Introduction

Aims

  1. Identify and record sites of frontier massacres of Aboriginal, Torres Strait Islander and non-Indigenous people across Australia from 1788 to 1930.
  2. Provide the first Australia wide record of frontier massacres that is comprehensive, based on a rigorous methodology, with well-structured data and a map, and providing the available evidence for each frontier massacre site.
  3. Inform public debate about colonial frontier violence.
  4. Provide open access knowledge to the public and invite contributions.

Methodology

The research team built on the approach to the investigation of massacre set out by historical sociologist Jacques Semelin. He considers that the ‘triptych of aggressor, victim and witness, comprises the “basic triangle” for investigating massacre’. The investigator must examine the pre-conditions leading to the event and include oral and written accounts of the event at the time and in the aftermath (Semelin 2005: 376). He also notes that evidence produced in the long aftermath is often more reliable than in the immediate aftermath. From Semelin’s work, the research team established a definition of frontier massacre and its common features.

Definition

Unlike ‘genocide’, there is no legal definition of massacre, or a ‘frontier massacre’. Most international scholars of massacre appear to agree that the minimum number of people killed to constitute a massacre is between three and ten people (Dwyer and Ryan 2012: xiv-xv).

In this project, a colonial frontier massacre is defined as the deliberate killing of six or more relatively undefended people in one operation.

The definition applies to the frontier massacre of Aboriginal and Torres Strait Islander people and Colonists. The number, six, has been selected because of the devastating impact on these people.

For Aboriginal and Torres Strait Islander people, as Barbara A. Mann points out, most hunter/forager people operate in groups of about twenty people. The killing of an undefended group of six or more of them in one operation, comprises 30% of the group, which Mann terms a ‘fractal massacre’. The impact is immediate and devastating. The survivors are immediately vulnerable to further attack, such as the kidnapping of women and children. They are impeded in their ability to hunt and forage for food, reproduce the next generation and carry out ceremonial obligations to ‘Country’. They are also vulnerable to introduced disease. (Mann 2013: 167-183) In the longer term they are often forced to join other Aboriginal groups, or surrender to Colonists.

For Colonists and other non-Aboriginal people, a frontier massacre of six or more undefended people can also have immediate and devastating impact. Frontier colonist communities were isolated, mostly male, and usually consisted of only a few families. Of the massacres of colonists included on the map, four of them include the survivors of shipwrecks on the Australian coast. Two others involve the slaughter of colonial men alone and the remainder include the killing of colonial families. In all but two cases, the frontier massacres of non-Aboriginal people generated reprisal massacres of extraordinary scale and impact.

Euphemisms for frontier massacre

While some frontier massacres were widely publicised, and some were recorded by police and government officials, in most cases a code of silence was imposed in colonial communities in the immediate aftermath. Frontier massacres were only referred to indirectly. According to The Queenslander, 1 May 1880, p.560, the ‘bush slang’ word ‘dispersal’ was often used as a convenient euphemism for ‘wholesale massacre’. Other euphemisms such as ‘clear the area’, ‘pacify’, ‘teach them a lesson’, ‘affray’, ‘collision’, or ‘fell upon’ were also used.

Frontier massacres are sometimes alluded to in placenames, such as Skull Creek, Waterloo Plains, or Blackfellows Bones Bore, and in others, the word ‘murdering’ such as ‘Murdering Gully’ appears. Places are also sometimes named after colonists who have committed frontier massacres. They include the town, ‘Bunbury’ in Western Australia, named after Lieutenant William Bunbury who was the key perpetrator in several frontier massacres in 1836-7. ‘Coutts Crossing’ in New South Wales, is named after settler Thomas Coutts who poisoned 14 Aboriginal people in the 1840s.

Aim and purpose of a frontier massacre

The aim of a frontier massacre of Aboriginal or Torres Strait Islanders is either to eradicate the victims or to force the survivors into submission. The purpose is to clear Aboriginal and Torres Strait Islander peoples from a particular area; to punish or prevent resistance such as killing of colonists and livestock, or; to prevent access to important sites such as food and water sources, or sacred and ceremonial sites.

The aim of frontier massacres of colonists is typically to resist colonisation and drive colonists away, or; to seek traditional justice or ‘payback’ such as for rape, abduction and murder, destruction of food and water sources or violation of sacred sites.

Common features

This project describes frontier massacres as either (1) reprisal; (2) opportunity.

A 'reprisal' massacre is carried out in response to a specific incident, such as the alleged:

An 'opportunity' massacre is not in response to a specific incident, though it may be done with a purpose in mind, or as the opportunity arises. Examples of opportunity massacres include:

Evidence

The evidence for a frontier massacre in colonial Australia is usually found in printed and archival sources, many of which are now available online. A complete list of sources consulted for the project, is available in the Bibliography under the Sources section of the Menu.

Australian newspapers on Trove comprise the major source for the project. More than 90 newspapers from every colony, state and territory were consulted. The newspaper references are listed under a separate heading in the Sources section in the Bibliography.

Newspapers often provide the first reports of a frontier massacre and provide reports of official inquiries into a possible massacre. They can also provide the voices of the attackers and survivors telling their story, long after the event.

Police and government reports sometimes provide clear accounts of how many people were killed, where, by who and for what reason. In other cases reports appear to be missing, neglect to mention events or use euphemistic language.

Massacres were often covered up, involved few witnesses, and were often not spoken of until long after the event, when fears of arrest or reprisal had long passed. It is sometimes the case that while it is clear a massacre occurred, some details can only be garnered from later sources and by inference. For these reasons, the most reliable sources are sometimes secondary sources which have gathered and considered all available information.

Other key sources of evidence:

Published sources:

Unpublished sources:

The evidence of most frontier massacres is compiled from a combination of published and unpublished primary and secondary sources.

Bibliography

This project uses the Harvard (UON) bibliographic style, except where some archives use specific styles.

The structure of the bibliography for Stage 4 has undergone some minor changes to make it easier to navigate. It comprises:

Data descriptions

FieldDescription
Site NameA unique name of the site of the frontier massacre. This is not necessarily the same as the official name of the place or nearby location. The name may have changed if the research team became aware of a more appropriate name.
Aboriginal Place NameThe name that Aboriginal or Torres Strait Islander people give to the place.
Language GroupThe language group of the Aboriginal or Torres Strait Islander people involved in the massacre. The names of language groups are from the AIATSIS information at AustLang, unless otherwise requested by Aboriginal or Torres Strait Islander groups.
ColonyThe name of the colony in which the event occurred at the time of the event (colony boundaries changed over time and may differ from present day state and territory boundaries).
Present State/TerritoryThe present-day state or territory where the massacre event took place.
Police District or Pastoral DistrictThe name of the Police District or Pastoral District at the time of the incident. Both are useful regional indicators for historical research.
CoordinatesThe geographical coordinates locating the site. This point is imprecise to around 250 metres and is a best estimate. It may also be inaccurate due to the vagueness of historical records, because the event took place over a large area, or to avoid desecration of the site.
LatitudeThe Latitude of the incident in WGS 84, rounded to 3 decimal places.
LongitudeThe Longitude of the incident in WGS 84, rounded to 3 decimal places.
Well Known DateThe date of the incident as described in sources, if a specific date can be established clearly.
DateThe estimated date when the incident, or series of incidents commenced. If we cannot be accurate to a day, the start and end date are the range within which the massacre occurred. Eg: if records indicate 'one day in late May' the date range will be 20 May to 31 May.
Attack TimeThe time of day of the attack if available: Daybreak, Morning, Midday, Afternoon, Evening, Night.
VictimsAn identification of the victims as either Aboriginal people or Torres Strait Islanders, Colonists or otherwise. In some cases more specific information about who the victims were is also provided.
Victim DetailsMore specific information about the victims if it is available, such as whether they were shepherds, warriors, women or children.
Victims KilledThe number of victims killed in the incident. Although numbers are often not exact, this is a single number so that it can be used in calculations. Conservative estimates are used. For example, if records indicate 6 to 10 people were killed, the map database records the lower number, 6. Numbers that have been published in sources are typically used as they are the best or only available evidence, unless there is a clear reason not to. In some cases where the only numbers available in sources are exceptionally large and made long after the massacre we have estimated somewhat lower numbers based on available information. Where it is clear a massacre occurred but there are no published numbers, numbers have been estimated by professional historians with expertise in frontier violence in the region at that time, using available information.
Victims Killed NotesAdditional information about the number of victims killed. For example, to note ranges of estimates, differences among sources, more detail on the number wounded or whether they were men, women and/or children.
AttackersAn identification of the attackers as either Aboriginal people or Torres Strait Islanders, colonists or otherwise. In some cases more specific information about who the attackers were is also provided. Note that magistrates, military, mounted police and native police are counted as 'colonists'.
Attacker DetailsMore specific information about the attackers where available, such as whether they were military, native police, or settlers.
Attackers KilledThe number of attackers killed in the incident. Although numbers are often not exact, this is a single number so that it can be used in calculations. Minimum estimates are used. Usually no attackers were killed despite resistance and defensive action.
Attackers Killed NotesAdditional information about the attackers. For example, to note ranges of estimates, differences among sources, or more detail on any wounded.
TransportThe mode of transport used by the attackers.
Motive‘Reprisal’ or ‘opportunity’. This is limited to whether the frontier massacre was in direct ‘reprisal’ for a specific incident, such as spearing of livestock, theft, burning crops, murder, etc or the attackers were acting on an ‘opportunity’ to attack, rather than in response to a specific incident. Motivations are often detailed, specific to the frontier massacre and may include a long series of events.
Reprisal For Death OfThe number or description of people who may have been killed in the immediate lead up to the frontier massacre.
Weapons UsedLists the weapons, if known, used by the attackers in carrying out the frontier massacre.
NarrativeThe narrative is a summary of the frontier massacre event compiled from the sources. They include the pre-conditions, and the date it took place. Where possible names of the victims and attackers are included, along with what happened in the immediate aftermath, how the incident was finally identified and any other relevant information.
SourcesA list of the sources compiled as evidence of the frontier massacre. Links are provided to online versions of sources if they are available. For a glossary of acronyms and a full bibliography see the Sources page on the Menu.
Corroboration Rating One to three asterisks indicate the level of confidence of the researcher in the evidence:
* Reliable evidence that a massacre occurred but more corroboration and detail welcome.
** Better evidence that a massacre occurred, such as 2 sources, or a detailed police report but further corroborating evidence or more detail welcome.
*** Corroborating evidence drawn from various and reliable sources with a relatively high level of detail.

Cartography

The full collection of sites is stored in the common standard datum of WGS84 and projected in the online map using WGS 84/Pseudo-Mercator (EPSG:3857).

Points showing massacre sites were located using a variety of sources and tools, cross checking sources against each other, nearby sites, old maps and within GIS systems. Each point has purposefully been made imprecise by rounding to 3 digits (approx. 250m) for 3 reasons:

The map and 3D terrain visualisation is implemented using the ESRI ArcGIS Javascript API.

Project stages and updates

Stage 5 is the final version of the website, map and data. There will be no further changes as part of this project. It represents the best information available to us, on all colonial frontier massacres in Australia that the research team been able to find sufficient evidence for. This should provide a useful resource for further research into individual massacres and colonial frontier violence at the regional and national scale.

With the release of stage 4 on 16 March 2022, the project was drawing to its conclusion, as it appeared that almost all sites that could be identified through the project had been. Stage 5 has involved a thorough review, initiated by the research team, of all data on the website, assisted by contributions from the general public, volunteers and peer reviewers. Any issues raised were checked by the research team, in some cases leading to further research. This resulted in wide ranging improvements to the information on the website, including more accurate locations and dates, better narratives and context, cautious estimates of victims killed, better citation of sources and proof reading corrections. In some cases closer research resulted in sites being added and some being removed; some sites previously thought to be two distinct incidents appeared to be descriptions of the same incident and have been combined into a single site; and some sites previously thought to be a single incident involved two massacres. The research has improved our understanding of this violent history.

Throughout the project there have been ongoing changes and additions. Data has been uploaded to the digital map and information on the site adjusted at regular intervals, with a stage number in the footer of each page. Ongoing maintenance included some interim corrections and some additions thereafter. See Updates and Changes for a summary of updates and changes.

Information on this website represents the best evidence available to the research team at the time. Future research may reveal more information that could affect these results. Finding, checking, organising and preparing information for frontier massacres across colonial Australia is a massive undertaking. From the beginning, the research team has invited, welcomed and acted on feedback, suggestions and corrections from the community. These contributions have been checked by the research team before inclusion. We thank everyone who has assisted the project in this way.

Findings

Summary

438 sites of frontier massacre, in which at least 10657 were killed, were included in stage 5. The number is indicative rather than definitive and may vary with future research. Stage 4 included 421 sites, in which 11,257 were estimated to have been killed. These number of people killed depends on estimates. This project has used conservative estimates. The increased number of massacre sites is primarily a result of more sites having been identified. The lower conservative estimate of the amount of people killed is primarily a result of some very large estimates on sites with a low corroboration rating being revised down. The only available sources providing numbers for these sites were long after the event, and out of keeping with numbers reported in other massacres in the region and at the time and so may be exaggerated. The most prominent cases of this involved death tolls in the hundreds, hence a few uncertain sites have affected the overall estimated death toll by hundreds. In some cases where the circumstances described make unusually high death tolls likely, relatively high estimates have been retained.

A massacre not appearing on the site does not mean that a massacre did not happen. The numbers estimated do not mean that there was not a higher death toll. The massacres, numbers and details are only what sufficient evidence could be found for as part of this project.

The map timeline indicates that frontier massacres of Aboriginal and Torres Strait Islander people spread steadily across southern Australia from 1794 to 1860 with notable peaks in the 1820s in Tasmania and the 1840s in NSW, Victoria, South Australia and Western Australia. From the 1860s the frontier shifted to Northern Australia. A massacre peak took place in Queensland in the 1860s to 1870s. There were two broad peaks around 1880 to 1900 and 1910 to 1930 in the Northern Territory and the Kimberley region in Western Australia. The average number of Aboriginal and Torres Strait Islander people killed in a frontier massacre increased over time from 21 to 27 after 1860.

Group massacres: There are at least 23 ‘groups’ of frontier massacres, comprised of at least 66 separate incidents. These are instances where it is difficult to determine whether they should be regarded as a single massacre at many locations, or a series of closely related massacres, for example where an expedition of colonists rides out and kills people at several locations, sometimes over several days before returning. Sometimes this is a consequence of ecology – in arid areas people are more spread out, and travel time is longer. Where possible we have described these as what they are – a ‘group’ of massacres. In some cases where it has not been possible to identify distinct locations in a ‘group’, these have been treated as a single massacre, such as at Arafura, and Slaughterhouse Creek, and the three Coniston expeditions which each involve multiple sites. The first ‘group’ was recorded in Tasmania in 1827 and then Gippsland in Victoria in 1843. They intensified in Queensland in the 1870s and 1890s and continued in Western Australia and the Northern Territory until 1928.

More detailed statistics are available on the Statistics page.

Weapons

Colonists and Native Police weapons

1794 - 1860: In this period the muzzle loading musket was the key weapon used by attackers. Most widely used was the Brown Bess Musket. Issued to British regiments which served on the frontier in the Australian colonies from 1788 to the 1840s, it was a smooth bore muzzle loading 0.75 calibre flintlock weapon that could fire 3 shots in 45 seconds over a range of 80 yards (73 metres). The main drawback was that it was loaded and fired from a standing position. Several versions of the musket were used in carrying out frontier massacres, including the bayonet that was fitted to the musket barrel and the carbine, which had a shorter barrel designed for use on horseback.

Before 1830, most perpetrators of frontier massacres of Aboriginal or Torres Strait Islander people were on foot. In the century following, 1830 - 1930, the horse became the vehicle of choice although on at least one occasion in the Northern Territory in the 1920s, camels were used. The horse was an effective weapon in driving Aboriginal people from their campsites, and in pursuing and cornering people.

The carbine was used by the native police in the Port Phillip District in the 1840s and in the Northern District of New South Wales in the 1850s. The Baker rifle which was used by some colonists on the frontier to 1860, was more accurate than the smooth bore musket and could fire over a longer range. Other weapons that were recorded in this period include swords, cutlasses, pistols, swivel guns, carronades and ships’ cannons.

Another weapon was poison – strychnine, arsenic and plaster of paris – either in freshly made damper or in flour used by Aboriginal people to make damper.

In some cases large numbers of Aboriginal people were corralled in preparation for massacre and were tied up with leather straps, rope or chained. In many instances their bodies were burnt and the bones crushed to hide the evidence.

1860 - 1900: Breech loading rifles became widely available after 1860. They fired over a longer range of 300 yards (274 metres) and could be loaded and fired from a prone position. According to Jonathan Richards, the Queensland Native Police were issued with British made Terry breech loading rifles in 1861, single shot Snider-Enfield rifles in 1874 and Martini Henry-Enfield rifles after 1884 (Richards 2008: 55-6). According to Chris Owen firearms used by police in the Kimberley in Western Australia in the 1880s 'were initially the single shot Snider-Enfield rifles, which fired enormous .577 cartridges, although by the late 1890s they were considered too old, complicated and prone to becoming clogged with sand.' The Winchester Repeating Rifle arrived in Australia in the 1880s and could fire many shots before reloading and ‘was the weapon of choice [in WA] through the mid-1890s. The side arms used were the Webbley revolver until they were replaced by Smith & Wesson colts' (Owen 2016:165). The Snider-Enfield was issued to police in the Northern Territory in the 1880s. Other weapons such as swords and cutlasses were phased out, but poison including poisoned alcohol and tobacco continued in use across northern Australia. In the 1890s in the Kimberley and the Northern Territory, Aboriginal people were also tied up with neck chains in preparation for frontier massacre. The horse remained the vehicle for carrying colonel attackers to frontier massacre sites across Australia.

1900 - 1930: The Lee-Enfield bolt action .303 repeating rifle with a firing range of 300 yards (274 metres) was first used by Australian colonial troops during the Boer War 1900-1902, and then by Australian infantry until the 1950s. The short barrel Lee-Enfield rifle and the Enfield revolver were widely used by police and settlers to carry out frontier massacres of Aboriginal people in northern Australia from 1918 to 1930. The horse remained the vehicle of choice for carrying colonial attackers to the frontier massacre site.

Aboriginal and Torres Strait Islander weapons

In carrying out the 13 recorded massacres of non-Aboriginal people, Aboriginal and Torres Strait Islander warriors were reported as wielding wooden and stone tipped spears, nulla nullas, large stones and rocks, waddies and hatchets. They also used fire to set alight huts and homesteads and smoke out their intended victims. The bodies of frontier massacre victims were sometimes mutilated and dismembered. The only known frontier massacres when Aboriginal or Torres Strait Islander attackers used firearms against their intended victims was in the Latrobe Valley in Victoria in the 1840s; and the slaughter of the Mawby family by Jimmy Governor and his brother Joe at Breelong in New South Wales in 1900. In each of these cases, the Aboriginal attackers rode horses. Most frontier massacres by Aboriginal and Torres Strait Islander people were carried out on foot.

Genocide

The purpose of this site is to present information about frontier massacres. Some frontier massacres are clearly acts of genocide, according to the definitions of the UN (Definitions Genocide: 2022) and leading historian of genocide, Jacques Sémelin (Sémelin: 2002). The UN definition of genocide requires both 'intention' and 'action'. In many frontier massacres the intention was clear. Readers should bear in mind that while some attackers openly declared their intention and actions, in most cases there is a strong incentive to cover up both the massacre and the intention.

Conclusion

The number and regularity of massacres between 1788 and 1930 shows that frontier massacres of Aboriginal and Torres Strait Islander people in Australia are a significant component of the violent colonisation of Australia.

The overwhelming majority of the victims of frontier massacre were Aboriginal people, killed by Colonists. Frontier massacres have a traumatic and enduring impact on Aboriginal and Torres Strait Islander communities and are remembered to this day in oral histories, paintings, petroglyphs, and dance.

Government agents of the early colonies and later states and territories, such as military and police, were identified as participants in around half of the frontier massacres.

The data and timeline reveal clear patterns in frontier massacres. The regions of particular intensity in frontier massacres indicate the shifting location of the Australian colonial frontier at particular historical periods. The map data reveals that frontier massacres were rarely isolated events. Rather they were often connected at a regional level, raising new questions about the causes and aftermath of these events.

While some frontier massacres are well documented, and some were openly admitted or were recorded in police and government reports, many were covered up and seldom spoken of in the colonist community. Very few cases were brought to court and of these only one of them, Myall Creek in 1838, resulted in the conviction and hanging of some of the attackers. After that, the practice of frontier massacre became more difficult to detect and the details more difficult to interrogate. As research into frontier massacres continues, more information can improve our knowledge and understanding. The Colonial Frontier Massacres map and website has raised awareness and informed public debate on this controversial topic and the research team hope it will remain a valuable research tool as we continue to learn more about Australia’s violent colonial frontier.

  1. Definitions: Genocide United Nations Office On Genocide Prevention and the Responsibility to Protect, accessed 27/07/2022 https://www.un.org/en/genocide-prevention/definition
  2. Dwyer, P G, and Ryan, L (eds), 2012, Theatres of Violence Massacre, Mass Killing and Atrocity throughout History, Berghahn Books, New York, pp. xiii-xxv.
  3. Mann, B.A. 2013: ‘Fractal massacres in the Old Northwest: the example of the Miamis’, Journal of Genocide Research, Vol.15, No.2, June 2013, p.172.
  4. Owen, C 2016, ‘Every Mother’s Son is Guilty’: Policing the Kimberley Frontier of Western Australia 1882-1905, UWA Press, Nedlands, Western Australia.
  5. Richards, J. 2008 The Secret War, A True History of Queensland’s Native Police, UQP, Brisbane, pp.55-6.
  6. Sémelin, J. 2002, ‘From massacre to the genocidal process.’ International Social Science Journal, 54: 433-442. https://doi.org/10.1111/1468-2451.00397