
































Microsoft Word - HOM6_04_OBrien_Deliberate_Confusions.docx


	
   History	
  of	
  Meteorology	
  6	
  (2014)	
   	
  	
  
	
  

	
  

17	
  

	
  

	
  	
  	
  
	
  
	
  
	
  
	
  
	
  
	
  

Deliberate Confusions 

 

Chris O’Brien 

Charles Darwin University, Darwin; 
Australian National University, Canberra, Australia 

 

Climate’s very complexities have been manna from heaven for climate change deniers. 
Reports of global atmospheric warming, the increasing of carbon dioxide in the 
atmosphere and record after record tumbling on continent after continent have met with 
a common refrain from the international chorus of deniers – ‘the climate has always 
changed’. But this is a deliberate confusion. It is myth number one on the Skeptical 
Science web site, expressed in the phlegmatic claim ‘the climate has changed before’.1 
It is a staple of climate change denial. The issue, however, is not change versus stasis; it 
relates instead to magnitude, rates of change and tipping points. Nevertheless, this 
spurious assertion has been effective in confusing the issue of anthropogenic global 
warming, at least in Australia. While a clear majority there understand that human 
activity causes global warming, this confusion has weakened the impetus for change 
and provided an alibi for business as usual from government and many (though not all) 
in the corporate and industrial sectors.2 Confusion has, for the time being, defused this 
issue. This is partly because many people know little about the intricacies of climate but 
know vaguely about climate change on geological time scales. More importantly, as 
Chapter Three of Mike Hulme’s Why We Disagree about Climate Change explains, the 
broader public both understands that disagreement is integral to science yet it does not 
grasp the robustness of the methods, data quality, and conclusions underlying scientific 
consensus.3 So, claims by deniers fall into this template of claim and counterclaim, 
which is widely seen as scientific debate. To some consensus looks more political than 
scientific.  

This paper considers another crucial set of factors that has enabled these ideas to 
make sense in Australasia at this historical moment. Until recently, the inherent 
variability of Australasia’s climate has been invisible in the formal scientific study of 
weather and climate. Yet such variability is an experiential hallmark of life throughout 
Australasia. One only has to live for several years in any part of Australia to experience 
seasons so different from one year to the next that they impress themselves on people’s 
memories. Usually tied to the drought/flood cycles of the El Niño Southern Oscillation 
(ENSO), this variability affects communities, regions and usually leaves visible marks 
on the economy of the entire Commonwealth of Australia. This variability creates 
individual memories; it shapes collective memories and spurs art, literature, and myth.  



	
   O’Brien,	
  Deliberate	
  Confusions	
   18	
  
	
  

	
   	
  

	
  

Poems such as Dorothea Mackellar’s My Country (1908) and paintings like 
W.C. Piguenit’s Flood on the Darling 1890 (1895) long resonated with lived experience 
in a manner that science tried for so long to do but could not. Farmers across Australia 
also understood this quirk of the continent’s climates. Only in 1969, when Jacob 
Bjerknes began to elucidate the interactions between sea and sky across the Pacific and 
Australasia, did it become possible to conceive of a rigorous, scientifically valid 
mechanism to explain what people across Australasia experienced. These ideas did not 
permeate to the broader public until long after the lethal 1982/83 El Niño and its 
monumental droughts and fires in south-eastern Australia. Both the concepts and the 
physical dynamics sparked intense interest among Australian climatologists, but as 
Australian environmental historian Don Garden has frankly noted, even he did not 
grasp the significance of ENSO until the mid-1990s.4 Knowledge about ENSO has 
slowly spread to the broader population since then. Indeed, Garden’s 2009 opus would 
have illuminated many still grappling with this complex atmospheric and oceanic flux.  

Variability has come back into sight with anthropogenic global warming. 
Studying aspects of the Dutch colonial encounter with the atmosphere above Indonesia 
and British efforts at deciphering atmospheric fluxes above northern Australia, this 
paper sketches how variability became invisible. After discussing implications of the 
‘rediscovery of variability’ and its ready, deliberate confusing with global warming, I 
will outline how a thorough understanding of historic variability will help signpost a 
greater variety of consequences of climate change. Far from confusing the issue, 
variability, at least in Australasia, can clarify. 

Seasons and Climates of Northern Australia 

A startling reality about northern Australia is that people in any locale experience 
common weather, but two climates and two seasonal regimes. In the tens of thousands 
of years that Aboriginal people have lived on, tended to and made home the vast lands 
of tropical Australia, they have developed elaborate concepts of climate and seasons. 
They link constellations of events on the land, in the skies and, where appropriate, in 
the waters. These ideas are comprehensive, nuanced and linked to particular places. In 
Indigenous cosmologies north Australia has a multitude of climates and seasonal 
patterns. Through extensive field work in the 1980s cultural anthropologist Deborah 
Bird Rose gleaned that rain is the kernel of Yarralin concepts about season and climate. 
Yipu is the generic term for rain among the Yarralin people of the Victoria River district 
of Australia’s Northern Territory. Living in what western science classifies as a 
monsoonal climate, about 100 kilometres from the coast, they use a set of words to 
distinguish phenomena related to common kinds of rain in the region: hot weather rain, 
cold weather rain, the first rain after extended periods of heat, even the smell of the first 
rains.5 In Indigenous cosmologies seasons are linked to ecological happenings, not the 
calendar. The Yarralin can reliably predict therefore when crocodiles will start to lay 
eggs, and associated weather, not from the time of year, but from observing the arrival 
of March flies.6  

Where rain is at the centre of Yarralin notions of season, the Gidjingarli, along the 
middle of Australia’s northern coastline, define the seasons according to their dominant 
winds.7 Geographer Stephen Davis has identified that Aboriginal people in far northern 
Australia typically recognise six major seasons each year. But the particular concepts of 
each season vary from one locale to another. Gupapuyngu speakers among the Yolngu, 
live along the coast of north eastern parts of the Northern Territory in and around 



	
   History	
  of	
  Meteorology	
  6	
  (2014)	
   	
  	
  
	
  

	
  

19	
  

	
  

Milingimbi. They classify their six seasons as: Dhuludur, the pre-wet season; Barra 
mirri, the growth season; Mayaltha, the flowering season; Midawarr, the fruiting 
season; Dharratharramirri, the early dry and birth time of sharks and sting rays; and 
Rarraandharr, the main dry season.8 Living across Bathurst and Melville Islands, 
ranging from 400 to 600 kilometres west of Milingimbi, the Tiwi experience a very 
similar climate and also recognise six annual seasons. Their seasons however comprise 
three different seasons of Jamutakari, identified by the predominant type of rain of each 
season; two seasons of Yirriwinari, one a part of the dry where cold weather and 
morning fog frequently recur, the other when large-scale burning is undertaken; and 
Tiyari, when humidity builds and thunderstorms begin.9  

Perhaps the most accessible account of Indigenous Australian seasons for 
northern hemisphere readers appears in Ben Orlove’s chapter ‘How People Name 
Seasons’, in his 2003 collection edited with Sarah Strauss, Weather, Climate, Culture. 
Two things strike the reader from his account of the six seasons reckoned by speakers 
of Gundjeihmi, in the western part of coastal Arnhem Land. First, the seasons are not of 
equal duration. Second, seasons are a recurrence of particular sequences in which 
constellations of certain kinds of weather and related ecological phenomena occur.10 
This account, however, evinces a widespread problem in Western discussion of 
Indigenous seasons and the calendars they relate to. Each of the seasons is drawn as 
coming at the same time each year, going at the same time each year and so always 
being of the same duration. Yet indigenous concepts of season allow for differing times 
of arrival and departure, some also include periods of transition between seasons. In 
their critique of Western-produced Indigenous seasonal calendars, linguists Robert 
Hoogenraad and George Jampijinpa Robinson observed that these calendars of 
Aboriginal knowledge typically, and erroneously, used the European month as an 
organising framework. These charts therefore risk forcing one system of knowledge 
into the parameters of another system – in this case traditional Aboriginal knowledge 
into European conceptualisations of time, season, and climate.11 That this has happened 
so commonly attests to the ontological force of western notions of season and weather 
linked to precise calendrical time and in particular, the calendar month.12 

When the British invaded north Australia in the nineteenth century, they imported 
their clockwork climate. Simply put, this is an understanding linked to calendrical time 
and bound up in millennia-old Western European ideas of nature as regular and orderly. 
Although there is ever-increasing documentation of the British learning much about the 
land from Aboriginal people,13 there is no evidence of Indigenous concepts or 
knowledge shaping European ideas about weather and climate in Australia. Even if 
small groups of newcomers are, in the future, found to have taken on Indigenous 
knowledge, this did not shape broader, authoritative, scientific understandings. Across 
northern Australia the parallel epistemologies of Indigenous knowledge and the 
European clockwork-climate have existed in parallel since Europe washed up on its 
shores.  

This clockwork-climate is the Wet/Dry dichotomy that typifies tropical regions in 
western knowledge. Darwin-based weather observer par excellence J.A.G. Little 
encapsulated this when he described (many times) the region’s climate as rhythmic such 
that ‘the changes of these seasons are so uniform that they may be predicted to almost a 
day’.14 Little’s quote not only featured in contemporaneous guides to the Northern 
Territory at the turn of the twentieth century, but also by distinguished geographer 



	
   O’Brien,	
  Deliberate	
  Confusions	
   20	
  
	
  

	
   	
  

	
  

Griffith Taylor in his authoritative work of 1918, The Australian Environment 
(Especially as Controlled by Rainfall). This idea of a wet/dry seasonal dyad regularly 
‘flipping’ in tropical Australia has a long pedigree dating to the late seventeenth century 
and Edmund Halley in British thinking. This concept was reinforced by William 
Dampier’s Discourse on Winds, which from 1699 accompanied the widely read 
narrative of his explorations – A New Voyage around the World. Outlining the winds of 
the Indian Ocean tropics, Dampier gives a strong sense of a mechanical regularity when 
he notes that: 

All shift in the shifting seasons, which are April and September, at one and 
the same time and to their opposite points…these shifting winds in the East 
Indies are called monsoons; one is called the East-Monsoon, the other the 
West-Monsoon.15 

This understanding pervaded eighteenth-and nineteenth-century British sailors’ guides 
and nautical manuals.16 Descriptions of the climate of Darwin and its hinterland in 
almanacs commonly identified the turning from wet to dry as March 31 and the reverse 
as October 1. During the early decades of the twentieth century, April 30 and November 
1 were more frequently identified as the dates the seasons flipped. This may well be an 
implicit acknowledgement of the region’s temporal variability, yet scientists, officials, 
and many European newcomers clung to the idea of a timely clockwork-climate, that 
seasons switched on a particular date.  

Seasons and Climates of the Maritime Continent 

For the Indonesian Maritime Continent, the difference between Indigenous and 
non-Indigenous knowledge is a little less stark. This is not to dispute that there is a 
wealth of place-based Indigenous knowledge across this archipelago of nearly twenty 
thousand islands. But many Indonesian coastal locales traded with each other, as well as 
Arab and Chinese mariners long before the Portuguese arrived. Some 250 language 
groups have been identified throughout the archipelago; Malay was adopted as a lingua 
franca around 1400. We would be remiss not to allow for exchanges and modifications 
of knowledge. With this rich history of encounters it is difficult to untangle long held 
indigenous understandings of people of a particular locale from those influenced by 
Arab, Chinese or indeed indigenous seafarers from other parts of the archipelago. 
Certainly, a trove of knowledge greeted the Portuguese in the early 1500s. By the time 
the Dutch and, to a much lesser extent, the English came, the Dutch already had 
detailed knowledge of the maritime routes and the weather and climate of the Maritime 
Continent. 

John Huyghen van Linschoten was an official in Portuguese Goa for five years in 
the late sixteenth century. Today, he seems somewhat like a late Renaissance/Early 
Modern Edward Snowden or Bradley Manning. Early European traders and later, the 
English and Dutch East India Companies, meticulously recorded details of ocean and 
weather conditions during their voyages. To best their competitors they jealously 
guarded the data they amassed - it was secret. This information could only be accessed 
by those who could use it to the financial betterment of the organisation that had 
accrued these archives. Even by the 1590s, a decade or so before the formation of the 
English and Dutch East India Companies, the Portuguese had acquired invaluable 
knowledge about wind and seasonal patterns across the Indian Ocean and the waters 
embracing the archipelago we now know as Indonesia.  



	
   History	
  of	
  Meteorology	
  6	
  (2014)	
   	
  	
  
	
  

	
  

21	
  

	
  

Over the century since Da Gama’s voyage, Portuguese had learned much about 
the monsoons on an oceanic scale. Before Arab pilots guided da Gama across the Indian 
Ocean, rather than around its shores, European knowledge of the winds and seasons 
was confined to the Indian Ocean littoral. The Portuguese traded out of Malacca (on the 
west coast of the Malay Peninsula) from 1511, and Ambon and Ternate during the 
following decade. During the 1550s the Portuguese presence on Timor, just a few 
hundred kilometres from present day Darwin, became decisive. Steadily and very 
quietly, van Linschoten copied many of the charts the Portuguese had drawn based on 
decades of voyages. He also took copious notes about the different sailing routes. Back 
in Holland he began publishing the four parts of his then famous, if now forgotten, 
Itinerario in 1595. Successive Dutch editions were published in 1596, 1604, 1614, 1623 
and 1644; the first English translation came in 1598, and there were multiple 
translations into French and Latin.17 The secrets were well and truly out. With this 
knowledge, the Dutch bypassed Goa and the contingencies of the monsoons; they went 
to Malacca or Java instead, taking advantage of more consistent wind regimes.  

Itinerario transmitted Portuguese knowledge of Indian Ocean climate and seasons 
– along its shores and over its waters. This became Dutch and European knowledge. 
Indeed, a fascinating question arises as to just how much this was to influence Halley’s 
thinking. Shaped by the needs of maritime travel and trade, Itinerario and its content 
likely provided a framework within which subsequent experience and knowledge has 
been interpreted – one in which seasons are seen as regular and operating to time. Let 
me cite one example of this knowledge. Outlining trade between the Mozambique and 
India, Van Linschoten remarks: 

They sayle from thence to India but once every year in the month of 
August till half September, because that throughout (the whole countries 
of India) they must sail with the Monssoyns, that is with the tides of the 
year, which they name by the windes, which blow certain months in the 
yeare, whereby they make their account to goe and come from the one 
place to the other, and the time that men commonly sayle between 
Mossambique and India is 30 days, little more or less, and then they stay 
in India till the month of April when the winde or Monssoyne commeth 
again for Mossambique.18 

This is a climate working to time. At this point, ‘India’ included what we now call 
Indonesia and so the reference to sailing with Monssoyns throughout the countries of 
India might be among the earliest references of this idea of climate for the Maritime 
Continent.  

John Splinter Stavorinus held the same idea of climate and seasons for the 
Indonesian Archipelago on the cusp of the nineteenth century. Stavorinus’ 
characterisation of the seasons there is conceptually identical. Given the nearly 200 
years’ experience the Dutch had accumulated throughout the Maritime Continent; given 
the volumes of data and personal accounts that had circulated back to the Netherlands, it 
is not unreasonable to consider that Stavorinus reflected prevailing Dutch thinking, as 
well as his own experiences. References to monsoons, good monsoons, bad monsoons, 
and departures timed for particular monsoons, abound in his 1798 opus Voyages to the 
East Indies. British sailors’ guides of the time (Capper, Horsburg etc.) draw on the same 
lexicon and evince an identical conceptual repertoire. In a piece that captures both 
general and local Dutch knowledge he states about the Southern Celebes that: 



	
   O’Brien,	
  Deliberate	
  Confusions	
   22	
  
	
  

	
   	
  

	
  

The seasons, known here, as throughout the east, by the denomination of 
monsoons are the same as at Java; the southeast monsoon being called the 
good and the northwest the bad one, the first always brings (at least in the 
country west of the mountain range) a clear sky and dry weather; the last 
is accompanied with violent winds and continual heavy rain; but to the 
eastward of the ridge, the exact contrary takes place … so that the 
different seasons of summer and winter are felt, at the same time, at no 
more than eight Dutch miles distant from each other.19 

Dutch and indeed, European knowledge was extensive. During the eighteenth century 
intricate interactions between land, sea, and sky came to be understood across vast 
swathes of Indonesia’s seas and coasts. Place came to matter. Expectation was not based 
upon latitude, as would be the case for Australia for at least a century. We see this in 
Stavorinus’ discussion of Amboyna. Latitude would decree a climate and seasonal 
regime identical to that of Java. Topography, however, has bequeathed monsoons that 
‘are exactly contrary here, to what they are along the islands of Java, Borneo, Bali, 
Lombok, Sumbawa, the west coast of the Celebes’.20 An interesting set of questions for 
future investigation arises: did the Dutch discover this themselves? Did this come from 
the Portuguese? Did Europeans learn this from Indigenous mariners of the region? 
Certainly the Dutch had experienced the climate long enough to discern these patterns. 
But this does not preclude knowledge exchange with other mariners across the 
archipelago. 

This idea of a climate with two timely annual seasons is very tidy, perhaps a little 
too tidy. Even the locales recognised as different are a mirror of what predominates 
throughout the region. Conceptually though, these understandings of climate are 
identical – the same framework, the same parameters, the same elements, the same 
inextricable link to the western calendar. This was not just the Dutch, or the British or 
the Portuguese notion of the Maritime Continent’s climate. This was the European idea 
of climate and seasons in the tropics. Symmetry, opposites, dualities, binaries; all of 
these suggest deep, potent cultural influences so taken for granted that they scarcely 
visible.  

Colonial Weather Watching 

 Fortunately, we can test the concept against data. Modern, quality controlled 
meteorological data was routinely gathered at both Batavia (Jakarta) and Darwin from 
the 1860s. Meteorology was a high priority for the colonial invaders of Australia. 
Indeed, as Richard Grove has shown, this was the case throughout India, the Caribbean, 
and the entire ‘colonial periphery’.21 The first observatory in Sydney was established by 
the first of the British colonisers in 1788, just two years after the English East India 
Company established the observatory at Madras (Chennai). A number of observatories 
were established and disbanded in Sydney and other Australian colonies, but by the late 
1850s observatories had been permanently established in Sydney, Melbourne and 
Adelaide. Darwin’s observatory began systematic weather watching in March 1869. 
Operating to British practices, with reliable equipment they set about casting a net of 
observatories, strung together by telegraph wire, across as much of the continent as 
possible. Until Federation in 1901, ‘Australia’ was a collection of separate British 
colonies. Until Australia’s Commonwealth Bureau of Meteorology was established in 
1908, each colony (later, state) had its own network of weather and astronomical 
observatories. This is not to say that standards changed at colonial and state borders and 



	
   History	
  of	
  Meteorology	
  6	
  (2014)	
   	
  	
  
	
  

	
  

23	
  

	
  

that data from one jurisdiction was incompatible with that from others. During three 
vital meetings between 1879 and 1888 the heads of Australia’s major colonial 
observatories set down common standards and practices for the observation of weather, 
the reliable measurement of weather elements, and the organisation and analysis of data. 
Much of this was settled at the First Intercolonial Meteorological Conference in 1879.22 
Participants also agreed to systematic and regular sharing of weather data. The 
establishment of international standards for observing and recording weather in 1876 no 
doubt aided standardisation across the Australian continent and ensured that Australia’s 
then fledgling observatories operated to international, scientific standards.  

From 1863 till 1911 the Northern Territory was administered by South Australia. 
Under the assiduous command of Charles Todd of Adelaide Observatory, Darwin was 
linked to both southern Australia and Java by 1871. Todd was both head of the 
observatory as well the body that created, maintained, and extended its network: Post 
and Telegraphs. It was these telegraphic lines between southern Australia and Java that 
tied Darwin in place. As these networks were being created for local knowledge and the 
colonial purposes of mastering new lands, interest among European scientists about 
larger regional, even global tropical dynamics also grew - an interest that cut across 
empires.  

With the establishment of observatories in Australia and on the Indian 
subcontinent no less a man of science than Alexander von Humboldt had written to the 
government of Netherlands India in 1856 urging the establishment of an observatory at 
Batavia to ‘promote our knowledge of the meteorological phenomena between the 
tropics’ and its use in conjunction with the meteorological and magnetic observatories 
of British India and Australia.23 Records began there in 1864. According to the Badan 
Meteorologi, Klimatologi dan Geofisika (BMKG) – In donesia’s chief agency for the 
observation and study of weather and climate – Batavia’s observatory was not the locus 
of Java’s first systematic weather observations. These began in 1841, when the chief of 
Bogor Hospital started his weather observations there.24 By 1878 there were 55 stations 
recording rain on Java, 22 on Sumatra, 2 on Bangka, 8 on Borneo, 4 on Celebes, 4 on 
Amboyna, 1 on Banda and 1 on Ternate.25 As a first grade observatory, Batavia 
recorded rainfall, wind, temperature, humidity, atmospheric pressure, atmospheric 
tension, magnetic readings, and lunar observations at every hour of the day. 
Observations were recorded accordingly. In Darwin observations were taken and 
recorded at three hourly intervals.  

This region was the locus of the one apparent nineteenth-century challenge to the 
concept of the clockwork climate ticking to the calendar year: the cyclical climate. 
Interest in this partly explains the assiduous interest in regular systematic weather 
observation throughout Australasia and India. So too does an interest in trans-colonial 
atmospheric dynamics. Both are an important aspect of the proliferation of observatories 
there during the mid- and late nineteenth century. When Swiss astronomer Rudolf Wolf 
identified an eleven-year sunspot cycle in 1852, climatic cycles appeared to have gained 
an explanatory mechanism. At the same time, William Stanley Jevons recognised and 
wrote explicitly of Australia’s many variable climates.26 By the 1870s numerous British 
scientists proposed links between sunspot cycles, the frequency of Indian Ocean tropical 
cyclones, the scale and intensity of the Indian summer monsoon, and cycles of drought 
and flood. By then Jevons was arguing that sunspot activity not only shaped Indian and 
Chinese agriculture, but also powered the global business cycle.27 According to his 



	
   O’Brien,	
  Deliberate	
  Confusions	
   24	
  
	
  

	
   	
  

	
  

logic, sunspots not only drove cycles of drought, famine, and flood – they also 
accounted for fluctuations in the global economy. In Australia, colonial meteorologists 
such as Charles Egeson and H.C. Russell recognised climate variability and proposed 
various astronomically-driven weather cycles.28 On reflection, this was more a 
modification of the idea of the clockwork climate than a direct challenge to it. The 
cycles they proposed were exact. Although these concepts of cycle undermined the 
understanding of weather repeating very similar patterns, year after year, the cycles, 
conceptually, still posited a distinct regularity the weather and climate. They were 
orderly. They allowed for prediction. They ultimately even allowed for the distant 
possibility of control.  

Decades earlier, weather watchers had actually observed synchronicity in weather 
cycles across different parts of the globe. Richard Grove has shown that by the 1810s, 
awareness had developed of simultaneous drought in India, the Caribbean, and the then 
embryonic colony of Sydney in 1791.29 Curiosity about a possible mechanism emerged. 
Following the 1877-79 drought in India, the Director of the newly established Indian 
Meteorological Office (IMO), Henry Blanford became interested in the possibility of 
simultaneous, anomalous weather elsewhere. Charles Todd was among the 
meteorologists who responded to Blandford’s request for weather data, especially 
barometric pressure observations. Again drought in India and drought in Australia had 
happened at the same time. As Mike Davis demonstrated in Late Victorian Holocausts, 
this drought was far more widespread and global drought happened again in the late 
1880s and between 1896 and 1903.30 Todd and his colleagues in NSW and Victoria – 
H.C. Russell and R.J. Ellery – had had concluded by 1888 that there is a climatic 
teleconnection between Australia and India, at least in respect to atmospheric pressure 
and cycles of drought and flood. Such thinking was not merely the province of science. 
It found a much wider audience in a very detailed and lucid discussion in the Australian 
press.  

Often referred to, an 1888 article in the Australasian with Todd, Russell and 
Ellery is most revealing. This was a detailed delineation of their understanding of 
drought cycles and synchronicity with other parts of the world. Here, Australia’s top 
three weathermen endeavoured to explain Australia’s climate, its variability, and 
relationship to the broader region. Moreover, they were doing so to the broader public. 
Written during the drought of the late 1880s, all three men explained Australian drought 
in terms of larger, more regional, even global dynamics. Although Ellery did not discuss 
India, a section of his contribution dealt with ‘drought cycles’.31 Under this very heading 
he dismissed the idea that droughts had an identifiable periodicity, but conceded a 
relationship between the earth's atmosphere and sunspot activity, which is cyclical. 
Ellery also averred that ‘heat at the equatorial regions’ influenced atmospheric fluxes 
above Australia.32 Russell likewise linked droughts in Australia to the failure of the 
monsoon over northern Australia.33 Later he spoke of how the development of ‘the 
meteorological system of Australasia’ will help elucidate questions about drought in 
Australia.34 Like Henry Blanford in India, Russell concludes that the relationship 
between ‘a continuous high barometer’ and ‘very little rain’ is strong in Australia.35 
Although he avowedly disputes links between drought and Wolf’s 11-year sunspot 
cycle, Russell reiterates his own proposal of a 19-year climate cycle in Australia, first 
circulated in 1876. Accordingly, he implicitly acknowledges Australia’s climate as 
inherently variable.  



	
   History	
  of	
  Meteorology	
  6	
  (2014)	
   	
  	
  
	
  

	
  

25	
  

	
  

Todd likewise discusses the periodicity of droughts.36 He, however, draws no firm 
conclusions. His contribution is distinguished by the attention he gives to the 
synchronicity of droughts in Australia and India – indeed one whole section of Todd’s 
piece appears under the title ‘Australian and Indian weather’. Todd’s conclusion about 
this teleconnection is worth quoting in full: 

Comparing our records with those of India I find a close correspondence or 
similarity of seasons with regard to the prevalence of drought, and there can 
be little or no doubt that severe droughts occur as a rule simultaneously over 
the two countries. The most remarkable instance of this was the disastrous 
drought of 1876, the year of the great Indian famine. This drought, it is said, 
has been traced from 30 deg. south in Australia to 60 deg.  north, over 90 
degrees of latitude and 100 degrees of longitude.37 

This is meteorology writ large. It goes beyond colonial borders. It far transcends 
the shores of the Australian continent. Todd’s scope encompasses not just the broader 
region but much of the tropical and temperate girdles of the world. Moreover, Todd’s 
global vision was being transmitted to the broader Australian public who otherwise 
would have no reason to conceive of simultaneous patterns of drought between 
Australia, India and far flung places such as Mauritius. Colonial meteorological 
networks were far from merely physical. This did not only apply to networks within the 
British Empire. Todd used weather data from Banjoewangie (far eastern Java) and 
Jakarta to place the Australian drought in context. For the former he noted that the 
north-west monsoon had been late in 1888, resulting in a 10 inch (250mm) deficit in 
rainfall for the calendar year. Similarly Jakarta’s 1888 rainfall was 14 inches or 350mm 
below the annual average.38 Humboldt’s vision – to some extent - was being fulfilled. 

Ordering Colonial Meteorology 

 This affords a great opportunity. With the region’s seasons defined as Wet and 
Dry, in accordance with the different monsoons, and presence of this reliable and self-
consciously systematic network, we can use rainfall data to test the concept against 
reality. Average rainfall figures for Darwin tell a story that fits the concept almost 
perfectly: annual drought, followed by linear increase to deluge, then a linear decline to 
drought, crossing thresholds of wet and dry in between (see Fig. 4.1). As Fig. 4.2 
illustrates, data for Batavia fits this pattern, albeit not as ideally. My work on Darwin is 
more advanced hence the disparity of the data sets; Java research is in progress. 
However, this data set for Batavia is lengthy enough for the purposes of this paper. 
Although I am using weather data and climate statistics, my analysis is historical, rather 
than statistical. Moreover, this was what people interested in Indonesia’s climate at the 
time had to think with. Unlike Darwin, Batavia does not experience annual cycles of 
drought. The middle of the year is drier than the rest, but rain – sometimes heavy – 
usually still falls.39 Between 1864 and 1882 the mean number of rain days between July 
and September was under eight; for January and February, it was over 20. So these 
‘Dry’ months would have seemed much drier, relative to ‘Wet’ months. 



	
   O’Brien,	
  Deliberate	
  Confusions	
   26	
  
	
  

	
   	
  

	
  

 

Fig. 4.1. Mean Monthly Rainfall, Darwin Post Office, 1869 – 1942. 

 

 

Fig. 4.2. Mean Monthly Rainfall, Batavia, 1864 – 1882. 

Means, crucially, were the grammar of weather narratives at the time. This has 
continued in Australasia until relatively recently, and continues among quasi-scientific 
lay approaches to weather and climate. The average is the essential value. The 
organising of the Batavia data in the earliest publications of the Observatory indicates 
just how intently this was so. Volume VI of Batavia Observatory’s observations, 
published in 1885, contains upwards of 180 tables of enumerated data.40 The 
overwhelming majority of these are tables of means of various weather phenomena for 
each month, for times of day, for times of each month. To give a sense of just how 
meticulous the observing was, and how diligently data were arranged, allow me to cite a 
number of the less obvious titles: Table 60 shows the ‘Relative Humidity of the Air and 
Mean Change from Hour to Hour’; Table 71 the ‘Mean Degree of Relative Humidity, 

0 

50 

100 

150 

200 

250 

300 

350 

400 

450 

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec 

R
ai

nf
al

l (
m

m
) 

0 

50 

100 

150 

200 

250 

300 

350 

400 

450 

Jan Feb Mar Apr May Jun Jul Aug Sept Oct Nov Dec 

R
ai

nf
al

l (
m

m
) 



	
   History	
  of	
  Meteorology	
  6	
  (2014)	
   	
  	
  
	
  

	
  

27	
  

	
  

Corresponding to 16 Points of the Compass’; Table 50 reflects the lunar preoccupations 
of the period – ‘Mean Temperature of the 24 Hours on 8 different Moon Phases’. 
Similarly, data were organised and distributed according to monthly means of different 
weather phenomena. Means then and since, have been the kernel of weather data and the 
discerning of climates. By the time meteorology emerged as a distinct discipline, 
averages and distributions of numerical values had acquired particular meanings. As 
Stephen Stigler has noted, with the work of Belgian astronomer Quetelet in the 1820s 
and 30s, the average came to be widely understood as normal. Above and below values 
were deviations from the norm – variations but not signifiers of variability.41 As weather 
and seasons were understood to repeat each year, the issue of variability between years 
and patterns on decadal rather than yearly scales made little sense. Variation was read as 
deviation, abnormality, and freak weather. Averages, as philosopher Ian Hacking has 
concluded, were not just values but a powerful episteme.42 For a long time they were a 
principal instrument of how science understood the atmosphere. 

As meteorology became more systematic, more statistical and more numerical, 
variability became invisible in Australasian science. The cycles proposed for both global 
and local weather collapsed. They failed because reality did not bare them out. With 
meteorology becoming more numerical, oceans of data from across the continent and 
region were collected. Basic statistical calculations helped order this immensity and 
provided knowledge that ministered to many queries and needs. Even the tele-
connections identified by Sir Gilbert Walker, the conceptual forerunner of ENSO, 
became peripheral. Science had become blind to climatic variability in Australasia. 

Histories from Numbers 

Why does this matter? Consider these rainfall graphs from Darwin and Batavia 
(Figures 4.3 to 4.10).43  

0

100

200

300

400

500

600

700

800

18
70

18
73

18
76

18
79

18
82

18
85

18
88

18
91

18
94

18
97

19
00

19
03

19
06

19
09

19
12

19
15

19
18

19
21

19
24

19
27

19
30

19
33

19
36

19
39

19
42

Ra
in

fa
ll 

(m
m

)

Fig. 4.3. January Rainfall, Darwin Post Office, 1869 – 1941. 

 



	
   O’Brien,	
  Deliberate	
  Confusions	
   28	
  
	
  

	
   	
  

	
  

 

Fig. 4.4. January Rainfall, Batavia, 1864 – 1882. 

 

0

100

200

300

400

500

600

700

18
69

18
72

18
75

18
78

18
81

18
84

18
87

18
90

18
93

18
96

18
99

19
02

19
05

19
08

19
11

19
14

19
17

19
20

19
23

19
26

19
29

19
32

19
35

19
38

19
41

Ra
in

fa
ll 

(m
m

)

Fig. 4.5. April Rainfall, Darwin Post Office, 1869 – 1941. 

0 

100 

200 

300 

400 

500 

600 

700 

800 

900 

1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 

R
ai

nf
al

l (
m

m
) 



	
   History	
  of	
  Meteorology	
  6	
  (2014)	
   	
  	
  
	
  

	
  

29	
  

	
  

 

Fig. 4.6. April Rainfall, Batavia, 1864 – 1882. 

Fig. 4.7. July Rainfall, Darwin, 1869 – 1941. 

 

0 

50 

100 

150 

200 

250 

300 

350 

400 

450 

1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 

R
ai

nf
al

l (
m

m
) 

0

10

20

30

40

50

60

70

18
69

18
72

18
75

18
78

18
81

18
84

18
87

18
90

18
93

18
96

18
99

19
02

19
05

19
08

19
11

19
14

19
17

19
20

19
23

19
26

19
29

19
32

19
35

19
38

19
41

Ra
in

fa
ll 

(m
m

)



	
   O’Brien,	
  Deliberate	
  Confusions	
   30	
  
	
  

	
   	
  

	
  

 

Fig. 4.8. July Rainfall, Batavia, 1864 – 1882. 

 

0

50

100

150

200

250

18
69

18
72

18
75

18
78

18
81

18
84

18
87

18
90

18
93

18
96

18
99

19
02

19
05

19
08

19
11

19
14

19
17

19
20

19
23

19
26

19
29

19
32

19
35

19
38

19
41

Ra
in

fa
ll 

(m
m

)

Fig. 4.9. October Rainfall, Darwin Post Office, 1869 – 1941. 

 

0 

50 

100 

150 

200 

250 

1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 

R
ai

nf
al

l (
m

m
) 



	
   History	
  of	
  Meteorology	
  6	
  (2014)	
   	
  	
  
	
  

	
  

31	
  

	
  

 

Fig. 4.10. October Rainfall Batavia, 1864 – 1882. 

 Historically, rainfall has been enormously variable from one year to another in 
both locales. The range in volumes for January, April and October in Darwin and for all 
months in Batavia is staggering – so much so that it is obvious that in both centres, 
Aprils and Octobers can be in one season one year; another the next. Using 50mm as a 
threshold between wet and dry, there were 23 dry Aprils and 39 dry Octobers in Darwin 
between 1869 and 1942. Or, about one in three Aprils and three in five Octobers 
qualified as dry. Between 1864 and 1882 in Batavia three, or one in six Aprils were dry 
and five, or one in three Octobers were dry. The April range during these respective 
periods was 1.3mm (1897) to 603mm (1891) for Darwin and 22mm (1872) to 406mm 
(1882) for Batavia. Octobers ranged from 0.0mm (1891, 1896, 1926 and 1941) to 
213.4mm (1914) in Darwin and from 0mm (1877) to 238mm (1867) in Batavia. Even 
deep into ‘The Wet’ both centres recorded large variations in historical monthly rainfall: 
Januaries in Darwin between 1869 and 1942 ranged from 68mm in 1908 to 711.3mm in 
1895. During the driest January from 1864 to 1882 – 1878 – 163mm was recorded in 
Batavia; 822mm fell during the wettest January of this period (1872). But it is around 
the middle of the year when differences between the two are most stark. Most 
obviously, Darwin’s driest months on average are June, July and August and Batavia’s 
are July, August and September. More importantly, drought does occur around Darwin 
in about 95% of years lasting between 4 to 6weeks on both sides of the southern Winter 
Solstice. Only twice during the 73 years from 1869 did at least one of the three driest 
months record over 50mm of rain in Darwin. In contrast, during the 18 years from 1864, 
this only failed to happen on one single occasion in Batavia.  

Averages cannot show this, only histories of various weather elements. As shown 
above, such histories draw out different patterns between places with seemingly similar 
climates. Of course, variability is not just an issue of volumes or quantities: timing is 
vital too. Daily records for Darwin from 1870 to 1942 demonstrate a climate at odds 
with the received understanding of two six-month periods of Wet and Dry each year. 
Twice rain fell each calendar month. On another six occasions, it fell during 11 calendar 
months of its respective year.44 Yet in 1896 no rain was recorded from 24 April until 16 
November; 1925 likewise saw one of the only other six- month periods without rain. In 
1925 rain was light, occasional and sporadic until December. The year 1926 seems to 

0 

50 

100 

150 

200 

250 

1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 

R
ai

nf
al

l (
m

m
) 



	
   O’Brien,	
  Deliberate	
  Confusions	
   32	
  
	
  

	
   	
  

	
  

have experienced two “dries”, one from 10 May till 10 September, the other from late 
September till early November. Moreover, daily records show that in many years 
rainfall does not increase and decrease in the steady, linear fashion that monthly means 
suggest.  

My research on Batavia has not advanced to analysis of daily weather data, yet. 
However, the monthly data set I have been using here yields worthwhile information 
about temporal patterns of rain.45 Between 1864 and 1882 there was just one year, the 
infamous global drought year of 1877, when no rain fell for at least three months. 1881 
was the only other year when rain was not recorded in a calendar month and in 1875 two 
months recorded just 1mm each. Only three of these nineteen years experienced three 
consecutive months with less 50mm of rain. Although July and August had by far the 
two lowest monthly rainfall means there were still five years when over 100mm fell in 
either July or August. Similarly, there were six years in which at least 25mm was 
recorded each month and two on which over 50mm were recorded. A sample of 19 
years is too small from which to extrapolate definite cycles. But it is enough to show 
that historical reality is significantly at variance with the dominant concepts of weather, 
climate, and season. Continuing research on Batavia’s weather history should give better 
shape to the nature of these differences and to the intricacies of the region’s atmospheric 
fluxes. In the early 1980s climatologists such as Neville Nicholls and John McBride 
began studying variability in Australasia. Until then a bizarre situation had developed 
such that data spoke of regularity, while the evolution of suggested variability. Now, 
amidst debates about climate change variability and other complexities, Australia’s 
climates are yet to be understood – and climate change can and will change the very 
variability of variable climates. 

Concluding Thoughts 

Variability re-emerged in Australasia as the world has been grappling with 
anthropogenic climate change. For a time this enabled the deliberate confusing of the 
two. Sorting out the confusions, however, is an opportunity for remarkable clarification. 
One cause of confusion is the imported understandings of the colonisers. New concepts 
of seasons for different locales across the region can be identified through study of 
historical data. For Darwin, this, unsurprisingly, would largely match the nuanced 
concepts and patterns discerned through the millennia long empirical observations of the 
locale’s Aboriginal peoples. With a better understanding of seasons and their turning we 
have many more signifiers of climate change: are seasons earlier, later, of different 
lengths? Are some seasons becoming more frequent or less frequent on decadal scales? 
Have some seasons acquired a different quality? We not only learn more about climate 
change but about the patterns that constitute climate itself. Data about quantities such as 
rain volume, temperature and humidity are vital in identifying both change and 
variability; but both happen across time, on various timescales. Indeed knowing about 
long term mechanisms of variability such as ENSO we can now, in the long term, look 
to see whether global warming is changing the dynamics, scale and timing of this 
mechanism.  

 Common to understanding both Australasia’s climate cycles and global 
warming is the need for this rigorous, empirically based reconceptualising of seasons. 
Our received ideas of season fail to account for much of what we experience. Also, 
climate change will alter the nature and timing of these seasons anyway. This re-think 
demands considering seasons as not merely anchored to annual cycles of Earth’s tilting 



	
   History	
  of	
  Meteorology	
  6	
  (2014)	
   	
  	
  
	
  

	
  

33	
  

	
  

on its axis, but also to the decadal scale fluxes in the great ocean of air we call our 
atmosphere. Finally in the West, we need to uncouple the seasons from calendar. or at 
the very least, integrate events in the physical world into our reckoning. In some places 
circumstances might even demand that we acknowledge that some seasons do not occur 
every year and, conversely, that some years do not include every season. 

The diligence of observers in the past has given us a treasure trove of 
meteorological material. With this we can learn so much more about weather, climate, 
and the relationships between them. Other cultures offer other ways of knowing about 
the dynamics of the atmosphere and, sometimes, their relationships to various 
ecological events. It is up to us to ask new and revealing questions to combat the 
confusions others would foist upon us for personal or political gain. 

 

	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  
References 
1 See home page ‘Skeptical Science’, https://www.skepticalscience.com, accessed: 17/2/2014. 
2 Opinion Poll, 21/1/2014, Essential Research, 
http://essentialvision.com.au/category/essentialreport/page/3, accessed: 17/2/2014. 
3 Mike Hulme, Why We Disagree about Climate Change, (Cambridge: Cambridge University 
Press, 2009), pp. 72-109. 
4 Don Garden, Droughts, Floods and Cyclones: El Niños that Shaped our Colonial Past, 
(Melbourne: Australian Scholarly Publishing, 2009), p. xiii. 
5 Deborah Bird Rose, ‘When the Rainbow Walks’, in Eric K. Webb ed., Windows on 
Meteorology, (Melbourne: CSIRO Publishing, 1997), p. 3. 
6 ibid., p. 5. 
7 Rhys Jones and Betty Meehan, ‘Balmarrk wana: big winds of Arnhem Land’, in Webb (ed.), 
op. cit., p. 15. 
8 Stephen Davis, ‘Documenting an Aboriginal Seasonal Calendar’, in Webb (ed.), ibid., p. 31. 
9 ibid., p32. 
10 Ben Orlove, ‘How People Name Seasons’, in Sarah Strauss and Ben Orlove (eds), Weather, 
Climate, Culture, (Oxford: Berg, 2003), pp. 130-31. 
11 Robert Hoogenraad and George Jampijinpa Robertson, ‘Seasonal Calendars from Central 
Australia’, in Webb (ed.), op. cit., pp. 34-35. 
12 For a detailed history of this relationship in western thinking, see author’s chapter, ‘Imported 
Understandings’, in James Beattie, Emily O’Gorman and Matthew Henry (eds), Climate, 
Science and Colonization: Histories from Australia and New Zealand, (New York: Palgrave, 
2014), pp. 195-212. 
13 The most recent comprehensive treatment of this appears in Bill Gammage, The Biggest 
Estate on Earth, (Sydney: Allen and Unwin, 2011). 
14 J.A.G. Little, 1902, quoted in Griffith Taylor, The Australian Environment (Especially as 
Controlled by Rainfall), (Melbourne: Government Printer, 1918), p. 70. 
15 William Dampier, John Masefield ed., Dampier’s Voyages: Consisting of a New Voyage 
around the World, Two Voyages to Campeachy, a Discourse of Winds, and a Vindication, in 
Answer to the Chimerical Relation of William Furnell, (London: E. Grant Richards, 1906), pp. 
246-47. 
16 I detail this in my essay a ‘Cultural History of the Monsoon’, currently under review for 
publication. 
17 Charles McKew Parr, Jan van Linschoten: the Dutch Marco Polo (New York: Thomas Y. 
Cromwell Company, 1964), p. xviii. 
18 John Huyghen van Linschoten, The Voyage of John Huyghen van Linscoten to the East 
Indies, Volume 1, ed. Arthur Coke Burnell, (London: Hakluyt Society, 1885), p. 33. 



	
   O’Brien,	
  Deliberate	
  Confusions	
   34	
  
	
  

	
   	
  

	
  

	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  
19 John Splinter Stavorinus, Voyages to the East Indies, Vol II, (London: G.G. and J. Robinson, 
1798), p. 179. Dutch mile – called an hour is about 3.5 English miles. 
20 ibid., p. 321. 
21 Richard Grove, ‘The East India Company, The Raj and El Niño, in Richard H Grove, Vinita 
Damodaran and Satpal Sangwan (eds), Nature and the Orient: The Environmental History of 
South and Southeast Asia, (Delhi: Oxford University Press, 1998) pp. 301-303.  
22 Minutes of the Intercolonial Meteorological Conference held at Sydney on the 11th, 13th, 14th 
November 1879, (Sydney: Government Printer, 1880). NB: Freestanding copies of these 
minutes can be found in the Bureau of Meteorology’s Libraries. Originally, these would have 
been published in the Parliamentary Papers of the Colonies of New South Wales, Victoria, 
South Australia for 1879. 
23 P.A. Bergsma, Observations made at the Magnetical and Meteorological Observatory at 
Batavia, Vol II, (Batavia: Government Printing Office, 1878), v. 
24 BMKG Indonesia, History of Meteorological, Climatological and Geophysical Agency, 
http://www.bmkg.go.id/bmkg_pusat/Profile/History.bmkg, accessed: 4/2/2014. 
25 Bergsma, op. cit., vii. 
26 William Stanley Jevons, Waugh and Cox’s Australian Almanac 1855, (Sydney: Waugh and 
Cox, 1855). 
27 Mike Davis, Late Victorian Holocausts, (London: Verso, 2001) p. 222. 
28 Charles Egeson, Egeson’s Weather System of Sun-spot Causality, Being Original Researches 
in Solar and Terrestrial Meteorology, (Sydney: Turner & Henderson, 1889); H.C. Russell, ‘On 
the Periodicity of Good and Bad Seasons’, Town and Country Journal, 13 June 1896, p28. 
29 Grove, op. cit, pp. 302-304. 
30 Davis, op. cit. 
31 R.J. Ellery, The Australasian, 29/12/1888, p. 1455. 
32 ibid., p. 1456. 
33 H.C Russell, The Australasian, 29/12/1888, p. 1456. 
34 ibid. 
35 ibid. 
36 Charles Todd, ibid. 
37 ibid. 
38 ibid. 
39 Observations made at the Magnetical and Meteorological Observatory at Batavia, Volume 
VI, 1885, (Batavia: Government Printer, 1885), p. 280. 
40 Observations made…at Batavia, Vol VI, op. cit. 
41 Stephen Stigler, Statistics on the Table, (Cambridge, MA: Harvard University Press), pp. 53-
63.  
42 Ian Hacking, Taming of Chance, (Cambridge: Cambridge University Press, 1990), pp. 105-
106. 
43 These graphs have been constructed using Meteorological Observations at Darwin Post 
Office from 1869 to 1942 taken by Adelaide Observatory to 1908 and the Commonwealth 
Bureau of Meteorology (Australia) from 1908 and from Observations at Batavia Observatory, 
1864-1882, op. cit. 
44 Adelaide Observatory and Commonwealth Bureau of Meteorology, Daily Rainfall, Darwin 
Post Office, 1869-1942. Unless otherwise cited, all data in this paragraph comes from this 
source. 
45 All data here comes from Observations made at…Batavia, VI, op. cit., p. 280. 


