The Ultimate Ecosystem Engineers

See allHide authors and affiliations

Science  30 Mar 2007:
Vol. 315, Issue 5820, pp. 1797-1798
DOI: 10.1126/science.1137740
Early strategists.

Human niche construction, including controlled burning, may date back to as early as 55,000 yr B.P. Shown is an Australian aborigine.

“Evolution helps those who help themselves” is the basic idea behind the concept of niche construction or ecosystem engineering (1). Many animal species attempt to enhance their environments, and humans have been trying to make the world a better place—for themselves—for tens of thousands of years, often with unforeseen consequences. We have long been the ultimate niche constructors in terms of our rich repertoire of ecosystem skills and the magnitude of their impact. Today, as our efforts at ecosystem engineering are beginning to attempt to reduce and reverse human modification of environments, interest is also growing across diverse fields (including archaeology, biology, climatology, genetics, and geography) in the origins of human dominance of Earth's ecosystems. The general concept of niche construction provides an important new window of understanding about how our distant ancestors, through their initial domestication of plants and animals, first gained the ability to significantly alter the world's environments.

Currently, research on domestication is carried out on two largely disconnected scales—at the level of individual plant and animal species to document the “what, when, and where” of domestication worldwide (2), and at a regional or larger scale, to identify causal “macro” variables (such as climate change and population growth) that may account for “why” human societies first domesticated target species (3). The theory of niche construction provides a link between research at these two different scales of analysis by offering insights into the intervening “how” of domestication—the general human behavioral context within which macroevolutionary factors forged new human-plant/animal relationships of domestication.

Niche construction or ecosystem engineering activities have been documented in a broad range of different animal species. Beavers, for example, through their tree-cutting, dam building, and pond-creating efforts, generate new landscapes for themselves and many other species. Such efforts at environmental modification are proposed to play an important if underappreciated role in shaping biotic communities and evolutionary processes (1).

Studies of human niche construction have usually concentrated on either a particular form of environmental modification, such as controlled burning of vegetation, or on human intervention in the life cycle of a particular target species. Only recently have regional-scale studies offered a fuller appreciation of the extent to which traditional resource-management strategies involve the coherent and integrated manipulation of a broad range of plants and animals and their habitats (4).

Documenting the overall niche construction strategies of past human societies, however, remains a difficult challenge. Archaeological evidence for the management of wild plants (sowing, burning, weeding, irrigation, transplanting, and mulching) provides widely scattered clues to the developmental history of integrated systems of human environmental management. Controlled burning of vegetation to maintain a preferred ecosystem state, for example, is documented throughout the Holocene (the last ∼10,000 years) in numerous temperate and tropical environments, and may have been present as early as 55,000 years ago in southern Africa (5). Yet in many world regions, the appearance of domesticated plants and animals in the archaeological record provides the strongest evidence for integrated strategies of ecosystem engineering.

A number of different aspects of our current understanding of the initial, worldwide domestication of plants and animals points to domestication taking place within a broader behavioral context of niche construction strategies. The development of such human-target species relationships was not a unique event, limited to a particular time or place. Eight to 10 environmentally and culturally diverse world regions have been identified as likely independent centers of domestication and agricultural origin (2). Each exhibits a distinct sequence of domestication of different species over millennia. Human societies thus domesticated a diverse array of species at different times and in developmental isolation across a broad range of ecosystems.

In addition, recent research indicates that the initial domestication of plants and animals in these independent centers encompasses a remarkable diversity of species-specific relationships reflecting a wide range of different forms of human intervention. In Asia, for example, the domestication of two utilitarian species—the dog (for hunting) and the bottle gourd (for containers)—by ∼12,000 years before the present (yr B.P.), did not so much involve deliberate human intervention as it did allow dogs and bottle gourds to colonize the human niche. In contrast, the initial cultivation of tree crops (such as fig by 11,400 yr B.P. in the Near East and banana by ∼10,000 yr B.P. in Southeast Asia) involved the intentional cutting and transplanting of branches, indicating a recognition of the long-term benefits to be gained by a commitment to sustained management.

The initial domestication of goats in the Near East by 10,000 yr B.P., and the subsequent domestication of other livestock species, also called for considerable sustained human intervention, but of a very different kind. Human preemption of herd dominance hierarchies, resulting in human control of animal reproduction, allowed the selective culling of immature males, producing an age and sex profile (that is, a majority of adult females and a few adult males) that is the hallmark of managed domesticated herds. In contrast, the initial domestication of seed-bearing plants (such as wheat, rice, maize, millet, and sunflower), including two species of squash that were domesticated in South America and Mexico by 10,000 yr B.P. (2), involved yet another form of human intervention in the life cycle of target species—the sustained planting of stored seed stock in prepared planting areas. Plants with starch-rich underground organs (including manioc, arrowroot, and leren), on the other hand, were brought under domestication in South America by ∼9000 to 8000 yr B.P. by replanting fragments of parent plants, paired with sustained selection for desired attributes (larger tubers and preferred starch types, for example) (2).

Evidence is also growing for early attempts at domestication that eventually proved unsuccessful. In both the Near East and the Americas, initial efforts at plant management and manipulation were often abandoned, for as yet unknown reasons, and the crops in question were not successfully domesticated until several thousand years later (6).

Whatever the exact mixture of macro-evolutionary forces that were in play, humans identified potential domesticates within the broader context of niche construction strategies through endless auditioning and experimentation with a long list of possibilities. Domesticates would not have been different, necessarily, from the many other managed species in either requiring a greater investment of labor, or constituting a greater intellectual challenge. What set humans apart was their recognized potential for open-ended expansion and ever-increasing returns.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
View Abstract

Navigate This Article