This was not inevitable. Had Australia rafted into the tropics at a different rate, had its initial floristic composition not included so many scleromorphs, had Acacia replaced the ancestral Eucalyptus as part of the original Gondwanic ark, had Australian soils experienced more general rejuvenation, had other species leaked across its borders, the story might have turned out differently. Instead fire seized the core, not merely recycling habitats and nutrients but diverting the whole biota into new evolutionary pathways from which recovery might be impossible. Fire integrated the elements of the bush, and anything that affected the presence of fire ramified throughout the bush. Above all, fire bonded the bush to humans. The bush could not be understood without its distinctive, singular fires.
But fire did not by itself render Australia a land of contrarities: fire is everywhere on the planet. The evolution of the Earth into a fire planet has paralleled the evolution of life. Marine life gave it an oxygenated atmosphere; terrestrial life packed its surface with fuels accessible to oxygen; lightning supplied spark to both life and fire. Since the time these elements first came together, the Earth has burned. Terrestrial life has evolved in the presence of fire. The earliest coal beds are laced with fusinite, the charcoal of ancient fires. Heracleitus was right, “This world … was ever, is now, and ever shall be an ever-living fire, with measures of it kindling, and measures going out.”2
Fire has been everywhere on the Earth’s lands. Huge conflagrations have swept the boreal forest on a cycle of centuries. Temperate forests burn on more irregular rhythms, with greater variations in intensity. Conifer forests throng like chemical kindling, ready to erupt into fire. Thick lenses of charcoal underlie the Amazon Basin. Fire on the order of 4 million hectares gutted the normally dripping jungles of Borneo after several years of drought. Tropical savannas burn almost annually. Fire invades temperate grasslands routinely, sometimes yearly. The grasses of wetlands burn. Tundra burns. Peat burns. Swamps drained by drought burn. Deserts, suddenly flush with life after rains, burn. Cropping out from the ice of Antarctica are coal seams charged with pyrofusinite. The fires in each of Australia’s climates can be matched, and often exceeded, by fires elsewhere or in the past.3
Likewise, the adaptations to fire regimes that Australian organisms display, while marvelous, have analogues among the flora and fauna of other continents. Warm-season grasses everywhere show a similar suite of traits to protect against grazing, drought, and fire. The longleaf pine of the American South experiences a “grassy” stage in its early life cycle that makes it susceptible to burning; if not burned, it is vulnerable to the fatal blue spot fungus as an adult. The seeds of Sequoia germinate best in warm, ashy soil, exactly the conditions that prevail after an intense fire. Several North American pines—lodgepole and jack, among them—feature serotinous cones that open only when heated; a flash fire through the canopy exactly serves this purpose, and the exposed mineral soil is quickly saturated with descending seeds; the forest recovers as an even-aged stand. Other organisms rally around these patterns—the endangered Kirkland warbler thrives only in jack pine of a certain age class. Evergreen sclerophylls exist in all Mediterranean-climate sites. The chamise of California, for example, prepares itself to burn as it ages. By age twenty it steadily increases the proportion of dead wood to live in its crown; virtually all new growth is available as fuel; a woody understory begins to form; the oily leaves, drained of moisture by seasonal or secular drying, blaze like blow torches.
The list goes on. And on. Combustion is too much a part of the biological fabric of the Earth for Australia to have created a monstrous incongruity of fire. The fire regimes of ancient Australia echoed those found elsewhere among its Gondwana cognates. In the deserts fire followed episodes of unusual rainfall. In a pattern that mimicked the central Australian fires of 1974–75 extensive fires swept the Kahalari in the early 1980s after a bout of exceptional precipitation. The border between grassland and desert has ebbed and flowed with the movement of storm tracks and monsoonal winds. Fires have crept into the Sahara with the encroachment of grasses; in Pleistocene times, when the region was much wetter than at present, fires burned grasslands in what is now sanddune and stone. The border of rainforest in eastern India, equatorial Africa, and Amazonia is all but universally inscribed by fire. Tropical savannas burn almost annually. More temperate grasslands burn only slightly less often, a shifting cultivation by fire of scattered trees, shrubs, and grass. The Mediterranean-climate regions of modern Gondwana are notorious as fire environments, fluffed with scleromorphic vegetation, kiln-dried under a baking sun, and subject to outbreaks of foehn and sirocco winds that drive fire before them like a flaming avalanche. Fire is everywhere, differentiated into regimes but never truly absent.
Yet Australia was incontestably different, and fire served as a catalyst for that difference. By the end of the last glacial epoch the island continent moved from the triumvirate that had governed Old Australia—impauperate soils, aridity, lightning fire—to the triumvirate that would shape its succeeding Australias—eucalypts, humans, anthropogenic fire. Old Australia felt that impact first in those regimes where the new elements most intersected, its savannas and its Mediterranean-climate lands. Its ever-living fire was a means of transition, the point through which, like a crystal lens reversing everything within its field of vision, the landscape inverted.
FIRE FUSE: THE AUSTRALIAN SAVANNA
Australia’s savannas link it with a global biogeography. Tropical savannas are extensive in Africa, South America, Mesoamerica, India, and Southeast Asia; temperate savannas, often enormous, border the pampas, veldt, steppes, and prairies of the major continents. The essential ingredients are grasses, a seasonal cycle that brings sufficient moisture to grow fuels yet passes through a dry phase, and fire. In the tropics and subtropics, the border between savanna and rainforest is abrupt and, except in droughts, unbridgeable by fire. In more temperate lands, wooded enclaves thrive in wet river bottoms or on the lee side of rocky outcrops, wherever fires cannot routinely penetrate. When fires are excluded, forest reclaims grassland. Apart from these generic traits, however, the Earth’s savannas reflect local biotic materials and their history.4
It is not clear how the savannas originated. Grasses first appear in the geologic record during the Eocene (50 million years ago). Associated forbs and composites materialize in the Miocene, roughly coincident in Australia with the onset of the Great Upheaval. To what extent this grass mélange constituted a grassland is difficult to determine; almost always there is a concurrent record of trees, either conifer or hardwood. Probably the associated grasses were elements of a forest understory. Then, as climates dried, as drought appeared in seasonal rhythm, as fire increased in prominence, grasses replaced trees as a dominant flora, and forests evolved into savannas. In some places grassy fuels massed; elsewhere they fingered outward into surrounding biotas. Regardless, they became an enormous fuse that carried fire throughout Old Australia.
What complicates this simple scenario is the origin of that critical fire. In historic and prehistoric times, the vast proportion of ignitions has been anthropogenic—so much so that many observers have even doubted the competence of natural sources. In part this reflects how massive human intervention has been: humans have preempted natural fire, or through their fire practices have restructured its vital core. In part, however, contemporary observations speak to a biota that now exists under a climate different from that under which it was created.
The savanna exists in dynamic equilibrium between a tidal climate that drives it alternately to grassland or forest. Lightning fire assisted, and anthropogenic fire arrested, those natural oscillations. The contemporary savanna is likely a human artifact, a biotic edifice sculpted by anthropogenic burning. Fire is not merely something that occurs in a savanna: it is a prime mover of savanna dynamics. To be effective, the burning must be regular and at short intervals for which, as Carl Sauer asserts, “man is the competent agent.” 5
Yet where humans are removed, where the climate is sufficiently ambiguous to support either regime in the absence of hominids, natural ignitions are evident and often potent. Lightning at the onset of the wet season can be effective at kindling large fires. When combined with drought, grass-fueled fires may break through the microclimatic barrier that segregates grassland