WS MoreOrLess: Are 95% of Terrorism Victims Muslim?

Summary

More or Less: Are 95% of Terrorism Victims Muslim?

BBC Radio 4 | Host: Tim Harford
Aired: January 19, 2015

---

Episode Overview

This episode investigates the often-cited claim that "95% of victims of terrorism are Muslims." Prompted by a listener's question after the Charlie Hebdo shooting, Tim Harford and his team dig into statistics, data sources, and the challenges of defining and quantifying terrorism victims—especially with regard to their religion. They speak with data experts and fact-checkers to unravel the nuance behind the numbers and explore both methodological and interpretive pitfalls.

---

Key Discussion Points and Insights

1. Origins of the Statistic (00:28 - 01:12)

• Listener Query: After the Paris attacks, a BBC interview with an imam referenced the figure that "95% of terrorism victims are Muslims."
• Producer Ruth Investigates: The number originates from a 2011 National Counterterrorism Center (NCTC) report, stating, "in cases where the religious affiliation of terrorism casualties could be determined, Muslims suffered between 82 and 97% of terrorism related fatalities over the past five years."
• Critical Caveat: The figure applies only "in cases where the religious affiliation...could be determined"—which may not represent the broader dataset.

2. How Terrorism is Defined (01:29 - 03:16)

• Global Terrorism Database (GTD): Producer Ruth turns to the GTD, the US Government's official source, run by the University of Maryland.
• Expert Interview: Erin Miller (criminologist/statistician, GTD) explains the difficulties, both emotional and methodological, in cataloging terrorist acts:

  “It can sort of do a number on your view of the world. But it's also fascinating and...important.”
  — Erin Miller (01:48)

• Definition Nuance: The GTD counts an attack as terrorism if it is:
  • Intentional, violent, or threatening
  • Perpetrated by non-governmental actors
  • Driven by political, economic, or religious goals
  • Intended to send a message to a broader audience, beyond immediate victims
  • Outside the context of legitimate warfare

3. Problems with Measuring Religion of Victims (03:24 - 04:22)

• Data Limitations: Media reports rarely state victims’ religious affiliations, complicating quantification.
• Country Focus: From 2004–2013, 50% of terrorist attacks and 60% of fatalities occurred in Iraq, Pakistan, and Afghanistan (majority Muslim countries), and 80–90% of attacks targeted domestic populations.

  "So the perpetrators aren’t attacking necessarily foreign entities in these countries."
  — Erin Miller (04:00)

4. Is 95% Accurate? (04:22 - 05:14)

• Expert Assessment: Erin Miller finds 95% "a bit high," but in the "right ballpark."
• Population Reflection: In countries where most attacks occur, the civilian population is mostly Muslim, so most victims are as well.
• Data Snapshot:
  • UK: 400 attacks (mostly non-lethal, Northern Ireland)
  • US: 131 attacks (fewer than 20 lethal)
  • France: 47 attacks
  • Iraq: 12,000 attacks (8,000 lethal) - (04:48)

5. Western Perceptions and Media Impact (05:14 - 06:20)

• Why the Statistic Surprises Westerners: Notable attacks like 9/11, 7/7 and Charlie Hebdo occurred in majority non-Muslim countries and dominated media coverage.

  "Those attacks were meant to grab attention and they succeeded. And while Muslims did die in them, most of the victims wouldn’t have been Muslim."
  — Tim Harford (05:40)

• Ethical Reflections: Erin Miller warns against turning victim tallies into "scorecards":

  "I think it’s a mistake to do that because these are conflicts that are sort of entrenched in deep geopolitical issues and religion is certainly a part of them, but it is not the only part."
  — Erin Miller (05:46)

6. Case Study: Boko Haram and Counting Casualties (06:20 - 08:59)

• Baga, Nigeria Massacre: The team notes the difficulties of establishing an accurate death toll after terrorist events when access and verification are limited.
• Fact-check Perspective: Julian Rademeyer (AfricaCheck) discusses wildly varying estimates (150-2,000 deaths), noting the impossibility of certainty without independent observers or verification:

  "The simple answer is that nobody knows...what is the scale? 2000 is a number that potentially at this point seems as if it may be too high. But again, we don’t really know."
  — Julian Rademeyer (06:59—08:59)

---

Notable Quotes & Memorable Moments

• On data limitations and media reports:

  “The tricky thing is Erin and her team rely on media reports, and she says they don’t often contain information about the religion of victims.”
  — Ruth (03:31)

• On defining terrorism:

  “Saying that something is terrorism or isn’t terrorism, I mean, this is something that politicians do all the time, purely for their own convenience.”
  — Tim Harford (02:13)

• On “scorecard” mentality:

  “It is tempting for many people to try and turn it into almost a scorecard...And I think it’s a mistake to do that...these are conflicts entrenched in deep geopolitical issues”—Erin Miller (05:46)

• On uncertainty in casualty figures:

  "Any numbers that are coming out of there are questionable...what is the scale? 2000 is a number that potentially at this point seems as if it may be too high. But again, we don’t really know...we’ll never know how many people have died up there."
  — Julian Rademeyer (07:59—08:59)

---

Important Timestamps

• 00:28 – Listener question introduces the claim (95%)
• 00:54 – Origin of the statistic from US government report
• 01:48 – Erin Miller describes working on the Global Terrorism Database
• 03:00 – Defining terrorism: criteria and examples
• 03:38 – Why religious data on victims is rarely available
• 04:00-04:22 – Data on attack locations and typical victims
• 04:22 – Erin Miller’s assessment of the statistic’s accuracy
• 04:48 – Concrete numbers: terrorism in UK/US/France vs. Iraq
• 05:40 – Discussion on how western perception is shaped
• 05:46 – Erin Miller on the error of turning fatalities into a religious “scorecard”
• 06:59 – Julian Rademeyer on uncertainty in Baga, Nigeria casualty figures

---

Conclusion: What Did We Learn?

The claim that "95% of terrorism victims are Muslims" is rooted in data, but comes with critical caveats:

• The actual proportion is difficult to confirm and subject to limitations, primarily because the religious affiliation of victims is often undocumented.
• Most terrorist attacks and fatalities occur in Muslim-majority countries, targeting local (often Muslim) populations.
• Well-publicized Western attacks, while horrific, do not reflect the majority of worldwide terrorism casualties.
• Attempting to reduce complex tragedies to simple numbers or "scorecards" erases underlying geopolitical, social, and contextual factors.

Final Thought:
Statistics on terrorism must be approached with caution and context; the numbers rarely tell the whole story, and the desire for simple narratives often misleads rather than informs.

Transcript

[A]

This is the short edition of More or Less first broadcast on the BBC World Service.

[B]

Thank you for downloading from the BBC. For details of our complete range of podcasts and our terms of use, go to bbcworldservice.com podcasts.

[C]

You're listening to more or less on the BBC World Service and I'm Tim Harford, loyal listener. Mark has been in touch with a question.

[D]

After the Charlie Hebdo shooting, I heard an imam interviewed on the World Service and one of the things that struck me was that he said that 95%

[B]

of victims of terrorism are Muslims, Unfortunately.

[D]

Is this the sort of statistic that you might be interested in investigating?

[C]

Well, it is, Mark, and thank you. We started to look to find out where this number comes from. Ruth. Ruth is our producer. Ruth, you found it in a US government document.

[A]

Yes, a report from the National Counterterrorism centre published in 2011. And it said that in cases where the religious affiliation of terrorism casualties could be determined, Muslims suffered between 82 and 97% of terrorism related fatalities over the past five years.

[C]

So the imam in Paris wasn't far off then?

[A]

Well, the problem is that qualifier in cases where the religious affiliation of terrorism casualties could be determined, we don't know whether that's most cases or just a few. And we don't know whether they're representative of other attacks or not.

[C]

Ruth, of course, does not give up easily.

[A]

I don't. So I went to the Global Terrorism Database, which has been going for 40 years and is now the official source of data for the US government. It's run by a group at Maryland University. And let's just stop for a minute to think about what their job involves because it is a bit depressing.

[E]

It can be trying, right? I mean, essentially we spend all day reading articles about devastating violence and it can sort of do a number on your view of the world. But it's also fascinating and it's also something that we think is important.

[A]

That's Erin Miller, a criminologist and statistician at the Global Terrorism Database. Every day she and her colleagues sift through thousands of media reports to determine where attacks have been happening.

[C]

But wait, what counts as terrorism? I mean, saying that something is terrorism or isn't terrorism, I mean, this is something that politicians do all the time, purely for their own convenience.

[A]

It is. So the Global Terrorism Database has a particular methodology. They label and attack terrorism and put it into this database if it's intentional, violent or threatening and the perpetrators aren't governments.

[C]

So the alleged hack by North Korea of Sony Pictures.

[A]

If it was in fact by North Korea, it wouldn't count because that would be a state act. And anyway it's non violent.

[C]

Ok, and what else? Because if someone pulls a knife on me and demands money, well, that's an intentional, violent, non governmental act. So there must be some other criteria that go into the definition.

[A]

That's right. There are other criteria, other considerations. Is there a political, economic or religious goal? Is one, is it an attempt to send a message to a wider audience, not just to the direct victims? Or is it outside the context of legitimate warfare activities?

[C]

So all this goes into making up the definition of terrorism. Let's go back to the claim that 95% of terrorism victims are Muslim.

[E]

It's a really interesting question. It's very difficult to know for sure and to put a specific number on it like that.

[A]

The tricky thing is Erin and her team rely on media reports, and she says they don't often contain information about the religion of victims.

[E]

But we do have some information that could be useful for considering this type of question. So take for example, the most recent 10 years of data that we have from 2004 to 2013. The top three countries, the three countries with the most attacks in the world, actually comprise about 50% of all attacks and 60% of all deaths from terrorist attacks, and they are Iraq, Pakistan and Afghanistan. So they're majority Muslim countries. And we know that about 80 to 90% of them are domestic targets. So the nationality of the target is the same as the nationality where the attack took place. So the perpetrators aren't attacking necessarily foreign entities in these countries.

[C]

So Does Erin think 95% is right?

[A]

Well, she thinks it's a bit high, actually, but that it's probably in the right ballpark. The story then is that most terrorist attacks affect domestic targets, and so casualties will to some extent reflect the makeup of the civilian population.

[C]

And since the hotspots for terrorist attacks are majority Muslim countries, it makes sense to say that Muslims are both the most likely attackers and the most likely victims.

[A]

And just to put the latest available numbers on this, over the 10 years from 2004, the UK suffered 400 terrorist attacks, mostly in Northern Ireland, and almost all of them non lethal. The U.S. suffered 131 attacks. Fewer than 20 of them were lethal. France suffered 47. But in Iraq, there were 12,000 attacks and 8,000 of them were lethal.

[C]

But that original claim that 95% of the victims of terrorism are Muslims, that claim is surprising to Westerners because when they think of terrorist attacks, they think of Charlie Hebdo, of the 77 London tube and bus bombs, the Madrid train bombs, and of course 9, 11. Those attacks were meant to grab attention and they succeeded. And while Muslims did die in them, most of the victims wouldn't have been Muslim.

[A]

Yes, although Erin Miller is a bit dismayed that we find the religious affiliation of victims so fascinating.

[E]

It is, I think, tempting for many people to try and turn it into almost a scorecard, trying to figure out, you know, which religious groups are more violent than the other religious groups and boil it down to this sort of grossly oversimplified keeping of score like, like it's a football game. And I think it's a mistake to do that because these are conflicts that are sort of entrenched in deep geopolitical issues and religion is certainly a part of them, but it is not the only part. And sort of boiling it down to that identity I think is somewhat mistaken.

[C]

Our thanks to Erin Miller from the Global Terrorism Database for taking us through the numbers. The work Erin Miller does isn't easy. The recent attack in Baga, Nigeria by Boko Haram shows you why. Although satellite images have now been taken showing widespread devastation across two towns where thousands of people lived in, it's still not clear how many people have been killed. The BBC has reported that estimates vary from 150 to 2000 deaths. Julian Rademeyer from fact checking website AfricaCheck has been looking into this. We asked him what he thought the true number of casualties might be.

[F]

The simple answer is that nobody knows. Something clearly quite catastrophic has happened there, given the eyewitness accounts. Survivors who've escaped the town have spoken about people being slaughtered like insects. One man told CNN today that he walked along an eight kilometer stretch of road and it had been littered with bodies. People have spoken about running for their lives and the refugees themselves have come up with various estimates. One man saying that he believed as many as 500 people had been killed. He'd seen several groups of villagers that had been killed at least, and the smallest group was five people. Other accounts as high as 3,000. The numbers are all over the place. What happened was that it gained. The initial reports of 2000 stemmed from a local government official who was interviewed by journalists shortly after these massacres were first reported and he listed the figure of 2,000. He wasn't in the town itself, he was much further away, but he based it on the reports of survivors. The Nigerian military initially agreed that certainly that this would have been one of the deadliest acts committed by Boko Haram. But more recently, they've tried to downplay the figures, saying that that the numbers of dead have not exceeded 150, including Boko Haram militants and soldiers and civilians. And the bottom line at this point is that there are no independent journalists, there are no independent monitors gaining access to that area. It's still under the control of Boko Haram. Any numbers that are coming out of there are questionable. And you know, the eyewitness accounts, again, these are people who are fleeing for their lives. They're not counting bodies. They stay relating stories of what they've actually seen. So we certainly know that something truly horrific has happened there. But, you know, what is the scale? 2000 is a number that potentially at this point seems as if it may be too high. But again, we don't really know. And in some ways it's probably like it. We'll never know how many people have died up there.

[C]

Julian Rademeyer of AfricaCheck speaking to us from Johannesburg. That's all we have time for this week. Please keep your questions and your comments coming in to more or lessbc code. Our website is BBC.com more or less.

[B]

There are dozens of different podcasts now available from the BBC, including news, documentaries, science, business, arts and sports. The details of them all go to bbcworldservice.com podcasts.