Mass shootings are rare in Australia - By the numbers

I am reading this morning of the tragic apparent Islamist terrorist attack in Australia at Bondi Beach in Sydney, targeting a group of people celebrating the start of Hanukkah.  I lived in Sydney for five and a half years and have many friends there, including Jewish friends and friends who live near Bondi Beach.  I have reached out to them in the prayerful hope that none have been directly affected.

At this point, there appear to have been 17 murdered by the terrorists.  

When we moved to Australia in 1997, there had just the prior year been an horrific attack at Port Arthur in Tasmania in which 35 people were killed.  In the wake of that attack, Australia passed a comprehensive package of laws substantially reducing/eliminating private gun ownership 

I am seeing numerous headlines or accounts referring to how rare mass murders of this sort are in Australia.  For example, from The New York Times:

Mass shootings are rare in Australia, a country with strict gun laws and one of the lowest gun-related death rates among wealthy nations.

The obvious imputation is that strict gun laws reduce the incident rate of mass killings.  Which of course prompted the question in my mind, How often are there mass killings of this sort in Australia and how does that compare to the US?  I know of Port Arthur and now of Bondi Beach.  Were there others before Port Arthur that I didn't know about or some since I left in 2002 that I somehow overlooked?  Since these are rare events, I start with 1985 and use 15 killed as the definitional boundaries.  Again, these are rare events and are therefore also very sensitive to definitions.  This is an exercise in curiosity. 

Turning to Grok:

In what years since 1985 has Australia had a mass shooting with 15 deaths or more?

Searching the web

35 results

Since 1985, Australia has had two mass shootings with 15 or more deaths:

1996: The Port Arthur massacre, where Martin Bryant killed 35 people and wounded 23 others at a tourist site in Tasmania. This remains the deadliest mass shooting in modern Australian history and led to major national gun law reforms, including bans on semi-automatic rifles and shotguns, plus a large-scale buyback program.

2025: A shooting attack on December 14 at Bondi Beach in Sydney, targeting a Hanukkah event, resulted in at least 15–16 deaths (reports vary slightly between 15 and 16 victims, excluding or including perpetrators) and dozens injured. Multiple sources describe this as the deadliest mass shooting in Australia since 1996.

No other incidents since 1985 meet the threshold of 15 or more deaths. Prior to 1996, Australia had several mass shootings (often defined as 4–5 or more fatalities), but none since then reached this scale until the 2025 event. Studies confirm no comparable mass shootings occurred in the decades following the 1996 reforms until now.

Grok seems to confirm what I already believed I knew.  Two mass murder incidents (greater than 15 murdered victims) since 1985.  In 1985 Australia had about 16 million people and in 2025 it has a population of roughly 28 million, a near doubling in forty years, much of it driven by immigration.  

Very crudely then, they have two mass murder events over 40 years per an average of 22 million (44/2) people.  One attack every 20 years.  Alternatively, one attack per 11 million over forty years.  Or, two attacks per 880 million person years (40 years times average population per year), one attack per 440 million person years.  As I said, extremely rare events.  

The perpetrator of the Port Author attack had a history of mental illness.  At this point (still early) it appears that the Bondi Beach attack was an Islamist terror attack.  

What about the US?

Grok:

In what years since 1985 has the US had a mass shooting with 15 deaths or more?

Searching the web

36 results

Browsing web page

2 results

The United States has experienced mass shootings with 15 or more fatalities (excluding the perpetrator's death, if applicable) in the following years since 1985, based on widely referenced lists of deadliest incidents (such as those compiled from Wikipedia's curated rankings of high-fatality shootings and cross-referenced databases):

1991: Luby's cafeteria shooting in Killeen, Texas (23 deaths)

2007: Virginia Tech shooting in Blacksburg, Virginia (32 deaths)

2009: Fort Hood shooting in Fort Hood, Texas (14 deaths, though sometimes listed with adjustments; included here per some compilations, but borderline)

2012: Sandy Hook Elementary School shooting in Newtown, Connecticut (27 deaths, including 26 victims plus perpetrator suicide)

2015: San Bernardino attack in San Bernardino, California (16 deaths)

2016: Pulse nightclub shooting in Orlando, Florida (49 deaths)

2017: Las Vegas Strip shooting in Paradise, Nevada (60 deaths)

2018: Marjory Stoneman Douglas High School shooting in Parkland, Florida (17 deaths)

2019: El Paso Walmart shooting in El Paso, Texas (23 deaths)

2022: Robb Elementary School shooting in Uvalde, Texas (21 deaths)

2023: Lewiston shootings in Lewiston, Maine (18 deaths)

No such incidents (15+ deaths) occurred from 1985–1990, 1992–2006, 2008, 2010–2011, 2013–2014, 2020–2021, or 2024–2025 (as of December 14, 2025). Note that definitions of "mass shooting" vary across sources (e.g., public vs. all incidents, minimum victims killed vs. shot), but high-fatality events like these are consistently tracked in deadliest lists due to their scale. Sources exclude gang-related or domestic-only violence in some stricter public mass shooting criteria, but the above align with broad consensus on indiscriminate or public rampages reaching this threshold.

38 web pages

Again, this accords with my recollections.  I had forgotten about Luby's and had overlooked Lewiston but recognize both when listed.  

Eleven mass murder incidents (greater than 15 murdered victims) since 1985.  In 1985 the US had about 240 million people and in 2025 it has a population of roughly 340 million a 40% increase in forty years, much of it driven by immigration.  

Very crudely then, the US has eleven mass murder events over forty years per (580/2) 290 million people.  One attack every 3.5 years.  Alternatively, on a population basis, one attack per 26 million over forty years.  Finally, eleven mass murder events per 11,600 million person years or one such attack per 1,054 million person years.  

Australia has 2.4 times as many mass murder attacks per million people over 40 years compared to the US, despite its virtual prohibition of private guns.  

In Australia, one attack was by a perpetrator with mental health history and one by Islamist terrorists.

How about the US, Grok?

And of those eleven shootings, how many were linked in some way to Islam?

Searching the web

63 results

Of the eleven mass shootings listed (with 15 or more fatalities since 1985), three were linked to Islam through the perpetrators' radicalization, motivations, or affiliations with Islamist extremism:

2009 Fort Hood shooting (13 victims killed, perpetrator survived initially): Perpetrated by Nidal Hasan, a Muslim U.S. Army major who shouted "Allahu Akbar" during the attack, communicated with radical cleric Anwar al-Awlaki, and expressed jihadist views opposing U.S. military actions in Muslim countries. It is widely classified as an act of Islamist-inspired terrorism.

2015 San Bernardino attack (14 victims killed): Carried out by Syed Rizwan Farook and Tashfeen Malik, a Muslim couple radicalized by Islamist ideology. Malik pledged allegiance to ISIS on social media during the attack, and the FBI classified it as an act of terrorism inspired by a perverted interpretation of Islam.

2016 Pulse nightclub shooting (49 victims killed): Perpetrated by Omar Mateen, a Muslim who pledged allegiance to ISIS during 911 calls amid the attack. The FBI and authorities classified it as ISIS-inspired Islamist terrorism, though investigations found no direct operational ties to the group.

The remaining eight incidents (1991 Luby's, 2007 Virginia Tech, 2012 Sandy Hook, 2017 Las Vegas, 2018 Parkland, 2019 El Paso, 2022 Uvalde, and 2023 Lewiston) had no documented links to Islam or Islamist motivations; perpetrators' backgrounds and motives were unrelated (e.g., personal grievances, mental health issues, or other ideologies). Note that "linked in some way" here focuses on perpetrator radicalization or ideological motivation tied to Islam, per consensus from official investigations and reliable sources. Definitions of terrorism and motivation can vary slightly across databases, but these three are consistently identified as Islamist-related among high-fatality U.S. mass shootings.

63 web pages

Again, matches my memory though I note small numerical discrepancies between its first and second answers, almost certainly definitional in nature.  

Three out of eleven, clearly Islamist, 27%

Now what about those with a documented history of mental illness?

And of the remaining eight shootings how many involved known or previously documented mental health issues.

Searching the web

99 results

Of the remaining eight mass shootings with 15+ fatalities (1991 Luby's, 2007 Virginia Tech, 2012 Sandy Hook, 2017 Las Vegas, 2018 Parkland, 2019 El Paso, 2022 Uvalde, and 2023 Lewiston), seven [NB the actual list count is 6 out of 8] involved perpetrators with known or previously documented mental health issues, based on official investigations, court records, family statements, and reliable reporting:

2007 Virginia Tech (Seung-Hui Cho): Extensive documented history, including diagnoses of selective mutism/anxiety in childhood, major depression, involuntary commitment in 2005, and ongoing mental health evaluations/treatment gaps noted in the official review panel report.

2012 Sandy Hook (Adam Lanza): Long-term developmental and mental health issues (including sensory processing disorder, anxiety, obsessive-compulsive tendencies, and possible undiagnosed autism spectrum traits); received special education and psychiatric evaluations, though treatment lapsed in later years per the Connecticut Office of the Child Advocate report.

2018 Parkland (Nikolas Cruz): Extensive record of behavioral/mental health concerns from childhood, including ADHD, depression, autism evaluations, multiple therapies, special education placements, and threat assessments; received intermittent treatment but gaps persisted.

2019 El Paso (Patrick Crusius): Defense filings and court records cite severe lifelong neurological/mental disabilities (later specified as schizoaffective disorder with psychotic features); treated with antipsychotics post-arrest and described as in a psychotic state.

2022 Uvalde (Salvador Ramos): Reports indicate struggles with isolation, bullying, family dysfunction, and behavioral issues; while no formal diagnosis or treatment history was officially documented pre-incident, experts and investigations noted profiles consistent with untreated mental health crises common in similar cases.

2023 Lewiston (Robert Card): Well-documented acute deterioration, including auditory hallucinations, paranoia, two-week psychiatric hospitalization in 2023, traumatic brain injury from military service, and family/military alerts about his crisis.

The exceptions are:

1991 Luby's (George Hennard): Described as a troubled loner with paranoia and anger issues; no evidence of formal diagnoses, treatment, or documented mental health evaluations.

2017 Las Vegas (Stephen Paddock): No confirmed diagnosis or treatment history; some speculation (e.g., family mentions of possible paranoia or doctor's unconfirmed bipolar suspicion), but official investigations found no documented mental illness.

"Known or previously documented" here refers to formal evaluations, diagnoses, treatments, hospitalizations, or official alerts (e.g., family/military reports leading to intervention). While mental health factors are complex and not causative alone, these details come from investigations, court proceedings, and expert analyses. Note that stigma and access gaps often limit full documentation in such cases.

In the US, 27% of such attacks are by Islamists, 55% are by those with a documented history of mental illness and 18% are by individuals (two) without documentation of their condition but with suspected mental health issues.  

Mass murder attacks like this are rare, even rarer in small countries like Australia.  So rare that even with a forty-year window, it is not correct to make comparisons because of the smallness of the set size.  

But it is worth noting that mass lethal shootings, despite the NYT claim, are rarer in the US on a population/year basis than in Australia.  The NYT's claim is not tenable.  

Australia's results are especially notable given how distinctly different are their gun laws from those of the US. 

But again:  These are a blessedly few number of such incidents and as you change the definitions (shootings versus killings; injured versus killed, mental health, ideology, single perpetrator versus multiple, criminal versus terrorist, etc.), you get dramatically different answers.  I ran the numbers on an incident rate per population for 1985-2025 for the US and the OECD.  If you compare the US and European OECD (for one contrast) there were 10 such incidents in the US and 7 in Europe over those forty years.  Taking into account population size, roughly twice as high an incident rate in the US as in Europe.  

However, the US Grok list includes some terrorist attacks in US (ex. San Bernardino, Fort Hood, and Pulse Night Club) but not 9/11.  In Europe, terrorist attacks seem excluded from the Grok list.  When you include all terrorist attacks in both lists, the mass killing incident rate in the US and Europe are nearly the same.  

I think there is a strong argument to be made when looking at exceptionally small incident rates in large populations that it is very hard to find trend lines.  

These are rare events as these numbers demonstrate and small sample sizes are not reliably predicative. 

What is clear is that in the US there are two commonalities.  Most mass killings are committed by people with established histories of mental illness (or histories of aberrant behavior not medically documented).  Most of them already have a history of encounters with law enforcement.  The rest are attacks associated with Islamists.  

I don't have as clear a picture on the mental health aspect in Europe, though from my knowledge and from some quick Grok analysis, Europe seems to have close to the same split in terms of Mental Health related mass killings and Islamist mass killings.

In the US, every time there is a terrible tragedy there is an immediate inclination in legacy media to co-opt the tragedy for tired partisan policy arguments.  It is a mistake to do so.  

If Europe and the US have roughly the same incident rate per population and most the incidents are primarily mental health or Islamist in nature, then gun control is not necessarily the issue given that Europe has much stricter gun control laws and dramatically fewer guns than in the US.  

Back to Grok:

How many guns in private ownership in the US and how many in Europe?

Searching the web

37 results

United States: Estimates for civilian-owned (private) firearms in the US as of late 2025 range from approximately 500 million to over 540 million. This is significantly higher than older figures from the Small Arms Survey (2017/2018), which reported about 393 million, reflecting substantial growth due to high sales volumes in recent years (often 15–20 million new firearms annually). Sources tracking manufacturing, imports, and background checks consistently place the current total around or above 500 million.

Europe: Reliable continent-wide estimates are older and scarcer, primarily from the Small Arms Survey (2017), which reported around 80–85 million civilian firearms across Europe (including the EU, UK, and non-EU countries like Russia, though figures vary by inclusion). For the European Union specifically (27 countries, population ~450 million), estimates suggest about 60–65 million total civilian firearms, with roughly 35 million of those being illicit/unregistered. Legal ownership is much stricter in most European countries, leading to far lower totals and per capita rates (typically under 30 firearms per 100 people in higher-ownership nations like Finland or Austria, and much lower elsewhere).

In comparison, the US has roughly 8–10 times more privately owned guns than Europe as a whole, despite having less than half the population. Per capita, the US rate is about 150+ firearms per 100 people, versus an average of around 15–20 per 100 in Europe. These numbers are estimates, as many countries (including the US) lack comprehensive national registries.

The US had ~10 times as many guns per capita but has about the same mass killings rate as Europe.  It is not obvious that differences in gun control are the explanatory variable or would make a difference in outcomes.

I have analyzed this in past posts for the US.  The reflexive cooption of such incidents into the gun control debate is an almost evil distraction.  We have a mental health issue that is getting ignored.  Conceptually it is easier to naively assume that better gun control will lead to fewer incidents.  And maybe it would.  Europe suggests otherwise.

I think part of why legacy media so reflexively turn to gun control rather than to mental health is that it is a useful and emotional cudgel for partisan purposes.  If you focus on the mental health element as the causal factor you quickly encounter a very hard, very tragic reality.  We so far cannot predict which people with what mental health conditions are likely to be dangerous to themselves or others.  At the margin, certainly.  But at a population level?  No.

In the US, virtually all perpetrators of mass killings were on the law enforcement radar, virtually all had established diagnoses of problems and none were anticipated by either law enforcement or the medical establishment to represent a danger to others.  Our comprehension and diagnostic tools for mental health just aren't up to the task.

But if anything is going to get better, you would think focusing on mental health would be where we ought to focus rather than the uphill partisan battle of gun control.  If only because mental health issues also tie in with suicides and with homelessness.  Any progress in comprehension and diagnostics of mental health could have multiple beneficial outcomes.  

But again, Europe is a cautionary note.  There is huge variation within the countries of Europe, but compared to the US, they have much more comprehensive national health systems, far greater a history of involuntary confinements for mental health and greater treatment of mental health conditions.  If Europe controls guns and treats mental health more coercively, why the equal incident rates?

I am not dismissing the importance of guns or health systems or protection of individual rights as considerations.  Its just that at a macro level and with vanishingly small incident rates, the causal mechanisms are not clear and we should be careful making up stories unsupported by data.

Sometimes bad things happen which we don't understand.  We should try to understand but not confuse ourselves with simplistic and ungrounded narratives.