Colonial censuses and musters
This article was contributed by Beth Wright who is undertaking a PhD on the History of the Australian Census at the Australian National University.

Users are warned that this article contains quotations from historical sources which contain language or views which, reflecting the authors' attitudes or that of the period in which the item was written, may be considered to be inappropriate or offensive today.

Although the first Commonwealth Census occurred in 1911, that was not the first census held on Australian soil. The colonies of Australia had been running censuses for many decades. And even before that, the early Governors had ordered musters of convicts and settlers which filled a similar need.

The Musters

The history of censuses in Australia can be traced back to the arrival of the First Fleet under Governor Arthur Phillip. At that time the British Navy used musters to check on the numbers of people embarking and disembarking from their vessels. This was a method borrowed from the British Army, who had used it to keep track of militias since the middle ages. What with deaths and injuries on sea voyages, sailors absconding and others taken on, musters filled a need to regularly check the number of people actually on each vessel. Given that Governor Phillip was a Naval Captain, it is not surprising that he applied naval methods to keeping track of people within the colony.

In the early days of the colony, mustering was primarily used to keep track of the convicts, who were distributed around the colony working for free settlers or on government projects. This evolved to a point where all non-military settlers were also required to muster. Various types of musters were developed, as mustering was not just about counting people but also about controlling convicts, overseeing the distribution of stores and enforcing discipline. One form of muster was gradually formalised into an opportunity to regularly count the convicts and settlers. Under Lachlan Macquarie’s governorship, from 1809 to 1821, these musters settled to become annual events.

Originally, people were called to a muster by bugle. Later, notices of the annual musters would be placed in public places and in the colonial newspapers. By 1810, the annual musters could take several days and require several muster points. Each year the process of the annual muster became more difficult as the colony grew in size. For many years, Governor Macquarie personally attended each point of the annual muster, even though the process took weeks once the colony grew. This was apparently an attempt to encourage everyone to attend, to increase the accuracy of the counts. Regardless, more and more free settlers realised that they could not be forced to muster and many refused to do so. And as the process of calling a muster changed, it is possible to see a gradual shift towards a census. By the 1820s, musters still continued (and could take over a month) but they were checked via door to door visits which obtained the same details. The last muster in New South Wales was in 1825 under Governor Brisbane.


Governor Lachlan Macquarie
by Richard Read, 1822
State Library of New South Wales PT/144

The musters of convicts and settlers collected basic personal attributes such as sex, convict status, ship arrived on, and year of arrival. Questions such as religion, age groupings, and marital status, were occasionally added. Settlers were also asked in the muster about the livestock they owned and the area of land cleared. As well as a tool for the control of convicts and the distribution of limited resources, musters produced the information which had to be reported back to England in the annual ‘Blue Books’. Gradually, the government of the colony grew to rely on the information collected by the musters, censuses and other statistical collections in their own right. They found it useful for governing and making decisions about the colony.

The First New South Wales censuses

The first New South Wales census was in 1828. The first census was quite controversial in the new colony as it was enforced by an Act of the Government and therefore (unlike the musters) was compulsory with a fine of up to £10 for false answers or for refusing to answer. The census was run by the Colonial Secretary as there were no government statisticians or even births, deaths and marriage registrars in the colony. A district constable, with a clerk to assist him, visited each house, where the householder supplied the information for all the people living in their house (including their servants and convicts). Literate householders completed the form themselves, otherwise the clerk filled in the form with information supplied verbally. No consideration was given to including Aboriginal people in that first census.

In 1828, the census of New South Wales included Norfolk Island and the colony of Moreton Bay. The censuses continued in New South Wales through the 1830s and 1840s. Before it became the separate colony of Victoria, the settlement at Port Phillip was covered by the New South Wales census. The first separate count for Port Phillip was from the 1836 census, when there were only 224 inhabitants. Five years later in 1841, Port Phillip had 11, 783 inhabitants.

The questions included in the early censuses in New South Wales were similar to those used in previous United Kingdom (UK) censuses or to questions used in the earlier musters. Religion was also included, despite it not being included in any previous UK census and in only one muster in the early 1800s. At the time, there were hot political debates about how best to divide the government resources between the various denominations both in terms of land and to enable them to educate the children of the colony. While these were significant reasons for the inclusion of religion in the census, concerns about the growing numbers of Catholics in the colony was also an issue in this period. At this time the religion question was mandatory, but through the second half of the nineteenth century all the colonies in Australia made the question optional. In the 1841 census, houses were counted for the first time, something which was also introduced in 1841 in the UK censuses.

One of the most controversial items in the early New South Wales censuses was convict status as it didn’t just account for present status but also determined if a person had ever been a convict. In a land where past convicts could become land and business owners, and even magistrates, no ex-convict wanted to report his convict status to inquisitive census collectors. Civil condition (i.e. convict status) was last reported in New South Wales in 1851, one year after convict transportation was formally abolished in that colony.

Early censuses in other colonies

Tasmania held its first separate census in 1841. Like the New South Wales census, it included name, age, sex, marriage, convict status, housing, occupation and religion. At first there were plans for an annual census but it was quickly realised that this was not practicable. This timing was revised to triennial but even this was not achievable given the logistics of running a census. The timing was therefore changed to whenever a census was proclaimed. The second Tasmanian census was in 1848 and the third in 1857.

South Australia held its first census in 1844, a decade after the British Government first passed the South Australia Colonisation Act 1834 which declared the lands of South Australia available to purchase for settlement. The questions were similar to the other colonies with the exception of convict status. Since South Australia was not settled with convicts, it had no reason to ask that question. As South Australia's free settlers included some fleeing religious persecution, the religious question was more controversial there than in the other colonies. In fact even though the question was technically voluntary in 1861, it was so controversial that a decision was made not to publish the results. In 1866, the process was much improved by allowing an ‘object’ option to the question.


Western Australia's first census was in 1848 and as in the other colonies questions included sex, age, marital status, housing, occupation and religion as well as livestock and crops. This census was also one of the first to count Aboriginal people. Although it only actually counted 541 who were working for the colonists, it also attempted to estimate (at 1,960) other Aboriginal people living in the immediately surrounding districts. At the time there were 4,622 colonists in total. The second census occurred in 1854, a couple of years after Western Australia become a penal settlement. Census takers were consequently somewhat preoccupied with a new data item - convict status. The convicts had added significantly to the colony and there were by then 11,743 colonists excluding Aboriginal people.

The Port Phillip district separated from New South Wales to become the colony of Victoria in 1851. With significant gold discoveries in Victoria at just this time, the population of the colony expanded rapidly and a census was urgently needed and hastily arranged in 1854. This census was directed by William H Archer, who had recently arrived from England with a background in statistics, a rare commodity in the colonies. Because of the popularity of the goldfields, the population of Victoria continued to rapidly expand and a second census was undertaken a mere three years later in 1857. In those three years, the population of Victoria had increased from 237,000 to 411,000.

Victoria was another state that attempted to count Aboriginal people in the census and the number was lower in each successive census. The Statisticians tended to state their reservations about the accuracy of the counts of Aboriginal people. In 1857, the report on the census noted

The returns shew [the number of Aboriginal people] to be only 1,768, of whom 40 only were to be found in the houses of Europeans, and 43 were living in tents; indicating that these last are employed by Europeans, shepherding or in other rural occupations. The returns, however, are necessarily very imperfect, for they are mainly based on the statements of persons in charge of stations, who gave the number of natives believed to be camped on the runs at the moment; and it is known that at the period of the census some of those who belong to this colony were engaged in an expedition towards the Darling; while others were in the Mallee Scrub, whither it is said to be their custom at that season to repair for hunting the emu and the wild turkey. (Endnote 1)

Later, the Statistician Henry Hayter commented on the fact that the Murray river was the state border but that many Aboriginal people used land on both sides of the river, meaning that they could be in different states at different times.

Coordinating the censuses of the 1860s and 1870s

From the early 1860s, the call to coordinate the censuses between the colonies resulted in meetings and more information flowing between the colonies. Social change also had an impact on the timing and procedures used for censuses in this era, with the gold rushes and colonial self-governments resulting in more censuses in some colonies and new procedures and new questions in others.

The censuses of 1861 were the first attempt at coordinating censuses between the Australian colonies. In fact, the British Registrar-General had requested that the colonies coordinate their censuses in 1851 but this had not been particularly successful. Only South Australia, New South Wales and Tasmania ran their censuses in 1851 and only New South Wales and Tasmania chose the same day. In 1861, all the colonies except Western Australia ran a census on the 7 April 1861, including Queensland which ran its first independent census. Western Australia’s census was held just one week earlier, on 31 March 1861.

In 1871, three colonies: New South Wales, Victoria and South Australia, coordinated their censuses to the same date, while Queensland managed to run one a few months later and Western Australia and Tasmania had run theirs twelve months earlier.

The censuses of 1861 and 1871 were attempts not just to coordinate the date but also the content of the census. Even on new questions, some colonies attempted to coordinate their actions. For example, the Victorian William H. Archer promoted the idea of including questions on illness and infirmity, which South Australia and New South Wales added to their censuses. A question asking each person’s relationship to the head of their household (an important forerunner for the development of family data) was another data item which was added for the first time in 1861 in New South Wales, Victoria, Tasmania and South Australia.



William Henry Archer
Courtesy of University of Melbourne Archives

The censuses of the 1880s and 1890s

Coordination between the colonies in statistics became more significant in the 1880s and 1890s, particularly for censuses. There was some experimentation with census topics and also with publishing the results, led by the new Government Statistician Henry Hayter in Victoria (Archer having moved on in 1874) and later by the New South Wales Statistician Timothy Coghlan.


Mr H. H. Hayter
Courtesy of La Trobe Picture Collection, State Library of Victoria.
IAN01/04/95/3.

All colonies held a census on 1 May 1881. There were few changes between 1871 and 1881 and most colonies asked similar questions. A disastrous event occurred in Sydney with the destruction of the census forms due to a tremendous fire that destroyed the Garden Palace Building where the forms were stored, in September 1882. All that was saved were those census tables that were fortunately at the printers when the fire occurred. As well as the 1881 census forms, records for most earlier New South Wales censuses were burned, together with some early records held by other government departments occupying the large wooden building - a great loss to family historians and other researchers.



The 1891 census was coordinated across the colonies, not only with a common day and an agreed core schedule but also with a common occupation classification for the first time. The classification was drawn up to replace that used since 1854, which had been designed primarily for use in health statistics and had been heavily criticised and amended in the United Kingdom. It was developed by the Tasmanian Statistician Robert Johnston and the New South Wales Statistician Timothy Coghlan.


Sir T. A. Coghlan
National Library of Australia
an22111591-v

Coghlan was appointed in 1886. While the Victorian Statistician Henry Hayter introduced greater reporting of census results, Coghlan made it into an art form, including a detailed and elaborate history of the New South Wales censuses in his 1891 census report. Along with these more detailed accounts, amusing incidents were sometimes reported, giving us a flavour of the difficulties and misunderstandings which occurred in this period.

Three ladies residing in Surry Hills [then considered an inner city slum area], whose language, spelling and character were alike no better than they ought to be, described their occupation in such a way that a singularly obtuse examiner of schedules rendered it as “mining pursuits undefined”. (Endnote 2)

The last colonial censuses in 1901

The last census run separately by each state of the newly formed Commonwealth of Australia was in 1901. At that time there was a clear and urgent need to attempt a fully coordinated census to ensure that the new Commonwealth would have comparable statistics. In particular population counts needed to be comparable in order to create fair and accurate population divisions for voting in federal elections. This led to a meeting of all the state statistical officers (including the New Zealand Statistician) in February-March 1900, one full year before the census, to arrange the coordination.

While the date, the form, the questions and the occupation classification were all standardised, unfortunately the results were not. The difficulty was in variations in tabular presentations such as calculations of groupings and subtle differences in who was included and excluded in the population. For example, while some states chose to exclude Aboriginal people in their tables, others included this population. As this was in the time before computers and even before mechanical tabulation, the tables could not simply be rerun in each state to make them comparable. Further, a complete picture of the nation had to wait until all states had tabulated and published their data and some states were much more timely than others. So despite genuine attempts at coordination, comparable census data eluded the states and hastened the introduction of a commonwealth statistical agency.

Endnotes

1. Victoria. Registrar-General's Office 1857,  Census of Victoria, 1857 : population tables / Registrar-General's Office  John Ferres, Government Printer, Melbourne. p. 7.
2. Coghlan, Timothy A 1894, Census of 1891, Statisticians Report: General Report of the Eleventh Census of New South Wales, Charles Potter Government Printer, Sydney. p 326.
Select Bibliography

ANU, Historical Census and Colonial Data Archive, Australian Social Science Data Archive, Canberra http://hccda.anu.edu.au/ accessed July 2011.

Baxter, Carol J (ed.) 1989, Musters of New South Wales and Norfolk Island 1805-1806, Australian Biographical and Genealogical Record (ABGR) in association with the Society of Australian Genealogists, Sydney.

Baxter, Carol J (ed.) 2000, Musters of New South Wales and Norfolk Island 1823, 1824, 1825, AGBR in association with the Society of Australian Genealogists, Sydney.

Byron, John 1884, New South Wales, Census of 1881, Thomas Richards, Government Printer, Sydney.

Census of the Colony of Queensland taken on 7th April, 1861, np, 1861, Brisbane.

Census of Victoria, 1871, 1874 General Report and Appendices, John Ferres, Government Printer, Melbourne.

Coghlan, Timothy A 1894, Census of 1891, Statisticians Report: General Report of the Eleventh Census of New South Wales, Charles Potter Government Printer, Sydney.

Collins, David 2004, An Account of the English Colony in New South Wales, Vol. 1, Ebook #12565, The Project Gutenberg.

Forster, C and Hazelhurst, C 1988, 'Australian Statisticians and the development of Official Statistics' in Year Book Australia 1988, Cat. no. 1301.0, ABS, Canberra.

Fraser, Malcolm A M (Registrar-General, Government Statistician and Superintendent of Census) 1904, Seventh Census of Western Australia 31 March 1901, Volume 1 – Superintendent’s Report, Government Printer, Perth.

Johnston, Robert M (Registrar-General) 1893, Census of the Colony of Tasmania 1891 – Parts I-VIII with Introductory Report, William Grahame Jnr, Acting Government Printer, Hobart.

Knibbs, George H 1917, Census of the Commonwealth of Australia Volume 1: Statistician’s Report, McCarron, Bird & Co., Melbourne.

Madden, R R 1848, Government Gazette of Western Australia, Perth.

Pike, Douglas 1957, Paradise of Dissent, South Australia 1829-1857, Melbourne University Press, Melbourne.

Sainty, Malcolm R and Johnson, Keith A (eds.) 1980, Census of New South Wales: November 1828, Library of Australian History, North Sydney.

Sholl, Charles 1854, Statistics of Western Australia, Statistical Returns 1848-1854, Perth.

Thumbnail Image acknowledgment

The thumbnail which appears on the table of contents page is a detail from "Melbourne from the eastern end of Collins street" by Robert Russell, 1841, Mitchell Library, State Library of NSW (ML57) which appears in this article.