51

Recently this map of terrorist attacks in Europe went viral and was reposted many times in social media. Does it present actual data about terrorist attacks?

Map of terrorist attacks

Background: The map gained popularity because of ongoing discussion about immigrants. Poland firmly opposes immigration from Muslim countries (North Africa, Middle East) and obligatory limits of immigrants European Commission want to impose. This country is also a big gap on this map. Some nationalists claim that there is no terrorism in Poland, because of their immigration policy.

Examples of usage:

Gallifreyan
  • 109
  • 5
GSPdibbler
  • 611
  • 1
  • 5
  • 8
  • 46
    The vast majority of those dots are in Asia/Africa. The dots seem to cluster in areas where there is well established conflict going on (Israel, Iraq, Syria, North Caucasus, Eastern Ukraine.) Since it isn't dated I have a feeling someone is including the Troubles and making Northern Ireland look much worse than it might otherwise be. Also, I can't find any references at all to actual terrorist attacks in Iceland, so they may using a VERY broad definition of "terrorist attacks". – DenisS May 31 '17 at 17:01
  • 12
    Also, as far as I can tell, Poland has no Muslim immigration ban. – DenisS May 31 '17 at 17:05
  • 15
    -1 Because this is clearly not about a map, but about the implied claim. Any answer would need to [take that into considerations](https://skeptics.meta.stackexchange.com/questions/3916/). Still, it makes it unclear what is actually being asked, which will attract bad answers, which makes it a bad question in my opinion. To make this a good question, it should instead directly ask about the topic the OP is interested in (if it's not a duplicate and generally on-topic here, and if there are sources to show notability; otherwise, politics.SE may be a better fit). – tim May 31 '17 at 18:24
  • 10
    The map may well be true, but than the time-scale has to be several decades. Lots of dots in Northern Ireland (and London). Those are no Muslim terrorist attacks, I don't think Ireland had any of those. Same goes for the Basque region. – DocM May 31 '17 at 19:15
  • 5
    The map appear to include The Troubles in northern Ireland which suggests it covers a long period of time, so an interesting comparison would be so show the non-terror related violent incidences for the same period. – dmckee --- ex-moderator kitten May 31 '17 at 20:16
  • @dmckee I don't think it includes the troubles (see my answer; the earliest mention of the map mentions that it only includes attacks from after 2001; if it would include earlier attacks, it would also show attacks in Poland). Note also that attacks in Ireland have [not stopped](https://en.wikipedia.org/wiki/Dissident_Irish_Republican_campaign) in 1998. – tim May 31 '17 at 20:24
  • 2
    https://en.wikipedia.org/wiki/Category:Terrorist_incidents_in_Austria None since 9/11, yet 4 dots as far as I can see – sgf Jun 01 '17 at 00:06
  • @Tim at least we know to check the info on Poland carefully. – Andrew Grimm Jun 01 '17 at 03:55
  • 1
    @dmckee In [2015 Europol data found half of EU terror incidents were in the UK](http://www.bbc.co.uk/news/uk-36845647), of those 103 the majority were related to Northern Ireland. The province is more lively than you'd think, it just rarely makes front page news when there's a bomb threat or someone shot. Especially when there's bombs going off in the wider middle east which kill dozens. So I don't see evidence to suggest it's pulling data from The Troubles, as 50 incidents are going to be densely packed in Ireland? –  Jun 01 '17 at 07:27
  • 5
    They seem to set the bar quite low on what the call an attack. Corsica is rather a hotspot; There they have a separatist movement that does things like burning down Parisien's holiday homes and blowing up municipal and governement buildings (out of office hours - usually). – Oscar Bravo Jun 01 '17 at 07:48
  • 16
    The map definitely takes liberties with the definition of terrorism. For example the Tuusula shooting (Finland) had no ideological motivation, similar to what I believe is referred to in the USA as an 'active shooter event'. – Nobilis Jun 01 '17 at 08:23
  • I'm curious which event the dot in northern Germany (Bremen? Oldenburg?) represents. – Martin Schröder Jun 01 '17 at 08:27
  • 1
    @MartinSchröder: It it's based on the GTD and supposed to be Bremen, it would be [this](http://www.start.umd.edu/gtd/search/IncidentSummary.aspx?gtdid=200108200002) event, which wasn't terrorism but a troubled man. – Martin Schröder Jun 01 '17 at 08:37
  • @MartinSchröder It could also be this [attack on a mosque](http://www.start.umd.edu/gtd/search/IncidentSummary.aspx?gtdid=201408300060) in Oldenburg. I'm also not sure if it is fair to characterize the arson attack on a politician as just the acts of a troubled man (I was unable to find any further information on it that describes the motivation in-depth). – tim Jun 01 '17 at 09:39
  • 4
    All those dots in North Spain are from ETA (now disbanded Basque independentist armed group), similarly the same can be said for all those dots in Norhtern Ireland, almost all (of not all) caused by the Irish conflict, now over. If there is a conclusion to be drawn from this map is that we have more in-house terrorism in Europe than Islamic one. – Ander Biguri Jun 01 '17 at 09:58
  • 1
    @tim: I lived in Bremen at that time, actually passed by the house minutes later and read all news articles. It wasn't terrorism. – Martin Schröder Jun 01 '17 at 10:00
  • 1
    @AnderBiguri This was pretty much my first impression when I saw the big clusters around Ulster and Euskadi – ManyRootsofAllEvil Jun 02 '17 at 06:04
  • The Corsicans are keeping busy. – Adrian Jun 03 '17 at 09:59
  • I would like to add that one way to show the map is evidently bogus is the fact that there was a (thankfully mostly unsuccessful) [bomb attack](https://translate.google.pl/translate?sl=pl&tl=en&js=y&prev=_t&hl=pl&ie=UTF-8&u=https%3A%2F%2Fwww.wprost.pl%2Fkraj%2F10041865%2FChcial-zabic-jak-najwiecej-osob-Sprawca-zamachu-we-Wroclawiu-oskarzony-min-o-terroryzm.html&edit-text=&act=url) by a Polish right-wing extremist barely a year ago in Wrocław. – mikołak Jun 10 '17 at 16:18

3 Answers3

96

tl;dr: The map is likely correct and based on the GTD from 2001 to 2014. As the map is not about Islamic terrorist attacks, but all terrorist attacks, no conclusion about Islamic attacks or refugees can be drawn from it.

Source of the Map

It is very likely that this map uses the Global Terrorism Database as a source, likely from 2001 to 2014, and there is no reason to doubt its accuracy.

The two dots in Iceland are the two terrorist attacks in Akureyri and Reykjavik in 2012 and 2014.

The dot in Portugal represents a terrorist attack in Lisbon in 2011. Note that this was not an Islamic attack, but an Anarchist attack.

We can assume that the map shows attacks after 2001, because before then, there are recorded terrorist attacks in Poland. This is the earliest occurance of the map that I could find, which mentions 2001.

We can also assume that the map is from before 2015, as the high number of attacks in Finland are not included.

A similar map can be seen here which visualizes the GTD data from 2011 to 2014.

The GTD also provides their own map, but it's for a 40 year span. Still, it shows a similar tendency as the map from the OP.

Quality of the map

Note that the map from the OP displays attacks as dots, which is a poor representation for what the map wants to express.

In countries in which terrorist attacks are concentrated on a specific location, it makes it appear as if there was less terrorism than there actually was, and for countries were terrorist attacks are spread out, it leads to the opposite effect.

The heatmap from the GTD linked above better visualizes the concentration of terrorist attacks.

About the GTD

Note that the definition of terrorism by the GTD does not just include major attacks or attacks with multiple fatalities. It also is not limited to a specific motivation (Islamic for example):

  • Criterion I: The act must be aimed at attaining a political, economic, religious, or social goal.
  • Criterion II: There must be evidence of an intention to coerce, intimidate, or convey some other message to a larger audience (or audiences) than the immediate victims.
  • Criterion III: The action must be outside the context of legitimate warfare activities.

Conclusions to draw from the map

We can't draw the conclusions proposed in the OP from this map for various reasons, among them:

  1. It is a map of all terrorist attacks, not just of terrorist attacks committed by refugees, and also not just Islamic attacks. This can easily be seen by the high amount of attacks in Ireland and Spain (specifically Basque Country), as well as the anarchist attack in Portugal. A considerable number of the attacks (even worldwide) in the GTD are non-Islamic.
  2. Even if we were to accept the map as showing attacks by refugees - which it does not - the map does not match the data of refugees by country (see also here; note the relatively high number of refugees in Sweden in 2014, and compare it to the very low number of refugees in Spain).
tim
  • 51,356
  • 19
  • 207
  • 177
  • 4
    +1, I feel like a lot of these kind of images and data-sets tend to break the world into a pre- and post- 9/11 world when using the data. Also the Iceland finding seems to have confirmed what I suspected in my comment on the question. They are using a very broad definition of terrorist attacks to generate this data. I wonder what this image would look like if you excluded A) Terrorist attacks outside of Europe B) Terrorist attacks that did not have any fatalities and C) Terrorist attacks by non-Islamic organizations – DenisS May 31 '17 at 20:57
  • 2
    I've got my doubts about the quality of the GTD database. Of the eight local incidents it lists, one was definitely a terrorist attack (attempted bombing of a parade), three look like armed robberies using a bomb as a diversion (each counting as two incidents), and one doesn't have enough information to classify. – Mark May 31 '17 at 21:43
  • The GTD map you link to seems to be for attacks in 2015 only, not 40 year time span. Might also be worth emphasising that the majority of attacks shown in the map had zero casualties and/or no evidence of any link to Islam or immigration – user56reinstatemonica8 May 31 '17 at 22:37
  • 28
    The number of attacks in Ireland doesn't match any recent timescale: it would need to go back to the 1970s for those dots to be credible. And terroriism in Ireland wasn't committed by immigrants but by natives so it doesn't exactly support the attached claim that immigration control stops terror (same is true for other concentrations, but Ireland is obvious for those who know history). – matt_black May 31 '17 at 23:38
  • 1
    You didn't answer OP's question. Is the map accurate? – BlueRaja - Danny Pflughoeft May 31 '17 at 23:59
  • There are comments on the question that say the map must go back well before 2001 because of the inclusion of a large number of attacks in Northern Ireland. Your answer says it must be after 2001 because of an absence of attacks in Poland. Can you address this point? – David Conrad Jun 01 '17 at 00:11
  • @BlueRaja-DannyPflughoeft First sentence: "It is very likely that this map uses the Global Terrorism Database as a source, likely from 2001 to 2014, and **there is no reason to doubt its accuracy**." (emphasis mine) You may want to clarify if you were driving at a more subtle point. – jpmc26 Jun 01 '17 at 00:18
  • 23
    Re. "the number of attacks in Ireland doesn't match any recent timescale." -- the GTD returns 140 incidents from 2001 to 2015 (most recent data). However, it seems relevant that of those, only 6 had injuries, only 1 had multiple injuries, and none caused a fatality. https://www.start.umd.edu/gtd/search/Results.aspx?page=1&casualties_type=b&casualties_max=&start_yearonly=2001&end_yearonly=2015&dtp2=all&country=96&charttype=line&chart=overtime&ob=GTDID&od=desc&expanded=yes#results-table – Larry OBrien Jun 01 '17 at 01:15
  • 1
    Re: being likely that the Global Terrorism Database is the data source, that is what [the reddit OP stated to be the case](https://www.reddit.com/r/europe/comments/4brazv/map_of_terrorist_attacks_in_europe_after_911/d1bnpxq/). – Xiong Chiamiov Jun 01 '17 at 02:44
  • The dots (at least 2 of the 3) in Norway must be Anders Breivik attack of 2011. – Akavall Jun 01 '17 at 05:36
  • The number of attacks in spain since 2001 seems a bit way off, there's been no terrorist attacks in Barcelona or Valencia since 11/9/2001. Even more, there are three unnotified attacks, two in Girona and one in Salou, all of them bomb-related and made by ETA back in 2001-2002. – CptEric Jun 01 '17 at 06:29
  • @DavidConrad People often aren't aware that terrorism in Northern Ireland is still a thing. According to [Europol data for 2015 alone](http://www.bbc.co.uk/news/uk-36845647) half of the recorded terror incidents in Europe were British, and most of them were related to Northern Ireland. So it's unlikely that it is drawing on data from The Troubles because that's about 50 dots to cram into Northern Ireland. –  Jun 01 '17 at 07:35
  • 2
    @BlueRaja-DannyPflughoeft I did answer it; there is no reason to doubt the accuracy. It just doesn't say anything about Islamic terrorism or terrorism by refugees. – tim Jun 01 '17 at 08:24
  • 1
    @DavidConrad I actually addressed that as a comment to those comments already. [Attacks in Ireland did not stop after 1998](https://en.wikipedia.org/wiki/Dissident_Irish_Republican_campaign), but continue to happen until today. – tim Jun 01 '17 at 08:26
  • 1
    @matt_black The attacks in Iceland are about a [bomb](https://www.start.umd.edu/gtd/search/IncidentSummary.aspx?gtdid=201201310025) near a government building in 2012 and about an attempt to [burn down](https://www.start.umd.edu/gtd/search/IncidentSummary.aspx?gtdid=201404160092) a Lutheran Church in 2014. – tim Jun 01 '17 at 09:27
  • @CptEric, [Valencia in 2003](https://www.start.umd.edu/gtd/search/IncidentSummary.aspx?gtdid=200305240003), [Valencia in 2007](https://www.start.umd.edu/gtd/search/IncidentSummary.aspx?gtdid=200711290016). – Peter Taylor Jun 01 '17 at 09:46
  • edit: saw the links. okay, didn't consider those attacks as terrorism, @PeterTaylor . – CptEric Jun 01 '17 at 10:14
  • 31
    I find it interesting that maps like these are used as anti-immigration propaganda. I checked the Global Terrorism Database, and ALL the listed terrorist attacks in Finland from 2015 are by white native Finns who are trying to burn asylum seeker reception centers. Can't really be racist and use that map to show how bad asylum seekers are. – shortyputting Jun 01 '17 at 10:36
  • @shortyputting the same will be the case (homegrown extremists, nothing to do with Islamist attacks) for all the attacks shown in Greece (and, as already mentioned, the vast majority of the ones in Spain - all bar one, probably - and Ireland). – terdon Jun 01 '17 at 11:33
  • 1
    In trying to track down the three dots for Scotland, I find that the only three incidents the Global Terrorism Database lists are the Glasgow Airport bomb of 2007, and two assaults on Muslim shopkeepers. Unless someone else can find other incidents - the Northern Ireland incidents tend to swamp the search. – DJClayworth Jun 01 '17 at 13:26
  • 1
    I'd say that calling some of these a terrorist attack might be hyperbolic. Here is why: I'm from Iceland, but couldn't easily recall any "terrorist attack" in Akureyri in 2014. One of the [sources](https://www.start.umd.edu/gtd/search/IncidentSummary.aspx?gtdid=201404160092) for the claim is [this](https://grapevine.is/news/2014/04/22/organised-attack-on-church-in-akureyri/) article from Reykjavik Grapevine, which had this to say on the matter: "An attempt was made to burn the doors to Akureyri’s Lutheran church, but the arsonists were not entirely successful." – steinar Jun 01 '17 at 13:52
  • 5
    @steinar that's kind of the point. The GTD uses a VERY open definition for "terrorist attack" than your average person. When you think of terrorist attack you think of 9/11 or the Manchester Bombing, not an attempt to burn a door to a church. Not that I fault the GTD for this as the site itself is very open about the details of a terrorist incident. Taking their data and aggregating it into a map which displays every terrorist incident as a dot is very deceptive. – DenisS Jun 01 '17 at 14:31
  • @Mark The D in GTD stands for database, thus the phrase "GTD database" is redundant. – Pharap Jun 01 '17 at 14:47
  • 2
    @DenisStallings The definition of terrorism is "the unlawful use of violence and intimidation, especially against civilians, in the pursuit of political aims.", which I think it's fair to say includes attempting to burn a church door (assuming the motivations were political, religious or idiological of course). Just because nobody died and it didn't cause much of an impact, does not make it any less a form of terrorism. I agree that it's easy to misrepresent their data, but the same could be said of any database. – Pharap Jun 01 '17 at 14:53
  • 1
    @Pharap that's my point. The textbook definition of terrorism and what people think of when someone says terrorism are two very different things. Data like this provided without context tries to conflate the two. – DenisS Jun 01 '17 at 15:13
  • 1
    @DenisStallings From the wording of your comment I thought you were trying to say that the avaerage person definition was more correct. (Particularly the phrase "When you think of terrorist attack you think of 9/11 or the Manchester Bombing, not an attempt to burn a door to a church."). – Pharap Jun 01 '17 at 15:30
  • 1
    It's worth emphasizing that the classification of "terrorist attacks" does not match at all with the public perception of terrorism. Some of these incidents are literally adolescents blowing up letter boxes - which I can't believe didn't happen in Poland. If no incidents were reported I'd doubt the accuracy of the reporting. – Peter Jun 01 '17 at 16:25
  • 1
    @Pharap oh definitely not :) I wasn't trying to comment at all on which one may or may not be more correct, just that what the average person perceives as a terrorist attack and what the GTD uses for its definition are extremely different. Someone bringing in those initial beliefs when viewing this map could be swayed into believing that "holy s**t, look at all these terrorist attacks!" when probably 99.9% of them don't fit within their definition of a terrorist attack. It's an apples to oranges comparison. – DenisS Jun 02 '17 at 14:36
15

The claim that there are terrorist attacks all over the world but not in Poland is wrong.

The claim that the map presents a set of data, however incomplete, is correct.

The accepted answer explains where the data comes from, it's the recorded incidents after 9/11 from the GTD. But I think it's also necessary to explore the accuracy of the data.

There are 2 issues with the accuracy:

  1. The incidents are not what the typical reader of the map is likely to suspect. A majority of incidents involved deliberate property damage with no dead or injured, and the incidents are often committed by unknown assailants. I am not saying these events don't classify as terrorism. It primarily includes events such as:

  2. While there are no dots in Poland, similar events did indeed happen in Poland. This is the result of a brief search:

There are also some stories that debatably don't classify, yet demonstrate that Poland does have a similar level of terrorism as some other European countries:

In conclusion:

The data contains an incomplete set of events that can reasonably be classified as possibly terrorist in nature. Notably missing from the data are events in Poland, which the claim and the question is about.

Peter
  • 1,407
  • 11
  • 19
  • Comments are not for extended discussion; this conversation has been [moved to chat](https://chat.stackexchange.com/rooms/75009/discussion-on-answer-by-peter-is-this-map-of-terrorist-attacks-in-europe-accurat). – Sklivvz Mar 23 '18 at 21:25
-1

No. The map seems to follow a vaguely correct pattern but there are obvious errors. For example hot spots like London are shown but when you look closer the attacks have the same density over the rural regions surrounding it.

Here is an alternative map from the university of Maryland. https://www.start.umd.edu/gtd/images/START_GlobalTerrorismDatabase_2015TerroristAttacksConcentrationIntensityMap.jpg Many spots allign which would imply that they cover the same period but there are spots that appear on the OP's map and not on MaryLands and vise versa.

Here is a discussion about what appears to be the world map in which the OP's map was taken. Including a discussion about individual points which are questioned. https://www.reddit.com/r/MapPorn/comments/3taz61/map_of_terrorist_incidents_since_911_37741893/

PStag
  • 107
  • 2
  • -1 Because the "obvious errors" aren't explained clearly, the map from "the same time period" clearly is labeled to imply it covers only the year 2015, and the most relevant part of this answer is link only. If these points are improved I'll revisit, remove the downvote. – Peter Jun 01 '17 at 22:43
  • @Peter The label of the map is misleading, it actually shows "terrorist attacks that occurred worldwide across 45 years of data" (as per GTD, which is the same source as the map in the OP). Otherwise, I agree though. Any obvious errors should be explained and supported with sources, and possibly relevant points from the reddit link should be handled the same way. – tim Jun 01 '17 at 23:03
  • @Peter if the OPs map simply covered a longer time period that it would include all of the points from the map by Maryland. – PStag Jun 02 '17 at 00:42
  • @tim in which case the answer needs at least one source that spells put what the linked map actually shows. – Peter Jun 02 '17 at 06:41