Cooked or misquoted studies are often cited that “show” that religious believers are happier and healthier and less insane than atheists.

“So what?”

“So atheism is bad for you, is what.”

Or so the argument attempts to go.

(Every study anyone has seen like that, please cite it in the comments [I already know about the study salad at Conservapedia, and any sharp bloggers out there who are psych majors could do us a solid by writing a good critique of that]. Likewise, if any blogs or websites have also tackled this myth [even if doing so after I posted this], please cite those articles in comments as well. Plus articles in magazines that discuss it, if you know any. I’d like to gather up a collection.)

Yes, one can reply that the truth is still the truth, and if you care about the truth, you should just deal with it if it sucks, rather than try desperately to live a lie just to be happy. There’s a whole line of debate one can follow down that route, but it generally ends with: bad epistemologies are always a net bad for you (and society), but you can only believe comfortable lies if you commit to a bad epistemology, the side effects of which will never be good for you or society overall–because bad epistemologies cannot protect you from harmful false beliefs (and even entail an increasing resort to harmful false beliefs in order to protect the harmless ones from being exposed).

So just suck it up and get over it. In practice, it’s all not as bad as you think (and it’s certainly better than all this eternal hell business, much less a god who by his total inaction manifestly doesn’t give a shit about anything). And all else being equal, you’ll find plenty to love about life anyway.

But in fact we needn’t take that grim tack anyway. Because the facts don’t support the premise in the first place. Atheism doesn’t make you less happy, healthy, or sane.

One common flaw that invalidates almost all of these studies (if not in fact literally all of them) is that they tend to compare the religiously devout with nonbelievers as a whole, who are statistically mostly people who don’t identify with any nonbelieving community or care much about atheism or humanism or any secularist cause or philosophy, but who just don’t have a specific belief in anything. But that’s a false comparison. The only meaningful comparison would be between the devoutly religious and devoted humanists, including organized atheists, or anyone intellectually committed to godless philosophies.

But no studies do that (so far as I know: please cite any that do in comments). They therefore don’t show the relative benefit of religious belief vs. atheist philosophy.

Let’s take a look.

Atheism and Depression

For example, in a public debate I was in many years ago, my theist opponent cited a study “proving” that atheism leads to depression and therefore was bad for us. When I got home and was able to check that study, I discovered that he’d snowed me (a common tactic theists use in debates: cite, even misquote, an obscure study we can’t rebut because we can’t check what it actually says before the clock runs out). What I found that study really said was not what the theist claimed. Surprise surprise.

To the audience my opponent read the sentence “religious belief […] was a significant predictor of lower levels of hopelessness and depression,” but he didn’t read the following sentence, that “there was also a small direct positive association of belief with depression.” In other words, fewer nonreligious persons ever get depressed in the first place. The study thus supports what I actually ended up saying in the debate: that only nonbelievers who lack “a cognitive framework for finding meaning in a negative event” are more likely to become depressed at all.

I would revise that statement today to mean remain depressed without treatment at all, since I now know that depression itself is predominately biochemical and thus not actually distributed in the population according to belief or nonbelief. You generally can’t choose to not suffer depression. You can only choose whether and how you treat and manage it. And your belief system and social support system will have a lot to do with that…neither of which needs to be religious. In fact, nonreligious systems are more likely to treat depression scientifically, which means more effectively. At any rate, the study my opponent cited found that you will be in a depressive state less often if you aren’t religious. Which means you are better off being an atheist. As the study concluded, “religious variables added significantly to the prediction of depression and to the prediction of hopelessness.” The exact opposite of what the theist claimed.

The only negative effect the study found for nonbelief was that when atheists were in depressive states, those states tended to be worse. But it also found the same outcome for strong religious belief, which also magnified the severity of depression. More importantly, the study did not distinguish people who were merely nonbelievers from people who have strong philosophical beliefs (e.g. intellectually committed atheists, humanists, naturalists). I suspect if it did, the negative effect of nonbelief would only show for the unreflective nonbelievers, and not the philosophical atheists and humanists.

The authors even acknowledged this defect in their study and called for more discerning instruments for distinguishing different kinds of philosophical framework (including those lacking “belief in a personal god”). As I wrote before, “I know first hand many atheists who have positive world philosophies, which help them cope very well with all manner of tragedies and problems.” No study can tell us anything about the effect of that kind of atheism unless it actually identifies and tracks those kinds of atheists. Keep your eye on any study, see whether it actually is looking at those atheists (us), and not just nonbelievers as an undifferentiated whole.

The authors also admitted that they could not prove the direction of causation, so it may be that it was severe depression that caused their correlated atheism–which would be an unhealthy atheism indeed, not reached by sound intellectual inquiry but mere emotional despair. And most of us are promoting not an atheism of despair, but an atheism of positive humanist values. So how does that kind of atheism correlate with the severity of one’s depression? Inquiring minds want to know. Because that’s the only finding that is actually useful.

There was one study that sort of almost did this, finding that the most religious and the least religious scored the lowest on measures of depression. But its sample size was so small as to render its results more or less useless. But generally, all studies related to this subject demonstrate no increased rate or severity of depression among intellectually devout atheists (such as humanists and other philosophical atheists), and even suggest a reverse correlation in some cases. In other words, not only has no one shown that philosophical atheism is bad for you, what evidence there is suggests it might actually be good for you.

Atheism and Health & Happiness

In tackling the bogus claim that atheism is bad for you on general measures of physical health and overall happiness, recent studies show religiosity in fact probably has no benefits once secular factors are controlled for. Having a social identity and engaging in regular socialization explain all the gains in health and happiness among the religious, and yet organized atheists have both. Again, in studies that purport to show negative effects for nonbelief, philosophically committed atheism is not distinguished from mere apathetic nontheism, even though it is likely to have effects comparable to religion (having belief structures and social networks of similar function).

As Victor Stenger reports:

A systematic review of sixty-nine studies of an initially healthy population showed (p < 0.001) that religiosity/spirituality was associated with lower mortality, but the association was negative for cardiovascular mortality. Furthermore, twenty-two studies of a diseased population showed no such effect (p = 0.19). So all they really found was that people who attend church regularly are healthier than those who don’t, and thus “organizational activity” (e.g., church attendance), and not religious belief, was associated with greater survival, and only in healthy populations. But that isn’t surprising. A lot of people are too sick to go to church. Moreover, none of these studies compared religious believers with philosophical atheists. If the merely apathetic unbelievers are separated from those actively pursuing a self-examined life, the difference from religious believers might vanish completely. The same authors found, for example, that merely having a positive mood and a sense of humor had the same or greater benefit as spirituality on mortality and health for all populations.

In his endnotes, Stenger points out that this review study’s authors even warn that they “found evidence of publication biases” which “indicates that results [like this] should be interpreted with caution.” In other words, they found evidence of an under-reporting of negative outcomes, preferring to publish only when the results make religion look good. Which inflates the evident effect beyond the actual. The studies they aggregated also do not fairly compare more philosophically devoted atheist population groups, but again simply lump all nonbelievers together.

But above all, they only found a correlation between church attendance and mortality benefit (in fact, the variable “religious/spiritual coping” had no results worth tabulating). Moreover, the authors observed that “the protective effect on mortality of Christian church activity or attendance…was quite similar to that of organizational activity in general, suggesting that this effect may not be restricted to Christian faiths alone,” or indeed religion at all. Which is telling because it brings us back to the original point I made about what’s wrong with studies like these:

[A] more recent study similarly found that previous studies showing life satisfaction increases with religious belief were only finding that friendmaking and social networking, not the belief itself, generate the effect: Chaeyoon Lim and Robert Putnam, “Religion, Social Networks, and Life Satisfaction,” American Sociological Review 75, no. 6 (2010): 914-33. Lim and Putnam concluded that “religious belonging, rather than religious meaning, is central to the religion-life satisfaction nexus” (926), which means atheists who feel they belong to a group or movement and participate therein will probably see the same benefits, and given the findings of the Chida study group [whom Stenger cites], this is as likely to be true for health and mortality.

(Quotes from The End of Christianity [ed. John Loftus], pp. 330 and 419, nn. 92-94. The Chida studies are Yoichi Chida, Andrew Steptoe, and Lynda H. Powell, “Religiosity/Spirituality and Mortality,” Psychotherapy and Psychosomatics 78, no. 2 [2009]: 81–90; and Yoichi Chida and Andrew Steptoe, “Positive Psychological Well-Being and Mortality: A Quantitative Review of Prospective Observational Studies,” Psychosomatic Medicine 70, no. 7 [2008]: 741–56. The latter’s finding has been challenged, by arguing that the degree of effect that positive mood has on mortality is very small, but large or small, it still found religiosity had the same degree of effect.)

There is another study often cited, which also illustrates the flaws in using these studies as an argument against atheism…

In Mayo Clinic Proceedings 76, No. 12 (December 2001), pp. 1225-35, we hear of a meta-study where once again correlation with higher mortality was only consistently found for frequency of church attendance, not religious belief (indeed, even whether a person is counted as “highly religious” was solely based on frequency of church attendance). Not only does this run afoul of the confounding discovered by Lim and Putnam (thus eliminating “religion” as in fact the actual variable, and leaving only social activity in general and having any social identity as the true causal factors), but it also runs afoul of the problem Stenger called attention to: sick people die more often but also, by being sick, attend church less often. Thus we expect higher church attendance to correlate with lower mortality even if church attendance isn’t causing that effect. Rather, variables correlated with higher mortality cause lower church attendance.

The same Mayo meta-study also found a positive correlation between Orthodox Judaism and lower rates of cardiovascular disease, but this far more likely has a genetic explanation (as Orthodox Jews are more racially insular than secular Jews) if not dietary or environmental (as Orthodox Jews tend to eat different diets and live and work in different social and physical environments than secular Jews overall). This can have no bearing on religion in general, much less theism. Because we know that no such effect has been found for any other religion, and in fact the more recent Chida group meta-analysis in 2009 found that religiosity overall actually correlates with higher rates of cardiovascular disease.

(Incidentally, the Mayo study conceals this fact with the cleverly worded sentence “of 16 studies examined in a recent review, 12 found that religious involvement was associated with less cardiovascular disease or cardiovascular mortality,” which conveniently doesn’t tell you the other 4 found the opposite result, and that when in fact a proper meta-analysis is performed on these and other studies, the effect not only disappears, but reverses. Always be skeptical when the number of studies is being compared, and you aren’t being told the size of the studies. Or the confidence levels and intervals, as one study can use very loose standards and another very strict ones, but when an effect is only seen under loose standards but not strict ones, you know the effect probably doesn’t exist.)

The Mayo group located one other study in which, apart from obvious medical variables, “participation in social groups and lack of strength or comfort from religion were the most consistent predictors of death,” but “participation in social groups” doesn’t even refer to religion (the study counted all “participation in social or community groups” as the same) and “religion” was primed and framed to exclude philosophical atheists from being measured. Thus, atheists who took comfort from their philosophical worldview or outlook (e.g. humanism, naturalism, scientism) and who would otherwise have reported receiving “strength and comfort” from it were not measured. So this study was incapable of discovering whether that had the same benefit or not.

Accordingly, “those without any strength and comfort from religion had almost three times the risk of death as those with at least some strength and comfort” does not mean atheists had three times the risk of death, since atheists might take “at least some” strength and comfort from their own belief system, and this study did not test for that (instead, it just tested for religious people who were getting no comfort from their religion). Similarly, “not feeling deeply religious was associated with an increased risk of death,” but this again lumped together people who had no philosophical outlook on life and people who do (only without religion: again, philosophical atheists, humanists, and so on). So this study, like so many others, simply didn’t test the effect of philosophical atheism on mortality. At all. It therefore cannot be used as an argument against it.

I needn’t continue. In these same ways the rest of the Mayo meta-study is multiply-flawed, conflating social activity with religious belief (note the repeated use of the phrase “religious involvement” as the correlating variable), and not even testing the effect of atheist philosophies (like humanism) on the measured outcomes, or philosophically contemplative atheism generally or even just being part of an atheist movement. None of its results therefore can be used as arguments against the equal or (for all we know) greater benefits such active or contemplative forms of atheism might afford in all the same cases.

Thus, when the Mayo authors conclude the only discernible reason religion can have its positive effects, after controlling for the (ultimately secular) variable of social activity and support, is that it “engender[s] positive emotions such as hope, love, contentment, and forgiveness and limit negative emotions such as hostility,” we can immediately conclude that atheist philosophies should therefore, by their own reasoning, be just as efficacious as religion (humanism, for instance, promotes all those same values). So on that account, the Mayo study actually confirms the benefits of organized atheistic humanism. Certainly, none of its results argue otherwise (since they don’t measure it).

Although even their reasoning can be questioned, considering the evident behavior of many religionists is routinely hostile, rarely forgiving or typified by contentment, and more often based around hatred and fear than hope or love (just watch Fox News). If religion does not in fact engender such positive emotional states any more than nonreligion, then the co-variable must be something else, such as the comfort generated by having a social identity (as discovered by Lim and Putnam), which is again something atheists and humanists can share, or the comfort (and healthy behavior) generated by having “a cognitive framework for finding meaning” in life (and in the face of life events), which is again something atheists and humanists can share.

Conclusion

Ultimately, there is no science that shows humanistic or philosophical atheism is bad for you in any way at all. It has no demonstrated negative effects on happiness, health, or sanity. And it doesn’t cause suicide. There is science to back the conclusion, however, that embracing a strong social identity and regularly attending community or social events (and being a part of something in general) is good for all three, as is having a good social support network (friends, family, a helpful community) and a self-reflecting cognitive framework for finding meaning (a worldview, a philosophical outlook).

But there is another conclusion to take away from this that I think is even more important: it is scandalous that we are excluded as a demographic from these studies. Religions (even specific denominations) get measured and studied, but organized and philosophical atheism is invisible. We get lumped in with just “everyone else,” a category so broad that in many cases it even includes religious people (such as those reporting “little or no comfort from their religion”), and certainly includes many people who don’t have a strong sense of identity, or a useful worldview, or regular ties to any community outside the home.

Atheists exist now as an identity movement, with networked communities in real space, locally and nationally, and online. We have good cognitive frameworks for finding meaning (e.g., humanism, naturalism, philosophy in general). We even have our own charities and educational institutions. It’s high time someone included us as a distinctive population group when looking for co-variables to human health and happiness.

Update: For more examples of misrepresented studies being used to fabricate bogus claims that atheism is bad for you, see this extensive fisking by Nicholas Lamar Soutter at The Atheist Codex.

§

To comment use the Add Comment field at bottom, or click the Reply box next to (or the nearest one above) any comment. See Comments & Moderation Policy for standards and expectations.

Discover more from Richard Carrier Blogs

Subscribe now to keep reading and get access to the full archive.

Continue reading