88

The Economist isn't a fan of Donald Trump as a presidential candidate.

In an analysis of his ascendence to the Republican nomination they try to discern some of the factors. As part of their list of factors driving Trump support they argue that terrorism has become a national bogeyman while quoting an odd statistic (my emphasis):

Terrorism—though it claimed fewer American lives last year than toddlers with guns—has become a national bogeyman.

This isn't an appropriate place to discuss the relative merits of Trump, but that is one unexpected statistic.

Is it true? Were there really more American deaths from gun-toting toddlers than from terrorism in 2015?

NB This isn't a question about politics. This is merely the context in which the statistic was quoted. The question is whether the statistic is true. Please keep the politics out of any answers.

matt_black
  • 56,186
  • 16
  • 175
  • 373
  • 23
    I worry this will come down to defining terrorism. We all agree the San Bernadino shootings was terrorism, but many claim the the Colorado Planned Parenthood shooting was not. I'm not entirely sure this can be answered apolitically, since (as the Economist points out) terrorism is becoming a political term. – Will May 08 '16 at 13:15
  • 12
    @Will There might not be a definitive answer, but we could give the answer using a range of definitions of terrorism. One classification would be *international* terrorism (excluding purely domestic issues such as the planned parenthood one). Another would include all acts designed to spread fear. Providing context would tell the broader facts without prejudicing the answer by political choice. – matt_black May 08 '16 at 16:03
  • 15
    Which category are shootings performed by terrorist toddlers counted in? – Knells May 08 '16 at 22:56
  • I won't make this an answer, but it seems true, depending on what your definition of terrorism is. The real problem is there is no solid definition of terrorism. But even in Tim's answer which seems the most comprehensive, the widest interpretation is only off by 10. – coteyr May 08 '16 at 23:49
  • 49
    If you don't want to talk about Trump, then [don't mention him in four out of your five paragraphs](http://skeptics.stackexchange.com/review/suggested-edits/22043)! – jscs May 09 '16 at 05:14
  • 1
    @JoshCaswell Until your comment we'd done a good job of avoiding the trump issue. – matt_black May 09 '16 at 07:16
  • 3
    For what it's worth, the version of the statistic I read was that more people were *shot* (no "killed" qualifier) by toddlers than were *killed* by terrorists (there may have been an "Islamic" qualifier, I don't recall). I don't know which statistic came first, if one of them was a bastardization of the other, nor the veracity of either. – Jason May 09 '16 at 13:29
  • Terrorism is often defined as the use of violence by a non-state actor to try to cause political change. Hence shooting a doctor at an abortion clinic would definitely be terrorism. – Paul Johnson May 23 '16 at 20:58
  • Related: [In 2016, Toddlers Have Shot More People in the US Than Muslim Terrorists Have](https://mic.com/articles/142348/in-2016-toddlers-have-killed-more-people-in-the-us-than-muslim-terrorists-have#.F4G1iPVmQ) Not the same claim as the question - 2016, not 2015. Shot, not killed. Muslims terrorists, not general terrorists. People in the US, not Americans. – Oddthinking May 24 '16 at 03:46
  • @Oddthinking That makes me wonder whether the Economist was misquoting the claim. – matt_black May 24 '16 at 08:15

2 Answers2

131

It depends on your definition of terrorism and your interpretation of the quote: is it about American lives anywhere in the world, or about death in America including non-Americans. For the purpose of this answer I'll assume the second interpretation.

If terrorism only includes Islamic terrorism, it would be true, although quite close. See for example the Snopes article on this question:

In 19 instances a toddler shot and killed themselves, and in two others, the toddler shot and killed another individual. That brings the total of toddler-involved shooting deaths in the United States in 2015 to 21.

By contrast, if we count both the Chattanooga shootings and San Bernardino as strictly terrorism, 14 Americans were killed in San Bernardino and five in Chattanooga. As such, 19 Americans were killed [...] in instances of Islamic terrorism in 2015.

Ultimately, even the broadest leeway led to the same mathematical conclusion. The meme was basically correct; more Americans were shot and killed by toddlers in 2015 than were killed by Islamic terrorists.

But your question isn't really about Islamic terrorism, but terrorism in general.

Inclusion of only two other events provides insight: the Charleston church shooting (white supremacist terrorism, 9 dead) and the Colorado Springs Planned Parenthood shooting (Christian terrorism, 3 dead) may be included, leading to 31 victims of terror attacks in the US in 2015, compared to 21 victims of toddlers.

This would make the claim as stated false.

tniles
  • 102
  • 8
tim
  • 51,356
  • 19
  • 207
  • 177
  • 16
    I’m not sure why you say it’s “close” (implying it’s *almost* correct): counting only Muslim terrorists, it *is* correct. For what it’s worth, other sources have explicitly referred to Muslim terrorists. I therefore assume that the omission in the economist was an error they introduced when copying the statistic. – Konrad Rudolph May 08 '16 at 18:17
  • 5
    @KonradRudolph Yes, I meant in that case it would be correct (but close; the numbers are almost equal). But that's really not the claim. Of course it's a question of definitions, but no reputable source defines terrorism as soley being islamic terrorism, so in it's stated form, it's wrong (but still somewhat close, which I would imagine is the whole point of people comparing these statistics). – tim May 08 '16 at 18:24
  • 14
    When I heard the claim about toddlers claiming American lives, I didn't consider the possibility of toddlers killing themselves. – Andrew Grimm May 08 '16 at 21:00
  • 5
    Americans were also killed by terrorist attacks in other countries, such as the ones in Paris as well as in Middle Eastern countries, such as Syria. So, even excluding military, the claim is still false if you include all Americans and not just those on American soil at the time. – reirab May 08 '16 at 22:07
  • 15
    @reirab "false if you include all Americans and not just those on American soil at the time" - Only if we assume that no Americans are shot by toddlers while abroad. – Knells May 08 '16 at 22:54
  • 1
    Removed a long debate about which deaths should be counted. Since that's an opinion, and not something that can ever help improve this answer, let's not debate it further in comments. – Sklivvz May 10 '16 at 08:41
  • 3
    @reirab - being killed by IS in Syria isn't exactly the same though, that's not a threat to "America" that's a threat to people in a warzone – Jon Story May 10 '16 at 14:45
  • 1
    I like this answer as it gives us the key information we need to judge several possible versions of the claim. And it does so clearly and without too much irrelevant discussion. – matt_black May 10 '16 at 15:47
  • "Statistics can prove anything" -- http://www.quotegarden.com/statistics.html . The big take away from a Q&A like this is that a keen observation (even a questioning) of assumptions, presumptions, and definitions can generate significant differences. – tniles May 11 '16 at 17:25
18

This is a start at an answer; better sources could help. This will be made into a Community Wiki so feel free to improve.

The source of the "toddlers with guns" claim might be a New York Times article dated May 5, 2016, entitled "One Week in April, Four Toddlers Shot and Killed Themselves". It contains the claim:

Last year, at least 30 people were killed in accidental shootings in which the shooter was 5 or younger, according to Everytown For Gun Safety, a gun control advocacy group that tracks these shootings, largely through news reports.

The Everytown website has a list of unintentional shootings by children 17 or under, each accompanied by a link to a news report. However, the list doesn't distinguish between fatal and non-fatal shootings, and moreover, it lists the ages of victims, not shooters. (In many cases, of course, the victim and shooter were the same person; i.e. children who accidentally shot themselves.) So we can't extract the desired information directly from that list.

Of course, the exact count also depends on how you define "toddler"; the "5 or younger" definition is arbitrary.

Wikipedia has a list of terrorist incidents in 2015. 5 incidents took place within the US, causing a total of 31 fatalities (not counting perpetrators):

Wikipedia is not a great source, but it should be a start, and the incidents themselves are well documented. In any case, the determination of whether a given incident should be classified as "terrorism" will be somewhat subjective, but Wikipedia's classification seems reasonable to me.

Community
  • 1
Nate Eldredge
  • 9,275
  • 5
  • 46
  • 48
  • It's also worth noting that the claim in the question specifically refers to the number of American lives taken, not to the number of lives taken _on American soil_. As such, you'd also need to include the Paris attacks (which claimed at least 1 American life,) as well as other terrorist attacks around the world. – reirab May 08 '16 at 22:14
  • 8
    @reirab as well as other toddler attacks around the world, otherwise results are inevitably skewed. – Knells May 08 '16 at 22:59
  • If perpetrators were not counted the toddler attacks would drop to a low figure. – KalleMP May 09 '16 at 05:47
  • 5
    @KalleMP I think perpetrators are excluded in an attempt to count the "victims" of an incident. I think everyone would consider toddlers that killed themselves victims, but the waters are a bit murkier in the case of deliberate shootings by adults. – Will Vousden May 09 '16 at 11:04
  • https://www.washingtonpost.com/news/wonk/wp/2015/10/14/people-are-getting-shot-by-toddlers-on-a-weekly-basis-this-year/ – DavePhD May 09 '16 at 18:15
  • 2
    Wikipedia defines toddlers between the ages of 1 and 3. CDC defines toddlers as ages 1 and 2. – Mark Lakata May 09 '16 at 21:48
  • 6
    It's quite rare for other countries' toddlers to even access guns. I guess if you count the Terrorist's toddlers too, then uh... I have no idea what those stats are. – Nelson May 10 '16 at 06:05
  • 1
    @WillVousden I'm not trying to discount the needless suffering. It is just pointless to say perpetrators excluded on one side of a statistic and included on the other side, sort of like "Lies, lies and statistics" double speak. The goal I think here is can they make something trivial by comparing it to something else that barely relates. Do we try eradicate medical care because more people die as a result than due to terrorism, actually we see less is spent on repairing it than on armaments (thinking USA here). This meme can be made to read correctly if we define or select words to suit it. – KalleMP May 10 '16 at 18:09