The problem with “Muslim religious extremists”

Whenever an Islamic extremist blows something up, or threatens to blow something up, or threatens to do any other act of terrorism, the headlines always include the term “Muslim extremist” or “Islamic terrorist” or some variation of those. Liberals have a problem with this, and it is often difficult for people who are not liberals to understand why.

The conversation goes something like this:

Conservative: An Islamic religious extremist blew up this building.

Liberal: When you say that, it seems like you’re demonizing the entire religion. It seems like you’re denigrating all Muslims.

Conservative: I’m totally not! I’m just telling the truth. Didn’t this person blow up a building? Yes. Wasn’t the person Muslim? Yes. Didn’t he even say that religion had something to do with why he did it? Yes! Everything I said is 100% true!

Liberal: Well, it still seems prejudiced to me.

Unfortunately, this is where the conversation usually ends or wanders off-topic, so the liberal “intuition” of prejudice is often never examined or explained. The problem is that an honest conversation about this topic has to delve deeply into semantics, the way words are used and understood, the psychology of conversation, and other heady abstract things. Especially on the internet or on television, ain’t nobody got time for that.

But for anyone who cares, this is my take on the explanation of the “liberal view” on this issue.



When a specific individual person does a thing, you can announce it in many possible ways. For example, if a 5’10” overweight Native American man with a beard robs a bank, then in theory the headline could accurately be any of the following:

Man robs bank

Native American robs bank

Overweight person robs bank

Bearded man robs bank

Average-height individual robs bank

…and so on and so forth. All of these are perfectly legit. All of these are completely true. But all of them send very different messages and have very different implications.

Why? Because people tend to assume, in both headlines and in normal day-to-day conversations, that when people communicate with each other they follow certain general rules to help aid in communication and make it go more smoothly. In the 1970’s, social psychologist Paul Grice called this the “cooperative principle” and came up with four general guidelines that people tend to assume about conversations unless they have reason to believe otherwise:

1) People don’t say things they believe to be wrong
2) People say as much as they need to to express their statement completely
3) People do not include irrelevant information
4) People try their best to be orderly and unambiguous (even if sometimes they fail)

Together these principles are known as “Grice’s Maxims” although I personally refer to these principles as “don’t be a dick to person you are talking to”.

The third principle has, in the decades since Grice, been further elaborated and studied in detail as “Relevance Theory” by Dan Sperber and Deirdre Wilson. But the main point that I want to highlight is simply that it is a well-known fact that people generally assume that the information you get from others (whether in headlines or in conversation) will not include a random detail or fact that is totally unrelated to the point of the communication.

Thus, it seems racist for the headline to read “Native American robs bank” because that headline implies that something about the person being Native American somehow caused or was otherwise relevant to the crime. The implication is that either Native Americans are just prone to robbing banks, or that the fact that the person was Native American somehow caused the person to behave in a criminal manner.

(Aside: A really astute reader will say, “In that case, isn’t it sexist even simply to say ‘Man robs bank’ instead of ‘Person robs bank’?”  The answer is: yes. And what I will call “hardcore liberals” will even point this out from time to time. The fact that “Man robs bank” seems more natural and less awkward than “Person robs bank” is really just a testament to how deeply ingrained gender differentiation and gender politics is in our culture. Why don’t more liberals complain about this? You gotta pick your battles, I guess.)



But this is where the issue of “Muslim religious extremists” gets tricky.  The obvious conservative response to everything I’ve said so far is that Muslim religious extremists usually cite their religion as one of the motivations, or even the main motivation, of their violent acts. Therefore, wouldn’t it be considered very relevant? Wouldn’t it therefore pass the “relevance test” to include the label “Muslim” or “Islamic” in the headline?

This is where there is really a deep and profound difference in world-view between today’s liberals and conservatives in the United States.

Liberals believe that the cause of violence does not originate from the fact that these people are Muslim, but rather from the fact that they are religious extremists.  To liberals, that is the real factor that leads to violence. After all, non-extremist Muslims do not commit acts of terror, just as non-extremist Christians do not commit acts of terror. On the other hand, extremist Christians do commit acts of terror, such as assassinating abortion doctors and going on shooting sprees in churches they deem to be “too liberal”. There are also out-and-out terrorist groups, such as the Ku Klux Klan, that overtly describe themselves as Christian and their motivations as based on their religion.

(Aside: You might say, as many do, that the Ku Klux Klan are not “real Christians”. That may very well be; but keep in mind that most Muslims also say that Islamic terrorist groups are not “real Muslims” either. So if it is logically correct and proper to not describe violence and threats by the KKK as “Christian violence” then it would also be logically correct and proper to not describe threats by Al-Qaeda or the Muslim Brotherhood as “Muslim violence”. You have to be consistent.)

Diagrammatically, a liberal sees this:

Muslim Extremism vs. Christian Extremism

When you look at it this way, it is very difficult to make an argument that  the label “Muslim” is truly relevant in a headline about terrorism, even if the terrorism happens to have been committed by Muslim extremists.

Instead of the headline “Muslim plans terrorist attack” or even “Muslim extremist plans terrorist attack”, the headline that properly follows the Gricean maxim of relevance would be, “Religious extremist plans terrorist attack”.

To a liberal, saying “Muslim extremist plans terrorist attack” is just as ridiculous as saying “Six-foot-four man plans terrorist attack”. It seems weird, because it seems like it’s targeting, and by implication, generalizing about a trait that a person has that is not really related to the news story.

Of course, at this point conservatives often try to bring up arguments about frequency or proportions, claiming that a higher percentage of terrorist attacks recently have been by Muslims (which is true) or a higher percentage of Muslims are extremists (which is false: Americans regularly under-estimate the total number of Muslims in the world).

Unfortunately for conservatives, arguments about frequency are actually completely irrelevant here. The issue is one of causality. One can have a long and nuanced conversation about the history and cultural factors that lead there to being more Muslim extremists right now in the world than Christian extremists, but it doesn’t change the underlying fact: the extremism is the problem. The specific religion is not the problem.

And because of that, it has no place in the way terrorist attacks are labelled in headlines.


5 views shared on this article. Join in...

  1. Logan Hartanian says:

    “The specific religion is not the problem.

    And because of that, it has no place in the way terrorist attacks are labelled in headlines.”

    – This is a fallacy based on emotion not fact. …..actually, if the religion, everywhere it is expressed, has radicalism and resultant death, at a consistent rate of approximately .5 % of the population, then one could honestly say that Islam causes radicalism, though the numbers are small.
    If some populations have no radicalism at all, then one could honestly theorize that Islam is taught differently in geographically disparate locations and the method of teaching MAY be the cause of radicalism.
    But to simply withdraw causality based on quantitatively low levels of stimulated response, would mean that causes of cancer and various other diseases are completely invalid as well, because the vast majority of cases, do not result in a disease response.

    • Greg Stevens says:

      Hi Logan! Thanks for your comment.

      “This is a fallacy based on emotion not fact. ….actually, if the religion, everywhere it is expressed, has radicalism and resultant death, at a consistent rate of approximately .5 % of the population, then one could honestly say that Islam causes radicalism…”

      No, this is mathematically incorrect for two reasons.

      First, you’ve got the simple problem of what is called “base-rate neglect”: If the places where religion X exists have a 0.5% rate of radicalism and death, and places where religion X is NOT PRESENT also have a 0.5% rate of radicalism and death, then the impact of religion X on radicalism and death is zero. The terms you want to read up on here are “conditional probability” and “Bayes Theorem”.

      Second, there is a problem that religion co-varies with other factors, such as regionality and political structure. If two separate values that are correlated with each other are also correlated with radicalism and death, then you cannot attribute the cause to one over the other. The term you want to read up on is “multicollinearity” and I’ve written about it here:

  2. R says:

    Interesting post, so clearly expressed 🙂

Pings to this post

  1. […] 1: Muslim extremists.  People are inundated with stories about Islamic religious extremists doing terrible things, and […]

  2. […] mechanisms behind why people understand sentences this way, in my article about the term “Muslim religious extremists.” The principle is the same: If you are criticizing “Muslim religious extremists” […]