Time and Complexity in Historical Ecology is a sustained argument—almost a manifesto—masked as a collection of articles. Subjects range from the Petén basin in Guatemala and Belize to the Andean shore, but the heart of the book is that fabled storehouse of biodiversity, the Amazon forest. Contrary to long-held belief, the contributors insist, all of these landscapes are not little disturbed or pristine, but “domesticated,” “cultivated,” “cultural artifacts that archaeologists can recover and recognize,” even “a form of the built environment.” Indeed, editors William Balée and Clark L. Erickson claim that these places cannot be understood by biologists without reference to “humans’ intentional, long-term, custodial influence,” because the environments are usually shaped, and in many cases actually created, by those humans.

Historical ecology, as the editors call this perspective, inverts the widely known adaptational model, in which cultures adapt to the resources of their environments, with the differing adaptations explaining the main differences between one culture and another. Instead of fitting themselves into ecological constraints, Balée and Erickson argue, human groups rapidly “transform most of those constraints into negligible analytic phenomena,” so that the environment becomes a “physical record of intentionality.”

To some extent, this perspective embraces the new ecology emblematized in Botkin's Discordant Harmonies. But rather than assigning a starring role to “chance and randomness”(Botkin 1990), the contributors to this volume see ecosystems as driven for millennia by human agency. Controversially, they explicitly reject normative terms like “beneficial” or “degrading” to describe that agency's environmental impact, because, Erick-son writes, “There is no ‘natural’ baseline or benchmark of pristine wilderness that should be used as a standard for comparisons…if humans played a major role in creating the very landscapes where biodiversity and nature are said to occur” (p. 246).

Instead of reading the human stories encoded in landscapes, Michael Heck-enberger writes, researchers have taken the “absence of robust historical knowledge”—the relative dearth of written documents from pre-Columbian societies—“as a lack of history at all… The history is ‘naturalized’ into an imagery of pristine wilderness and primitiveness” (p. 312). One need only look at a Green-peace calendar to see what he means.

The articles in this volume exemplify diverse ways of reading these landscapes. David G. Campbell and his team contribute a clear quantitative study demonstrating that the array of tree species in the Guatemalan Petén still bears the stamp of the Maya more than a thousand years after their ninth-century “collapse,” even though the region has been thinly inhabited since then. Peter W. Stahl argues that the distribution of prehistoric small-mammal remains—especially those of generalists with broad niche requirements—indicates where forests were disturbed in past centuries. More than 90 percent of western Ecuador's Jama River valley, thickly forested as late as 1920, is now cleared. But Stahl's examination of more than 85,000 archaeofaunal specimens from the region “strongly suggest[s] a prevalence of unstable edge environments and forest fragments…spanning at least 3600 years” (p. 145). The Jama's dense, nearly unbroken canopy, in other words, was a recent phenomenon, an ecological novelty no more than a few centuries old.

How did the contemporary tropical forest come into existence? The research is far from definitive, but articles by Erickson, Heckenberger, William M. Denevan, and especially Eduardo G. Neves and the late James B. Petersen provide some clue. The Amazon basin has been inhabited for at least 13,000 years (Roosevelt et al. 1996). Gradually those early inhabitants turned from foraging to “landscape management,” Neves and Petersen suggest, beginning by scattering useful palms from half a dozen genera, especially Bactris gasipaes(peach palm). This low-intensity landscape management “may well have made a substantial impact,” but it is hard to distinguish today from nonhuman processes because it represents “the cumulative outcome of individual, small-scale interventions.” Perhaps 3000 years ago, though, there was “a radical shift in economic and social patterns in Amazonia,” and some societies began “high-intensity” landscape management. The result, according to Denevan, was the creation of landscapes of “semi-intensively cultivated fields intermingled with fruit orchards, managed fallows, house gardens, and brief bush fallows, with semi-permanent settlements, some numbering thousands of people, surrounded by zones of modified forest manipulated by hunting and gathering activities” (p. 154)—an intricate system that both created and exploited the swaths of rich soil today called terra preta do índio.

Among the most impressive examples are the Llanos de Mojos (Mojos plains) in eastern Bolivia. Roughly twice the size of Ohio, the region is seasonally flooded with up to a meter of water for months at a time. Beginning perhaps 2000 years ago, Erickson argues, its original inhabitants humanized much of the landscape by erecting 10,000 settlement mounds, “thousands of linear kilometers of causeways and canals,” hundreds of still enigmatic circular ditches, countless earthen fish weirs, and thousands of hectares of raised fields. The goal was to create “millions of linear kilometers of rich terrestrial aquatic ecotones or edges in what was previously a relatively homogeneous, flat environment.” This “patchwork of artificial landforms,” to Erickson's mind, was “as productive and sustainable and probably equally species-rich as the forests that exist there today.” (I have twice visited the area with Erickson, once accompanied by Balée; both times I was powerfully struck by the evidence of landscape domestication.)

This entire system, and others like it, fell apart in the 16th century. Between 1500 and 1700, European disease, slavery, and war killed 90 percent or more of the native population. With no hands on the tiller, so to speak, landscapes throughout the hemisphere went feral, which is not the same as going “wild.” The tropical forest we seek to protect today was in significant part the accidental byproduct of this horrific loss.

Driven by population loss and fear of enslavement, many groups gave up agriculture in favor of foraging; Loretta A. Cormier tracks one such group, the Guajá in eastern Amazonia, who survived by “patch-to-patch movement from old fallow field to old fallow field,” living off the landscape domesticated by their predecessors. Others continued to farm, Denevan notes, but in a new way: slash-and-burn. Although ecologists have both celebrated slash-and-burn as adaptive and vilified it as destructive, they have commonly regarded it as an age-old practice. Stone axes were such inefficient instruments of land clearing, Denevan writes, that “long-fallow shifting cultivation was probably difficult, even with the girdling and burning of tree trunks.” Once patches of forest were opened up, farmers had an incentive to keep using them, with short fallow times to reduce weeds. By contrast, metal axes are up to 60 times more efficient (Carneiro 1979)—ideal for populations that want to move quickly in and out.

Many ecologists have come to accept these ideas, at least in part, but they remain anathema to most conservation groups and are a source of puzzlement to land managers, for whom the vision of pre-Columbian wilderness is a useful benchmark. For these people, Time and Complexity will provide little consolation. The task of understanding and conserving the lowland Neotropics, it suggests, falls into the purview of anthropology, archaeology, geography, and the other human sciences.