The certainty of the “New Atheists”

From the excellent Jonathan Haidt:

…I took the full text of the three most important New Atheist books—Richard Dawkins’ The God Delusion, Sam Harris’s The End of Faith, and Daniel Dennett’s Breaking the Spell and I ran the files through a widely used text analysis program that counts words that have been shown to indicate certainty, including “always,” “never,” “certainly,” “every,” and “undeniable.” To provide a close standard of comparison, I also analyzed three recent books by other scientists who write about religion but are not considered New Atheists: Jesse Bering’s The Belief Instinct, Ara Norenzayan’s Big Gods, and my own book The Righteous Mind(More details about the analysis can be found here.) 

To provide an additional standard of comparison, I also analyzed books by three right wing radio and television stars whose reasoning style is not generally regarded as scientific. I analyzed Glenn Beck’s Common Sense, Sean Hannity’s Deliver Us from Evil, and Anne Coulter’s Treason. (I chose the book for each author that had received the most comments on Amazon.) As you can see in the graph, the New Atheists win the “certainty” competition. Of the 75,000 words in The End of Faith, 2.24% of them connote or are associated with certainty. (I also analyzed The Moral Landscape—it came out at 2.34%.)

There is more here, and for the pointer I thank Eric Auld.


I'm a Jehovah atheist and an Allah atheist. About the Flying Spaghetti Monster I'm more of an agnostic.

I wonder where the works of Christopher Hitchens on Mother Teresa would rank on the Certainty Scale? Truth is, the modern day militant atheists are not really any different from the rest of us: they just want to increase their page counts and audience views of the articles they write. It's hard to stay relevant for long in the public eye (I admire TC for doing so).

The thing about New Athiests that bugs me is that they will be cruel to people who are harmless, or even good. They are indiscriminate. As an example,

‏@RichardDawkins - 22 alleged examples of creationist low intelligence. Is it a joke? It's hard to believe such stupidity really exists.

Many of those are simple and heartfelt, and not even opposing science or evolution. But they're all stupid, sez cool kid atheism.

I'm not a fan of the New Atheists nor their style, but reading those slides...

I'd rather have Dawkin's shrill tenor than the ignorance of those slides.

To give one example, "BANG it happened" girl sounds like she could pass physics. She hasn't set her god in opposition to science. Or as another, there is nothing wrong with answering "are you scared of a divine creator?" It is useful introspection. I mean, sure, "sunset" girl needs some help, but she isn't everyone there.


But... they brought those signs to an academic debate. Specifically instructed to show those signs to the other side.

These aren't Hallmark cards in the grocery store.

I'm not saying the people are evil or idiotic or unworthy of life. But... that stuff is junk.

It never really was a debate, and I pretty much agree with Switek's dismissal. It was a meet-up, but as I say, I don't think all "creationist" comments were terrible, especially given that these people (a) probably have limited science education, and (b) are probably heavily inculcated in an anti-science view.

I guess expecting the glass to be empty, I was surprised to find it half-full.

Alexei, they did not.

They were at this event. Some guy walked around and asked them to put their best argument in Sharpie on a pad of paper in the form of a question. Then he took their picture in order to pull out from the answers the ones that he thought would best illustrate how stupid the people he disagrees with are.

I might question their intelligence only because they should have been smart enough to tell him no.


I think if you delved into the field you might be surprised that the hard core creation science supporters are neither a nor b. They are putting forth several alternative theories, and it's not your grandmother's creationism any more.

I don't find the arguments convincing, but what I have found after sincerely and openly asking about this stuff is how helpful it is in forming a critical eye in several fields. True scientific inquiry dies when people are afraid to question the current wisdom, and I think this will probably be the contribution creation science advocates make, reminding the culture of the importance of doubt and challenge. I'm sure some will think this dubious, some ironic.

Marie, scientists and the scientific public have a lot of interesting discussions under the broad category of evolution. Creationists, in contrast, no matter how liberal, want a yes or no framing. As much as I am sympathetic to lay people, I think that when they approach science with that reduction not much productive can come. Any interesting question about the origin of the species is just a chink to exploit.


Your experience sure might be different than mine, but one of the first things a friend (who does have a science degree, not sure what field) explained to me was that creation science believes some definitions of evolution are obviously accurate.

As someone who previously had believed creationism was monolithic and knee-jerk anti-evolution in all senses, this opened a window for me.

But you may have more interaction with the field than I've had.

To be honest, those creationists I meet tend to be 6 million year old earth types, or excuse for the dinosaur types.

The scariest thing about the slides. . .Only one person used "there" correctly. "How do you explain a sunset if (sic) their is no God?" and "...reject (sic) their being a God...".

I think if you're going post signs all over the internet, check the spelling and grammar of each. Also, when enumerating, the digits one through nine are written out unless a larger number also appears in the same sentence.

In this day and age, that seems trivial, but if you're trying to appear intelligent, don't make stupid grammatical errors.

John Personna: I beg to differ to your statement: I don't think Atheists are cruel to people who are harmless... I don't know where you live or who you hang out with. Historically, it has been those believers who have done the most damage to people who didn't share their beliefs look at the Inquisition for god's sake. BTW I do agree that those who call themselves creationists have a low IQ. Give me a break, in this day & time who would believe such nonsense? I'm agnostic btw.

It's nothing personal. Dawkins' profession is science education. If you insist on not just remaining ignorant of basic science but also parading your ignorance in public, you are working to counter his life's work and so you might expect a bit of a smackdown. I don't think Dawkins et. al. have much intention of trying to convince hard-core creationists that they are wrong because there is little evidence that this can be done, at least through logic and persuasion. Instead, the idea is to reach the people who don't know the science that well but who also haven't made up their minds. Any of the arguments these people scribbled on evolution can be easily refuted by picking up an elementary textbook or popular audience book on evolutionary biology.

Dawkins wasn't cherry picking those, so I don't see how you can call him "cruel" for that. Maybe if the person collecting those had sought out idiot creationists to publicize you could call him that.

Not to be too pedantic, but Jehovah/יְהֹוָה, Allah/‏الله ‎, and YHWH/יהוה‎ are all the same the same diety, worshipped by strictly monotheistic religions which agree without any reservation that this diety is the one they all share. Admittedly, Christianity, a not exactly strictly monotheistic faith, complicates this picture, though it too claims to be composed of worshippers of the same diety as Jews and Muslims. Though only as part of a trinity which is, and is not in some way not possible for mortals to comprehend, the same as Allah and YHWH.

That the strictly monotheistic followers of Allah and YHWH find this to be ridiculous should go without saying, but when there are those unaware that Jehovah and Allah are just different names used for the same diety, it is clear that pure monotheism still has a lot of work ahead of it.

And rereading that comment, it occurs to me you just might be aware of that, while still being a devoted follower of the Trinity, in all its majestic mystery and distinction from pure monotheism.

@PA- of course a learned skolar like you does realize that the early Jews were not monotheists but henotheists ( Passages about how YHWH is the best of the competing gods are still found in what the Christians call the Old Testament. Later on, the Jewish and Christian writers tried to expunge this more and more, and make it sound like there was only monotheism. And let's not forget the concept of a heaven did not occur in Judaism until the Maccabees in the 2nd B.C.E.

I'm not an expert on Islam, but I understand the whole flap over The Satanic Verses was to do with Mohammad, years into his movement and having made little ground, playing footsie with ideas of polytheism in order to attract adherents, which was later whitewashed after the movement had established itself.

Do you mean monolatrism?

@PA: while in Greece I was impressed by the Islamic Sufi Bektashi Order ( ) which worships icons of the Virgin Mary and Jesus Christ, seemingly in contradiction to the Koran. I could not quite figure out this synthesis of two religions but it was pretty cool, kind of like those Muslims in the Persian Gulf that pray upright.

Much too pedantic; how can "strictly monotheistic religions" be said to agree? Agreeing is not a circumstance that an abstraction such as a religion can accomplish. To agree, you need people. And their nature is often as not to disagree.

I have a special Allah who is not Jehovah or Yahweh. Please don't be an infidel towards him.

Of course, He could be a schizophrenic, in the style of the hackneyed, old jokes.

Dawkins is at least as much anti-clerical as he is atheist, a distinction far too many Americans seem no longer able to recognize.

It is quite possible to have no belief in what various faithful people believe without caring in the least that those people have such beliefs.

Unlike anti-clerics -

'Anti-clericalism is a historical movement that opposes the clergy for reasons including their actual or alleged power and influence in all aspects of public and political life and their involvement in the everyday life of the citizen, their privileges, or their enforcement of orthodoxy.[1] Not all anti-clericals are irreligious or anti-religious, some anti-clericals have been religious and have opposed clergy on the basis of institutional issues and/or disagreements in religious interpretation, such as during the Protestant Reformation.

Anti-clericalism in one form or another has existed through most of Christian history. Some philosophers of the Enlightenment, including Voltaire, attacked the Catholic Church, its leadership and priests claiming moral corruption of many of its clergy.'

Admittedly, the wikipedia entry also gives a bit of insight into why most Americans are unfamiliar with the broad concept, even if Quebec offers a local example of it in practice - 'In Quebec itself, the Quiet Revolution of the 1960s broke the hold of the church on provincial politics. The Quebec Liberal Party embraced formerly taboo social democratic ideas, and the state intervened in fields once dominated by the church, especially health and education, which were taken over by the provincial government. Quebec is now considered[by whom?] Canada's most secular province.'

Oops - social democracy rears its head again, even if the statement that Quebec is now the most secular province is unsupported. (Those ethnic Catholics are simply different, at least according to a certain commenter here.)

Dawkins is fiercely opposed to those beliefs which claim the right to determine what is and what is not acceptable in scientific inquiry - such as teaching evolution.

I think it's very accurate to note Dawkins as anti-clerical. His wrath, whether you agree with it or not, is very personal and directed at human agents, not ideas. In this I think he shows a consistency in his atheism, as opposed to some (few) atheists who seem to be arguing with God.

This kind of corpus-based word counting has been popular in linguistics for over a decade, but it is limited if it doesn't involve a qualitative component.

I haven't looked at this research in depth, but just counting how often a text has words like "always" and "certainly" won't necessarily describe the text's confidence as much as it will describe the text's attempt to convince an audience. At least in some cases. This may just wind up describing the genre of the text rather than the confidence of the author.

Again, I haven't looked closely at the study, but I think it's a tad irresponsible to quote this section in a blog. The right-wing books could lack those adverbs because they're not trying to convince the world of a hypothesis as much as they are trying to affirm a pre-conceived belief.

I don't have a problem with the tool itself, but a bit of an issue with how it was used. He used LIWC (Linguistic Inquiry and Word Count), which has done work to validate their categorizations. See for more details. "Certainty" is a subcategory of Cognitive Processes, under Psychological Processes.

This is more appropriate for larger texts like books, but I don't like the fact that only Certainty was included. I would have liked to see the other cognitive subcategories (Insight, Causation, Discrepancy, Tentative, Inhibition, Inclusive, and Exclusive) just for comparison, plus one or two negative emotion categories (anxiety and anger, maybe).

I also agree that including texts that target (what I assume to be) very different audiences is not helpful. Stick to the religion books, and try to include more than one book per author.

What I'd love to know is scores of some other random works. Say, Freakonomics or The Great Stagnation or Friedman's works or Ayn Rand's.

Does anyone have access to the program?

Or Conscience of a Liberal, or General Theory of Employment, Interest and Money.

I agree - without qualitative analysis of the works (or at least a better explanation of the empirical underpinnings), seems of limited use. It's like when journalists talk about the number of hits on Google for some word or topic as evidence of something - it seems like a very crude tool.

What results could come simply of counting the numbers of question marks and interrogative constructions in respective texts?

I certainly never would have though of this, would I? Every other commenter has undeniably overlooked this angle. I suppose it's always something, isn't it?

How'd I score?

Of course, the arguments of the "New Atheists" also have the feature of being most likely to be correct.

@JT: "Of course, the arguments of the “New Atheists” also have the feature of being most likely to be correct."

Of course, but it's a cheap form of "correct" IMO--> correct by construction. It's like when your girlfriend is right about something because of a tautology, but you *know* you are right overall ;-) This is because the New Atheists are using an old trick in the Scientific Method: Falsificationism ( So, like a weather vane always being 'right' about which way the wind is blowing, the New Atheists are also "right", but in a cheap, unsatisfactory way.

Proof of this is found in the world of scientific hypothesis. The really cool stuff--the creative stuff of science--is done in the world of hypothesis first, not falsifiable fact, then through experiment shown to be "true" (i.e., falsifiable fact). So in a way religion is like this: it deals with metaphysics (hypothesis) rather than falsifiable facts. That said, I would not condone every religion this way, for example one that believes in killing another if they don't agree with you, but in general I think religion 'gets it right' more often than the New Atheists realize.

BTW none of the above is necessarily original thought: I'm just blogging what various deep thinkers besides myself have concluded.

According to the Irrational Atheist, the New Atheists fail at their own standard of reason. They aren't even right about their own arguments, let alone religion.

Paul Feyerabend, anyone? (Or have all dangers of scientism receded already?) Or: why does evolutionary biology seem at least a century behind astronomy and astrophysics?

According to anyone with good sense, the guy who wrote "the Irrational Atheist" is a raving dunce.

Stumbled onto that guy's blog once. Yeah, he's a moron.

You should read Bruno Latour's "Rejoicing" it makes some similar arguments.

To share one discussion I had with a young New Atheist, he explained why all religions are wrong because they rely on a proof of God, I explained that some religions support and even welcome agnostics, and he .. paused a bit before declaring all religions wrong because they all rely on a proof of God.

(It is not even true that all religions rely on a provable god, but the fact that there are agnostic religions kind of shoots the New Atheist narrative to hell. So to speak.)

Some examples of such religions? Genuine question.

NB: This doesn't change the fact that your new atheist was still wrong. Religions CAN'T rely on a proof of God since there isn't one. They rely (quite explicitly in most cases) on Faith...

Reformed Jewish congregations can have agnostic members, for one example. I was reading something by a rabbi who wasn't sure how many he had, but he made clear that they were welcome.

There are certainly religious groups who welcome atheists or agnostics and contain a lot of good wisdom and beliefs.

But that doesn't change the fact that their underlying supernatural claims are unjustified and usually contradicted by multiple lines of evidence.

I think you made an error there, or perhaps the kind of switch I accuse New Athiests of making all the time. Do you reserve the right to define "god," and then judge people on your definition, not theirs? I mean how is a purely spiritual god "contradicted by multiple lines of evidence?" I can only suspect that you are substituting a god that is easier to disprove.

Any definition of god involves the supernatural. A purely spiritual god has the issue of defining the spirit and dealing with the little we do know about the brain and the origins of consciousness. I'm not saying a spiritual ground isn't on much stronger ground than others and you could make a strong argument, but you've still got the big problem of a lack of supporting evidence.

The error atheists make, in my experience, is they think atheism is the opposite of religion. That's a category error. Atheism is a mass movement, under the same tent as every religion and political movement. The main attribute shared by all of them is an obsession with their competing religions. Atheism, a phenomenon of the Christian world, does not spend much time with Muslims, but it obsesses over Christianity.

Are you kidding? The new atheists are so obsessed with Islam you'd think they were hired to do the ideological work of the war on terror. Where do you think Hitchens got the title for "God is not great"?

I was not born last week so my sample size may be larger than your sample size. My three plus decades of atheist data does not include a lot of Islam, but there's still time as I expect to be around at least another three decades.

Have you lived in an Islam dominant nation? It'd be kinda pointless to be an atheist living in Rome obsessed with Buddhism.

True, but atheism as an intellectual movement never got going amongst the mohammedan or sub-Sahara populations either. Organized disbelief has been with us since the Greeks, at least, but the flowering of atheism as we know it is a European phenomenon starting in the Enlightenment. It is wholly dependent on Christianity to exist. That's not to say it will not adapt and make a fetish of Islam. Richard Dawkins probably has another book or two in him.

Well, there's a few atheist movements in India and those sure don't focus on Christianity. Mostly Hindu targeting.

It's a European phenomenon like you say probably because Europe in general has been more enlightened than both the Hindu & Islamic world. It's like asking why is quantum mechanics such a Western world obsession & not so much in sub-Sahara? Or asking why do Western charities obsess with Alzheimer's & not amoebic dysentery?

Rahul 1, Z 0

I know Iranian atheists and agnostics, they're concerned about Islam. Their societies are less tolerant so they're a bit more quiet about it (though it's not as repressive as you'd expect) but they're out there and will become more vocal and prominent as their societies liberalize.

Atheism as you've seen in Western society is generally concerned with Western religion, mostly Christianity with a bit of Judaism.

None of the atheists I know consider themselves part of a "mass movement". I am an atheist, but I have never read a Dawkins or other "New Atheist" book (nor do I have a desire to do so). There are no atheist churches where atheists congregate every week. I hadn't even heard of this "New Atheist" concept until I read this post. Sure, call me uninformed--but that's precisely my point. One can be an atheist without being part of any "movement".

Atheism is just simply not a movement or religion. It's a lack of religion. It's no more a movement than the lack of belief in Santa Claus is. The fact that there are fewer people who believe in the Santa Claus than there are who believe in God has no bearing on how you "classify" non-believers.

On the other hand, maybe you don't mean that "atheists" make this error, but that certain people (i.e., Dawkins, Harris, and Dennett) do? If, on the contrary, you mean ALL atheists, then I have a lot of counterexamples for you.

You may not be that kind of atheist, but there is "a thing"' growing out of‎ and etc.

This is my view and my experience. I don't read the atheist books. I don't need anyone to convince me. And Hell, maybe we need God to keep the lower classes in order. I'd prefer old-time buddhism, but whatever works.

It's the richer folks who are the believers. Maybe it does work.

Actually, what you wrote is a category error as well. If you insist on comparing atheism to religion, the proper comparison would be atheism v. monotheism or atheism v. polytheism not, say, atheism v. Christianity. Is there such a thing as "monotheist identity" or a "monotheist community"? Not really. Despite the efforts of a few atheists to mimic religious communities, it is hardly a necessary element of rejecting religion.

No, it is a reading comprehension error on your part. I put atheism in the mass movement bucket, right next to religion.

Well, you put it in the wrong bucket is what Ricardo (and the rest of us) is saying.

How does their algorithm score a sentence like, "I don't know what the right answer is, but I'm certain those guys have got it wrong, just like they always do"?

I guess you are suggesting a potential flaw in the metrics, but I don't really see it. That sentence still contains an expression of absolute certainty.

Jacob and bliksem do a better job below of explaining what I was trying to get at. Much of Harris' book was about calling out religious beliefs for having too much certainty.

The other thing I was trying to point out is that not every expression of certainty is closed-minded or unscientific. You don't have to know the right answer in order to identify some propositions that are definitely wrong. Indeed, much of science proceeds by identifying error, proving with certainty that it is wrong, and concluding that the truth must lie among the conjectures that cannot be excluded. The hallmark of reason is not uncertainty per se, but rather careful distinction between conclusions that are certain, and those that are contingent or probabilistic.

So, to summarize I'm none too convinced by the methodology. Use of certainty words can indicate closed-mindedness, or it can reflect the sensible exclusion of error. Conversely, a plain statement like "Those guys are destroying the country" expresses a great deal of certainty without using any trigger words.

Based on the word list for "certainty" most academic journal articles would do horribly. e.g. absolute, confidence, distinct, fact, proof....

I'm not sure whether he's measuring anything that means something. Is the word list merely biasing the score against scientific / quantitative work? doesn't give much detail, but you have a high score for "Overall cognitive words" which makes sense given what you're trying to demonstrate.

We really can't conclude much from such a short text. Over the course of a book, or multiple articles, we can find patterns with varying levels of validity.

> How does their algorithm score a sentence like, “I don’t know what the right answer is, but I’m certain those guys have got it wrong, just like they always do”?

And how they deal with sentences like "those guys always seem so certain...", which does not tell one anything about the author's epistemic over-confidence.

All in all, it would be rather useless to simply count the tokens of words like "always" or "certainly" unless the text analysis program can also reliably identify e.g. reported speech, etc etc.

I agree with you bliksem. Simple word counting seems like a very bad metric. Sam Harris speaks quite a bit about the unshakable certainties of religious faith and adherents, both in the "fanatics" who are willing to commit violence, and importantly for this metric, the moderates who doubt the certainties of faith. This would bias the count upward if you thought you were only measuring how "certain" the author is.

Further, the nature of the books would already indicate that certainty should be more important here, as they discuss in a large part, what we CAN be certain of scientifically and what we CANNOT be.

Dawkins' criticism of Christians and Christianity is based on fact but often childishly gleeful and annoying.

Though, I can accept his righteousness on this easier than I can on 'climate change,' where his arguments are certainly not based on empirical fact.

(Maybe he's not as logical as he thinks, and more a lifelong fan of Science akin to a lifelong fan of the Bears or Broncos.)

That's not the way I'd shape it. There are actually only a subset of religions who are in conflict with science. Many, perhaps most, religions have moved to a self-consistent spiritual realm. Those religions don't need to comment on evolution or climate change, because that's not what they are about. Or if evolution or climate change are real, who are we to second-guess God?

Now stepping back to those religions which do decide to take science head-on, they are kind of asking for it. Crazy theories like "fossils are from a planet Satan ruled that later crashed into the Earth" are just asking for it. I mean, planetary collision? Surely you have some astronomical or geological evidence for that? No? You've moved on ...

Where I fault the New Atheists though is that they are indiscriminate and lump anyone "religious" as a evolution or climate change denier. Not all are. Many religious people have the subtlety of a scientist on the confidence we can assign to various aspects and details of climate and evolution.

Dawkins, along with most atheists I know, has consistently maintained that not all religions are equally wrong.

And yet he was willing to shotgun all those signs (link above) as equally "stupid."

You mean all the signs from people prescribing to Ken Ham's specific version of creationism?

Yes, all those people from that specific belief are very very wrong. That in no way conflicts with what I said or what Dawkins said in that interview.

A second analysis looking for words connoting doubt and uncertainty would seem appropriate, it compare internal relative frequency. Breaking the Spell is an odd choice too; from what I remember it was mostly about how little understand about religion as a phenomenon and imploring researchers to pursue a rigorous science of it.

There seems to be a tendency that atheists are more ardent in professing their non-belief than deists.

That position seems difficult to justify in this reality. There seems to be more examples of religious theory being used to condone harmful actions.

If we're looking to all of history, there's a massive sample bias in that statement. Secular philosophy has only existed for a small fraction of human history, yet even in that short span secular beliefs have a record of "condon[ing] harmful actions" that I would stack up against any religious equivalent. See, e.g., Communism, Nazism, Scientific Racism.

That argument seems to be a confusion of type. When we say that religious beliefs condone a given action, we point to actual authorities from that religion who literally reference some religious scholarship as a justification. Are you making the same argument for secular beliefs, or are you instead referring to beliefs held by people who are secular?

Nazism was motivated by religion (I think Catholicism to be specific), not secularism and definitely not atheism (nor evolution).

Scientific racism was unfortunate but more an example of existing racism being expressed in scientific terms.

I think the only one that could be connected to secularism is communism because it came from a line of philosophy that excluded religion, but I find the causal connection to be very tenuous (as opposed to religious motivated harmful philosophy where the connection is explicit).

Hitler was not secular

Alternative hypothesis:

The "New Atheists" use the same number of words associated with certainty per argument as other authors, but write more clearly and concisely, with less fluff.

God Delusion - 464 pages. Selfish Gene - 364 pages. Greatest Show on Earth - 496.

Null hypothesis established.

were they all double spaced ? with the same margins ?

New Atheists are be able to express a high level of certainty with respect to his or her "opinion" (or for that matter, Christians, Muslims, Economists!!) because those opinions are not subject to any market test. Therefore, I am of the opinion that if they w(c)ould be held to a high level of potential financial liability for getting their opinions wrong, just like lawyers (the level of confidence here being directly proportionate to the enforceable and verifiable consequences of being wrong), it is more likely than not they would couch the robustness of their conclusions in terms like these:

--"Not frivolous;
--"Reasonable basis";
--"Realistic possiblity of success':
--"Substantial authority";
--"More likely than not: and

That is, of course, all of these terms are preceded by a large list of assumptions and exclusions.

Ever heard of Pascal's wager?

'Pascal's Wager is an argument in apologetic philosophy which was devised by the seventeenth-century French philosopher, mathematician, and physicist Blaise Pascal (1623–1662). It posits that humans all bet with their lives either that God exists or does not exist. Given the possibility that God actually does exist and assuming the infinite gain or loss associated with belief in God or with unbelief, a rational person should live as though God exists and seek to believe in God. If God does not actually exist, such a person will have only a finite loss (some pleasures, luxury, etc.).''s_Wager

Admittedly, Pascal's wager was probably not intended to cover religions that encompass reincarnation, while possessing a very different concept of 'god/s' - Hinduism and Buddhism come to mind, though it might still be applicable.

Taoism seems utterly indifferent to Pascal's wager, however - but then, Taoism normally appears utterly indifferent to most things that Abrahamic religions consider crucial.

"...and assuming the infinite gain or loss associated with belief in God or with unbelief..."

That's quite a large assumption. Sounds like Pascal was a lawyer.

The idea of eternal damnation for non-believers was likely introduced into some religions for precisely that reason. Without consequences for being wrong, opinion certainty is a cheap commodity.

Pascal's wager fails because

a) there are many many mutually exclusive religions (just listen to Ken Ham last night), if you choose the wrong one you're screwed anyways

b) you're assuming a god that punishes honest atheists and rewards cynical believers

c) you're assuming the cost of an incorrect belief is small enough to justify the minuscule odds of being right.

Why is an atheistic book compared to a political one? Why not compare it to a Ray Comfort book or something?

Hmmm. What does this textual analysis make of Dawkins' chapter entitled "Why There Almost Certainly Is No God", in which he goes to great length explaining how he cannot be absolutely certain.

If he is not certain, then why is he using so much certainty language (never mind claiming his opponents have a "delusion")?

I haven't read the book, but maybe he's often talking about the lack of certainty? That would necessarily involve using lots of "certainty language", right?

For sufficiently small values of P I'm comfortable using the word 'certain'.

There are also many specific claims religions make which are falsified with very high certainty.

Dennet's "Breaking the Spell" scored high on the certainty statistic, but still find it odd that is lumped in with "The God Delusion" (I haven't read Harris' book).

All Dennet claims to do -- and all he does -- is muse about how religions come about as human institutions. I think I even points out that there can be one religion for which the answer is "because it's true, and we have some way of knowing it". Unlike Dawkins he bends over backwards to be polite to believers. I find it telling that Christian writers feel he is attacking them.

I expected more anger from anthropologists and historians dismissing him as a dilettante philosopher.

Dawkins is not a leading thinker for atheism, but of a fairly marginal denomination within atheism.

His mission is not to convince committed theists or agnostics of anything, but to win over adherents from within the atheists that already exist.

He is best compared to someone like Jack Chick.

Or, in the opinion of a friend who earned her biology PhD at GMU (I know her doctoral adviser was also an Englishman, and I believe he was also the department chairman in the later 1980s), he is similar to Galileo, in the sense that a certain rabid group of believers reject his science, without even bothering to actually know that science at all. Or rejecting it out of hand, of course, which does not make their wilful ignorance more attractive.

That's a joke? Take on Godwin's Law?

I get the thrust, but I'm not certain this is the way to go about telling the story. My sense is that lightning rod issues will generate a higher baseline of certainty claims in public as people employ more rhetoric in the absence of evidence. What I'd like to see is a relative analysis to establish baseline certainty about climate change, fetal personhood, the minimum wage, and also some innocuous things that have less politial or ideological weight such as ... i'm strugglng because all unknowns I can think of seem to be political these days. But something like that. Then, let's evaluate works relative to neutral subject matter.

I'd love to put Krugman through that kind of analysis.

This all reminds me that I need to finish Nagel's Mind and Cosmos....

Logical positivism is simply the official religion of our society now. Science and reason are the beginning and the end - the alpha and the omega - of all things. We value only those things that can be represented and proved to be empirically and theoretically valuable in abstract terms. Yet the rates of mental illness and suicide climb steadily. We seem to be grasping at something more desperately than ever while our insecurity about that thing secretly grows and grows.

That is because we are cheaply and poorly defining science and reason.

We live in an era that deifies them, yes, but also an era where you could stop most folks on the street and say, "My border collie is a dog. Border collies are dogs. Therefore all dogs are border collies, so your poodle is not a dog", and half of them would scratch their heads and walk away feeling like they'd learned something that day, and go buy cat food.

If you put it on a CNN segment, they'd forward it to all their friends and they'd all start buying cat food, too.

And if Dr. Oz was the one interviewed saying it, it would go viral.

Bumper sticker I saw this a.m.: "If it isn't an Airdale (sp.?), it is just a dog."

I didn't get your point. Not everyone understands logic? Humans have a herd mentality?

Again, true, but what's new? I doubt you'd say the population 100 years ago was any more immune to bad logic or herd behavior?

I don't think too many people, scientists included, hold to a verifiability theory of meaning.

I bet they don't, but I bet they think all their friends do!

Too bad Wendell Berry doesn't comment on this blog.

"rates of mental illness and suicide climb steadily. "

Is this really true? Do we have data on this? Or is it just better measurement & more expansive definitions?

Haidt is too full of TED-talk for my taste, but his work is instructive, just not in the way he concludes. I am referring, in particular, to The Righteous Mind, a book that liberals won't read but definitely should. As for his word counting, it reminds me of Gematria, the technique of interpreting the Bible by assigning numbers to the letters and words in it and then adding up the numbers.

Some of the weaknesses of arrogant people are their misunderestimations of their limitations.

You never can tell. You may go to Heaven. Or, you may go to Hell.

Kid's rhyme:

Go to A.
Go to B.
Go to C.
Go to D. . . . . .

My reaction to The God Delusion was similar to what Haidt is saying. Dawkins is a marvelous writer and a top-notch scientist. I've read ALL of his science books (and all of Dennett's works except for Breaking The Spell) and I think they are among the top thinkers of our age, but I felt The God Delusion was far and away Dawkins' least compelling effort. I think he wrote it as a catharsis- getting a lot off his chest after having to deal with the stupidity of Creationists for decades.

I find this very weak, for several reasons. Most notably, certainty is one of the major *subjects* of all three New Atheist books listed. You might expect words like "certainty" to crop up more often for this reason alone, even if that certainty is being criticized. I would also expect that words like "doubt" and "uncertainty" would be disproportionately common in these books, for the same reason. (So should we simultaneously conclude that the authors are especially doubtful?)

I do appreciate the irony though, whether real or fictitious - The promoters of epistemic humility indulging in atypically confident rhetoric.

Meh, I see the point he's trying to make, but that's some pretty weak NLP sauce. For instance, Haidt's article itself would score pretty high on its own metric. Without context, maybe all the New Atheists are bemoaning the professed certainty of others, not expressing certainty themselves.

Right. A more refined predicate analysis would be more appropriate, comparing sentiments between one corpus vs another.

Agnostics are beyond Atheists as Atheists are beyond Theists.

Are you agnostic about unicorns? Minotaurs? Santa Claus? Or, from Sagan's "The Demon-Haunted World", are you agnostic about my claim to have a fire breathing dragon in my garage?

I suppose one could claim to be technically agnostic about them all (and hence about everything), but then that reduces agnosticism to a pretty meaningless concept. Is agnosticism beyond adragonism?

But is possible to believe in a religion that requires no belief in the supernatural at all? Religion isn't simply a Western construct, after all - Buddhists and Taoists both come to mind as 'believers' that do not require anything apart from the world around us, and a perspective on how humans are to view it. One that is distinctly different from anything in the Abrahamic tradition, of course.

No orbital teapots are required in their respective cases. Starting, in the case of Taoism, with the idea that words are not the same as what they describe.

If you strip out the supernatural elements, it's just philosophy.

Why is there something rather than nothing? Unicorns don't figure in here.

Sure they do. The point is simply that it is not always better to reserve judgement. Perhaps you believe in unicorns, or are agnostic about them, in which case it is a bad example. But most of us reject unicorns.

Maybe what you really meant is that you think the evidence for theism is better than the evidence for unicorns (and maybe in particular that the existence of the universe is evidence for theism). But consider this, as much as I don't believe in unicorns, I put *even lower* odds on a band of unicorns who created the universe.

Q: Why is there something rather than nothing?

A1: Because unicorns exist.
A2: Because God exists.
A3: Because God, in the form of a band of unicorns, exists.

A1 has to be > A3, since A3 is a subset of A1.
A2 has to be > A3, since A3 is a subset of A2.

But, other than the A3 subset, I don't see how A1 has any bearing at all on the question.

Willem of Occam chimes in and says: what's all this unicorn business?

A2, IMO, is the only meaningful non-zero probability answer to the question.

To Brian Donohue below (regarding whether the existence of the universe is evidence for the existence of an entity which created "God"!) -

I agree that "unicorns exist" is not a meaningful explanation of why the universe exists. But I think it is on equal footing with A2 ("God exists"), at least pending further specification of what you mean by "God" (which btw may destroy A2 > A3). You are simply renaming the problem. Why does "God" exist? If you are willing to say without explanation that God exists, then why aren't you willing to say without explanation that the universe exists?

This might be uncharitable, though. Maybe you have some more specific idea of god in mind which sets it apart from unicorns and answers my objection. If so, you should be able to win the following game. 1. Define some entity G and ascribe properties to this entity. 2. Explain why this entity exists in terms that would not apply to the universe. 3. Argue that the existence of G explains the existence of the universe.

If you can do this, I'll admit that you're onto something. My prediction is that you'll fail on step 2, but it would be fun if I'm wrong. Otherwise, I'll stick with Willem, and pick B1 over B2:

B1: The universe exists.
B2: The universe exists, and also some other entity exists which created it.

(Comments on starting my game: Note that "G=entity which created the universe" does not work, because it fails at step 2 (we haven't explained why G exists). Also "G=entity which created the universe, and does not need an explanation for its own existence" does not work, because if that move is available to us we can play it on the universe instead (failure at step 2 again). Also "G=entity which created both the universe and G" does not work unless you can specify further properties of G which justify the self-creation move, but which properties are not shared by the universe (failure at step 2). Etc. I have tried hard at this game and always lost, but maybe Brian or someone else can succeed where I've failed. Or maybe you want to reject my game and argue for a better one. If so it'd be helpful to specify what game you are playing and why you think it is better.)

Can I ask you a question?

Why is there something rather than nothing?

Extra points for brevity.

A unicorn willed it into existence.

I'm afraid that the "unicorn" argument is a "given" one. I mean that in the sense that New Atheists believe (like the question of gods itself) that there is one question, and one answer. In fact, not so much.

Do I believe in unicorns? In what sense? As physical beings from earth's present or past? I can ask for evidence of that, and look at the fossil record. Of course, what if you mean spiritual unicorns, or unicorns on Kepler-61b? Those I have no way to test, and remain completely agnostic.

Of course, a well-schooled "new atheist" will insist that there is only one kind of unicorn, the testable kind, and that god is like that. A huge logical fail.

Can someone run the analysis on his article?

Ah I'm only getting around 0.5%. He's a good boy.

I think there's something of a connotation here that because the new atheists are so convinced, it means they are likely more close-minded. I don't think that's the right view of things.

We could ask people to write books about how there are no unicorns, no mermaids, about how the religious beliefs of the Greeks or Romans or Native Americans are false, and I think these books would contain a lot of words of certainty. It's because they have compelling arguments on their side and there are no compelling arguments on the other side.

Religion, generally speaking, is like this, but most people don't recognize it so most people think people's beliefs about the subject should contain more uncertainty. So far as I can tell, they're simply wrong. I believe when you set about carefully thinking about what are appropriate standards of evidence in the world -- what are the appropriate means of discriminating between different theories of the world -- you will find these standards / means tell you that you shouldn't believe any religion.

Since there are probably religious people who read this, I'll outline what I think this "careful thinking" is, cut-and-pasting from a quick note I wrote to someone else recently (though space limitations mean you're gonna have to spend time convincing yourself).

I claim the correct rule for discriminating theories is to choose the one that's simplest, because this is implicitly / subconsciously what we all do all the time. For example we believe that what we observe reflects reality, as opposed to being some Matrix-like hallucination or something, because that's simpler. We believe the patterns we have observed in the past should repeat themselves in the future because that's simpler. We believe the patterns we observe here should also hold there because that's simpler.

These are the prejudices we employ from the instant we wake up to when we go to bed, when for example we assume that hitting the appropriate button on the alarm clock will turn it off, as it has in the past, as opposed to detonate a bomb, which is another logical possibility. When we trust the floor will be there when we step out of bed, as opposed to a pool of sharks. We are always, implicitly, trusting in the regularity of the world.

Science simply extends this, gathering more and more evidence from more and more situations and developing a language for expressing patterns and understanding their complexity. We are all instinctively scientists but professional scientists are trained to do science with the greatest rigor, mindful of all of the evidence and all the tools for discerning complexity (like for instance mathematical consistency).

And what else can you do? Once you abandon the idea that you should prefer the simplest theory, what becomes your standard for choosing one understanding of the world over another? I could tell you that the world was created today, while you were sleeping, just as it was when you woke up, and you only remember it being there before because those memories were created in your brain when you were created in your sleep. On what grounds can you rule this theory out, other than it is unnecessarily complicated?

Likewise with religion: on what grounds does one accept one religion over another? People somehow convince themselves that their religion gives a simpler description of the world than other religions (given pieces of evidence like the scriptures they are familiar with and feelings that they have) but somehow failing to get that if simplicity is the standard then they should throw out their religion too. The simplest description isn't that all scriptures but one have "natural" explanations, the simplest description is they all have natural explanations.

I don't think it's simplicity that draws people to religion; it's completeness. Haidt's article points to the reality that people prefer a feeling of certainty. Religion gives a sense of certainty about the world around us.

Regarding the appeal of simplicity, many religious beliefs are intensely complex and nuanced. They appeal because the nuance, refined over time, answers otherwise unanswered questions.

As a religious person, this is what frustrates me most about new atheists; they often don't respect the nuance of religious belief. Perhaps the most popular rhetorical tool (used by some commenters on this very article) is to reduce religious belief to belief in unicorns and Santa Claus. A belief in unicorns has nothing to say about why bad things happen to good people. Being a good boy so I can get presents from Santa is nowhere nearly as sophisticated and thoughtful as religious beliefs about why I should care about a total stranger. The nuance matters because it explains and motivates.

If there is a meaningful, enduring appeal of simplicity in religion, it's the simplicity on the other side of complexity. The kind of simplicity appreciated by people who have spent significant time working through the complex ideas and harmonizing them. (It's akin to the joy you see in Feynman's face when he describes the beauty of the scientific world.) This level of expertise on a subject can't be easily explained, so it might look foolish or incomplete or hopelessly complex to others.

Many varieties of nuance are undeserving of respect.

I agree, but you can't know which deserve respect without understanding them first.

In my opinion, saying "Your belief in God is just like believing in Santa Claus," is much like saying, "How can there be global warming if it's so cold today?" Both use a simplified straw man instead of addressing the complete concept.

Indeed instead of the multitude of concepts.

That is a very good analogy.

I've read enough nuance. Augustine, Aquinas, Descartes, Locke, Kierkegaard, Hegel, &etc. It's really not worth the bother, qua validation of religion and God. Certainly plenty of use for thinking about other kinds of issues though, which are incidentally but not necessarily related to religion.

I've read enough nuance. Augustine, Aquinas, Descartes, Locke, Kierkegaard, Hegel, &etc. It's really not worth the bother, qua validation of religion and God. Certainly plenty of use for thinking about other kinds of issues though, which are incidentally but not exclusively related to religion.

You say "Religion, generally speaking," but don't we know that just over half of scientists (51%) believe in some form of deity or higher power?

The reality would seem to be that there is not a binary choice of religion and simple explanations, vs science and complex views.

As you climb the science totem pole, you find fewer and fewer scientists who are religious. My personal opinion is that the religious ones haven't given the question enough thought.

I could have / should have emphasized that the "you should believe the simplest theory" is of course not necessarily the path to truth. And this is the caveat that allows a lot of smart people (including those scientists) to comfort themselves that their religious beliefs are not stupid. However I think these people fail to recognize that they are adopting a fallacy of "special pleading" when they take this route: the "believe the simplest theory" philosophy is really the only standard we have -- if you give that up, there is really no guidance as to what you should and shouldn't believe -- and I think this isn't more appreciated just because you have to be smart and thoughtful to really fully appreciate it. (And just because someone is a scientist doesn't mean they are smart and thoughtful on this question... there are similar dynamics as with politics that prevent lucid analysis on these questions.)

Isn't that a plea for us to believe that the best scientists support your argument? An argument from authority without their actual support attached?

Religious doctrine can be pretty complex:

REDEMPTION, n. Deliverance of sinners from the penalty of their sin, through their murder of the deity against whom they sinned. The doctrine of Redemption is the fundamental mystery of our holy religion, and whoso believeth in it shall not perish, but have everlasting life in which to try to understand it.

- Ambrose Bierce

"Certainty: the sure epistemic means for attaining approximate, partial, or tentative epistemic ends."

This IS the standard of "certainty" under discussion here, correct?

A disease which the patient and his friends frequently mistake for deep religious conviction and concern for the salvation of mankind. As the simple Red Man of the western wild put it, with, it must be confessed, a certain force: "Plenty well, no pray; big bellyache, heap God."

- Ambrose Bierce

It is folly to think that someone is wise just because they equivocate all the time. Being "certain" of your beliefs is a good thing if you actually have sound evidence or a reasonable prior to back those beliefs up.

Equivocation, uncertainty, and risk are actually three different things.

A very erudite and interesting discussion. But - pointless. Unless claims of religious truth can be submitted to an empirical test (at least potentially) they are meaningless and discussing them is a waste of time. (See "Language, Truth and Logic" by A. J. Ayer.)

For followers of an economics blog, you're all surprisingly caught up with logical arguments and what these people think. But what is the correlation between certainty words and books sales? I'm willing to bet it's positive - people are happy to pay to have a public figure confirm their opinions, but they're not so willing to part with their money for a book they (gasp) might not agree with. Authors know that and leverage that. All of the New Atheists derive large portions of their income from these book sales, and so it makes completely perfect economic sense for them to overstate, simplify, and include as much emotion and passion as possible in their words. What's right or wrong is as irrelevant to them as it is to Beck or Coulter - they're all completely capable of convincing themselves that what makes them money is right - that's a key human characteristic!

You are convinced the tooth fairy does not exist. I am not sure. Therefore, I win (the meta-argument).

Pretty disappointed to see this superficial "analysis" recommended by Mr Cowen. I'll keep visiting the blog, though...

God is charge. All powerful. Is everywhere. All matter is charge. Charge is undefinable. God is undefinable. We are all charged. Spirit is emotion. We all have personal charge and a personal spirit. So God does exist and spirit does exist. One can describe God with mythological or scientific both describe charge and emotions. Good luck defining God or charge.

Spirit is emotion. It was once called, "fiery tongues" because emotion makes one bold when speaking. School spirit is emotional. One can have a spirited horse. Ones spirit is portrayed in an attitude. One can possess an encouraging or an antagonistic attitude. One can be helpful or a bully. One can be kind or mean. A "holy spirit" is kind an antagonistic attitude is mean. Morals are whether we choose to be encouraging or antagonistic. Man and the animals have an antagonistic nature. An antagonistic nature can be called, "a sinful nature". Man can choose a path of encouragement, a less self-centered manner of navigating thru life. Both encouragement and antagonism are real powers that move people and can be used to manage affairs.
When God is defined as charge some mythological describers say its not enough. They desire a more personal God. They feel charge is not personal enough. However, charge is personal. Whose to say charge is not the thought of God. We are made of charge and all matter is made of charge. What could be more personal than what we are made from.
How these concepts are explained changes with language and the words available to describe the energy of an encouraging and antagonistic spirit. Emotional spirit is shown to exist and be real power because it causes people to move and change.

Comments for this post are closed