A recent literature review by Doctor Clare Allely, at the University of Glasgow has been attracting a lot media attention. According to Google News there are press reports in Britain, Europe, America and China, with more to come no doubt.
The Washington Post proclaims a “Significant’ statistical link between mass murder and autism, brain injury” in its headline. The Daily Mail has “Recipe for a serial killer? Childhood abuse, autism and head injuries are more common in murderers, study claims.”
In contrast, the paper itself, Neurodevelopmental and psychosocial risk factors in serial killers and mass murderers and the accompanying press release from Glasgow University are a model of academic restraint. Dr Allely is quoted as saying
“It is crucial to note that we are not trying to suggest that individuals with ASD or previous head trauma are more likely to be serial killers or commit serious crime. Rather we are suggesting that there may be a subgroup of individuals within these groups who may be more likely to commit serious crimes when exposed to certain psychosocial stressors.
“Research on mass and serial killing is still very much in its infancy. New research is urgently required to understand the mechanisms underlying these extreme forms of violence so that preventative strategies can be developed. We would recommend that in future, all serial or mass killers who are apprehended should be thoroughly assessed using standardised tools for investigating neurodevelopmental disorders including ASD and head injury.”
From the paper’s abstract:
“Our findings tentatively indicate that these extreme forms of violence may be a result of a highly complex interaction of biological, psychological and sociological factors and that, potentially, a significant proportion of mass or serial killers may have had neurodevelopmental disorders such as autism spectrum disorder or head injury. Research into multiple and serial murders is in its infancy: there is a lack of rigorous studies and most of the literature is anecdotal and speculative.”
Unfortunately, however nuanced the language, it cannot hide the fact that this paper is seriously flawed.
Others have commented on the flawed methodology, the over reliance on the internet, books and news reports as sources of data because there are so few peer reviewed studies to call on. (See for example John Elder Robison) Then there is the conflation of serial killers and spree killers, two distinct categories with different psychological profiles. Most striking is that the serial killer takes care to select their victims and tries to evade detection. The spree killer ‘selects’ victims at random and their killing spree is usually the prelude to suicide. They make no plans to avoid detection. These two categories appear to have been chosen because the perpetrators are mass murderers who have each killed at least three people. Other murderers were excluded from the analysis. If we assume for the moment that data was reliably collected this was a missed opportunity to ascertain whether autistic people are over represented in all murder statistics and not just mass murders.
But was the data reliably collected? The headline figure is that 67 out of 239 mass murderers were autistic (28.03%). But only six of these had “a definite diagnosis of ASD.” Twenty one were “highly suspected” of being autistic and thirty three were “possible/probable” autistics. Hang on. Do the math. 6+21+33=60. So what about the other seven? This is a really basic error. It should have been corrected during editing but it was not. The authors also state that all foreign language (non English) documents were excluded from their search and then go on to quote a Norwegian language website as the primary evidence for three of their ”definite” autistics, when, in fact, it only reports that Anders Breivik, the Norwegian mass murderer, might be autistic.
So what about these six definite diagnoses of autism?
Martin Bryant went on a killing spree in Australia in 1996. He had learning difficulties and may have had autistic traits. But a diagnosis of Asperger Syndrome was ruled out by a forensic psychiatrist hired by his defence lawyers. My thanks to Paula C. Durbin-Westby for the link to the full psychiatric assessment. Having read it I think that he may have met modern criteria for ASD. But we will never know and I do not understand the basis on which Allely et al. assign Bryant a “definite diagnosis.”
Robert Napper is a serial killer and rapist who was detained indefinitely at Broadmoor, a UK facility for the criminally insane, in 2008. Napper is a paranoid schizophrenic who witnessed his father physically abusing his mother until their divorce when he was ten. He was sexually assaulted when he was thirteen by a family friend and underwent a personality change. He was diagnosed with Asperger Syndrome after his arrest. Allely et al. rely on Murderpedia.org for this information.
Wolfgang Zaugg is referenced to the Norwegian language website, dagbladett.no which does not refer to him at all. A Google search found an entry on Murderpedia which reveals him as an immigrant to Sweden who experienced racism as a child. As an adult he became a Swedish citizen and an accomplished businessman. When his business collapsed and he could not pay his gambling debts he financed his lifestyle through bank robberies. He was involved with far right terrorists and embarked on a killing campaign against other immigrants, perhaps to prove his Swedishness. There is no mention of autism.
Nicky Reilly is also referenced to the Norwegian language website, dagbladett.no which does not refer to him at all. He is autistic. He has a diagnosis of Asperger’s Syndrome. But he never killed anybody. He converted to Islam and was recruited as a suicide bomber by extremists who exploited his autism. The only person hurt in his suicide bomb attempt was himself. Nicky survived and is serving an eighteen year sentence for terrorism. Even if he had been successful he is not your typical serial killer or spree killer.
Ragnar Nilsson is also referenced to the Norwegian language website, dagbladett.no which does not refer to him at all. There is no English reference to him in Google searches. My limited Swedish leads me to suspect that he is the triple murderer in this story.
Cary Stayner claims he was sexually abused by an uncle when was eleven. This was when his brother was abducted by a paedophile and kept a prisoner for seven years. Cary was diagnosed with OCD, psychotic disorder and schizophrenia following his arrest for the murder of four women. According to Murderpedia there was a history of mental illness and sexual abuse in the family. But no mention of autism.
These are the six “definite diagnoses of autism.” We can rely on two, maybe three of the diagnoses. What about the cases where “ASD was highly suspected?” Top of the list is Seung-Hui Cho, who carried out the Virginia Tech massacre in 2007. Lee, Lee, and Ng (2007), the best resource listed for Cho regards any diagnosis as “speculative.”
“So we ask if he met criteria for a diagnosis based on Diagnostic and Statistical Manual of Mental Disorders-IV (DSM-IV) or International Classification of Disease-10 (ICD-10)? Or was he of a criminal antisocial or psychopathic mind? Unfortunately without having interviewed him or having access to his records, we cannot say for sure. We could speculate that he was depressed with delusional thoughts, and perhaps had undiagnosed Asperger's disorder (a mild variant of autism), or was taking illicit substances. But we do not have enough evidence to be certain of a definitive psychiatric condition that could account for his extremely violent behaviour.”
Jeffrey Dahmer is also included. He even gets his own case study. Yet Dahmer was subject to rigorous forensic psychiatric assessment that failed entirely to mention any putative autistic diagnosis. As Emily Willingham points out.
“According to mental health professionals who personally diagnosed serial killer Jeffrey Dahmer, he had a personality disorder. That didn’t stop the authors of a recent paper attempting to link autism and mass murderers, serial killers, and other homicidal maniacs from listing Dahmer as “highly suspected” of having an autism spectrum disorder, along with 61 other people who were never diagnosed with one, including Timothy McVeigh, Terry Nichols, and Dylan Klebold.”
Let us call a halt. If only two out of six ”definite diagnoses” can be confirmed what chance is there that any of the probable or possible diagnoses will stand up to scrutiny? There are compelling factors to do with family histories of abuse and mental illness that are sufficient to explain why these people became mass murderers without recourse to a retrospective diagnosis of autism.
I am also curious as to why the authors examined head injury in tandem with autism. Autism is neurodevelopmental. It is a result of atypical brain development in utero caused by an, as yet unknown, complex set of interactions between genes and environment. These create a predisposition to certain kinds of behaviour in which some of the determining parameters are intelligence and language development. The ameliorating factors include early identification and the quality of early intervention. Brain injury usually results from brute force trauma later in life. It can undo or delay development. But its relation to autism is never explained by the authors.
In conclusion, the authors found two definite cases of autism in 239 cases of mass murder. This is consistent with the rate of autism in the general population. For some reason they included another fifty eight possible cases that do not stand up to scrutiny. Then they get the maths wrong and proclaim their sixty as sixty seven. And from this we are led to the conclusion that there is a possible subset of autistic people who, given the right amount of psycho-social stressors might turn out to become mass murderers, So we should test all future mass murderers for autism, as the authors suggest, in the hope of identifying this subset and curing them? And should we be curing them of autism or of the propensity to become mass murderers? If so, how?
Having read more than I ever wanted to about serial killers and mass murderers in preparing this blog post I would suggest a simpler solution. Let us work together to reduce the number of children who are bullied and abused, especially those who suffer because their neurological difference is misunderstood. In this way we will increase the sum of human happiness and decrease the likelihood of future serial killers and mass murderers, whatever their neurology.