Learning so little from so much

AS you ponder this year's census form, spare a thought for Matthew Gregson, Australia's first statistician.

A clerk who had pleaded guilty to "feloniously embezzling bills of exchange and other money", and on arriving in 1824 he was promptly employed in the office of the colonial secretary, where his talent with numbers was desperately needed.

His task was to compile the badly overdue Blue Book, an annual summary of the colony's situation demanded by the governor's masters in London.

The Blue Book was a priority because the penal colony of NSW was born statistical. The settlement's establishment had been controversial.

No less an authority than Adam Smith had denounced colonies as a "showy equipage" likely to impoverish the mother country, while Edmund Burke had thundered against the cost of transportation when high and rising public debt meant "frugality and economy" were indispensable. Little wonder the "Heads of a Plan" for the new colony insisted transportation was to be "reciprocally beneficial" to the convicts "and to the state". And little wonder the governor was subjected to a novel reporting requirement in the Blue Books, to which "the most unqualified publicity" would be given and whose design had "been approved by the most rigid economists".

Exhaustively documenting the colony's progress, the Blue Books attempted to reduce the costs of governing across the oceans. But they also matched the spirit of the times, described by John Stuart Mill as an age which to be "rationally assured" felt it "necessary to examine and weigh an immense variety of facts": and nowhere more so than in public affairs.

The six censuses conducted in NSW from 1828 to 1851 reflected that demand. And the advent of responsible government did nothing to quell the statistical urge: indeed, the opposite.

The recently founded colony of Victoria took the lead. Booming gold rush revenues allowed it to recruit an English actuary, W. H. Archer, who from 1859 implemented a statistical system such that "England has nothing so complete, nor has any other country".

Victoria's priority was to attract a growing population by using overseas borrowing to accelerate investment. But foreign lenders, who had no prospect of visiting their debtor, required facts "recorded with correctness and impartiality".

Meeting that need was a key function of Archer's statistical register, which became a recognisable year book in 1861 and attained its modern form in 1874.

Inter-colonial rivalry also played a role. The 1861 census had been co-ordinated across the colonies; the energetic governor of NSW, Hercules Robinson, military engineer and administrative reformer, urged the colonies to use that and other data to compare their progress. As of 1873, the Victorian year book therefore included a section on "Australian statistics", trumpeting Victoria's stellar performance.

It took NSW time to react, but its appointment in 1886 of T. A. Coghlan as "government statist" more than righted the balance. Coghlan's year book, "The wealth and progress of New South Wales", set an international benchmark in economic and social measurement.

Interest in Coghlan's statistics grew as the Australian colonies came to be seen as an experiment in economic and social reform. The Economist had long reported on the colonies' profligate public works and unsound finances. But it was the early industrial relations legislation that proved most controversial. Denounced in Alfred Marshall's Principles of Economics (the bible of the new economics) as "a specious promise", Australia's wages regulation was the subject of the first substantive article in Britain's prestigious Economic Journal.

The crisis of the 1890s and federation only increased the prominence of statistics. While the political fault lines were primarily economic, the race issue, on which Labor campaigned strongly, framed the debates leading to the first commonwealth census in 1911.

George Knibbs, the newly appointed commonwealth statistician, was an ardent eugenicist, who shared the widespread view that "we are guarding the last part of the world in which the higher races can live and increase freely".

But that refuge was threatened by pressure from "lower races" and by the combination of a falling birth rate and degeneration "first, in bodily strength, then, in moral strength" due to urbanisation.

Commonwealth statistics therefore had "to distinguish Chinese and other coloured races, so that it may be possible to separate them from the general population" while examining whether "the social parasites of our civilisation" were "perpetuating their defect".

The premises were abhorrent. But the census's statistical achievement, embodied in the 1300 pages of the 1911 Commonwealth Year Book, was magnificent. A pattern was set: with federation reducing the competitive pressures that had disciplined the colonies, bad policy flourished; but so did an internationally unrivalled output of statistics. No country persistently learned so little from so much.

Entrenched intolerance of critics helped. When Griffith Taylor, the explorer and eminent geographer, questioned the claimed mental inferiority of Asians, he was hounded out of the country for denying the "science" of racial eugenics. Rather, the consensus, expressed by the nation's leading scientists, was that Australia was falling behind Germany, which since 1933 had "introduced measures to improve the eugenic quality of the nation".

But at least sceptics such as Taylor could see the data. Although freedom of information laws were unheard of, there was the Australian tradition of delegating fact-finding to royal commissions and statutory bodies that comprehensively disclosed their reasons. That governments could hide from independent scrutiny the evidentiary basis for the National Broadband Network, or the modelling for the carbon tax, would have seemed profoundly offensive.

All that was to come. And Gregson, struggling to prepare the Blue Books for "the most unqualified publicity" in Westminster's parliament, could hardly have imagined it. But it was not only in statistics that Gregson pioneered. For on obtaining a conditional pardon in record time, and with numeracy skills still scarce as hen's teeth, he resigned his post, only to be lucratively reappointed as a contractor. At least in that respect, the more things change, the more they have stayed the same.