Australian Institute of Criminology

Skip to content

Centenary Article - Crime in twentieth century Australia

Dr Adam Graycar

Canberra: Australian Institute of Criminology, January 2001

This article was originally published:

  • Graycar, A 2001, 'Crime in twentieth century Australia', in Australian Bureau of Statistics 2001, Year Book Australia, 2001, ABS cat. no. 1301.0, pp. 477-495.

Adam Graycar has been Director of the Australian Institute of Criminology (AIC) since November 1994. He has had long experience in policy making, management and research and research management at the most senior levels in Australia. He has worked as an academic at Flinders University and the University of New South Wales and has been head of various agencies within the Government of South Australia, including appointment as Australia's first Commissioner for the Ageing (1985-1990). He is the author of numerous books and articles in criminology and social policy, and has two doctorates from the University of NSW. He is a Fellow of the Academy of Social Sciences in Australia, a Fellow of the Australian Institute of Management, and is Adjunct Professor of Social Policy, University of Queensland and University of Hong Kong.


Australia was a less violent society at the end of the twentieth century than it was at the end of the nineteenth or eighteenth centuries. Violence and theft have been part of human interaction for all of recorded history, but it is not easy to compare statistics on the extent of violence and theft across the span of the twentieth century.

Crimes that have existed across the span of the century such as homicide, burglary, assault, theft, robbery, rape and kidnapping have seen variations in definition in legislation and varying levels of attention from police and courts. Crimes such as motor vehicle theft were inconsequential at the beginning of the century, but significant at the end of the century, while crimes that involve child abuse or domestic violence were more likely to be regarded as private matters at the start of the century, yet at the end were firmly within the criminal justice sphere.

Crimes such as superannuation fraud, health insurance fraud, theft of telecommunications services, electronic vandalism and varieties of computer hacking, credit card fraud, Internet child pornography, electronic funds transfer crime and electronic money laundering were not on the criminal horizon at the start of the century. However, nude or even topless bathing, or homosexual acts between consenting adults, brought criminal sanctions, while public drunkenness comprised more than half of all offences brought before the Magistrates' courts in the early years of the twentieth century, and this persisted until the middle of the century.

While we can focus on significant differences, some similar threads run right through the century. Alcohol-related crime was a predominant cause of criminal justice involvement in 1900, while today it is substance abuse in general, but alcohol still remains a major component in criminal activity. As noted later, many of our prison population in 1900 were incarcerated principally as a result of their mental state (although many of them would have been counted under petty crime or vagrancy), and in 2000 this situation has changed very little. In 1900 young males contributed significantly to criminal activities, and at the end of the century this continues to be the case.

What mattered 100 years ago and what happens today are very different. One hundred years ago there was great concern about drunkenness, gambling and 'Chinese opium dens', whereas today concerns such as cyber crime, the international trafficking of drugs and their consequences in Australia, domestic burglary and (family) violence against women are prominent in crime discussions.

At the beginning of the century Australia's Aboriginal population was not counted in official statistics, nor did Aboriginal people feature in crime statistics. A lot of Indigenous justice was 'extra-legal' (i.e. administrative, informal or traditional) and, except for the most serious matters, did not find its way into police or court statistics. At the end of the century Indigenous people have high rates of involvement in criminal justice both as victims and offenders, but there is no way of comparing these rates with offender and victimisation rates 100 years ago.

Over the century the criminal justice system has become much more complex. In 1900 people confronting the system faced one of three types of sanctions: absolute or conditional discharges, fines, or imprisonment, while today there is a multiplicity of sentence types and diversion processes. In addition the structure of the system has seen a more cooperative underpinning. In 1900 all criminal justice service delivery was in the hands of the States, and there was hardly any inter-jurisdictional law enforcement or administration. Today there is a significant and growing federal presence in law enforcement (e.g. Australian Federal Police, National Crime Authority, Australian Customs Service) and in some aspects of justice administration (e.g. legal aid). There is substantial cooperation between States and Territories, evidenced by the establishment of CrimTrac and the formation of six National Common Police Services. Furthermore transnational and cyber crime is emerging as one of the most significant issues for this century, but its activities are not constrained by the borders that were set in place at the beginning of Federation.

As the twentieth century opened, the Year Books were describing falls in crime. Four categories only were reported: offences against the person, those against property, drunkenness and 'other offences' (table C8.1). Three things stand out. First, there are enormous variations around the States. In 1900, for example, property offence charges against persons arrested or summonsed per 1,000 population were: New South Wales, 4.93; Victoria, 2.97; Queensland, 5.21; South Australia, 1.60; Western Australia 9.86; and Tasmania 3.91. This level of variation applied across the offence range. The second feature is the steady decline in violent and property crime. The third is that charges of drunkenness in 1900 were three times as high as charges of property crime and five times as high as charges of offences against the person.

C8.1 CRIME IN AUSTRALIA - 1890 to 1902
YearOffence against a person Offence against property Drunkenness Other offences
Number Rate per 1,000 Number Rate per 1,000 Number Rate per 1,000 Number Rate per 1,000
1890 16,907 5.43 16,795 5.39 48,201 15.48 86,064 27.64
1895 10,450 2.98 14,773 4.22 38,895 11.11 76,982 21.99
1899 10,469 2.85 15,426 4.20 46,983 12.78 80,071 21.78
1900 10,319 2.75 15,764 4.21 52,286 13.96 87,147 23.26
1901 9,975 2.62 15,273 4.01 54,412 14.30 88,031 23.13
1902 9,597 2.49 16,565 4.29 50,618 13.10 87,137 22.57
Source: Victorian Year Book, 1903, pp. 304-305.

These figures bear no resemblance to figures produced by the Australian Bureau of Statistics at the end of the century (table C8.2). The only similarity with the 1903 data is that there is still significant variation around the jurisdictions (see graphs C8.3 to C8.9 below).

C8.2 CRIME IN AUSTRALIA - 1993 to 1999
Year Robbery Assault UEWI(a) Other theft
Number Rate per 100,000 Number Rate per 100,000
1993 12,765 72.25 n.a. n.a. 381,783 2,160.98 n.a. n.a.
1994 13,967 78.23 n.a. n.a. 379,505 2,125.51 n.a. n.a.
1995 14,558 80.56 101,267 560.36 384,908 2,129.89 490,084 2,711.88
1996 16,372 89.41 114,156 623.44 402,079 2,195.87 521,762 2,849.49
1997 21,305 114.96 124,500 671.80 421,569 2,274.79 530,881 2,864.63
1998 23,801 126.95 130,903 698.23 434,376 2,316.93 563,482 3,005.57
1999 22,590 119.11 133,602 704.45 415,600 2,191.37 610,276 3,217.85
(a) Unlawful entry with intent.
Source: Year Book Australia (1301.0), years shown.

The decline in crime was noted in the first Commonwealth Year Book, yet no adequate explanations were found. The Year Book commented:

"Causes of Decrease in Crime - The statistics given show that there has been a considerable decrease in crime throughout Australasia...The deterrent effect of punishment, in respect of many offences, notably drunkenness, vagrancy, petty larcenies, etc., appears to be almost negligible. In general, punishment has declined in brutality and severity, and has improved in respect of being based to a greater extent upon a scientific [penological] system, though in this latter respect there is yet much to be desired…Part of the improvement may no doubt be referred also to the general amelioration in social condition that has taken place during the last fifty years"
(Commonwealth Year Book No.1, 1908, pp. 760-761).

Reporting on the 1999 Recorded Crime Statistics, the Australian Bureau of Statistics (Media release 82/2000, 28 June 2000) noted that robbery was down for the first time in seven years, and the report itself showed declines in most crimes: homicide (including attempted murder); sexual assault; robbery; blackmail; burglary; and motor vehicle theft. There was, however, an increase in assault and 'other theft'.

The first Commonwealth Year Book lamented the fact that Australia's statistical collections were not uniform. It noted "without uniformity there is no safety in statistics" (Commonwealth Year Book 1908, p. 7). It also noted that "comparisons are valueless unless the data compared are of the same type, it by no means infrequently happens that aggregates are formed from, or comparisons are made with, dissimilar data" (Commonwealth Year Book 1908, p. 7).

This warning is still made at the end of the century, and it has only been since 1993 that Australia has had crime statistics that can in any way be described as uniform. Even so, they cover nine sets of offences, and are confined to reports to police.

Few of the items reported at the beginning of the century can be reported on in similar terms at the end of the century. The first Commonwealth Year Book has tables of persons charged before magistrates 1901-1906, together with numbers of convictions and committals. These tables cannot meaningfully be replicated today. About 40% of the cases are for drunkenness. Different offences are now on the statute books, and some terminological changes have taken place.

In 1988 the Australian Institute of Criminology published, as a Bicentennial project, a substantial volume entitled Source Book of Australian Criminal and Social Statistics, 1804-1988. Section 6 on Cases Processed at Magistrates' Courts covers 260 pages of tables and goes back to earliest times, but the tables all end in the 1970s. The book notes that most jurisdictions "have introduced substantial changes in the processing and classification of charges which have resulted in irreconcilable discontinuities in the data system" (p. 228).

Other statements in early Year Books cause much pondering. For example the 1903 Victorian Year Book claimed "of the offenders who are reported as having committed offences, generally about 50 per cent are arrested, 38 per cent are summoned, whilst about 11 per cent are still at large at the end of March of the year following that in which the offence was reported" (Victorian Year Book 1903, p. 299). We cannot replicate data such as this in 2000 for we do not know th, p. 299). We cannot replicate data such as this in 2000 for we do not know the ingredients behind the statement, but what we do know at the end of the century is that clear-up rates are not nearly as high as reported in 1903.

Writing in 1999 two authoritative analysts state "the majority of reported property crimes do not result in a convicted offender being charged" Accurate data on charge rates is generally not available, but most estimates for property crime charge rates are no higher than 10 per cent" (Freiberg and Ross 1999, p. 52).

In 1993 the Australian Bureau of Statistics published the first national collection of uniform crime statistics. Prior to that there were collections of crime statistics in each State and Territory. These were not good indicators of actual levels of crime, as what was included and excluded reflected changing interests and priorities over time, as well as inconsistent recording methods.

Homicide, states and territories - 1993-1999

Assault, states and territories - 1995-1999

Sexual assault, states and territories - 1993-1999

Robbery, states and territories - 1993-1999

Enlawful entry with intent, states and territories - 1993-1999

Other theft, states and territories - 1995-1999

Motor vehicle theft - 1993-1999

Graphs C8.3 to C8.9 give us one picture only - that of crimes recorded by police. Other types of information about crime tell us about victims of crime, about offenders (those who commit crime), and about the police, courts, and correctional systems. Records over time give us little information about victims (though the ABS has conducted four national victim surveys); offender information comes from prison records (though only a very small proportion of offenders go to prison); police statistics found in annual reports reflect police priorities and judgements about where to place resources; court data gives us some indication of the nature and volume of offences deemed serious enough to come this far into the criminal justice process.

When examining our statistics it must be remembered that not every crime is reported to the police, not every crime that is reported is recorded, not every crime that is recorded is investigated, not every crime that is investigated is cleared ('solved'), not every crime that is investigated yields a suspect, not every suspect is apprehended, not every apprehended person is charged, not every charged person is brought before the courts, not every person brought before the courts is convicted, and not every convicted person is imprisoned. (It has been estimated by Satyanshu Mukherjee that in New South Wales in 1996 "one offender was sentenced to a term of imprisonment for every 156 crimes of break and enter, motor vehicle theft, assault or robbery" (Mukherjee 1999, p. 75).

So then, taking a snapshot at any time, or taking a series of data over time on any activity (e.g. court appearances, prisoner numbers, etc.) does not tell us about crime in Australia, but rather about that particular activity or segment.

There are, however, a number of interesting features which characterise crime over the century, and they are described below. At the beginning of the century executions took place in our prisons (there were 55 executions in the first decade of the twentieth century). The last execution in Australia was in 1967.


Homicide covers offences of murder and manslaughter. It is one offence for which data are generally and consistently available. Nearly all homicides are reported and police always respond to homicide. Homicide has been part of human behaviour and criminal codes for a long, long time.

In 1915 in Australia, the homicide rate was 1.8 per 100,000 population. In 1998 it was 1.6 per 100,000. During the intervening years it hit a low of 0.8 in 1941 and a high of 2.4 in 1988.

The annual homicide rates in Australia from 1915 to 1998 are displayed in graph C8.10 below. There was a long-term decline during the first half of the twentieth century, with the homicide rate being the lowest during the period of World War II (1939-1945). The rate then increased substantially to a plateau of about 1.5 per 100,000 population in the 1950s and 1960s. An upward trend occurred during the 1970s, reaching the level of around 2.0 per 100,000 population at the end of that decade. Since then, the rate has remained relatively stable, except for two temporary fluctuations in the 1980s. One of those temporary fluctuations resulted in the highest homicide rate recorded in Australia (rate of 2.4 per 100,000 population in 1988). This is more than double the rate observed in 1950.

Trends in homicide, Australia - 1915-1998

While there has been remarkable stability in the twentieth century, and while the differences among the States are negligible, this has not always been the case. During the first hundred years of settlement, homicide rates were much higher and only fell to 'modern' levels after about 1880. Today there is little difference in homicide rates between any of the Australian States and Territories (except the Northern Territory), but in 1900 Queensland and Western Australia had much higher rates (by a factor of 3 to 10 times) than the older States.

Australia's current rate is similar to that of many northern European countries and Canada and New Zealand, but double that of Japan, Norway and Ireland. The USA has a rate about four times that of Australia, and countries like Russia, Mexico and Croatia have rates much higher than the USA. Why this is so is a real mystery to criminologists, anthropologists, psychiatrists and psychologists. What is known, however, is that homicide was a much less frequently occurring activity in the twentieth century than in previous centuries.

For the last decade of the century the Australian Institute of Criminology has monitored every homicide in Australia. The dataset shows that homicide in Australia was characterised by the following features.

There were 3,150 homicide incidents over the decade, averaging 315 per year, a figure that did not fluctuate much.

Just under two-thirds of all homicide incidents (60.2%) occurred in residential premises. Nearly half of all homicide incidents occurred on Friday, Saturday, or Sunday, and over two-thirds of homicide incidents occurred between 6pm and 6am.

Eight out of 10 homicide incidents can be characterised as 'one-on-one' interactions between the victim and the offender, though there have been, on average, 15 multiple fatality incidents per year, resulting in approximately 39 victims per year.

There were 3,386 victims of homicide. Across the 10-year period, rates of victimisation have remained relatively constant, fluctuating between 1.7 and 2.0 per 100,000 population. Some 63.2% of victims were male and 36.8% were female. Females were killed at an average annual rate of 1.4 per 100,000 population, whereas males were killed at an average annual rate of 2.4 per 100,000 population. There has been a stable pattern of gender differentiation, with a ratio of 3 males killed for every 2 females. Male victims were more likely to have been single at the time of the incident, whereas female victims were more likely to have been married or living in a de facto relationship. Females were more likely to be killed as a result of a domestic altercation, although this proportion has declined in recent years. Males were more likely to be killed following an alcohol-related argument.

Victims of homicide were more likely to be killed with a knife or other sharp instrument than any other weapon. There was a declining trend in the proportion of victims killed with a firearm, with an average of 81 victims killed per year with a firearm.

The highest age-specific victimisation rate for females was for children less than one year of age (average rate of 2.6), whereas the highest victimisation rate for males was for young men between the ages of 24 and 26 years (average rate of 4.3). Indigenous persons were on average 8.1 times more likely to be victims of homicide than non-Indigenous persons.

Approximately 9% of all homicide victims were aged under 15, and this proportion has remained quite stable each year since 1989. Biological parents, usually the mother, were responsible for a majority of child killings in Australia. Very rarely are children killed by a stranger.

There were 3,481 offenders of homicide - 87.2% were male and 12.8% were female. Males consistently exhibited higher rates of offending than females, with a ratio of about 7:1. The median age of male offenders was 27 years and the median age for female offenders was 29 years. Male offenders were more likely to be single, whereas female offenders were more likely to be married or living in a de facto relationship at the time of the incident.

Between 1996-97 and 1998-99, just under 2 out of 5 male offenders and just over 1 out of 5 female offenders were under the influence of alcohol at the time of the incident. Approximately 6% of homicide offenders in Australia committed suicide during or following the homicide incident.

Eight out of 10 homicides occurred between people who were known to one another. Females were more likely to be killed by an intimate partner, whereas males were more likely to be killed by a friend or acquaintance, but under 2 out of 10 homicides occurred between strangers. Approximately 13% of all homicide incidents occur in the course of other crime, such as robbery and sexual assault. One in 10 homicide incidents occurred in the course of robbery, and 3.7% occurred in the course of sexual assault. This relatively low rate of homicides committed in the course of another crime is a factor which differentiates Australia's homicide rates from those of many other countries.

While only 13% of homicides were committed by females, women who kill tend to kill men. Women are more likely to kill (in descending order of frequency) husbands, ex-husbands, de facto partners, and lovers, followed by children and other relatives. Very few women kill strangers.

In Australia, between 1 July 1989 and 30 June 1999 there were 13 mass-murder incidents (where the number of victims was 4 or more), resulting in the death of 94 persons, though in the two most recent years of the century Australia recorded no mass-murder incidents.

Understanding homicide involves some fundamental neurological and sociological risk factors. Looking across many nations, from a policy point of view things like expanding the number of police, giving them better technology, setting longer prison sentences, imposing or abolishing the death penalty have had no effect on the homicide rate, which has remained fairly constant in most countries (Mouzos 2000).

Prisons and prisoners

In 1900 there were 4,755 prisoners in Australia, with a rate then of 126.28 per 100,000 population (table C8.11). On 1 December 1999 there were 20,455 people in Australia's prisons. This equates to a rate of 107.85 per 100,000 total population. Throughout the century the imprisonment rates were lower than they were at the beginning of the century and at the end (graph 8.12).

C8.11 PRISONERS, Daily Average Number and Rates per 100,000 Population - 1900 to 1999
Year Prisoners (no.) Total population (no.) Rate per 100,000 total population
1900 4,755 3,765,339 126.28
1905 4,459 4,032,977 110.56
1910 3,603 4,425,083 81.42
1915 3,777 4,969,457 76.00
1920 2,793 5,411,297 51.61
1925 3,295 6,003,027 54.89
1930 4,620 6,500,751 71.07
1935 3,767 6,755,662 55.76
1940 3,578 7,077,586 50.55
1945 4,271 7,430,197 57.48
1950 4,338 8,307,481 52.22
1955 5,787 9,311,825 62.15
1960 7,446 10,391,920 71.65
1965 8,243 11,505,408 71.64
1970 10,208 12,663,469 80.61
1975 8,683 13,968,881 62.16
1980 9,752 14,807,370 65.86
1985 10,504 15,900,566 66.06
1990 13,280 17,169,768 77.35
1995 15,973 18,187,697 87.82
1998 19,002 18,851,187 100.80
1999 20,455 18,966,788 107.85
Source: Australian Demographic Statistics, December 1998 (3101.0); Source Book of Australian Criminal and Social Statistics 1804-1988; Corrective Services, Australia (4512.0).

The number of people in prison at any time is a fraction of the flow of people through Australia's prisons. While the prison population stood at 20,455 on 1 December 1999, there were 10,038 sentenced offenders entering prison receptions in the December quarter 1999. In crude terms (because we do not know how many prisoners completed terms of less than three months during the quarter) about half of the prisoners in December 1999 were not there in September 1999. In 1999, 11% of prisoners were serving terms of less than six months (compared with 8.5% of prisoners a decade before). Table C8.13 summarises the number of prisoners in Australia's prisons in 1999 by length of sentence.

Prisoner population

C8.13 PRISONERS IN AUSTRALIA, Sentence Length - 1999
Aggregate sentence no. %
Under 1 month 136 0.8
1 to less than 3 months 502 2.9
1 to less than 6 months 1,258 7.3
Less than 6 months 1,896 11.0
6 to less than 12 months 2,008 11.7
1 to less than 2 years 2,531 14.7
2 to less than 5 years 4,037 23.5
5 to less than 10 years 3,768 22.0
10 years and over 2,100 12.2
Life and other indeterminate 826 4.8
All sentenced prisoners excluding periodic detainees 17,166 100.0
Source: Australian Prisoners 1989; Prisoners in Australia 1999.

During 1998-99 there were, in addition to the prison population, 55,253 people (average daily number) serving community correction orders. Almost half (27,177) were serving supervision orders (probation or parole), 10,347 were performing community service, 15,892 were serving fine options and 496 were in home detention (SCRCSSP 2000, p. 775).

Prisons involve substantial financial outlays. The value of prison assets in Australia is $2.008b, the average assets per prisoner being $109,005 (SCRCSSP 2000, p. 786). Recurrent costs in secure prisons across Australia average $158.95 per prisoner per day or approximately $58,000 per year, while community corrections average out at $6.45 per offender per day. Data at this level of detail are not readily available for the early part of the century.

Far more people move through the prison system in a year than are indicated in annual prison holding statistics. In essence, the 'flow' is much larger than the 'stock'. Many people have very short sentences. For example, more than a third have sentences of under a year. As a result the daily average number of prisoners cannot be calculated from the annual number of prisoners.

In 1903 the daily average number of prisoners in Victoria was 907 males and 141 females (table C8.14). In that year there was accommodation for 1,845 males and 533 females, so the male prisons were operating at half capacity while the female prisons were operating at one quarter capacity. At the end of the century there is much more efficient use of assets! The Victorian Year Book of 1903 proudly described the decrease in crime over the previous decade (p. 302).

While the daily average for males was 907, the total number of males received into the prisons in 1903 was 5,952, while for females receptions totalled 1,490 for a daily average of 141. In the years before 1903 the prison numbers in Victoria showed a steady decline, so that the number in 1903 was 26% less than in 1895.

C8.14 PRISONERS IN CONFINEMENT, Victoria - 1895 to 1903
Year Males Females Total
1895 1,208 216 1,424
1896 1,143 194 1,347
1897 1,044 182 1,226
1898 1,065 223 1,288
1899 1,020 215 1,235
1900 981 204 1,185
1901 951 200 1,151
1902 943 170 1,113
1903 907 141 1,048
Source: Victorian Year Book 1903, p. 314.
Year Males Females Total
1998 2,253 117 2,370
1999 2,371 143 2,514
2000 2,593 154 2,747
Source: Corrective Services, Australia, March Quarters (4512.0).

By the end of the twentieth century the average daily prison numbers for Victoria were 2,593 males and 154 females in secure custody (table C8.15), and a further 334 people in open prison custody. Over the century the population of Victoria increased from 1,196,213 to 4,741,468 - a four-fold increase, while the male prison population increased by a factor of 2.7, and the female prison population remained constant.

There had been an Australian boom in prison buildings from 1850 to 1880. There was, in the last decade of the 20th century, another prison building boom, and these new prisons replaced the 100-year-old prisons which housed more than half of all prisoners until the late 1980s.

There are significant variations around Australia in the rate of imprisonment. As can be seen from table C8.16, the Northern Territory stands out, while Western Australia and Queensland are well above the national average. South Australia and Tasmania are well below the national average, while Victoria and the ACT have by far the lowest rates.

Jurisdiction Rate per 100,000
New South Wales 147.4
Victoria 84.2
Queensland 190.5
South Australia 114.8
Western Australia 208.3
Tasmania 105.4
Northern Territory 445.8
Australian Capital Territory 75.7
Source: Adapted from ABS 4512.0, March 2000 p. 7.

There is considerable debate about what it is that causes prison populations to rise and fall. There are no clear explanations for variations between jurisdictions. Freiberg and Ross (1999, p. 53) note that studies which attempt to correlate crime rates with imprisonment rates have proved to be contradictory and confusing, though Victoria and Tasmania with relatively low crime rates also have relatively low imprisonment rates. Queensland, as the graphs above show, is a relatively low crime State, yet its imprisonment rate is very high.

In 1997 the State with the highest crime rate (Western Australia) had 1.7 times as much crime as the lowest ranked State (Victoria), yet the jurisdiction with the highest imprisonment rate (Northern Territory) had six times that of the lowest (Tasmania) (Freiberg and Ross 1999, pp. 61-62).

Other explanations for variations in prison rates can be found in demography-the prison population is overwhelmingly males in early adulthood; masculinity rates-the higher the rate the higher the imprisonment; the size of Indigenous populations-enormous over-representation, but insufficient to explain inter-jurisdictional differences; and prison capacity- a bed built is a bed filled, though as we saw above, in 1903 Victoria was operating at less than half capacity.

Official statistics at the beginning of the century did not deal with Aboriginality nor ethnicity.

Of the 20,455 prisoners in December 1999, 3,916 were Indigenous people (19% of the Australian prison population). The rate of imprisonment for Indigenous persons was 1,727 per 100,000 adult Indigenous population. The ABS (4512.0, December Quarter 1999, p. 5) reports that at the end of the century the Indigenous rate of imprisonment was 15 times the non-Indigenous rate.

In 1987 a Royal Commission into Aboriginal Deaths in Custody was established. This Royal Commission found that the over-representation in prisons and in their contacts with the criminal justice system was a significant factor in Aboriginal deaths in custody.

The Australian Institute of Criminology monitors all deaths in prison and police custody. From 1980 to 1999 there were 846 deaths in Australian prisons. Of these 393 (or 46.4%) were suicides. On average there have been 42.3 deaths per year over the two decades. Of the 846 deaths, 132 were of Indigenous people-15.6% of the total. Some 44% of Indigenous deaths were by suicide compared with 47% of non-Indigenous deaths (Dalton 1999; Dalton 2000).

One cause of death in prison which no longer occurs is that of execution by the state. In 1900 there were 10 executions (seven of which were in Western Australia) and in 1901 there were eight (five in Queensland and three in New South Wales). The numbers declined to between two and six for every year until 1923-the first year in over 100 years in which there were no executions. There were between one and three executions in 26 of the next 44 years (until abolition of the death penalty in 1967).

The 10 and 8 executions in 1900 and 1901 were down considerably from the peak years of 1826 to 1830 when the annual numbers of executions were 75, 59, 39, 19 and 80 (and this was in New South Wales and Tasmania only!).

While the ABS has conducted an annual census of prisoners since 1994 (and the AIC for several years before then) we know little more about prisoners than their demographic characteristics and their offences. Of the many immigrant groups in Australia nearly all have rates of imprisonment lower than that of the Australian-born population. In 1998 there were four groups that had imprisonment rates higher than those of Australian-born: those born in New Zealand and Oceania; Turkey; Viet Nam; and Lebanon (Carcach and Grant 2000).

Prison numbers have increased sharply in Australia in the 1990s, and one line of argument is that changes to our mental health system, and in particular the move to deinstitutionalisation, have put out onto the streets many who would otherwise not be at risk of minor offending and victimisation.

The conditions in prisons today are significantly better than a century ago. In the early 1890s in New South Wales those serving sentences of up to 14 days were to receive only bread and water during their term of imprisonment. Offenders serving sentences from 14 days to six months were to alternate on a weekly basis between bread and water and normal rations, while those serving sentences in excess of six months would be kept on bread and water for the final 14 days of their terms.

In addition to these dietary restrictions, offenders serving sentences of less than six months were required to sleep on a plank bed without a mattress. In 1894, with the Government's authorisation, the then Comptroller-General of Prisons reinstituted the gag in order to check the use of what he termed "blasphemous and frequently vile and filthy language". Intensive criticism saw the Government withdraw this authority in the following year (Grabosky 1977, p. 99).

A new Comptroller-General, Fredrick W. Neitenstein, was appointed in 1896 and his concern was largely with reform and rehabilitation rather than punishment, deterrence and retribution. The new century began with Neitenstein lamenting the fact that his prisons were warehouses for the cast-offs of, people with substance abuse problems, mental illness, intellectual impairment and low levels of education and employment. A century later, these characteristics describe many in Australia's prisons today.


Much of the writing on Australian history of the 18th and 19th centuries focuses on drunkenness. In Peter Grabosky's seminal history of criminal behaviour in Sydney, there are the same number of index entries for drunkenness as there are for theft, rape, robbery, murder, homosexual behaviour, and domestic violence put together (Grabosky 1977).

The first Year Book of the Commonwealth of Australia (covering the years from 1901 to 1907) had several tables on drunkenness. About 40% of all convictions in Magistrates' Courts in the first years of the century were for drunkenness. From 1901 to 1906 the number of convictions fell from 57,212 to 45,843-though convictions as a proportion of cases remained constant at around 90%. It can be seen that a great deal of police and court resources at the beginning of the century were devoted to drunkenness. So too was a great deal of statistical endeavour. The Year Books of the various States all devoted several pages to drunkenness. The first Commonwealth Year Book noted:

"The convictions for drunkenness taken by themselves are not an altogether satisfactory test of the relative sobriety of the inhabitants of each State, forasmuch as several important factors must be taken into consideration. The age constitution, for example, of the people is by no means identical in each State, Western Australia having by far the largest proportion of adult males. The avocations of the people affect the result, since persons engaged in strenuous callings are, on the whole, more likely to indulge in alcoholic stimulants than those employed in the less arduous ones. The distribution of the population is also a factor, it being obvious that the likelihood of arrest or summons for drunkenness is greater in the more densely populated regions, and lastly, so also is the attitude of the police and public generally in regard to the offence" (Commonwealth Year Book 1908, p. 762).

It also pondered on the treatment of drunkenness as a crime:

Though the problem of the correct method of dealing with dipsomania is by no means an easy one, it seems fairly clear that the present plan of bringing offenders before magistrates, and subjecting them to the penalty of imprisonment or fine, has little deterrent effect, as the same offenders are constantly reappearing before the courts. Further, the casting of an inebriate into prison, and placing him in his weakened mental state in the company of professional malefactors, doubtless tends to swell the ranks of criminals and certainly tends to lower his self-respect, and examination of the prison records in New South Wales some years ago disclosed the fact that over 40% of the gaol population had commenced their criminal career with a charge of drunkenness…. With regard to drunkards, however, Captain Neitenstein, the Comptroller of Prisoners in New South Wales, advocates the entire abandonment of the system of repeated fine or imprisonment in favour of a course of hospital treatment" (Commonwealth Year Book 1908, p. 762).

Concern with drunkenness did not abate, and the 1951 Year Book carried tables showing numbers of cases of drunkenness, numbers of convictions, and convictions per 10,000 population. The plaintive warning of the first Year Book (above) seems to have fallen on deaf ears, as it was repeated almost word-for-word half a century later in the 1951 Year Book (and presumably annually in the intervening Year Books!).

Towards the end of the century public drunkenness has been decriminalised, and this allows police to devote time to other priorities. This is not to say that drunkenness is not a problem. The context has changed.

In the first half of the century drunkenness was seen as a crime in itself. Today it is seen as a precipitating factor. Significant numbers of those charged with assault and other crimes against the person are affected by alcohol. At the end of the century there was concern with the drinking habits of young people, and binge drinking in particular. In 2000 the Australian Institute of Criminology completed a monograph entitled Alcohol, Young Persons and Violence in which topics addressed included: An indicator approach to the measurement of alcohol-related violence; Alcohol and disorder in the Australian community; Alcohol use and violent behaviour among youth: Alcohol and homicide in Australia; Reducing alcohol-related harm in and around licensed premises; and many others (Williams 2000).

At the end of the 20th century, substance abuse was a matter of considerable public policy concern. Although people were no longer arrested and processed for drunkenness, there was strong evidence of illicit drug use related to the commission of property and violence crime. The Australian Institute of Criminology conducts a national project entitled Drug Use Monitoring Australia (DUMA) in which people detained by police are urine tested and a drug history taken (not by police, but by researchers). Alcohol is not tested for in this project.

The first DUMA report analysing data collected in 1999 showed that of those charged with violent crime 70% tested positive to any drug, 58% positive to cannabis, 18% to opiates, and 12% to amphetamines. The figures were higher for those charged with property offences. Of these detainees, 86% tested positive to any drug, 66% positive to cannabis, 43% tested positive to opiates, and 13% to amphetamines. The rate of testing positive to opiates was as high as 78% in one Sydney police lockup (Makkai 2000).

This project demonstrates the link between illicit drug usage, especially opiates, and property crime at the end of the century.


Two features stand out when comparing police at the beginning and end of the century. Of Australia's 43,048 sworn police officers in 1999, 7,089 or 16.5% were women (table C8.17). At the beginning of the century none were women. While women are not well represented at senior levels, between one-quarter and one-third of constables, probationary constables, recruits and cadets are women.

C8.17 POLICE SERVICES, Composition - At 30 June 1999
Rank Males (no.) Females (no.) Persons (no.) Females (%)
Senior Executive(a) 81 2 83 2.4
Chief Superintendent(b) 65 2 67 3.0
Superintendent 392 17 409 4.2
Chief Inspector 131 2 133 1.5
Inspector 843 28 871 3.2
Senior Sergeant 1,724 73 1,797 4.1
Sergeant(c) 7,064 472 5,536 6.3
Senior Constable 15,063 2,691 17,754 15.2
Constable(d) 8,896 2,992 11,888 25.2
Probationary Constable(e) 899 485 1,384 35.0
Recruits 406 178 584 30.5
Cadets 57 31 88 35.2
Police Aides(f) 150 79 229 34.5
ACPO/Special Constable(g) 43 4 47 8.5
Other(h) 145 33 178 18.5
Total sworn officers 35,959 7,089 43,048 16.5
(a) Includes Commissioner/Chief Commissioner, Deputy Commissioner, Assistant Commissioner, Commander. (b) Includes Commanders in AFP and Northern Territory. (c) Includes Sergeant First Class in Western Australia, Sergeant qualified and unqualified in Tasmania. (d) Includes Constable First Class in New South Wales and Western Australia, Confirmed Constable in Victoria, Constable qualified and unqualified in Tasmania; Probationary Constable, Constable, Constable First Class and Senior Constable in the Northern Territory. (e) Includes Constable not confirmed in Victoria. (f) Includes Police Auxilliary in the Northern Territory. (g) Includes Special Constable unpaid in Western Australia and Aboriginal Community Police Officers (ACPO) in the Northern Territory. (h) Victoria includes reservists and Protective Security Officers (PSO).
Source: The Composition of Australia's Police Services - 1999.
C8.18 POLICE OFFICERS, Number and Rate per 100,000 Population, Australia - 1900 and 1999
Jurisdiction no. Rate per 100,000 population
New South Wales 2,107 154.8
Victoria 1,480 123.7
Queensland 880 178.2
Western Australia 519 288.4
South Australia 363 100.2
Tasmania 250 144.6
Northern Territory .. ..
Australian Capital Territory .. ..
Australia 5,599 148.4
New South Wales 13,471 210.1
Victoria 9,777 207.5
Queensland 7,519 214.1
Western Australia 4,850 260.6
South Australia 3,592 240.6
Tasmania 1,076 228.8
Northern Territory 876 454.2
Australian Capital Territory 663 213.8
Australia(c) 43,048 227.0
(a) From Source Book of Australian Criminal and Social Statistics 1804-1988. (b) Rates calculated using (ABS) Population by Age and Sex, Australian States and Territories, June 1999 (3201.0). (c) Includes AFP.
Source: Source Book of Australian Criminal and Social Statistics 1804-1988; Population by Age and Sex, Australian States and Territories, June 1999 (3201.0).

The second feature is that at the end of the century there were 94,676 private police (security guards and others)-twice as many as there were public police in Australia (Prenzler and Sarre 1998).

As table C8.18 shows, the proportion of public police per 100,000 population has risen by 53% over the century (from 148.4 to 227.0). There were large rises in Victoria, South Australia and Tasmania, but interestingly the rate fell in Western Australia. Modern communications and transport now mean that fewer police can cover a larger area.

A simple description of police is that they are mainly a "body of people patrolling public places in blue uniforms with a broad mandate of crime control, order maintenance and some negotiable social service functions. They are supplemented by non-uniformed adjuncts concerned primarily with the investigation and processing of criminal offences and sundry administrative tasks" (Maguire, Reiner and Morgan 1994, p. 716).

At the end of the century one of the important features in police recruitment was education. A significant proportion of new recruits are university graduates. At the beginning of the century the major qualification to become a police officer was whether one was physically big enough. At the end of the century the issue was whether one was smart enough. A lot of police work involves sitting in front of a screen because police are information dependent-their value lies in how they obtain, process, encode, decode and use information. A lot of this comes from the application of sophisticated technology. The whole structure of our information technology requires an analytical mind, and the ability to learn fast. Crime solving and dealing with evidence such as DNA typing, biochemical assay, automated database fingerprint matching, accident reconstruction, arson analysis, weapons innovation, not to mention non-coercive persuasive techniques, such as mediation, hostage negotiation and rape counselling, all requires skills that need to be upgraded regularly.

In addition to these highly sophisticated skills, many police work with complex human relations issues. Unlike many professionals, police do not work in surroundings they control. Many work where the rest of us would not want to go: dark back alleys, domestic situations that have got out of hand, and venues that are physically and socially uninviting. They work in the seamy side of life, with sleaze, lust, perversion, greed, rage and malice-not the sort of thing professionals would choose as their working context.

Young men and women, mostly under 35 years of age, continually deal with life situations that require either a very strong education in the social and behavioural sciences, or long years of life experience. The increased numbers of private security and law enforcement officers is a demonstration of the complexity of policing and the willingness of certain parts of the community to pay an additional premium for a level of protection they deem appropriate for their interests. Private law enforcement covers not only the armed guards who escort payroll or those who monitor entertainment venues, but also the many who deal with issues of computer security, hackers, cyber crime and a range of sophisticated investigators.

In 1993 Australia's Police Commissioners laid out a vision for the end of the century:

"By the year 2000", they said, "we will have a safer and more secure community. The focus will be on

  • A partnership approach to policing
  • An increase in community confidence in police
  • A reduction in the incidence, effects and fear of crime
  • An increased level of community safety".

Who knows what vision they will set for the end of this century?


While fraud is an old crime, technological, social, demographic and economic developments have brought about changes in the form fraud takes and how it is perpetrated. In simple terms, fraud involves the use of dishonest or deceitful conduct in order to obtain some unjust advantage over another (Smith and Grabosky 1998).

Fraud is often regarded as a modern day crime, but it has been around for a very long time. Plotting the trend of the extent of fraud over this century is not a useful exercise, as many frauds are not recorded, and any attempt to suggest that the number of frauds per 100,000 population has risen or fallen is based on shaky data.

Some attribute the beginnings of fraud to the Industrial Revolution, but there seems to be little reason for thinking that it was not around before then, as any sort of trade or dealings between people seems to offer the opportunity to commit fraud.

In the early days of colonial Australia, the shortage of currency and the resultant use of promissory notes, sometimes thousands of pounds worth, gave opportunity for fraud. Other simpler cases include one Emily Syster, who was a 'begging letter imposter' some time in the early 1840s. She wrote to various prominent figures, including the Bishop, under different guises-as an old officer, a pregnant woman, a ruined tradesman. The fraud was eventually detected through her handwriting. Other examples include false insurance claims, such as claiming for the theft of four watches when in fact only two were stolen (Smith and Grabosky 1998).

In the early days of settlement in Australia, fraud was facilitated by communication methods as it is today-only in those days it was its slowness rather than the electronic rapidity of today. When one Benjamin Boyd borrowed money in England ostensibly to establish the Royal Bank of Australia in 1839, it was very difficult for his creditors to keep track of how their money was being used. Boyd must have been quite a persuasive person as he was permitted by the other directors of the Royal Bank to set off for Australia with the money that had been invested in the bank, and was granted a loan for his own use. Very little of the money was in fact used in banking operations. Boyd spent most of the money on establishing a shipping business, investing in whaling works, the setting up of a wool washing dam on Sydney Harbour and the building of Boyd Town on the southern coast of New South Wales. Apart from being able to conceal his business failures through false reports, Boyd's fraudulent activities were assisted by the length of time it took for communication to take place between one side of the world and the other (Sykes 1988).

By the end of the twentieth century, expanding computer literacy has increased the number of prospective fraud offenders, while new technologies allow easier and cheaper access to a much larger pool of prospective victims. At the end of the century, in addition to old fashioned scams we have seen evidence of Identity-related Fraud, Internet Fraud, Credit Card Fraud, Advance Fee Fraud, and Nursing Home Fraud, to mention just a few types.

These few types provide an indication of the range and complexity of dishonesty offences that come to the attention of the police in Australia. There are, of course, many others that are either not discovered or not reported. These cases also show that fraud is invariably complex, and not possible to be described in detail here.

However, many fraud types involve an international element, and rely upon modern technologies for their commission. The proliferation of electronic funds transfer systems has enhanced the risk that such transactions will be intercepted and funds diverted. Most of the large scale electronic funds transfer frauds which have been committed have involved the interception or alteration of electronic data messages transmitted from bank computers.

It would be reasonable to say that most of the fraud types that existed at the beginning of the century were still with us at the end, but new technologies, and processes of globalisation, have vastly expanded the range of frauds that exist at the end of the century.

Restorative justice

At the beginning of the century most infractions were dealt with by the courts, and the data in early Year Books document this in very great detail. By the end of the century our court statistics were not very useful, and throughout Australia processes were in place to divert many offenders (particularly young offenders) from the courts to other forms of adjudication. This will have a significant impact on future court statistics.

Courts in many jusrisictions are overloaded, and the theory being tested is that diversion is an opportunity for a second chance for many young people for whom a court appearance and the subsequent consequences may be counterproductive.

The process of restorative justice involves bringing together victims and offenders, and others who may have an interest in a particular offence, to deal collectively with how to resolve the impact of the offence, and to chart a path for the future.

In addition to the victim and the offender we might see family members and other members of their communities who may be affected or who may be able to help prevent a recurrence of the offence. These meetings are facilitated by a mediator who helps bring about a collective resolution. The goals of meetings are to heal the relationship between the victim and offender, provide restitution and healing for the victim, reassure the community, and encourage acceptance of responsibility and healing for the offender through apology and reparation.

Restorative justice seeks to 'make things as right as possible' for all parties involved. Offenders are encouraged to learn new ways of acting and participating in the community, as soon as the safety concerns of the victim, community and offender are satisfied. Restorative justice seeks maximum voluntary cooperation and minimum coercion, from all parties involved.

There are many types of restorative justice practices.

Diversion is the channelling of an offender or suspect away from the criminal justice process. This may take the form of a caution or warning, diversion at the pretrial stage for resolution of the case by some informal procedure, or alternatives to conviction or sentence following a trial.

Another restorative justice practice is family group conferencing, which can be conducted in a variety of ways. All include the young offender, the family, and the victim in the conferencing process, and others who may be deemed appropriate or desired. During the meeting the participants are encouraged to express their feelings about the offence. Agreement is then sought on appropriate forms of restitution and reparation.

An alternative to this approach is victim-offender mediation, in which only the victim and offender meet in the presence of a mediator. Although family members may be present they are not active participants in the meeting.

Around Australia at the end of the twentieth century, research is being undertaken to assess the impact and usefulness of these various approaches.

Diversionary programs introduced in Aboriginal communities in Australia show potential. The continued over-representation of Indigenous people in custody, and particularly Aboriginal young people, has led to the development of the Local Justice Initiatives Program. The program recognises that it is the members of Aboriginal communities themselves who are best placed to plan and implement effective strategies to address these problems in ways appropriate to their particular needs and circumstances.

Concluding comment

We live in a world of rapid and sometimes unfathomable change. It would be trite to list the technological developments that have changed our lives, and criminal activities. What is obvious is that our ability to deal with social change lags badly behind our ability to deal with technological change.

Anticipating crime of the future and dealing with it is no easy feat. It was not very long ago that we could not have imagined crimes like credit card fraud, superannuation fraud, computer hacking etc., nor have we had the haunting spectre of the possibility of having genetic predictions, prior to birth, of an individual's likelihood of growing up violent. Globalisation, the continuing movement of products, finance, people, firearms, illicit drugs, plants and animals, and information, will test us severely. What we know is that, however we anticipate the future, it is not going to be a continuation of the past.


The author wishes to thank Diana Nelson, Research Assistant, Australian Institute of Criminology for assisting with the preparation of this article.


  • Australian Bureau of Statistics 1998, Corrective Services Australia, March Quarter 1998 (4512.0).
  • - 1999, Corrective Services Australia, March Quarter 1999 (4512.0).
  • - 2000, "Robbery Down First Time in 7 Years-ABS Figures", Media Release 82/2000, 28 June.
  • - 2000, Corrective Services Australia, March Quarter 2000 (4512.0).
  • - 2000, Corrective Services Australia, December Quarter 1999 (4512.0).
  • Carcach, C. and Grant, A. 2000, "Australian Corrections: Main Demographic Characteristics of Prison Population", Trends & Issues in Crime and Criminal Justice, No. 150, Australian Institute of Criminology, Canberra..
  • Dalton, V. 1999, "Aboriginal Deaths in Prison 1980 to 1998: National Overview", Trends & Issues in Crime and Criminal Justice, No. 131, Australian Institute of Criminology, Canberra.
  • Dalton, V. 2000, "Australian Deaths in Custody & Custody-Related Police Operations 1999", Trends & Issues in Crime and Criminal Justice, No. 153, Australian Institute of Criminology, Canberra.
  • Freiberg, A. and Ross, S. 1999, Sentencing Reform and Penal Change: The Victorian Experience, The Federation Press, Sydney.
  • Grabosky, P. 1977, Sydney in Ferment: Crime, Dissent and Official Reaction 1788 to 1973, Australian National University Press, Canberra.
  • Knibbs, G.H. 1908, Official Year Book of the Commonwealth of Australia, No. 1 1901-1907.
  • Maguire, M., Reiner, R., Morgan, R. 1994, The Oxford Hand Book of Criminology, 1st Edition, Clarendon Press, Oxford.
  • Makkai, T. 2000, "Drug Use Monitoring in Australia: 1999 Annual Report on Drug Use among Adult Detainees", Research and Public Policy Series, No. 26, Australian Institute of Criminology, Canberra.
  • Mouzos, J. 2000, "Homicidal Encounters: A Study of Homicide in Australia 1989-1999", Research and Public Policy Series, No. 28, Australian Institute of Criminology, Canberra.
  • Mukherjee, S. 1999, "Bringing to Justice" in G. Newman (ed.) Global Report on Crime and Justice, Oxford University Press, New York, pp. 65-88.
  • Mukherjee, S., Scandia, A., Dagger, D., and Matthews, W. 1988, Source Book of Australian Criminal and Social Statistics 1804-1988, Bicentennial Edition,
  • Australian Institute of Criminology, Canberra.
  • Prenzler, T. and Sarre, R. 1998, "Regulating Private Security in Australia", Trends & Issues in Crime and Criminal Justice, No. 98, Australian Institute of Criminology, Canberra.
  • SCRCSSP (Steering Committee for the Review of Commonwealth/State Service Provision) 2000, Report on Government Services 2000, AusInfo, Canberra.
  • Smith, R. and Grabosky, P. 1998, Taking Fraud Seriously-Issues and Strategies for Reform, Institute of Chartered Accountants in Australia Fraud Advisory Council, Sydney.
  • Sykes, T. 1988, Two Centuries of Panic: A history of corporate collapses in Australia, Allen & Unwin, Sydney.
  • The Composition of Australia's Police Services-1999, Australian Institute of Criminology Website:
  • Victorian Year Book, 1903, Office of the Government Statist, Melbourne, 1903.
  • Williams, P. (ed.) 2000, "Alcohol, Young Persons and Violence", Research and Public Policy Series, No. 35, Australian Institute of Criminology, Canberra.
  • Wilson, R. 1951, Official Year Book of the Commonwealth of Australia, No. 38, Commonwealth Bureau of Census and Statistics, Canberra.