't used to thinking of the late nineteenth and early twentieth century as a golden age of systematics. The usual story is that after the grand voyages and explorations of the late eighteen and early nineteenth century, natural history was gradually eclipsed by the newly ascendant sciences of the lab. This is true-but only half the truth. Lab culture did not cause systematics to wither away; it continues to flourish. As is often the case, "decline" was only relative and more a matter of perception and attitude than achievement. It was in the early twentieth century that inventories of most vertebrate groups became essentially (say, 90 percent) complete. For birds this point was reached around 1900; for most mammals, circa 1940; for bats and insectivores, by 1950; and for North American snakes and turtles, the 1920s.
The roster of collecting expeditions by American institutions alone is impressive. The U.S. Biological Survey under its founder and chief C. Hart Merriam fielded dozens of parties per year from the late 1880s to the early 1910s, mostly in western North America. A dozen state natural history surveys were also active, most of these small shoestring operations but a few systematic and sustained (e.g., in California, Michigan, and Nebraska). Research museums organized expeditions both in their own regions and abroad: the Museum of Comparative Zoology (Harvard), the Museum of Vertebrate Zoology (University of California), and the University of Michigan Museum. However, it was the larger civic museums that were organizers of expeditions on a truly grand and global scale: most notably the American Museum in New York, the Field Museum in Chicago, and Philadelphia's Academy of Natural Sciences but also midsized museums in cities like Brooklyn, Milwaukee, and San Francisco. The National Museum in Washington was barred from underwriting in-house expeditions, but its curators routinely took part in expeditions organized by federal agencies that had that right.
The scale of museum expeditioning in its heyday is remarkable-all the more so because it is now so little remembered. Between 1887 and 1940 the American Museum dispatched some 206 sponsored expeditions in vertebrate zoology, plus over 200 more in other field sciences (archaeology, paleontology, anthropology). The yearly average in zoology was 3.8 in 1891-1901, 8.5 in 1906-25, and 13.9 in 1926-40. In the peak years 1929 and 1930, 30 and 27 expeditions were dispatched at a cost of $283,000 and $207,000 respectively. In vertebrate zoology, the Field Museum organized 72 expeditions (1894-28), and the Academy of Natural Sciences sent out 57 (1889-1930). In addition to these official, named, and sponsored expeditions there were uncounted unofficial research trips by curators. Overall, thousands of expeditions must have been launched at a cost of millions or tens of millions of dollars (Kohler 2006, 117-123). If there was a big science before the era of Big Science, systematics was it.
Taxonomic field practice in this period was also distinctive: it was a survey mode, quite unlike the practices of earlier and later eras. Aiming at a complete inventory of the species of entire regions, survey collecting was both extensive and intensive: extensive in its geographic reach; intensive in the way parties combed every nook and cranny of each locale, crisscrossing and revisiting until every resident species was found and recorded.
Survey collecting was also collecting en masse. It sought, not a single "typical" pair of specimens per species, but large series that represented the full range of intraspecies variability. Expeditions were planned, organized, and (for their time) capital intensive. Whereas collections of tens of thousands of specimens were once thought large, survey collecting produced collections ten or a hundred times as large, and prepared by standardized methods. Survey was rigorous and exacting; and it made systematics if not an exact then an exacting science.
Systematics in the survey mode required an elaborate infrastructure of museums, with their dedicated study collections and elaborate exhibits, and of social networks of well-to-do patrons and local participants. Working in a survey mode also put new demands on its practitioners. To succeed in the business, systematists had of course to be schooled in basic sciences, adept in the arcana of diagnosis and naming, and at home in the vast and sprawling taxonomic literature. But since they were now both field and museum naturalists, they had also to have practical knowledge of how to organize and lead complex field parties and manage sponsors and the media. And there was the curatorial side of tending museum collections, which required skill in managing, fund-raising, and museum politics, plus knowledge of the specialized techniques of constructing naturalistic dioramas en masse.
What circumstances in the survey period enabled systematists to create such an elaborate and exacting science and to produce a great wave of species discovery? In my book I set forth a model of three interlocking elements: an environmental element, which gave systematists physical access to relatively unaltered natural areas; a scientific element, which gave them the intellectual incentives to undertake collecting in the demanding survey mode; and a cultural element-wide public interest in nature and natural history-which drew big private money into building museums and collections and into expeditioning on the grand scale.
First the environment. The era of survey collecting coincided almost exactly with a distinctive and fleeting phase in the settlement history of North America that I have called the "inner frontiers." The linear, westward-moving frontier was famously declared closed in the 1890 Census and by Frederic Jackson Turner in his celebrated essay of 1893. There were by then so many pockets of settlement scattered about the West that no distinct line separated settled from unsettled land. However, pockets remained in the interstices of road and railroad nets that were sparsely settled and ecologically lightly altered, yet also relatively easily and cheaply accessible to hunters, tourists, vacationers, and-most significantly for us-naturalists and collectors.
This landscape of inner frontiers was created by Americans' restless mobility and devotion to land speculation, as well as by what William Cronon has called the logic of capital.3 To preempt rivals, railroad companies built lines into areas in which there were not yet enough residents to produce an actual profit. State and federal giveaways of cheap homestead land likewise encouraged rapid hop-skipping settlement, and speculators would sit tight on large parcels of land and wait for rising prices. The political economy of manifest destiny thus created a landscape of intermingling wild and settled terrain. In the East, abandonment of marginal agricultural land created similar mosaics of fields and second-growth forest.
It was not that naturalists had never before been able to visit the wide West: of course they had. The point is that they had not previously been able to operate there in a survey mode. Transfrontier areas had been accessible only to individual explorers, who lived off the land, or to state-sponsored expeditions on the grand scale. In either case, visitors were limited to rapid transects and opportunistic collecting. Not so in the inner frontiers: there, cheap and relatively comfortable transport enabled naturalists to linger in safety and collect in depth. In inner frontiers expeditions could stay in continual touch by telegraph and rail with home bases, ship out tons of specimens cheaply and securely, and be resupplied en route. They could make repeat visits if necessary, to make species inventories as complete as possible.
The inner frontiers were in hindsight a once-only bonanza of species discovery waiting to happen. Here were large tracts made roughly known by earlier explorer-naturalists who had skimmed the cream of species that were easy to find and catch but who had not lingered-because they could not-to get those that were rare, shy, or elusive. Inner frontiers invited survey. Arguably, survey was possible only in such places. It was a (scientific) form of land use as characteristic of the place and time as bonanza hunting or land speculating. No wonder environment and scientific practice overlapped almost exactly in time.
But physical access to places of opportunity does not explain why scientists would see and take advantage of the opportunities. Naturalists had also to have intellectual incentives to visit inner frontiers, not just casually on working vacations, but systematically, as members of organized surveys. Investment in travel and organized fieldwork had to pay off scientifically-and to be seen in advance to pay off-so that surveys could be planned and patrons secured. Opportunity derived not just from nature but also from changes in taxonomic science.
In fact, systematists could be confident that survey collecting would be worth the trouble, because they already knew that more complete collections made for better and less error-prone taxonomies. This was apparent by the mid-nineteenth century, when the first moves were made toward a more intensive field practice. Naturalists first began to collect systematically and in depth in the 1850s: among them (in the United States) Spencer Baird, who made use of national railroad and boundary surveys in the American West, and Louis Agassiz and Joel Asaph Allen at the Museum of Comparative Zoology. These pioneers in modern collecting were ardent empiricists who took as a matter of faith that denser empirical evidence would make systematics a more exact and secure science. And they were right.
The chief cause of taxonomic error was making designations on the basis of too few specimens or of specimens from a too restricted locale. Without full knowledge of species ranges, for example, it was easy to mistake as new to science species that were in fact already known but simply outside their normal ranges. Likewise, systematists unaware of how variable species can be were prone to designate as new species forms that were only extreme variants of species already named and described. The result was a confusion of false and redundant names.
The conventions of Linnaean taxonomy made such errors not just an individual embarrassment but, in effect, a labor tax on the entire community of classifiers. To keep order in the Ark, systematists had to publicly correct every mistake and to reduce every redundant name to the dustbin category of synonymy. Thus the cost of fixing individuals' mistakes was borne by all. (Experimentalists have an easier time of it: their blunders and follies are buried forever in the dark, bottomless bog of uncited and forgotten literature.)
But if the first mass collecting made it clear just how error-ridden the taxonomic literature was and why, it also made clear that the problem could be solved by more of the same-by forming deeper and more comprehensive collections. What Baird, Agassiz, Allen, and others had begun as individuals to do became standard practice for all self-respecting and responsible systematists. Darwin's theory of speciation by variation and selection was for some (e.g., Allen) a further incentive to adopt a data-intensive practice. But the chief intellectual incentive for systematists to collect in a survey mode was not theory but improved empirical practices of diagnosis and revision.
The rewards of survey collecting were real and immediate, in terms of quality of work and personal credibility. Those with abundant empirical data got credit not just for the new species they described but also for correcting other taxonomists' mistakes; and they were less at risk of being themselves publicly corrected and discredited. Survey collecting could make less organized modes of gathering seem not just inferior, but irresponsible. If inner frontiers provided the opportunity for systematists to collect more widely and intensively, it was the intellectual and career rewards of improved practice that provided the impetus to exploit those opportunities in an organized way.
Yet the argument is still incomplete. However powerfully drawn by nature's opportunities and the scientific rewards of a data-intensive mode of work, systematists still had to have money and organizational backing to travel and collect en masse-which few individuals possessed. So most were dependent on financing from national or civic museums or well-to-do private patrons. And that was, at first, a problem.
For one thing, museums at the time did not employ curators to collect; curators were expected to stay put and curate. Like amateur naturalists, curators collected during their vacations or in their spare time. The modern role of curator-expeditioner had to be invented for survey to become a general museum practice. Nor did museums at the time engage in active, planned collecting or sponsor in-house expeditions. Collections were assembled from random gifts, purchase, or exchange of duplicates. The rewards of collecting en masse were hardly as immediate or compelling for museum directors and their bourgeois patrons as they were for scientists. Understandably: in their view, museums existed to inform and entertain the public. What benefit was there for museums or the museumgoing public in paying curators to go on junkets to exotic places and fill museums with vast numbers of nearly identical specimens?
Yet in the 1890s and early 1900s museum officials and patrons began to do just that, and the reason they did is the third, cultural, element of our model: popular interest in nature and the culture of outdoor recreation.
It is well known that cultural conceptions of nature were changing dramatically in the late nineteenth century.4 The prevailing utilitarian view of nature as a warehouse of "resources" to be turned as quickly as possible into money was tempered by alarm over the damage wreaked by all-out, use-it-or-lose-it exploitation, as well as by growing awareness of the recreational values of undisturbed nature. In the United States, preservationists like John Muir and Theodore Roosevelt and institutions like the Sierra Club and the Audubon Society sought to restrain the all-out "war" on nature and to have choice parts of the inner frontiers preserved forever for recreational use.
In literature and the arts, meanwhile, a sentimental and anthropomorphizing view of nature, exemplified by romanticized animal sculpture and nature fables in which animals were endowed with human virtues and vices, gave way to a more naturalistic view of animal life. The nature essay, invented by John Burroughs in the 1870s and widely popularized in the 1890s, was art designed to be scientifically accurate as well as emotionally pleasing. This same ideal of uniting art and science also inspired the invention of the habitat diorama, which occurred in American (and Scandinavian) museums around the turn of the century (Wonders 1993). Diorama builders went to extraordinary lengths to make these objects not just visually beautiful but also true to nature and to the realities of animal behavior and ecology. Conceptions of nature that had been purely cultural-economic, aesthetic-thus became naturalistic: not science exactly, but congruent with a scientific view of nature.
In addition, new and active forms of outdoor recreation evolved in the late nineteenth century that afforded not just new ways of representing nature but also new ways of personally experiencing it that combined pleasure with a naturalistic or scientific interest. Among the more important of these outdoor recreations were sport hunting and fishing, camping and outdoor vacationing, rural perambulating and mountain climbing, summer cottaging in lake and mountain districts, and buying "abandoned farms" as summer vacation homes. These novel cultural practices, I argue, were the soil in which the practices of survey expeditions flourished.
I call these activities "new," but of course they were not quite that. Hunting and fishing, for example, had long been pursued as sport by rural gentry, and by farmers and working folk to put food inexpensively on their tables. Well-to-do families had since the eighteenth century toured rural areas in search of sublime or picturesque sensations. What was new in the late nineteenth century was the meaning these activities acquired when they were taken up by the striving white-collar middle classes. It was these newcomers who made subsistence or leisure pursuits more naturalistic and sciencelike.
Attracted by the pleasures of outdoor recreation but uneasy with the association of leisure with aristocratic idleness and proletarian moral disorder, middle-class nature-goers altered the meaning of these pleasures to square with their values of discipline and self-improvement. This they did by pursuing outdoor vacationing as a kind of work, which they took to be improving and essential to the work of making money and getting ahead. They worked at play, as the historian Cindy Aron (1999) put it, turning leisure into physical and mental recreation (see also Bailey 1978).
The more strenuous forms of outdoor recreation were especially valued by promoters of middle-class vacationing for their physical and moral benefits, as were also instructive pursuits like birding or amateur naturalizing. These afforded an experience of nature-going and learning like those that John Burroughs evoked in prose, carried out in environments like those depicted in museum dioramas. A quasi-scientific interest in nature transformed outdoor recreations into suitably moral and improving activities for middle-class family vacationing.
It is no accident that enthusiasm for these culturally charged modes of nature-going coincided with the period of inner frontiers and with the survey mode in systematics. Inner frontiers, wild yet accessible, were ideal places for the strenuous but not too arduous or risky (or expensive) forms of family vacationing-working at play. This moralized form of outdoor recreation, like natural history survey, was a form of land use specific to inner frontiers.
Nor was it happenstance that middle-class nature-going also sustained survey science. For one thing, it enlarged the pool of sympathizers with-and potential recruits to-survey science. The practices of recreational hunting, camping, and naturalizing were identical in many respects to those of survey science: for example, knowing where to find, recognize, and observe (or catch) animals; and how to travel cross-country and live outdoors in safety and relative comfort. In effect, recreational nature-going constituted a pleasurable and unwitting apprenticeship to the work of scientific collecting.
Of course only a tiny fraction of recreationists ever became career naturalists. However, recreational nature-going also predisposed its devotees to take part indirectly in survey science, by supporting natural history museums or by underwriting surveys and expeditions. Outdoor culture provided wherewithal and created infrastructure: that was the vital connection. Well-to-do families could contribute to survey science by sponsoring expeditions-and, if they so desired, by going along as hunter-collectors-without having to devote their lives to doing science. Recreational nature-going made survey systematics culturally familiar and understandable to the sector of the public that had the inclination and the means to participate in some way large or small. Those who took an active part in expeditions found them a pleasurable and educational blend of science and vacation: it was perhaps the ultimate way to work at play.
Evidence of the connection between expeditioning and nature-going is everywhere. C. Hart Merriam subsidized his own large-scale collecting as an amateur naturalist before he learned to use the politics of government patronage to scale up a personal obsession into a national faunal survey. Annie Alexander, who founded and underwrote the Museum of Vertebrate Zoology and its program of field collecting, was a wealthy sugar heiress and an avid big-game hunter. She needed a public repository for her specimen-trophies (because of legal restrictions on private hunting) but quickly found a higher aim in the scientific program of her director, Joseph Grinnell. Alexander Ruthven ran the University of Michigan Museum and its program of expeditions with the modest financial support of a group of local amateur naturalists and collectors.
Civic museums likewise depended on the patronage of outdoorsy families for their global collecting projects. It was not the prospect of aiding science that first attracted patrons, though; rather it was the desire to assist in building habitat dioramas. It turned out that state-of-the-art dioramas could not be built with materials present in museum storerooms. Their spectacular illusionistic effects required fresh specimens of the highest taxidermic quality, as well as accessory material (dirt, rocks, branches, plants) and paintings or photographs gathered or made in the very places that the dioramas were to depict. And these materials had to be gathered by curators and preparators, because only they knew exactly what was required. If few museum patrons could see the point of vast scientific collections, many were eager to pay for diorama expeditions.
Once in the field, of course, curators and systematists also collected en masse for study collections, and it was not long before expeditions were as much or more for science than for exhibit making. In this way museums were drawn into the business of expeditioning, and curators' professional job descriptions were enlarged to include fieldwork. It was dioramas, initially, that connected museum science to the culture of outdoor recreation. Combining pleasure and science, they embodied the ideal of working at play and afforded museumgoers virtual-and sometimes actual-trips to the inner frontiers where systematists labored to record and order nature's abundance.
The culture of nature-going, conjoined with an exact science of systematics in a landscape of inner frontiers, produced the third wave of species discovery. Inner frontiers gave systematists physical access to places of species abundance, and supported a naturalistic culture of outdoor recreation, which afforded systematists the means to pursue their opportunities and to reap the professional rewards of improved practice. A dynamic of change, which started small and might at any point have stalled, developed into a sustained wave of discovery.
The Bigger Picture
Because this model was devised for the particular case of North American naturalists, the question will arise, Does it apply as well to others? British, German, French, Scandinavian, and Russian naturalists were active namers of new species, and many were also active field collectors. But did European governments, museums, and botanical gardens organize expeditions on a continental or global scale, as Americans did? Were they drawn by their own or distant inner frontiers? And were they, as Americans were, sustained by a culture of middle-class nature-going? (The science, we may assume, was much the same everywhere; Linnaean systematics has always been a markedly transnational science.)
There are as yet no answers to these questions: the research has not been done. We do know that museums in America and Europe had different operating principles. The British Museum, for example, preferred buying specimens on the open market to in-house collecting. And the Paris and Berlin museums built large collections from specimens gathered by state military or navigational expeditions. Networks of colonial administration were also sources of varying importance for the imperial powers. But civic museums in Germany never developed a system of in-house expeditioning. Nor was there a custom in continental countries of regional natural history surveys (Nyhart and Burkhardt pers. com. [on Germany and France, respectively]; McOuat 2001). But these are bare beginnings of a comparative history. We have no systematic data on collecting and expeditioning by Europeans, and no idea if differences in collecting practices gave the science of systematics distinctive national styles.
Likewise with the cultural element. Some European countries did have a culture of outdoor recreation, especially Britain and Scandinavia (Bailey 1978; Wonders 1993; Allen 1994; Frykman and Löfgren 1987, chap. 2). These (Sweden's in particular) resembled the American mode; but there were also differences. For example, sport hunting was generally not a mass activity in Europe, and feeling for nature seems more strongly linked to the politics of consolidating nation-states. The French were more attached to their domesticated agricultural landscapes than to ones more wild, in contrast to Canadians, Swiss, or Swedes (Nyhart 2009; Applegate 1990; Green 1990; Zeller 1987; Wonders 1993; Zimmer 1998). Whether or not customs of outdoor recreation had the same effects on science in Europe as in the United States remains to be discovered.
On the environmental side as well, little systematic work has been done. But the historical geography of Europe is well developed, so it would not be hard to do. In general, European landscapes in the western core were more intensely settled than those of North America and their flora and fauna more depauperate. But there were accessible frontiers in the far north and east (Scandinavia, Siberia, inner Asia) and in isolated pockets elsewhere. And the westernmost imperial powers-Britain, Spain, France, the Netherlands-had inner frontiers in their transoceanic colonies: they were just not contiguous as in the case of the United States, the Russian empire, and Scandinavia. So national differences may prove to be less stark than we might expect. One thing we can say for sure is that comparative history is in order.
Other lines of comparative inquiry lead from the survey period to earlier and later episodes of species discovery. If the science of one episode was specific to its environmental, scientific, and cultural circumstances, then it is reasonable to suppose that it would be the same story with others. We don't know this, because we don't have the factual evidence; but as a working hypothesis it seems a good way to begin. We would look for a distinctive practice of gathering and sorting, sustained by some combination of geographic access, scientific opportunity, and cultural infrastructure.
There is reason to think that developments in taxonomic science were as vital to earlier and later episodes of heightened activity as they were for the survey era. In the case of the first, eighteenth-century, wave of species "discovery" it was Linnaeus's system of naming that created a scientific opportunity: to refashion a world of competing and incommensurable local systems of classifying into a single, simpler, and universal one. This change in practice gave a second life to older literatures of natural history and materia medica, and data mining became a characteristic practice for a generation of Linnaeans. At a time when travel was costly and arduous, libraries were an accessible and rewarding field for discovery. Intellectual rewards lay in reconceiving what was already "known."
Of course species were also sought in nature. Linnaeans did travel, and those with access to state patronage traveled a lot. And an active international network of local collectors sent specimens to the metropoles of Linnaean science to receive scientific identities (but a project to train a cadre of traveling collectors failed, because unseasoned and inexperienced Europeans too often fell victim to accidents and disease) (Koerner 1999, 147-148).5 Correspondence and exchange networks constituted the characteristic mode of transnational science in Enlightenment Europe, and natural history collecting was in that mode (Lux and Cook 1998; Harris 1998).
Enlightenment science was also distinguished by its intimate connections with economic production and trade. Lisbet Koerner has shown just how involved Linnaeus and his circle were in mercantilist projects of acclimatizing exotic plant species to northern Europe (Koerner 1999). Here we see both the geographic element-in efforts to remake human and natural biogeography-and the cultural (economic) infrastructure of data gathering. What the culture of outdoor recreation was to the survey period, the political economy of expanding global trade was to the Linnaean. So we see the silhouette of a distinctive Linnaean taxonomic science, shaped by circumstances of science, physical environment, and culture.
Likewise with the second, or "Humboldtian," episode of species discovery. Here again we see a change in taxonomic practice opening up intellectual opportunities: in this case, the invention of natural systems of classification that relied not on single characters arbitrarily selected, as in the Linnaean system, but on whole organisms and their habits and ecology (Stevens 1994). Rewards thus lay in revising the provisional assignments of an artificial classifying system into the more permanent ones of a natural system. This recurrence of data mining, in a new form, may partly account for the bulge of "new"-that is, newly robust-species descriptions in the 1830s and 1840s.
But natural systems of classification depend for their proper working on abundant and comprehensive data, in a way that the simpler artificial systems did not.6 And getting such data-including knowledge of animals in life-entailed fieldwork. So in the decades following the end of the Napoleonic Wars we find travel becoming a more regular feature of taxonomic practice. Exchange networks remained active; but collecting was increasingly carried out by travelers, both recreational and official, from the European hearth to far-flung parts of the world.
Active on-site collecting was made possible in part by the industrial and economic revolution in global transport-that is obvious. Steamships and railroads, and a denser infrastructure of colonial extraction and administration, made distant outbacks accessible even to individual naturalists. Military doctors and officials in far-flung outposts were major contributors to species inventories. The French and Russians were especially open to combining diplomacy with exploring and natural history collecting. In the 1830s the Russian, Prussian, and French (but, oddly, not the English) consulates in Rio de Janeiro were all staffed by naturalists (Swainson 1839, 391-392; Shearer 2009). Growing imperial rivalries among European states made exploration an affair of state and turned explorers and explorer-naturalists into national celebrities.7 Systematics was thus enmeshed in the modern political economy of European expansion in its more invasive, coal- and steam-driven, colonizing phase. The science was the incidental beneficiary of these larger changes, as it would be later of the new middle-class culture of nature travel.
In its mobility and long-distance spatial projection, exploration resembled natural history survey; but in most respects, it was distinctly different. Exploration was an activity not of inner frontiers, as survey was, but of distant and sometimes dangerous frontiers that were hard to reach and were occupied by peoples who did not welcome European intrusions. Exploration was meant to be adventure, to afford experience of the exotic, to draw out heroic virtues-which survey decidedly was not. And as noted above, whereas survey was carried out by numerous midsized expeditions, exploration was more typically done either by single collectors traveling fast and light and gathering opportunistically or by large parties in extended, multipurpose expeditions on self-sustaining ships or in overland caravans.
Moreover, whereas survey expeditions were for science, the purpose of exploration was more commonly economic or strategic (commerce, conquest, showing the flag). Itineraries and pacing were dictated by commercial or strategic, not scientific, needs. Collectors for the specimen trade sought what connoisseurs would pay for: namely, rarities. The market afforded no incentives to gather large series of common species. And in large exploring expeditions collecting was often a sideline, pursued by naturalists as best they were able. Since exploring parties tended to traverse, not linger, naturalists typically could collect widely but not comprehensively or in depth, as survey parties did. Exploratory science was, in short, extensive but not intensive; opportunistic, not planned; not systematic but catch-as-catch-can.
A mode of fieldwork that produced lots of new species relatively quickly but not in large series was in fact well suited to a Linnaean taxonomy and, at first, to natural systems of classification as well. Both were improved by an abundance of species, and for both a deep knowledge of intraspecies variability was a mixed blessing (because it could make categorical boundaries less distinct). Cream skimming thus encouraged a typological conception of species, and a conceptual world of types in turn legitimated an exploratory mode of field practice. Collecting a few representative or "type" specimens for each species was what mobile explorer-naturalists could do, and collections assembled in this way made type species seem natural and real. The only reason to collect in depth was to acquire duplicates for exchange; only later did its advantages for taxonomic science become clear.
This is all somewhat speculative, of course; and it begs the question of when and how exploratory field practice gave way to in-depth survey. One thing is clear, however: it was not a sudden or a simple replacement. The process was more a layering of practices, followed by gradual divergence. Collecting in depth was already fairly common by the 1850s: in the United States, by Spencer Baird and his network of collectors, whom I mentioned earlier; and in Britain, by Joseph Hooker and his network of botanical collectors centered on Kew Gardens and the British Museum (Endersby 2005). And even if traveling naturalists did not collect in depth, firsthand experience of creatures in their natural abundance could open their eyes to the variability of species. Nor did exploration just wither away in the late nineteenth century. Quite the contrary. But its directions and purposes changed: from natural history collecting to physical and human geography; and from places of abundant life to extreme and inhospitable places (deserts, high peaks, the poles) that were poor in flora and fauna but rich in opportunities for heroic display. Natural history collecting thus developed practices distinct from exploration that were then amplified and regularized as intensive survey.
Here again, a combination of environment, science, and culture seems a promising frame for understanding the systematics of the age of steam and empire. It's no more than a model, mind; how well it will stand up to empirical research remains an open question.
A long view of the history of systematics suggests still other lines of inquiry. It is clear, for example, that collecting and discovery are patterned not just in time but in space as well. There is a historical geography of systematics that has been little studied. We know in a general way that there have been hotspots of discovery in various parts of the world, which have come and gone as different regions became accessible or interesting to Europeans. Explorers and collectors often favored areas where their countries were active in commerce or colonizing and where they could count on special access and diplomatic or logistic support. The question is, how often? Science followed flags; but it also followed scientific and biogeographic opportunity. Naturalists were no less drawn to places where biodiversity was rich and scientific problems most enticing, whatever their color on geopolitical maps.8
These are hypotheses; but there is evidence that they are sound guides to empirical research. As a test, I charted the 526 rodent species that received their currently accepted identities between 1758 and 1869, tabulating their type locales against the nationalities of the naturalists who described and named them. Geographic patterns of activity stand out clearly in the data (table 1.1). In the discovery of Eurasian species, Germans, Russians (or Germans in Russian employ), and Scandinavians predominated. Americans skimmed most of the cream of North American species, but British and Germans contributed substantially. In South America, Germans described more than twice as many new species as did British and French naturalists. But the British had a virtual monopoly in South and Southeast Asia and Australia, splitting East Indies species with Dutch naturalists. In East and Central Africa, discoveries were pretty evenly divided between British, French, and Germans; while in South Africa, the British predominated over Germans, with the French a distant third.9 (Americans were virtually absent outside the home territory-in sharp contrast to their presence worldwide after 1870.)
It is clear from this preliminary mapping that naturalists in any given country were most active where national trade or colonizing gave them an advantage of access and public support but that scientific interest figured as well. German traveling naturalists seem to have been particularly drawn to places of scientific promise (inner Asia, South America, Africa). It stands to reason, since Germany was by then the leading scientific power in most fields but not (yet) a commercial or colonial one. Historical research in this period could well begin with these singularly footloose and productive German naturalists (e.g., Haffer 1992, 122-128).
It is also clear that patterns of global activity in systematics have changed over time. For example, Bruce Patterson has shown that in the case of Neotropical (i.e., South and Central American) mammals there have been striking long-term shifts in the centers of scientific activity. Continental systematists predominated between 1758 and 1850; British and (increasingly) Americans, between 1850 and 1950; and Americans and (increasingly) South American naturalists, between 1950 and 2000 (Patterson 2000, 194, fig. 3).10 Whether these shifts reflect changes in commercial or strategic presence in the region or diverging national trends in scientific interest remains to be established.
How geopolitical and scientific imperatives balanced out is a key question when examining a historical period during which newly modernized taxonomic disciplines were strongly linked to commercial and imperial venturing. The history of systematics is a human biogeography as well as a history of ideas and practices, and we want to discover the principles of systematists' global dispersals. We need to map their taxonomic achievements and explain how that map came to be and why. How deliberate was its creation? Did systematists stake out national spheres of influence and leave other areas to their competitors? Or did all compete in the same strategic regions? To what extent did scientific practices draw upon those of commerce and colonizing?
How Periods End
Another general question is, What causes episodes of heightened species discovery eventually to wane? The survey model suggests a symmetrical view. If changes in geography, science, and culture create opportunities and incentives for discovery, then further changes in these elements must also cause episodes of discovery to wane. That's the hypothesis.
Some such combination seems to be what in the 1940s brought the period of survey collecting to a close. To be sure, the Great Depression and World War II helped-the one cutting off funding for expeditions, the other making travel of any sort temporarily impossible. But that can't be the whole story. Expeditioning continued through the 1930s, just more discreetly and frugally than before; and when the fighting stopped and prosperity returned (to the United States) after 1945, survey expeditioning did not resume.
What closed the survey era in North America, I have argued, was structural change in the conjunction of environment, science, and culture that had sustained survey science for some fifty years. On the environmental side automobility, interstate highways, and urban sprawl squeezed inner frontiers into enclaves. On the cultural side, outdoor recreation morphed into commercial mass tourism, shedding the intellectual imperatives that turned some recreationists into patrons of survey science. And scientifically, as species inventories neared completion, survey collecting encountered diminishing returns.
It was not a simple matter of natural limits: there were then and still are species left to discover. Limits lie rather in the relation between what nature offers and how naturalists use what is offered to make science and careers. Systematists in the late 1940s could have gone after the last 10 percent of vertebrate species, for example. But their quarry were too few, too scattered, or too evasive to warrant the expense of survey expeditions. It was more cost-effective to rely on individuals and serendipity-which is essentially what happened.
Survey scientists could in theory also have moved on to insects and other invertebrates, many groups of which are little known. But it is hard to see how survey practices could be applied to groups with astronomical numbers of species many of which occupy tiny local ranges.11 What purpose would be served by total inventories; and how would the public be induced to pay for them? Sampling may be a more appropriate practice in such cases; but would sampling of small locales deflect scientific interest from species inventory to ecosystem dynamics?
Or will the immediate future of systematics be shaped less by what collectors find in nature than by what systematists find in existing collections by applying new methods from the physical sciences-DNA fingerprinting, cladistics, cybernetic data sifting? Perhaps the vast accumulations of specimens and written records from the age of survey will prove to be the next inner frontier (or indoor frontier) of taxonomic science. Perhaps the Linnaean practice of data mining will recur in new ways. As these matters are well beyond my ken, I will leave them as questions-loose ends for someone else to pull on.
What I can say with confidence is that the practice of species survey was specific to its environmental, social, and scientific world. And it seems likely that the practices of Linnaean and Humboldtian systematics were as well to theirs. Doubtless it will be the same with whatever world and practices are coming next.
I have focused thus far on the methods and circumstances of species discovery. However, there are other issues in the history of systematics, no less interesting and significant, having to do with the community of systematists and their customs and with comparisons of systematics as a science with other natural sciences. As this is largely uncharted ground I will just point briefly to some issues that I find especially intriguing.
To start I would like to go back to my earlier observation that systematics is preeminently a science of data management. This will seem obvious for contemporary cyber-systematics-because in our computer age informatics is a familiar category. But systematics has always been a science of data management, from Linnaeus on up. It is, we might say, the type specimen of informatic science. Its communal customs and practices can be understood as tools for managing abundant things (specimens) and categories of things (subspecies, species, genera, families, etc.) that unmanaged would overwhelm the mind.
Many sciences are data-rich, of course. But no other science has anything like the number of categories that systematists must create, tend, prune, arrange, and police. There are millions of organic chemical compounds, for example, but not all that many categories to put them in. They constitute a vast and ever-expanding encyclopedia (Beilstein's Handbuch der Organischen Chemie), not a taxonomic classification (Gordin 2005).12 Likewise, accelerators, geophysical sensing technologies, and gene-sequencing machines produce amounts of data beyond the capacity of human brains to manage. But the game in these sciences is to sift rare signals from discardable noise or to form pictures out of pixels-not to order data into permanent categories. Systematics is special in its superabundance of categories, and what we want to know is how that superabundance has shaped the science and its community of practice.
Systematic biology is also singular in that it has remained preeminently a classifying science. Other thing-y or data-rich sciences have classified at some time in their histories-for example, chemistry, ethnology, ecology, and archaeology. But in these sciences classifying has always proved to be-by necessity or choice-a passing phase (Kohler 2007). Only in systematics have inventorying and arranging remained central and defining activities. Why this was so, and how this fact has affected systematists' culture and community relations are questions historians need to address.
Other features of the culture and customs of systematics are no less distinctive. Of all the sciences systematics has by far the deepest time horizon: that is, the point where working knowledge becomes (mere) history and ceases to be a tool of active use. In experimental sciences that point is typically reached in a few years or even sooner. Systematists, in contrast, operate in a living tradition that goes back 250 years and includes everyone who ever named a species. This is so because of the peculiar conventions of taxonomic nomenclature, especially the rule of priority (which gives precedence, in cases of redundant names, to the first one published) and the rule that rejected and redundant categories, once published, must be preserved forever as "synonyms"-a category of names that are not things but might be mistaken for things if not publicly labeled as phantoms or mistakes. Thus every act of diagnosis, naming, and revision requires systematists to actively engage with the taxonomic literature, and with their scientific predecessors, all the way back to Linnaeus.
Equally distinctive of taxonomic science is the social diversity of its practitioners. In principle anyone can describe and name a species, however expert or inexpert he or she may be. Of course all would-be describers must submit to exacting communal rules of describing and naming. And as the practices of diagnosis have become more exacting and quantitative, systematics has gradually become a game for the credentialed and experienced. Yet systematics remains open in principle to lay participation in a way that few sciences are (observational astronomy is another); and that openness affects the science in ways that have no parallel in the laboratory sciences.
Of course all modern sciences regulate traffic across boundaries with lay publics; but for systematists, because that boundary is unusually open, traffic control is especially tricky. Neither simple exclusion nor free access is an option. Systematists must engage lay naturalists while limiting their ability to disrupt the intricate practices of classifying. Typically they have steered lay participants away from formal taxonomy and toward observing, census, and ecological or life history study (Barrow 1998; Secord 1994; Drouin and Bensaude-Vincent 1996).
Systematics is also distinctive, and perhaps unique, in its relations to other sciences. Practices of naming and classifying are designed to serve the biological ends of systematic biology, obviously; but they also serve other life sciences, whose practitioners may not be interested in species or classifying but need to know (at least) what species they are working with. Systematists are thus stewards of nature's order for all the life sciences. And that fact makes for an ambiguous economy of reward and credit.
All scientists engage in housekeeping, of course, to keep order in the jungle of organic chemical compounds, or in collections of mutant strains of flies or mice or phage, or in databases and DNA and protein archives. But they do this for themselves, and the costs and benefits of housekeeping are assessed by and accrue to those who do it. Housekeeping for oneself is a necessity and brings no dishonor; in contrast, housekeeping for others is seen as low-caste labor and brings little respect. So systematists get no kudos for services to those outside their own ranks. Indeed, their status suffers, because consumers of their taxonomic favors experience the relation as unequal and so perceive those who serve as servants.
Of course, classifying is not mere housekeeping or cataloging (stamp or butterfly collecting is the time-honored gibe): the stereotype is wrong. Systematics is a biological discipline, and classifying gets at the principles of nature's workings just as effectively in its way as comparison and experiment do in theirs. It is an exacting and creative process-as any classifier knows. The puzzle is, why do outsiders persistently mistake what systematists do as routine maintenance and not creative science? The answer, I think, lies again in the social relation of systematists to other biologists, who see only the formalities of naming and ranking (because these are what they use) and not the science. So they take what systematists do as the familiar routine maintenance that they perform for themselves.13 The galling logic of credit and reward (or lack of them) is probably inescapable.
At the same time, systematists' deep living past and broad sense of stewardship may make their science a more humane and humanly engaging one than the sciences whose practitioners labor only in the immediate present and only for themselves. We know that laboratory sciences differ greatly in their "moral economies"-that is, in their customs of mutual obligation and reward. Fruit fly geneticists, whose little commensal species is a cornucopia of productive things to do, have for a century sustained a culture of generosity and mutual aid. In contrast, scientists who engage in winner-take-all quests for biopharmaceutical substances tend to develop competitive cultures of every man for himself and the devil take the hindmost (Kohler 1994, 13-14, 98-106, 233).14 Practices and material cultures shape communal values in varied ways: just how varied, we don't yet know. Systematics, with its exceptionally developed system of communal rules and obligations and its Noachian moral economy, is a premier site for further study.15
There is another way in which systematics may be more central to the history of modern science than the stereotypes allow. We believe it is normal for modernizing sciences to abandon description and classification as they acquire theory and experiment (a vestige perhaps of the old belief in universal linear progress). But it seems just as likely that the real bedrock of modern science is neither theory nor experiment but an intensive and exacting empiricism: a capacity to generate, organize, and manipulate ever more abundant factual knowledge. All modern sciences have that in common, however different their sites and modes of practice. Scientists who do not go to nature to record and gather its abundance create artificial environments (labs) and methods (experiment) for making facts even beyond what nature's cornucopia spills forth. Theory and experiment are tools for making and handling facts. And facility in dealing with empirical and categorical abundance is also what science has in common with those other quintessential institutions of modernity: industrial capitalism, the nation-state (Chandler 1977; Scott 1998).16
If this is so, then systematics, though it may be low in the prestige hierarchy of sciences, should be seen as the exemplary modern science-because it most fully embodies the defining characteristics of modern society generally. The notion flies in the face of common wisdom, I know-but for just that reason it might be worth pursuing.
The distinctive features of systematists' communal life must derive partly from the fact that systematists cultivate that part of nature that is most abundantly categorical: the part produced by that cunning bit of evolutionary machinery that Darwin was the first to grasp. Nature's endless diversity and abundance keeps them at the work of sorting and arranging. (Practitioners of sciences that generate their own facts at will can leave their past behind and give up classifying as unneeded and unmodern.)
But natural endowments do not explain everything. In the natural history of the sciences, history matters as much as nature. It makes a difference where and by what historical paths sciences happen to take shape. Systematics was born of an Enlightenment vision of universal and comprehensive, yet thing-y and particular, knowledge. And it evolved into its modern form in a society of economic superabundance that lives by producing, accumulating, ordering, and managing things and facts (e.g., Bowker and Star 1999; Dumont 1980 ). Perhaps humans are by nature collectors and categorizers; culturally we certainly are. We are Homo categoricus, and systematics is the science in the doing of which we are most ourselves.
I am grateful to Jim Endersby and Peter Stevens for helpful criticisms of a penultimate draft.
1. The first section of this chapter is drawn from chapters 1-3 and 7 of this book.
2. Pre-1800 data are lumped because rates per decade would simply reflect the publication of major treatises. It does not follow from these results, of course, that a periodic pattern also obtains for plants and invertebrates.
3. On the logic of capital, see Cronon 1991, esp. chap. 2.
4. The following argument is summarized from Kohler 2006, chap. 3; see also Anker 2007.
5. The Russian Imperial Army operated a successful school for explorer-naturalist diplomats (Shearer 2009).
6. A suggestive account of this change is Swainson 1839, esp. 188-235.
7. The literature on travel and exploration is too vast even to begin to cite. On explorers as celebrities, see Riffenburgh 1994.
8. For example, I have seen little evidence that American museums worked in Central and South America for consciously geopolitical reasons: accessibility and biogeographic interest were what mattered. However, Spencer Baird's activities in the 1850s were clearly part of America's imperial expansion in the far West.
9. Data were taken from Honacki, Kinman, and Koeppl 1982. I chose rodents as a test case because they were not popular with amateurs and show a markedly periodic pattern of discovery. Namers of species were of course not necessarily their collectors; some of the most productive (e.g., Linnaeus and keepers of the British Museum) stayed put and reaped the rewards of what collectors gave or sold their institutions.
10. Patterson infers nationality from the geographic locations of type specimens and the journals in which descriptions were published.
11. Inventories of groups of insects that are attractive to humans and live in densely humanized areas (e.g., beetles and butterflies in Britain) are nearly complete. And I have it on good authority that arachnologists feel that a complete inventory of world spiders is within reach (Norman Platnick pers. com.). On the other hand, noncelebrity groups in inaccessible places are still barely known.
12. The arcane rules of naming organic compounds, though not officially a classifying system, do in some ways operate as one.
13. For another example of this social logic, see Shapin 1989. The social relations of physicists to chemists, or chemists to biologists, or biologists to behavioral scientists are less fraught, since in these cases it is highly valued theories and reductionist strategies of explanation that are trickled down.
14. The ur-source of the idea of moral economy is Thompson 1991a, 185-258; also Thompson 1991b, 259-351.
15. We might fruitfully compare species inventories and classifications to natural resource commons, which historically have been intricately regulated (and remarkably effective and stable) systems of usufruct rights for diverse communities. The literature on commons is vast, but a good point of entry is Ostrom 1990.
16. For a similar argument about early modern science, see Shapiro 2000; Cook 2007.