Atlantic article pretends that atheists are really religious by lying about the data

August 29, 2019 • 10:30 am

When I saw this article from last year’s online Atlantic, I thought it was going to push the usual guff: “atheists are religious because they adhere fervently to the doctrine of No God, with no proof of their (non)beliefs.” But no, it wasn’t that. It was worse. In fact, the title is an arrant lie in at least two respects, and a distortion in another.

So how did author Sigal Samuel (a staff writer at Vox and former religion editor of The Atlantic) come to this conclusion? By distorting and misreporting the results from a 2018 Pew survey on the attitudes of Christians in Western Europe. That survey involved estimating the religiosity of Americans and Europeans by using standard questions like “Do you believe in God with absolute certainty?” and “Do you pray daily?”

Unsurprisingly, they found that Americans were more religious than Western Europeans. From Pew:

. . . Americans, overall, are considerably more religious than Western Europeans. Half of Americans (53%) say religion is “very important” in their lives, compared with a median of just 11% of adults across Western Europe. Among Christians, the gap is even bigger – two-thirds of U.S. Christians (68%) say religion is very important to them, compared with a median of 14% of Christians in the 15 countries surveyed across Western Europe.

Well, we’ve known this for a long time.

Second, as Samuel reports, there’s a difference between “nones” in America and “nones” in Europe:

. . . the researchers found that American “nones”—those who identify as atheist, agnostic, or nothing in particular—are more religious than European nones. The notion that religiously unaffiliated people can be religious at all may seem contradictory, but if you disaffiliate from organized religion it does not necessarily mean you’ve sworn off belief in God, say, or prayer.

Is there a deficit of neurons here? Lots of people who believe in a Higher Power don’t identify as members of a particular church. We all know some of these people.

Here’s what Pew says, affirming Samuel’s statement:

But even American “nones” are more religious than their European counterparts. While one-in-eight unaffiliated U.S. adults (13%) say religion is very important in their lives, hardly any Western European “nones” (median of 1%) share that sentiment.

Again, no surprise. Remember that “nones” aren’t all atheists, but simply a grouping term for people who don’t consider themselves affiliated with a formal religion. Atheists are only a small proportion of “nones”. And yes, you can still believe in God and be a “none”—you just don’t align yourself with the Catholic Church, Judaism, Islam, or any formal religion. Given that Americans are on the whole considerably more religious than Europeans, why is it a surprise that unaffiliated Americans are more religious than unaffiliated Europeans?

But here’s the result that got Samuel to her clickbait headline. As Pew said:

Similar patterns are seen on belief in God, attendance at religious services and prayer. In fact, by some of these standard measures of religious commitment, American “nones” are as religious as — or even more religious than — Christians in several European countries, including France, Germany and the UK.

And as Samuel tells us:

The third finding reported in the study is by far the most striking. As it turns out, “American ‘nones’ are as religious as—or even more religious than—Christians in several European countries, including France, Germany, and the U.K.”

“That was a surprise,” Neha Sahgal, the lead researcher on the study, told me. “That’s the comparison that’s fascinating to me.” She highlighted the fact that whereas only 23 percent of European Christians say they believe in God with absolute certainty, 27 percent of American nones say this.

Note the Pew statement (my emphases) “by some of these standard measures of religious commitment, American “nones” are as religious as—or even more religious than — Christians in several European countries, including France, Germany and the UK.

Two points here. First of all, “nones” aren’t all atheists, especially in the U.S. So Samuel has erred mightily in her headline, saying “atheists are sometimes more religious than Christians” when she means “nones are sometimes more religious than Christians”. Atheists, by definition, aren’t religious—at least according to the criteria Pew used for “religious”. And, of course,  her headline, even if corrected, doesn’t hold true for all European countries (Pew mentions three; I can’t be arsed to find the country-by-country data).

Here’s Pew’s table that’s apparently the basis for Samuel’s breathy conclusion:

The comparison we want to make is with Western European Christians (dark red dots in middle column) with “nones” in the U.S. (grayish dots in right column). It turns out that using the criteria “religion is important in my life” or “I attend religious services at least monthly”, American “nones” aren’t as religious as European Christians, belying the headline. (The difference is greatest for churchgoing, with 31% of European Christians going to church at least monthly compared to 9% of American “nones”.) And the “higher religiosity” of American nones than of European Christians isn’t impressive for the other two criteria: a difference of 2% in “praying daily” and 4% in “believing in God with absolute certainty.”

And if you compare European nones with European Christians, the “nones” are less religious—by a long shot—for every one of the four indices of religiosity.

So that is the lie, and Samuel should have known better. But telling the truth would have spoiled her headline: it would have had to be “Americans who don’t consider themselves affiliated with a church are, according to some criteria for religiosity, more religious than Europeans who identify as Christians.” That’s not very exciting, is it?

And given the secularism of Europe, and the fact that many who identify as “Christians” do so in a cultural rather than religious way, just as I identify as being a Jew, it’s not surprising that American “nones” are sometimes more religious than cultural European Christians. That’s a second contributor to the distortion in Samuel’s headline: that many Christians (she means European Christians) are really atheists and therefore don’t pray, go to church, or believe in God at all, much less with absolute certainty.

The Pew report has some interesting data; look for the table of how many European “nones” (as opposed to church-attending Christians or non-practicing Christians) think that science makes religion unnecessary (hint: it ranges between 53% and 69%.

One surprising result: a substantial proportion of the European “unaffiliated,” including those who are religious and those who aren’t, believe that they have a soul (see graph below). Such is the power of dualism. Perhaps some of it comes from the dualism inherent in many forms of free will. (I’ll get my coat.)

All in all, the headline really has the import of “Dog bites man” rather than the other way round. I guess the Atlantic doesn’t vet their headlines very well. And the rest of Samuel’s article is pretty much boilerplate reporting. It’s not worth reading once you find out that there’s little new here except some serious distortion.

h/t: Enrico

25 thoughts on “Atlantic article pretends that atheists are really religious by lying about the data

  1. The only conclusion you can reasonably make from the Pew data is that Americans are both far more religious AND more likely to identify as Christians than Europeans. This is hardly surprising.

  2. Isn’t the primary difference between religious people who go to church and those who do not, money. I don’t mean having it, I mean not wanting to give it up.

    Let us not forget that most religions have pretty strong requirements for money. The pressure can be pretty strong. So being unaffiliated is the cheap way to go.

    1. I don’t think Jesus ever said that being a believer required an entry fee, did he? (I could be quite worng about that, it’s six decades since I was coerced into reading any part of the Bibble 🙂


  3. “. . . a staff writer at Vox. . . .” Well, there’s the first problem. The report and questionnaire clearly indicate that Pew asked respondents if they were “atheist, agnostic, or nothing in particulary,” but group all those together as Nones, except in one chart where they are disambiguated. When they compare Western Europe to the US, they make no distinction. I would think the first thing to have done would be to remove the atheists from the nones before the comparison. I don’t really think Pew should lump those three groups together in that way for analysis. Clearly, what they have shown is that Nones are religious in many cases, which puts them in a different category from atheists (mostly).

  4. Just the other day, I found this: “End-of-life preferences for atheists” and in my estimation it’s just as confused and duplicitous (unwittingly? – I attribute it to woeful ignorance) as the Atlantic article.

    It’s well-intentioned but clueless, and it takes spirituality for granted with the “threefold model of spiritual care “intrapersonal, interpersonal, and natural interconnectedness to assess the appropriateness of potential interventions for a group of atheists”, even for atheists (some of us reject the concept of spirituality – can’t they do this without injecting spirituality into it, for me it’s like the camel’s nose under the tent or the vacuum salesman with a foot in the door)

    And what is meant by this “Further clarification of the use of the term “atheist” is also necessary to ensure that the inclusion of all individuals with nontheist beliefs are represented in future research efforts.”

    “The inclusion of all individuals with nontheist beliefs are represented in future research efforts.” The tacit assumption in this paper is that nontheism = atheism, which seems to be what many believers think. This is equivalent to “nones.”

    Wikipedia defines nontheism thusly, that is, after issuing this crucial caveat at the beginning of the entry “Not to be confused with atheism.”

    “: “Nontheism or non-theism is a range of both religious[1] and nonreligious[2] attitudes characterized by the absence of espoused belief in a Godor gods. Nontheism has generally been used to describe apathy or silence towards the subject of God and differs from an antithetical, explicit atheism. Nontheism does not necessarily describe atheism or disbelief in God; it has been used as an umbrella term for summarizing various distinct and even mutually exclusive positions, such as agnosticism, ignosticism, ietsism, skepticism, pantheism, atheism, strong or positive atheism, implicit atheism, and apatheism. It is in use in the fields of Christian apologetics and general liberal theology.” I concur with this definition.

    Much certainly needs to be done for atheists in end-of-life care but an atheist needs to define the terms, not some confused theist or nontheist. I’d like to see hospice services for atheists.

    The onset of death is a time when one is most vulnerable and who needs people shoving religious tracts at you, praying by your bedside, and expecting a deathbed conversion — or condemning the dying person for not believing when they’re “going to meet their maker.” Dog save me from the Goddists!

    1. “ietsism”

      I had not seen this term before. From Wikipedia: believing that “there must be something undefined beyond the mundane and that which can be known or can be proven”.

      I think this is very common. Without knowing anything at all about what they think it is, people still think this.

      I wonder if you can be “anti-ietsist”? I would be that, since once you go beyond what can be known, you have lost me.

      Wikipedia also notes that how ietsists are classified in a religious/not-religious split can significantly affect the results.

      1. With respect to “letism,” which “[you] think…is very common. Without knowing anything at all about what they think it is, people still think this.”

        I agree with you but I wonder how certain other vague nontheistic and theistic conceptualizations, including “spiritual but not religious” differ significantly from letism.

        I was lost right away when I began to look up the definitions of plethora of nontheistic
        -isms listed in the article Seems like nearly everybody wants to split hairs and define themselves as -ists of some particular -ism that’s just a hairsbreadth different from another -ism work up some theory, then inflate incomprehensible concepts with incomprehensible words and phrases, with lots of neologisms — all of which I’ll call “gasbagging.” Indeed, this isn’t any different from the plethora of religions and their dogmas, infinitely splitting off from some existing religion or sect or coming into being de novo.

        I may not understand much of this gobbledygook (same for theology) but I do know that whatever it is, I do not give a tinker’s damn about it.

      2. I think a lot of people, nominally religious and otherwise, are ietsists to some degree.

        That is, they don’t really believe in any specific supernatural phenomena, but they have a vague feeling there must be *something* beyond the day-to-day existence. A “this can’t be all there is” feeling.

        Some of them go on to astrology or pyramid power or crystal healing or other forms of woo. I think many others, when confronted with any specific set of woo-ish concepts, can see at once that those particular ideas are absurd; but still, they feel that there might be something undefined or indefinable lurking ‘out there’.


  5. . . . believe they have soul as well as a body. . .

    and what practical consequences follow as a result of soul possession, actually? Do you become James Brown?

    Religion as a practical matter is a communal behavior. Sure sects fight about words and which funny hats you should wear, but at the end of the day, its about communal rituals, holidays, communal prayers, burial ceremonies and communal taboos. If you do all that stuff together, it doesn’t actually matter if you say you are atheists or polytheists or monotheists or non-theists.

    To put it another way, is there one iota in New Atheism that would prohibit someone from being Shinto? [Christians care a lot about doctrine–it’s what they like to disagree over–but that doesn’t mean doctrine actually matters that much to practicing Christianity, nor does it matter much to other traditions.]

  6. I’m surprised at the high numbers of ‘religiously unaffiliated non-believers’ in Belgium and the Netherlands believing in a soul (40 and 43%). I know quite a few dozen unaffiliated non-believers in each of those two countries, but not a single one believes in a ‘soul’.
    There is something weird going on there in this PEW poll, the questions? The categorizing? Or something else I didn’t think of?

  7. Atheism is not a vaccine that protects you from irrational beliefs. I personally know atheists who believe in: the power of the free market to create utopia, veganism will will create utopia, luxury communism will bring about utopia, intersectional feminism… you get the idea.

  8. I just took a survey and for “religion”, since all the religions were listed, and it seemed they just meant if I’m an atheist, I picked atheist.

    But just as I clicked the “next” button, I saw – disappearing from view – the checkbox for “no religion”, as well as a write-in box.

    That really got on my nerves. So I share my experience that others might be forewarned about poor survey writing.

  9. I don’t know if it’s true, but I read a couple of times that writers of an article are often not responsible for the headline, which is written by editors.

    1. Well, for every popular article I’ve written, and there are a fair few, sometimes I write the headines and sometimes the editor does. But if I have big objections to the headline, which you get to see before publication, I would make them known, and we work out something joint acceptable. The headline here I would have fought against strenuously, as it’s a lie, and any decent editor would have changed it.

  10. The problem with surveys is that the participants answer on the basis of how they want to appear to society, not on how they actually behave. So these ‘studies’ are really about people’s display behaviour, not their belief systems.

    There’s no getting around this, other than having microrobots follow everyone around for months or years. And that sounds a bit silly (are smart phones doing this already?.


  11. I’m curious what the exact phrasing of the question was, if it was asked in English, and, if not, what the translation for the word ‘soul’ was. In colloquial English ‘soul’ is usually assumed to be something that survives death. I can see it translating to something more akin to a nonphysical self such as ‘ego’, however.

    1. And in some ancient philosophies at least, souls were held to be material (e.g., in both Epicureanism and Stoicism). (Of course, in the latter case, they held the bizarre view that it was both material and sometimes survived death of the body)

  12. I can’t say I’m surprised, though (as an Englishman) I am rather gratified.

    It is ‘well-known’ that the C of E is mainly a cultural/social institution; and if you visit the countryside in France or Italy there are churches and shrines everywhere – cultural relics. So a large amount of social life is still organised around the church, I think.

    It’s reminiscent of that map on this site a couple of days ago – ‘countries that have God in their anthems’. It’s a hangover from when God and anthems were fashionable, nobody can be bothered to change it.


  13. I’m curious how many of the ‘nones’ have had some exposure to 12 step programs.

    The first point I would make is that 12 step programs have high attrition or turn over rates. So while 12 step programs don’t have large memberships per se, there are maybe as much as 10 fold amount of people who have been exposed to 12 step programs. Court mandated attendance for an example.

    Second point, when I was a 12 stepper I would have considered myself a ‘none’, yet a believer in a higher power.

    This is anecdotal of course. I’m not unique, though. Others would have reached similar conclusions as I. I can say (anecdotally, again) that I’ve known many who have reached the conclusion of ‘spiritual, not religious’.

    I hypothesize that 12 step programs are a driver of the ‘spiritual, not religious’ subset of ‘nones’. Perhaps the main driver.

    In the case I was unclear, I’m an atheist now and no longer a stepper.

Leave a Reply