The recent murder of Zena Campbell, a 21 year old trans identified male in Wellington, has led to a series of media articles which highlight the unique vulnerability of trans people to violence and murder.
National Council of Women (NCWNZ) Gender Equal spokesperson Dr Gill Greer says “the murder is an example of the kind of transphobic violence directed towards trans people and particularly trans women.” [….] Speaking globally,she noted trans women in particular appeared to be a target of violence.
"The Human Rights Commission highlighted some of these issues in a report back in 2008 but we've not made enough progress in the past 10 years," she said.
"We know trans people experience high levels of violence of all kinds; trans young people are more likely to be bullied at school, and more likely to attempt self-harm and suicide. We want to see a more inclusive New Zealand society where all genders are treated equally."”
Green MP Jan Logie echoed these ideas at a candlelight vigil for Zena Campbell, stating that “a lot more work needs to be done about violence towards New Zealand’s gender minorities.”.
A quick scan of transgender advocacy internet sites in New Zealand reveals similar types of claims. In a speech for the November 20th New Zealand version of the ‘Trans day of Remembrance’, an annual ceremony marked all over the world to remember the trans victims of transphobic violence, Judy Virago had this to say
Trans women around the world are slaughtered weekly, just for being who they are. It’s scary. And it’s not our fault. I refuse to accept this.
By the 30th of October this year, 24 trans women, most of them women of colour, had been killed in the US alone. Rates of violence against trans people are so offensively high that there is even a wikipedia page that records reported unlawful deaths of transgender people by year and location. There are 52 people currently listed on this page for 2017.
In a Salient (Victoria University student magazine) article from 2015, transactivist Charlie Prout claims that “transgender people experience violence, sexual assault, murder, and verbal assaults at much higher rates than the general population.”
Are these claims of unique vulnerability to violence, murder and other forms of assault true? In what follows I will restrict my attention to murder rates for transgender people. I will not attempt to discuss or question the various forms of oppression, discrimination and prejudice faced by transgender people which do not involve murder. In looking at this issue critically, it is not my intention to minimise or dismiss the various forms of harms done to transgender people by a variety of means. I think that murder of Zena Campbell was a reprehensible act against a vulnerable person, and ought to be condemned. The question is whether this death served the interests of a broader narrative around murder rates, and if the assumptions behind this narrative are true.
So what is the rate of murder for trans identifying people in New Zealand? The most authoritative source for international statistics appears to be the Trans Murder Monitoring Project, a research project sponsored and run by the charitable organisation Transgender Europe. According to the data shown on their site , there has been just one murder of a trans identifying person in the ten year period between 2008 and September 2017, and just five other such murders in the entire region of Oceania (Australia, Fiji, New Caledonia, New Zealand, Papua New Guinea):
The research methodology section of the site contains the following disclaimer about the nature of the data provided:
When using the data presented by the Trans Murder Monitoring project, please note that the data presented is not comprehensive, for the reasons outlined below, and can only provide a glimpse into a reality which is undoubtedly much worse than the numbers suggest:
1. The collected data show only those cases which have been reported. There is no data and no estimates available for unreported cases.
2. The data presented here does not include all reported cases worldwide, but only those which can be found on the Internet, along with those murders that reported to us by local activists or our partner organizations.
3. Due to the dozens of languages used in the Internet, the variety of terms used to denote trans and gender-diverse people, and the myriad numbers of web pages to search through, it is simply not possible to find all reports shown on the Internet.
4. Finding reports of murdered trans and gender-diverse persons in particular is also problematic, as not all trans and gender-diverse people who are murdered are identified as trans or gender-diverse.
As I will go on to note when I turn to look briefly at the case of Brazil, there are many solid reasons for thinking that murders of transgender people in regions of the world where there is widespread institutional corruption and lack of recognition of transgender people would be under-reported and frequently unrecognised. Given the social prominence of transgender activism in New Zealand over the past ten years, and the amount of coverage and institutional support surrounding the recent death of Zena Campbell, I strongly doubt that these concerns apply to the New Zealand context. I think it highly unlikely that a trans identified person other than the two documented instances (Richard Milton ‘Diksy’ Jones in 2009, and Zena Campbell in 2018) has been murdered in the past ten years.
How does this compare to the national rate of murder for the country as a whole? New Zealand is a relatively safe country with a low homicide rate. Using police data from the period between 2007 and 2014, alongside census figures from 2006 and 2013 for overall population statistics, I worked out that New Zealand has a murder rate of 1.17 murders per 100,000 people each year. (Wikipedia quotes a rate of 0.91, apparently based on just the year 2014 – I believe my estimate for the period in question to be more accurate).
The hard part is to figure out what the trans identified population of New Zealand is. The census does not record this sort of data, and there are numerous issues and questions which surround the definition of ‘transgender’ which complicate the picture further. The best I could think to come up with is three estimates:
- 16,000: Roughly the number of people in NZ who report living with a same sex partner
- 28,000: Based on the US figure of 0.6% of total population
- 56,000: Based on an NZ school survey where 1.2% of high school students identify as trans
Here are the results based on my three trans population estimates for the ten year period up to 2018:
Trans population estimate
Expected deaths per year
Ten year period
0.117 * 1.6 = 0.1872
0.117 * 2.8 = 0.3276
0.117 * 5.6 = 0.6552
Of course there are big sources of possible variation and error here because we are dealing with such small numbers, if we went to the trouble of calculating confidence intervals there would be fairly wide margins of error for all of these estimates. Having said that the empirical data of 2 trans murders over a ten year period fits in fairly well with the overall average murder rate. If the larger population estimates are true, the the trans murder rate appears significantly smaller than the overall rate.
How does this analysis compare with other western countries with similarly low overall murder rates? Is New Zealand a strange and unusual outlier, or is it quite typical in its low trans murder rate? The answer is a definitive ‘very typical’:
From the UK:
“There is no evidence within the recorded data from the last 9 years that transgender people are murdered at significantly higher rates than average.”
“Applying the FBI's 4.7 homicides per 100,000 means we'd expect to see ~38 trans murders a year if the rates were identical to that of the general population (4.7 * 8). That's three times as many trans deaths as were actually recorded in 2014. Keep in mind that if you use a higher trans prevalence rate, you'd expect even more murders. Doubling the prevalence rate from .3% to .6% means you'd expect there to be ~76 trans murders per year.
If you adjust those numbers for race and gender it becomes obvious how some trans women are obviously at higher risk. Blacks and Hispanics are about 29% of the general population, and women are about half the US population, which means ~15% of the trans population should be TWOC - and those black and Hispanic trans women account for more than 90% of all trans murder victims.
Notably, white trans women were killed less often per capita than cis white women. Yes, there are issues with applying one white trans death per year against the white trans population, and yes, there are other forms of violence (in addition to murder), but the point is that transphobic murder appears to be an issue almost exclusively affecting trans women of color.”
“Meanwhile, back in the real world, the latest stats we had from police in Canada regarding hate crimes (in a nation of 35 million) showed an annual total of 186 “hate crimes” based on sexual orientation, with 120 of those involving violence, and with 40 or so involving “serious violence.” To put this “epidemic of hatred” in perspective, environment Canada reports an average of 174 people struck by lightning each year in Canada, with almost all of those being very serious in nature. Thus, it would seem those worried about hate crimes in Langley should also regard this threat from above just as seriously, maybe even more so. By Dirks’ admittedly very, very liberal standards, we must also conclude that we are having an epidemic of lightning assaults on Canadians too, although admittedly it is not clear how many lightning strikes specifically target LGBTQ people — but we can be sure that queen of all bigots, Mother Nature, whom stubbornly refuses to recognize men as women, is no doubt specifically targeting trans people and other members of the LGBTQ community.”
What if we look at the single biggest trans – killing country in the entire world, Brazil? In the ten years up to September 2017 the TMMP records 1071 trans deaths from Brazil. The South American region as a whole is by far the most deadly for trans identifying people:
The first thing that needs to be recognised and noted is that the caveats mentioned by TGEU surely do apply to Brazil. The very reputable and solid looking Brazilian LGBT advocacy group Grupo Gay da Bahia makes similar and consistent observations:
“For the database coordinator of this research, systems analyst Eduardo Michels of Rio de Janeiro, "the underreporting of these crimes is evident, indicating that those numbers represent only the tip of an iceberg of bloody violence, since our databank is based on articles published in newspapers and over the internet. Unfortunately, this type of information is seldom provided by the more than 300 active Brazilian LGBT nongovernmental organizations (NGOs). The real numbers of deaths of LGBT people in the country must certainly exceed the estimates herewith provided; this is especially true in more recent years since police officers and police inspectors have been increasingly and blatantly ruling out the possibility of homophobia being a factor in many of these killings of homosexuals."
With this caveat duly noted, the problem for the statistics on trans murders is that they still appear too small for us to be in any way certain that the trans murder rates are higher than the general rate for the whole population. Brazil is a huge country with many regional and demographic complexities I have not studied in any depth. But doing the math on the gross totals does not help the case for the idea that trans people are uniquely vulnerable to murder. Here are my calculations for what it’s worth:
The Trans Murder Monitoring Project contains this data for the 2008 – 2017 period:
The hard part is estimating the size of the trans population. Failing to find any solid estimates online, I will use the 0.6% of total population figure from a recent US survey. Census data for the population of Brazil over the 2008 – 2017 period gives a rough and ready population estimate of 200 million. This lets us guess the number of Trans Brazilians as 1,200,000.
Now for the homicide rate. Wikipedia says :
The 26.74 means homicides per 100,000 people, taken from 2015 data.
It also says:
So the 26.74 per 100,000 is a conservative estimate.
If we use these figures, the expected number of trans murders in one year in Brazil should be 26.74 * 12 = 320. That is using the conservative homicide rate. Using the TGEU data above, we get an average of 106 trans murders per year. This means that even if there were two unreported or unrecognised murders of trans people for every reported instance, the actual murder rate for trans people would be about the same as the rate for the population as a whole.
Another way of going about this is to use the aggregate LGBT population figures and make a similar comparison. This Al Jazeera article from 2015 states: that “the number of homophobic and transphobic killings in Brazil increased from 1,023 in 1995, to 1,243 in 2003, according to Brazil’s first gay rights group Grupo Gay da Bahia.”
There are stats on LGBT population as a whole for Brazil, which are probably a lot more reliable than my rough and ready estimate of the trans population.
20,000,000 LGBT people at a murder rate of 26.74 per 100,000 gives 200*26.74 = 5348 expected deaths. Again, this is using a conservative death rate estimate.
In this case for the LGBT murders to equal the national average rate, there would have to be around four unreported or unrecognised murders for every recorded instance. Brazil is a country riven by corruption, violence and prejudice. So it is surely conceivable, even likely, that there is a such a vast iceberg of unreported LGBT / trans murders. But the available reported data just does not support the thesis that trans or LGBT people are more susceptible to homicide. The number of murdered trans people in Brazil is high because Brazil has a large population and an extremely high homicide rate. There may be hundreds of unreported cases, and it could be that transphobia is a motive in many of these killings. But the data as reported simply does not support this speculation.