by Bill Gammage
Of course this is true of history, otherwise there would be only one history per topic. Bowman states the obvious because he assumes that it is not so of science, not so of ‘objective’ research. He made this clear in 2004: ‘while “fire-stick farming” is now a generally accepted concept, little effort has ever been made to test it. Indeed, while there is abundant anecdotal evidence of Aboriginal use of fire to attract kangaroos and maintain kangaroo populations, little empirical evidence exists.’4 Historians would find this laughable.
Bowman lists ‘isms’ (capitalism, feminism etc) to which history is subject, then quotes Stephen Pyne: ‘natural science builds on data banks; the humanities, on values’.5 What do data banks build on? One might say that they accumulate objectively verifiable answers. But where do the questions come from? They build on values, by what is seen to matter at the time in some social or personal context. Recall the head measurers, and how priorities change in wartime—if values don’t build data banks, why did planes progress in six years from barely flying to racing from England to Australia, and in another six from biplanes to rockets? Why didn’t the Chinese develop guns once they’d invented gunpowder? Objectivity is a research method, not a philosophy.
‘It is unacceptable’, Bowman states, ‘to make inferences about vegetation based solely on an uncritical acceptance of historical records that have not been checked in the field.’6 To the extent that is true, so is the reverse. He continues, ‘historical records concerning vegetation characteristics may be erroneous’,7 supporting this non-claim by noting that von Mueller’s 1867 tallest tree height was disproved in his lifetime. Mueller was a botanist, a scientist reporting contemporary objective data, not a historian using historical records or building on values. Bowman continues that when Mueller was challenged by theodolite and surveyor in about 1889, he ‘distanced himself . . . saying that he had had to “trust his memory”’.8 Mueller’s distancing is from a historical record, was subjective, and does not objectively prove him wrong, but Bowman accepts it, thereby crippling his case for the relative value of historical vs scientific research. He may be wrong anyway: in 1874 Mueller took a visiting naturalist to the fallen giant, needing neither theodolite nor surveyor. It measured 146 metres, supporting Mueller’s initial claims.9 A scientist might see objective data verified by repetition in this, even a data bank, but Bowman concludes, ‘Obviously there is a need to make objective assessments of landscape change rather than to rely solely on historical records.’10 Yet he quotes and relies solely on a historian’s use of a historical record.11 Time has made an objective measurement ‘anecdotal’.
Bowman cites nine Bradford Hill criteria used to infer causality. Most are sensible and verifiable by analogy or experiment, though one is merely ‘logical’ and another ‘plausible given current knowledge’. ‘Very few ecological theories about Australia’s environmental history’, Bowman writes, ‘would pass these exacting criteria unscathed.’12 That’s true—that’s why they are theories—but most are proffered by scientists, not historians. Fortunately neither will be deterred from a healthy curiosity about them, or cease to benefit from the stimulus they generate. History makes no bones about this, and put this way I hope that disciplines which produced the Origin of Species and the Theory of Relativity might agree, even though neither theory now passes those criteria unscathed.
Bowman concludes, ‘Practitioners of environmental history must accept that their studies are politically charged and that their findings are bounded by great uncertainty.’13 The first claim I accept, the second depends on the evidence. I think this obvious of any research; it seems Bowman does not. The history discipline above all others welcomes dabblers, but it is a delusion to think it not a discipline, not capable of refinement by training in skills, methods and values. I can only suggest that scientists wanting to use historical records study history.
Anthropologists apart, scientists rarely take the subjective implications of Aboriginal culture and values into their research. Many admire Aboriginal knowledge and skill, and especially in northern Australia many use it, but they don’t seek Aboriginal mindsets which inform what a historical source says, or which integrate land and people in the Aboriginal way. I have not found a single scientist who takes into consideration the fundamental Aboriginal conviction that people risk their souls if they do not manage every inch of ground they are responsible for. How far this conviction impinges on any land at any time is subjective and variable, but decisive. Science rarely admits culture: culture is subjective, which science has made a subjective decision to deny.
Natural factors must be considered, but do not always explain. For example soil type can explain a change in plant species (a change in place), but not why a species grows now where it didn’t then (a change in time). Many scientists focus on the first, I on the second. In the best history of its kind I know, Mick Allen writes of state forests in central New South Wales, ‘The mosaic of vegetation described by the explorers as they passed through this district quite probably has as much to do with variations in soil type as anything else.’ Here a good forester makes an assumption, but it is not proven, and it cannot explain why, as Allen notes, bordering tree species have since invaded grassland the explorers saw.14
Ron Vanderwal and David Horton similarly confuse a change in time with a change in place. They state of the southwest Tasmanian coast,
The current vegetational mosaic . . . reflects the mosaic of generally poor soils which have created relatively static boundaries. Fire plays some role in maintaining this mosaic. Bushfires occur today, but their frequency may have been greater when Aborigines inhabited the area, a hypothesis that receives some support from the apparent recent incursion of closed forests into grassland areas.15
Have those closed forests spread on the same soil or not? Yes or no, why spread now but not then? Perhaps this puzzle led Vanderwal and Horton to mention fire, but they do so very cautiously, suppressing the most sensible possibility, that fires could only have been more frequent then than now if Aborigines lit them (picture 39). Was their vision obscured by a subjective urge towards a natural explanation?
In the north, BA Wilson suggested of Melville Island’s boundaries between open forest and ‘treeless’ plains ‘that the deep sandy soils of the plains have been subjected to heavy leaching of nutrients, which is a primary factor associated with changes in vegetation across the forest–plains boundary. Fire is implicated as a secondary agent.’16 Do the forest–treeless boundaries follow soil boundaries or not? If they do, why is fire an agent at all? If they don’t, why is soil mentioned at all? And while strictly the plains Wilson discusses are treeless, they carry acacias, banksias and other shrubs. Why not more? Isn’t it possible that Melville Islanders worked with the country, burning more where trees find growing difficult and less where they find it easier, so that without their fires there might have been more trees on one or both sides of a boundary? Cloaking that possibility by vaguely naming ‘fire’ may be scientific, but a historian would consider human intervention.
Rod Fensham and Jamie Kirkpatrick investigated whether a want of soil moisture or chemicals explains why Melville Island’s woodland is so open. They found no universal correlation, instead noting, ‘the relative dominance of Eucalyptus miniata and Eucalyptus tetrodonta is changing in a mixed forest on Melville Island’, and concluding, ‘The coincidence of random historical events and the phenology and life-stage of the tree species at a site may determine some of the structural and compositional patterns at the local scale in the tropical savannah forest.’17 Phenology is the ‘study of the times of recurring natural phenomena’. Fensham and Kirkpatrick are saying that tree life cycles and chance or randomness explain why this woodland is as it is and was. Tree cycles alone poorly explain a change in species composition, while chance is the last and least useful historical explanation, because in the end it clarifies nothing. To prefer it to possible human intervention carries a liking for natural explanation a long way.
; In a thoughtful article, Kevin Mills puzzles at the ‘close proximity’ of grassy eucalypt forest and rainforest in the Illawarra (NSW). He considers soil, nutrients, rain shadow, wind, frost and aspect, finally offering a tentative explanation in soil and rainfall boundaries. Yet his Yarrawa Brush map shows both eucalypts and rainforest across his key soil boundaries, and of rainfall he concedes, ‘The location of the eucalypt–rainforest boundary in the west cannot only be due to a decrease of rainfall. If this were so, the boundary would be located 8 kilometres or so further west.’ He does mention ‘possibly fire’ and ‘perhaps’ fire, but not as a general cause, and not 1788 fire.18 I suggest that his well-researched vegetation anomalies show people working with and against the country in 1788.
Even scientists who know the importance of fire might prefer not to mention Aborigines. Jennifer Read found that in Tasmania phosphorus deficient soils retard Native Beech growth and canopy, but that high phosphorus concentrations occur on some Buttongrass plains, heath and eucalypt forest, which therefore should be growing Beech. She concludes, ‘the results are consistent with Jackson’s (1968, 1983) hypothesis that the absence of rainforest from some low nutrient soils may be influenced more by fire frequency (via the interactions among soil nutrients, vegetation and fire) than directly by soil nutrients’.19 Does this go far enough? At least on those higher phosphorus soils, how did those other plants dislodge rainforest in the first place? Read doesn’t say. She suspects chance fire, but not 1788 fire. Why is chance the better explanation, random as it is and inconsequential as Jackson found it?20
Darrell Kraehenbuehl has written a good account of Adelaide’s 1788 vegetation, the most detailed for any Australian capital.21 He recognises that Aboriginal fire ‘would inevitably have had an impact upon the ground-storey plant species, even perhaps converting some woodland areas and forest to tall shrubland and grassland’.22 He then ignores this, even when detailing such vegetation anomalies as the grass around Black Forest and the Peachey Belt. He thus describes what the land might have looked like without 1788 fire, not how it was in 1836.
On the Monaro (NSW) Alec Costin mapped a correlation between basalt soils and treeless plains. He suggested that there was ‘insufficient soil moisture’ and too much cold air drainage to carry trees.23 This is the common view. As George Seddon put it, ‘most ecologists who have studied the Monaro are convinced that the basaltic soils and cold-air drainage from the surrounding ranges down to the relatively lower plateaux are enough to inhibit natural tree growth without any other agency’.24 Trees on or beside the Monaro plains, scientists say, are on granite, not basalt.
On the Dawson, rarely frosty, Leichhardt puzzled at why trees grew well on some basalt soils but not at all on others of similar consistency (ch 1). Parallel questions rise about the Monaro. Was there in 1788 a good correlation between treeless basalt and treed granite? Are other valleys of similar size and altitude treeless or frost stricken? Are trees returning naturally to basalt grasslands despite these causes?
Mark Currie first described the Monaro. His account supports Costin. From about Michelago on 3 June 1823 he went south ‘through a forest country, and near several stony ranges, to a rather extensive plain, which proved to be the commencement of a very long chain of down country’. Near Bredbo next day he ‘Passed through a chain of clear downs to some very extensive ones, where we met a tribe of natives . . . we learned that the clear country before us was called Monaroo, which they described as very extensive.’ Across the Bredbo on 5 June he ‘observed a continuation of downs to the southward clear of timber’. On 6 June he rode south ‘say forty miles’ to about Billilingra Hill or east of it, and saw grass all the way. On a map he marked ‘Extensive downs clear of timber’.25
Yet not all, and not only, downs were treeless, nor all granite hills treed. In July 1844 Robinson saw plains with trees and hills without:
The immense Downs with their undulating grassy surface stretched out before me as far as the eye could scan, a Park of great magnitude and beauty studded with copses of Banksia, Casuarinae, Mimosa, Shrubs, and small belts of Eucalyptus with bare and isolated mamillary shaped and flat-top’d hills.26
Further, the forests were open, so burnt regularly. Sometimes miles off, Currie saw ‘hills for the most part stony’, therefore open: before the 2003 fires they were dense forest with few stones visible. On or near Currie’s route near Bredbo, Lhotsky ‘passed all the way forests of blue and white gum, but remarkably destitute of any underwood’,27 while Bunbury found the district ‘covered with fine grass; it is part plain and part fine undulating hills with groups of trees sprinkled all over them in a most picturesque manner . . . it bears a most park-like appearance’.28 ‘The forest ground is thinly timbered with the white gum, and gently undulated’, Stewart Ryrie wrote from his Monaro station in 1840. The granite ranges east of Bredbo were ‘all bare of timber (except in the gullies), and . . . covered with species of heath and coarse wiry grass’, while the country south and west was ‘undulated, with rich flats between the hills, some of which are lightly timbered, others perfectly clear’. North of Cooma he ‘came to open downs . . . about six or seven miles long by five broad . . . travelled through a mile of open forest and came out upon more open downs called Belaira [Monaro], and travelled across them for some miles to the Murrumbidgee’. West of Cooma he found country ‘beautiful in appearance, consisting of open forest ranges, well clothed with grass to the very summit, and large meadows lying between them’.29 These are familiar patterns, showing 1788 fire on ridges, hills and flats. Why not the rest of the downs?
The patterns were similar from Yass south. The Yass district, which no-one thinks is the treeless Monaro, was ‘for the most part open forest, with luxuriant pasturage . . . The “Plains”, or more properly speaking, extensive downs, are destitute of trees.’30 At Lake George and Canberra the plains were grassy, the hills usually scattered trees and grass (ch 9; pictures 51–2), but in 1954 Lindsay Pryor wrote, ‘there is little doubt that the Canberra Plains, Ginninderra and Tuggeranong were treeless and therefore true climax grassland’.31 These plains carry naturally regenerating trees now. As a result, ‘the Monaro’ has retreated south. In 1820 it began at Canberra, as a ‘beautiful clear plain . . . very extensive. Rich Soil and plenty of grass.’32 For Costin in 1954 it began south of Michelago,33 but today grassland there is intersected by low hills dense with eucalypts, there are patches of natural regeneration, and native tree plantings are flourishing. In short, while Monaro trees may grow more readily on granite than on basalt, there is no evidence of a fixed tree–grass boundary along a basalt–granite divide, and there is evidence of 1788 fire patterns. Ground where trees find it hard to grow is precisely where people working with the country might burn most intensively.
In cold country soil steps back, and treeless valleys become frost hollows. Costin put the prevailing view of their cause:
The dominance of dry tussock grasslands along many treeless valleys of the Monaro is also related to a high frost incidence determined by cold air drainage. The absence of trees is apparently due to the inhibition of seedling development, since it has been shown that trees can be grown successfully in these areas provided the seedlings are protected from frost during their early life.34
Millennia ago the Monaro was not frosty, and presumably trees protected seedlings then. How did it get clear? The question is also pertinent to open high valleys ringed by Snow Gum, Black Sallee and similar eucalypts, whose seedlings survive metres of snow. It is hard to see how frost might kill so completely where frost and snow in turn do not. Equally telling, trees are now returning, edge-invading hollows from forest both higher and lower. Nursery Swamp near Canberra lies in a grassy valley flanked by sheltering hills. Stockmen once used it to nurse calving cows, hence its name. Today tea-tree is capturing the grass: it has spread visibly in recent decades. Not far away, Smokers Flat is described as ‘treeless due to cold air drainage from the surrounding slopes’.35 Tea-tree is reclaiming it too, and euca
lypts are invading its edges.
Another ‘frost hollow’, Orroral Valley, was grass when Europeans came, began reverting to forest soon after, then was cleared by axe. Several portion plans between 1858 and 1870 state, ‘good open forest’ or ‘open undulating forest. Ringbarked.’ In 1998 Brian Egloff noted eucalypts ‘regenerating on the fringes of the grassland’. Across the river from Orroral homestead, an undulating grass plain is flanked on the east by steep timbered hills and on west by a narrow tree belt along the slope edging the swamp. North a creek blocks it; south, swamps, hills and rocks steadily narrow it until it meets a steep bank which blocks views further south. Just over the bank the ground drops sharply away. This is a wallaby trap, not a frost hollow, burnt from forest but in every direction terminating in obstacles. Prey could be driven to suit the wind, and always be at a disadvantage.36 Trees are now reclaiming the plain’s edges.
John Banks, who had unsurpassed knowledge of Canberra’s high country eucalypts, observed that even if frost did inhibit one tree species, this would merely let others in.37 Further, rarely is a valley floor entirely treeless, but where it is, so are adjacent slopes for a short distance up. This is a classic trap pattern in and out of frost zones. People lived at Nursery Swamp for ‘at least’ 3700 years before 1788,38 and as Hovell saw at Micalong swamp, they burnt hollows: ‘Here we found some little good grass, as the Natives had burned the old grass, some short time previous.’39 ‘Frost hollow’ carries natural explanation an unnaturally long way. ‘Fire hollow’ is a better term.
Some landscape scientists are interested in forest thickening. In 1991 EH Norris, PB Mitchell and DM Hart used three 1870s small area maps plus sources dated 1914, 1938 and 1970 to test claims about changing tree density in the New South Wales Pilliga.40 They gave Aborigines short shrift: ‘No research has been done on the occupation of the forest area by the Kamilaroi aborigines, but the few sites that have been recorded are only short distances from main creeks and it seems likely that they rarely visited the forest core.’41 I recall Blainey’s observation that most Australian goldfields were found near roads, and Aboriginal Law that people never leave any land long unchecked. Norris et al are venturesome in making their claim simply because they failed to find contrary evidence. Yet the claim lets them dismiss the possibility of human intervention in the forest core, so clearing the way for a natural explanation. They make an unproven and improbable assumption to lay down a supposedly objective baseline.
