On Sunday, March 22nd of this year, a large swath of the population in Quebec was woken up at 4:25 as cell phones lit up and screamed. An Amber alert had been broadcast. Less than听, the two missing children were thankfully found, unharmed, and the alert was cancelled.
Amber alerts cast wide nets in an effort to find abducted children and teenagers, but each time one is transmitted in Canada, we go through a predictable cycle. Many complain that these alerts cannot be turned off or muted, even in sleep mode, and that waking people up in the middle of the night to help them find a specific car that was last spotted 200 kilometres away is unnecessarily disruptive鈥攚ith some even calling 911 to complain. Meanwhile, law enforcement officials remind us through the media that these alerts are only used when necessary and that they save lives.
Any subject tied to the safety of children is bound to evoke the deepest emotions. In fact, the crime that led to the creation of Amber alerts鈥攁nd to their name鈥攚as monstrous enough that it鈥檚 no wonder it rallied influential people into building a system that could potentially prevent this kind of crime.
But warranted emotions aside, do Amber alerts even work? What little data we do have point to a system that has likely been oversold to us.
Every missing child does not get an Amber alert
Her name was Amber Hagerman, and in January 1996 she was riding her bicycle two blocks from her grandparents鈥� home in a working-class neighbourhood of Arlington, Texas. 鈥淭he streets are always full of kids playing,鈥� a neighbour told听The Associated Press,听飞丑颈肠丑听. 鈥淭hey aren鈥檛 now.鈥�
Amber was grabbed by a man in a pickup truck, according to a witness, and she vanished. She was 9 years old.听
Her body was found听听in a creek bed by a man who was walking his dog. Her throat had been cut. To this day, her murder remains听.
It鈥檚 easy to understand how such an unspeakable crime catalyzed change. It was a fellow Texan who contacted a local radio station and proposed that broadcasts could be used to recruit ordinary folks into finding missing children. Eventually, the idea snowballed into the Amber alert system in the United States鈥攁s well as the national sex offender registry. The word 鈥淎mber鈥� (often spelled 鈥淎MBER鈥�) is used in honour of Hagerman, but it has retroactively been turned into an acronym for 鈥淎merica鈥檚 Missing: Broadcast Emergency Response.鈥�
The idea migrated to other countries and to media platforms beyond just radio stations: televisions, electronic signs on the highway, and eventually cell phones.
Amber alerts are only used for a small minority of child disappearances. In fact, an听听revealed that an Amber alert had been broadcast for only one out of every one thousand missing children in 2021. Parents whose child has disappeared may wonder why theirs doesn鈥檛 warrant an Amber alert, but the criteria, which vary slightly from one jurisdiction to the next, are narrow: the victim must be under the age of 18 and be known (or strongly suspected) of having been abducted. There must be sufficient information about the victim, captor or the vehicle used, and often there has to be a risk of serious injury or death. Expanding this to any child that goes missing would result in a daily influx of alerts and undoubtedly cause alert fatigue: people would stop even looking at them.
Agencies in charge of Amber alerts will often tout very high success rates; these victories, however, can be built on a fallacy. If a child is safely found听补蹿迟别谤听an Amber alert, it can too easily be argued that the child was found听产别肠补耻蝉别听of the Amber alert. It鈥檚 important to remind ourselves that police officers do not sit idly by as soon as an Amber alert is declared. They continue investigating (possibly with renewed gusto inspired by the urgency of the alert), and this mundane police work can lead to the localization of the missing child.
You may wonder how easy it can be to identify the right stranger that kidnapped a child without recruiting thousands of pairs of eyes. After all, Amber Hagerman鈥檚 abductor was never found. The truth is that the majority of child abductions are done by a parent or relative. What little Canadian data we have on Amber alerts鈥攖his lot coming from听听apparently published in 2016鈥攔eveals that 54 of 92 Canadian Amber alerts that year involved kidnapping by a parent or relative, with an additional 11 implicating instead a non-relative who was known to the family or to the child. How many were caused by strangers? Twenty-two, or just under a quarter.
The situation is similar in the United States, where one researcher and his team have single-handedly been poring over Amber alerts, trying to suss out what is fact and what is fiction.
Success = 1 in 4
听If you dive into the meager scientific literature on Amber alerts, you will recognize one name pretty quickly. Timothy 鈥淪kip鈥� Griffin is an associate professor at the University of Nevada, Reno, and he has published almost every paper out there interrogating the effectiveness of Amber alerts.
In the U.S., just like in Canada, 鈥渟tranger danger鈥� type abductions are a minority. A child is much,听much听more likely to be taken by one of their parents in the middle of a messy custody dispute than to be abducted by a child murderer.
As for the percentage of Amber alerts in the U.S. that played a role in the successful recovery of the missing child, Griffin鈥檚 research, based on publicly available information, shows听听and听听that the number hovers around 25%. That鈥檚 one in four. Three quarters of Amber alerts do not appear to help. That isn鈥檛 to say that, in those cases, the child is never found; 补听听published a decade ago and looking at 448 Amber alerts in the U.S. revealed that over 95% of the children had been recovered alive and nearly 90% recovered alive听and听without physical harm, sexual abuse, of withholding of needed medical care during the abduction. Even when Amber alerts don鈥檛 trigger a helpful tip, the child is usually found.
But because a child鈥檚 safety is at risk, we may argue that even a low success rate is acceptable if it prevents a child from being murdered. The twist is that the issuance of Amber alerts is often too slow to thwart these very rare abominations. According to听听from the Washington Office of the Attorney General and the U.S. Department of Justice, three quarters of the children who are both abducted and murdered are killed within the first three hours of the kidnapping. That earlier听USA Today听investigation revealed that in 2021 less than a third of Amber alerts had been triggered within three hours of the child going missing. In fact, on average, children who are recovered after what police deems a 鈥渟uccessful Amber alert鈥� are found听. If the priority is to prevent the death of a kidnapped child, Amber alerts would need to be issued much faster.
I asked Captain Benoit Richard of the听S没ret茅 du Qu茅bec, our provincial police, about the effectiveness of Amber alerts in Quebec, since American data may not represent our own situation. He told me he was unable to answer my question as the SQ did not have studies or statistics readily available on this question. However, he was present during the last three Amber alerts and told me that direct citizen action had led to the recovery of the child in all three cases.
In Professor Griffin鈥檚 published work, he argues that Amber alerts share characteristics with 鈥渃rime control theatre,鈥� where law enforcement agencies generate the appearance of controlling crime while failing to do so. In听听with the officer who manages the Province of Ontario鈥檚 Amber alerts, the latter countered Griffin鈥檚 arguments by pointing out that Amber alerts are about more than preventing murders: they鈥檙e aiming to prevent or minimize any criminal harm to the child, and assessing the captor鈥檚 intentions as the situation is unfolding is often impossible. 鈥淲e can鈥檛 waste time on what-ifs,鈥� he said.
There have also been cases where the abductor, seeing their face on the Amber alert, decided to surrender to the cops, thus bringing the child back to safety. But one could also easily imagine a similar alert turning a domestic dispute over custody into a high-stakes, traumatizing public hunt.
Examples can be marshalled to back either position, but in an emergency, predictions fly out the window. And while we can look at the disappointing data emanating from the research literature, nobody wants to be the politician who suggests that money is being wasted on preventing the killings of children.
If Amber alerts only work some of the time, is there a better way of deploying them without irritating a segment of the population?
The screaming phone quandary
In the United States, a person听听Amber alerts in their cell phone鈥檚 settings; in Canada, this is not possible. And when an alert is broadcast, we are treated to a shrill, screeching听, each combining听, blasting from our phones for eight seconds.
Following public complaints, the CRTC, which developed our national public alerting system with government and the media,听听ten cell phones from five major manufacturers in 2020 to see how they behaved when an Amber alert was sent. The only way to prevent the deafening sound regardless of the brand of the phone was to switch it to 鈥淎irplane鈥� mode; this also blocks any phone call from coming in, which is not an option for people caring for vulnerable relatives. Some of the phones tested blocked the sound in 鈥淒o Not Disturb鈥� mode, while others did not. Many of the phones produced alerts above 90 decibels when the phone had previously been set to a 鈥渕iddle鈥� volume (halfway between muted and the loudest it can be). Ninety decibels is the equivalent of a subway train or a shouted conversation. It听will听wake you up.
I鈥檓 left with one question: why do Amber alerts on our phones need to scream at us? Why antagonize a large portion of the public when you need their help to find a child in need? A听听conducted by the CRTC revealed that roughly one out of every seven Canadians wants the option of turning these alerts off completely from their phones. Might they change their mind if the alert was silent or only accompanied by the ping of a text message? Why wake people up like there is a tornado warning when a person who is not out and about is unlikely to help spot a car?
I asked about the reason for the mandatory alarm bell to the organizations responsible for the broadcasting of Amber alerts in Quebec and Canada. The CRTC, which works at the national level, told me that Public Safety Canada was better placed to answer my question. (PSC has not answered my request at the time of publication.) Provincially, the Ministry of Public Security did not directly answer my question, instead letting me know (as the CRTC found out a few years ago) that cell phones differ in whether or not they allow you to block this sound.
Cell phone manufacturers should also be questioned. In the CRTC鈥檚 testing, neither the iPhone 7 nor the iPhone 8 blocked the sound in 鈥淒o Not Disturb鈥� mode. As the iPhone is听, I reached out to Apple but have yet to hear back.
鈥淲here鈥檚 my Amber alert?鈥�
It is difficult to judge how effective Amber alerts really are and if they are tailored correctly, seeing as they are universally understudied and that different jurisdictions can use slightly different criteria to trigger them. It would be wrong to say that they never help find a child: we have many reports of Amber alerts helping in individual cases. The claim that they regularly save lives, however, seems exaggerated. Most child abductees are kidnapped by a parent or relative who is highly unlikely to end their lives. Some abductions that trigger an Amber alert are incidental, as a thief steals a car and realizes there鈥檚 a backseat passenger. And because Amber alert eligibility goes up to age 18, some are borderline runaway teenagers presumed, by their parents, of having been kidnapped by their romantic partner.
Overselling the benefits of an Amber alert can backfire. As Professor Griffin argued in his debate, a silver bullet withheld from a parent will appear cruel. 鈥淚 automatically get my kid back if there鈥檚 an Amber alert,鈥� one might reason, 鈥渟o where鈥檚 my Amber alert?鈥�
It鈥檚 a clich茅 to conclude that 鈥渕ore studies are needed,鈥� but given the paucity of data on Amber alerts鈥攚ith the vast majority coming from a single researcher鈥檚 team鈥攚e do need more research.
As for the blaring siren that commandeers our phones when such an alert is launched, I have to ask: is it needed?
Take-home message:
- Amber alerts are not sent for every missing child; they are mostly limited to abductions where the child or teenager鈥檚 safety is thought to be at serious risk.
- The effectiveness of Amber alerts has not been studied much, but the data we have now indicate that only about a quarter of them have a direct effect on the child being found.
- In Canada, Amber alerts on cell phones produce a loud, eight-second alarm noise, which had led to complaints. While some brands of cell phones will block the sound in 鈥淒o Not Disturb鈥� mode, others do not.
