WEBVTT
00:00:02.189 --> 00:00:06.070
Welcome to Reasonable Doubts,
00:00:06.070 --> 00:00:11.379
your skeptical guide to religion.
00:00:11.380 --> 00:00:27.030
(Music)
00:00:27.079 --> 00:00:30.709
You're listening to Reasonable Doubts,
the radio show and podcast for those who
00:00:30.710 --> 00:00:32.118
won't just take things on faith.
00:00:32.118 --> 00:00:35.119
I'm Jamie Beahan
and for this special episode
00:00:35.119 --> 00:00:38.539
of Reasonable Doubts, we're featuring a
lecture I gave in February
00:00:38.539 --> 00:00:41.979
to the Grand Traverse humanists in
Traverse City, Michigan.
00:00:41.979 --> 00:00:45.479
The lecture was entitled;
“Does religion make us better?",
00:00:45.479 --> 00:00:49.558
a critical review of the religious
pro-sociality hypothesis.
00:00:49.558 --> 00:00:52.981
Longtime listeners will no doubt
recognize many of the studies
00:00:52.981 --> 00:00:54.640
talked about in this lecture.
00:00:54.640 --> 00:00:59.938
A previous RD-extra and our episode “The Skeptics Toolkit to Psychology of Religion”
00:00:59.938 --> 00:01:01.808
discussed these findings.
00:01:01.808 --> 00:01:05.259
But this lecture is a bit different,
mostly in the trivial
00:01:05.259 --> 00:01:09.439
and, I'm sure, disappointing fact that I
am presenting the findings rather than
00:01:09.439 --> 00:01:11.759
our resident doctor professor Luke Galen.
00:01:11.759 --> 00:01:16.319
But I think this lecture has some merit
in that it finally collects a wide range
00:01:16.319 --> 00:01:20.758
of studies, discussed over several years
on the show, into one convenient place
00:01:20.759 --> 00:01:24.959
hopefully making it easier for fans of
the show to review the information
00:01:24.959 --> 00:01:29.868
or share with a friend and please do
share. This is important research and I
00:01:29.868 --> 00:01:33.569
know doctor Galen would agree with me
in thinking that it hasn't gotten as
00:01:33.569 --> 00:01:36.919
much attention as it deserves.
So you'd be doing us
00:01:36.919 --> 00:01:41.110
and the cause of skepticism
a great favor by sharing this lecture
00:01:41.110 --> 00:01:45.689
on whatever blogs or social media you
frequent and never underestimate
00:01:45.688 --> 00:01:48.209
the power of good old
word-of-mouth sharing either.
00:01:48.209 --> 00:01:52.889
And please visit doubtcast.org to share
any comments or questions or feedback
00:01:52.890 --> 00:01:54.968
you may have about the episode.
00:01:54.968 --> 00:01:58.849
So be sure to tune in next week for
the Doubtcasters review
00:01:58.849 --> 00:02:02.739
of the New Christian propaganda film
“God's not dead”.
00:02:02.739 --> 00:02:07.054
Should be a good one. Until then, take
care and keep doubting.
00:02:07.054 --> 00:02:16.519
(Music)
00:02:16.519 --> 00:02:19.075
(Applause)
00:02:19.105 --> 00:02:21.489
Thank you for coming and thank
you for the privilege of
00:02:21.520 --> 00:02:24.870
allowing me to speak to your group. My
name is Jeremy Beahan.
00:02:24.870 --> 00:02:29.539
I teach World Religions and Introduction
to Philosophy along with a handful of
00:02:29.539 --> 00:02:32.169
other subjects
at Kendall College of Art and Design.
00:02:32.169 --> 00:02:35.299
I'm also the producer and cohost
00:02:35.300 --> 00:02:38.380
of the Reasonable Doubt Podcast which
00:02:38.379 --> 00:02:43.189
at its peak was the top atheist podcast
on iTunes for several years,
00:02:43.189 --> 00:02:46.740
won the People's Choice
podcasting award
00:02:46.740 --> 00:02:51.689
for best religious inspirational podcast
which was - (Laughing) - different.
00:02:51.689 --> 00:02:54.829
People look at me funny
when I mention that.
00:02:54.830 --> 00:03:00.070
I'm speaking tonight on the issue of
“Does religion make people better?”
00:03:00.070 --> 00:03:02.210
and we're approaching
this not so much from a philosophical
00:03:02.210 --> 00:03:05.120
perspective, as you usually
hear this question grapple with,
00:03:05.120 --> 00:03:08.653
but we're approaching this
from an empirical standpoint.
00:03:08.653 --> 00:03:11.146
What can science actually tell us
00:03:11.146 --> 00:03:15.719
about how religion affects morality.
The subtitle here's a skeptical review
00:03:15.719 --> 00:03:17.930
of the religious pro-sociality hypothesis.
00:03:17.930 --> 00:03:21.159
So, that might take some explanation.
00:03:21.159 --> 00:03:24.169
You might guess from that subtitle that
this is going to be a bit have been
00:03:24.169 --> 00:03:26.919
informationally dense talk tonight.
00:03:26.919 --> 00:03:29.479
But I don't have to tell you
that in our culture
00:03:29.479 --> 00:03:32.740
there's an overwhelming assumption
among the general public
00:03:32.740 --> 00:03:36.759
that being religious is necessary
to be a happy and ethical person.
00:03:36.759 --> 00:03:40.069
We have plenty of preachers
and pundits and ordinary people
00:03:40.069 --> 00:03:44.489
reminding us daily that without God
society will quickly de-evolve
00:03:44.489 --> 00:03:49.239
into wickedness and anarchy.
What you may not be familiar with
00:03:49.239 --> 00:03:53.239
as much however is the growing body
of social psychology research
00:03:53.239 --> 00:03:56.830
that at first glance actually
seems to support this notion.
00:03:56.830 --> 00:04:01.520
The more technical term for the
hypothesis that religion makes us good
00:04:01.520 --> 00:04:05.320
is known as the religious
pro-sociality hypothesis.
00:04:05.319 --> 00:04:09.259
My task tonight is to present you with
an overview of this research
00:04:09.259 --> 00:04:13.699
and to acquit you with the tools
necessary to think critically about it.
00:04:13.699 --> 00:04:18.009
Because as we're about to see the
religious pro-sociality hypothesis
00:04:18.009 --> 00:04:19.620
really does indeed
have some support.
00:04:19.620 --> 00:04:24.030
But when we look at the evidence
more closely, we're going to
00:04:24.029 --> 00:04:26.919
discover little devils
within the details.
00:04:26.919 --> 00:04:30.219
But first I have to
give you a quick disclaimer:
00:04:30.219 --> 00:04:32.639
I don't get any credit or blame
00:04:32.639 --> 00:04:36.659
for what I'm about to say this evening.
This is not my research
00:04:36.660 --> 00:04:39.930
that I'm reporting on, this is actually
doctor Luke Galen's research.
00:04:39.930 --> 00:04:44.039
He is a professor of psychology of religion
00:04:44.039 --> 00:04:46.339
at Grand Valley State University and
00:04:46.339 --> 00:04:51.359
almost all what I'm going to be drawing
from tonight comes from his paper
00:04:51.360 --> 00:04:55.830
in the Psychological Bulletin of
the American Psychological Association
00:04:55.829 --> 00:05:01.279
called: “Does religious belief promote
pro-sociality, a critical examination”.
00:05:01.279 --> 00:05:04.309
How did I get involved in this topic?
00:05:04.309 --> 00:05:06.550
He put me in charge of
00:05:06.550 --> 00:05:09.618
writing up a summary of his research,
00:05:09.618 --> 00:05:13.859
kind of distilling pages
upon pages and pages of review
00:05:13.860 --> 00:05:17.639
into something coherent
that the average consumer
00:05:17.639 --> 00:05:22.019
could actually understand.
So that was my task writing up
00:05:22.019 --> 00:05:25.469
his review and free inquiry,
since Luke Galen
00:05:25.470 --> 00:05:28.608
doesn't like their leave the house
too often or interact
00:05:28.608 --> 00:05:32.129
with ordinary human beings.
(Laughing)
00:05:32.129 --> 00:05:34.239
He kind of appointed me
to be his spokesman.
00:05:34.239 --> 00:05:37.159
He jokingly refers to me
as Galen's Bulldog.
00:05:37.160 --> 00:05:40.750
I guess I'm Thomas Henry Huxley
to his Darwin.
00:05:40.750 --> 00:05:45.100
So I've been glad to have
the opportunity to do interviews
00:05:45.100 --> 00:05:49.390
and talk to groups like this
about this research because I think it
00:05:49.389 --> 00:05:50.979
needs to get out there.
00:05:50.979 --> 00:05:53.619
All right,
before we go any further
00:05:53.620 --> 00:05:56.689
let us define
what we mean by pro-sociality.
00:05:56.689 --> 00:06:01.500
I hate that word already.
I am barely into this lecture
00:06:01.500 --> 00:06:05.240
and tired of saying it,
00:06:05.240 --> 00:06:09.920
but the term pro-social refers
to any kind of positive social behavior
00:06:09.920 --> 00:06:15.129
and this runs the gamut from generosity
in the form a charitable giving
00:06:15.129 --> 00:06:19.509
or time spent volunteering to personal
qualities perhaps
00:06:19.509 --> 00:06:21.879
such as positive personality traits:
00:06:21.879 --> 00:06:26.550
being helpful, being honest and there's
actually an impressive array of
00:06:26.550 --> 00:06:30.680
scientific studies that support this
hypothesis, that try to show
00:06:30.680 --> 00:06:35.930
that the religious exhibit greater
pro-sociality than the non-religious.
00:06:35.930 --> 00:06:39.919
In effect this has even become the subject
of a number a popular books.
00:06:39.919 --> 00:06:42.120
One you may have heard of
00:06:42.120 --> 00:06:45.890
is “A Friendly Letter to Skeptics
and Atheists” by David Myers
00:06:45.890 --> 00:06:50.689
or more recently “American Grace, How
Religion Divides and Unites Us”.
00:06:50.689 --> 00:06:54.459
So the general public is being told
that the data are in
00:06:54.459 --> 00:06:59.359
and religion makes you happy, happier,
healthier and more helpful.
00:06:59.359 --> 00:07:05.089
That this is a conclusion that is not
just philosophy or religion. It's science.
00:07:05.089 --> 00:07:07.539
In fact even some atheists
are getting in on this.
00:07:07.540 --> 00:07:10.980
A seemingly overwhelming case
for the pro-social effects
00:07:10.980 --> 00:07:15.480
of religion has been enough to
convince people here like Jessie Bering,
00:07:15.480 --> 00:07:19.640
an atheist psychologist and actually a
pretty good author.
00:07:19.639 --> 00:07:24.729
Enough to convince him that religion
is beneficial, at least for others.
00:07:24.729 --> 00:07:28.939
In a recent Slate article
entitled *“Don't trust the godless”(,
00:07:28.939 --> 00:07:33.430
Jesse Bering confessed; "Even as an
atheist, I have more confidence
00:07:33.430 --> 00:07:37.079
in religious people
and now science is backing me up.”
00:07:37.079 --> 00:07:40.720
A fuller quote is up here; "This is a
difficult confession to make
00:07:40.720 --> 00:07:45.879
because on the surface I'm sure
it sounds wildly, wildly hypocritical.
00:07:45.879 --> 00:07:50.850
Still here it goes; "I trust religious people
more than I trust atheists."
00:07:50.850 --> 00:07:53.306
Trustworthiness is a different thing
altogether from intellect
00:07:53.306 --> 00:07:56.682
and I suppose
I'm the ever so social pragmatist
00:07:56.682 --> 00:07:58.220
in my dealings with other people."
00:07:58.220 --> 00:08:02.710
So pretty serious claim,
if you get even atheist psychologists
00:08:02.709 --> 00:08:04.932
saying; "Don't trust the godless".
00:08:04.932 --> 00:08:07.115
Before we go any further though
00:08:07.115 --> 00:08:09.379
we're going to have to look at
what are the kind of methods
00:08:09.379 --> 00:08:12.532
that are used in pro-sociality research.
00:08:12.532 --> 00:08:15.685
We're going to see a variety of different
00:08:15.685 --> 00:08:19.240
experimental setups and methods
for conducting this kind of research.
00:08:19.240 --> 00:08:23.240
This would include self-reports,
what people say about themselves and
00:08:23.240 --> 00:08:28.330
third-party ratings of individuals,
laboratory tests of behavior,
00:08:28.329 --> 00:08:32.240
lab studies of economic games
– we’ll talk about those more later -
00:08:32.240 --> 00:08:34.300
priming studies, where people are presented
00:08:34.300 --> 00:08:37.039
with the religious concept subconsciously
00:08:37.039 --> 00:08:40.889
- usually where they will not realize
they've been primed by the concept
00:08:40.889 --> 00:08:44.819
and then we'll see what happens -
and also spirituality scales.
00:08:44.820 --> 00:08:49.419
Scales are meant to design, to detect
the level of one's spirituality
00:08:49.419 --> 00:08:51.799
and then we compare their behaviors on that.
00:08:51.799 --> 00:08:53.939
What I'm going to try to highlight
00:08:53.939 --> 00:08:56.560
is some other pitfalls
that researchers face
00:08:56.559 --> 00:08:59.859
in each of these types
of research methods.
00:09:00.359 --> 00:09:03.239
Let's start with the top report data;
00:09:03.240 --> 00:09:05.529
"Will being religious
make you a better person?"
00:09:05.529 --> 00:09:08.379
Well, the fateful
certainly seem to think so.
00:09:08.379 --> 00:09:11.759
When asked to give an assessment of
their own character and values,
00:09:11.759 --> 00:09:13.990
religious individuals
tend to report being...
00:09:13.990 --> 00:09:18.370
having a more grateful disposition;
they rate themselves as more helpful;
00:09:18.370 --> 00:09:22.139
they claim to value forgiveness
more highly than the non-religious;
00:09:22.139 --> 00:09:26.419
And many studies actually take
these self-reports at face value.
00:09:26.419 --> 00:09:29.279
The fact that believers
think they're more moral
00:09:29.279 --> 00:09:32.699
is actually taken as evidence
that they do exhibit
00:09:32.700 --> 00:09:38.390
these pro-social traits. Big question is:
"Should we take believers at their word?
00:09:38.389 --> 00:09:42.230
No. Not if their evaluations
are based on a self-serving bias
00:09:42.230 --> 00:09:46.360
rather than a realistic assessment
of their own character.
00:09:46.360 --> 00:09:51.400
Self-report data tend to be
unreliable by its very nature.
00:09:51.399 --> 00:09:56.289
People are prone to forming positive
illusions about themselves. We all do it.
00:09:56.289 --> 00:10:00.305
We tend to inflate our responses on
questionnaires as a result
00:10:00.305 --> 00:10:02.701
to make ourselves look better.
00:10:02.701 --> 00:10:06.179
Sometimes this is just concern
over our own personal self-image.
00:10:06.179 --> 00:10:08.896
Social psychologists call this
self-enhancement.
00:10:08.896 --> 00:10:12.043
Or sometimes we want to make
a good impression with others
00:10:12.043 --> 00:10:15.230
or good impression for our group in particular.
00:10:15.230 --> 00:10:18.190
This is sometimes referred
to as impression management.
00:10:18.190 --> 00:10:21.070
While this is a widespread tendency
00:10:21.070 --> 00:10:24.580
and it's by no means restricted
just to the religious.
00:10:24.580 --> 00:10:27.950
What's interesting is, this tendency
might be more pronounced
00:10:27.950 --> 00:10:31.230
in those who have
a strong level of religious belief.
00:10:31.230 --> 00:10:35.710
Highly religious people tend to view
themselves as better than others, generally.
00:10:35.710 --> 00:10:39.430
Even better than
other religious individuals.
00:10:39.429 --> 00:10:43.599
And they also evaluate themselves more
highly than non-religious individuals
00:10:43.599 --> 00:10:46.510
on attributes that have absolutely
nothing to do with religion.
00:10:46.510 --> 00:10:49.870
So for example they might
score themselves higher
00:10:49.870 --> 00:10:53.609
on measures of intelligence
or being a good worker.
00:10:53.609 --> 00:10:59.090
Things that do not seem immediately
related to their religious morality.
00:10:59.090 --> 00:11:02.649
Those high in intrinsic religiosity
actually have been shown to have
00:11:02.649 --> 00:11:05.919
a higher degree of self-enhancement
and impression management.
00:11:05.919 --> 00:11:11.589
Just one example: if you prime a
Christian with self-esteem primes,
00:11:11.590 --> 00:11:16.750
you'll see them actually rating themselves
as living up to Christian principles
00:11:16.769 --> 00:11:21.259
more often than their fellow believers.
If however you do the reverse
00:11:21.259 --> 00:11:25.409
and you offer up an assessment that
questions their high self-esteem
00:11:25.409 --> 00:11:29.230
or make them write about something
that they don't like about themselves,
00:11:29.230 --> 00:11:31.569
those who are high
in intrinsic religiosity
00:11:31.569 --> 00:11:36.328
- that means the level of belief -
they are more likely
00:11:36.328 --> 00:11:39.879
to resort to self-deception
as a compensating strategy.
00:11:39.879 --> 00:11:42.919
Also highly religious people
are particularly
00:11:42.919 --> 00:11:47.209
concerned with presenting themselves
as moral persons
00:11:47.210 --> 00:11:50.060
and particularly threatened
when that self-image is challenged.
00:11:50.060 --> 00:11:53.510
So, I guess the big question is;
"Why do researchers even rely
00:11:53.510 --> 00:11:58.240
on these self-reports some of the time?"
Well, because at least in some cases
00:11:58.240 --> 00:12:01.736
these positive self-assessments are actually corroborated by others:
00:12:01.736 --> 00:12:05.032
their family, their colleagues, their peers.
00:12:05.032 --> 00:12:08.430
So, third-party evaluators
rate religious individuals
00:12:08.429 --> 00:12:10.829
as being nicer, more cooperative
and highly altruistic
00:12:10.830 --> 00:12:15.300
and empathetic as well.
To some this is proof
00:12:15.299 --> 00:12:20.029
that the self-reports are not
self-delusion, they're not moral hypocrisy.
00:12:20.029 --> 00:12:23.209
They are correct assessments
of their character.
00:12:23.210 --> 00:12:26.810
But I think we can still
be a little skeptical here.
00:12:26.809 --> 00:12:29.700
When we're talking
about a predominantly religious society,
00:12:29.700 --> 00:12:32.500
where about 80 to 95%
of people are religious
00:12:32.500 --> 00:12:35.980
and around 75% are at least
nominally Christian,
00:12:35.980 --> 00:12:39.600
it's a good bet that a significant
proportion of those subjects, families
00:12:39.600 --> 00:12:41.629
and peers are also religious,
00:12:41.629 --> 00:12:44.660
meaning there's a possibility
of in-group bias at work here.
00:12:44.660 --> 00:12:48.360
And actually there is some evidence
to support that.
00:12:48.360 --> 00:12:52.529
In-group favoritism is a well-studied
phenomenon in social psychology.
00:12:52.529 --> 00:12:56.539
Again, this is not just religious
people here, this is all of us.
00:12:56.539 --> 00:13:00.329
It is natural for individuals to
derive self-esteem from the groups
00:13:00.329 --> 00:13:01.838
they’re associated with.
00:13:01.839 --> 00:13:05.070
It's natural to provide
a positive image to the public
00:13:05.070 --> 00:13:08.920
for those who share their identity.
So, consistent with the predictions
00:13:08.919 --> 00:13:10.770
of social identity theory,
00:13:10.770 --> 00:13:15.200
we see believers tend to show more
favoritism towards other individuals
00:13:15.200 --> 00:13:17.360
and speak more poorly of non-religious
00:13:17.360 --> 00:13:21.279
and this even includes
those from different religious groups.
00:13:21.279 --> 00:13:24.870
Often the favoritism
- and here's the key point here -
00:13:24.870 --> 00:13:28.320
often the favoritism is extended
to other religious individuals,
00:13:28.320 --> 00:13:30.990
regardless of whether or not
they behaved well or poorly,
00:13:30.990 --> 00:13:34.700
are still be reviewed more favorably,
00:13:34.700 --> 00:13:38.220
even when they've been up to no good.
I'll give you an example here
00:13:38.220 --> 00:13:43.170
of when sometimes believers will rate
religious individuals more highly
00:13:43.169 --> 00:13:47.699
than non-religious individuals, even when
they exhibit the exact same behaviors.
00:13:47.700 --> 00:13:52.089
I was a part of this study
which was published in 2011.
00:13:52.089 --> 00:13:56.160
I was a participant and so in front of a
camera I wore two different T-shirts,
00:13:56.159 --> 00:13:59.409
I wore three actually:
just a plain white T-shirt
00:13:59.409 --> 00:14:03.250
then in the other condition
I wore a Jesus fish T-shirt
00:14:03.250 --> 00:14:06.708
and then third condition
I wore a Darwin fish T-shirt.
00:14:06.708 --> 00:14:10.659
And then I read
the exact same script each time
00:14:10.659 --> 00:14:14.139
which was I was presenting myself
as a college student
00:14:14.139 --> 00:14:18.819
who was using my spring break to help
in disaster relief organization
00:14:18.820 --> 00:14:20.770
and talking
about my positive experiences.
00:14:20.769 --> 00:14:24.439
No mention of religion or anything else.
What we found in this study
00:14:24.440 --> 00:14:28.240
was that people rated me
as more likable,
00:14:28.240 --> 00:14:31.750
more intelligent, more trustworthy
and more kind,
00:14:31.750 --> 00:14:34.750
and more moral overall
00:14:34.750 --> 00:14:39.480
when I was wearing the Jesus fish.
So exact same behaviors but a subtle cue
00:14:39.480 --> 00:14:44.160
that I might be religious makes
people evaluate my behavior better.
00:14:44.160 --> 00:14:47.419
Most studies that rely on peer-rated
ratings do not adequately control
00:14:47.419 --> 00:14:52.819
for this tendency for in-group bias
and that's the problem.
00:14:52.820 --> 00:14:56.780
Ideally researchers would ensure
that participants are completely unaware
00:14:56.779 --> 00:14:59.289
of the religious identity
of those they're rating.
00:14:59.289 --> 00:15:02.969
Then we have more reason to trust
their evaluations as accurate,
00:15:02.970 --> 00:15:06.720
but this sadly is rarely the case.
However in studies
00:15:06.720 --> 00:15:09.870
that do, and most studies
where the raters
00:15:09.870 --> 00:15:13.509
are actually aware of their targets
religious identities
00:15:13.509 --> 00:15:15.869
a clear bias emerges
and that suggests
00:15:15.869 --> 00:15:18.970
an in-group bias
when the people know they're religious,
00:15:18.970 --> 00:15:23.360
they rate them more positively.
There is an interesting twist to all of this:
00:15:23.360 --> 00:15:27.909
non-religious individuals do not appear
to rate their fellow non-believers
00:15:27.909 --> 00:15:31.919
as any more pro-social than they do the
religious. For some reason this
00:15:31.919 --> 00:15:37.049
in-group bias doesn't seem to be affecting
the non-religious to the same degree.
00:15:37.049 --> 00:15:40.809
Now should the fact
that non-religious people also rate
00:15:40.809 --> 00:15:45.139
the religious highly, indicate that
these judgments are based on a clear
00:15:45.139 --> 00:15:48.659
added assessment of their character?
Are atheists really going to have
00:15:48.659 --> 00:15:53.049
a pro-religious bias? Actually it's quite
possible that they would
00:15:53.049 --> 00:15:56.929
if their judgments have been swayed
by a strong religious,
00:15:56.929 --> 00:15:58.458
pro-religious cultural stereotype.
00:15:58.458 --> 00:16:01.948
And again, there is evidence to suggest
that's the case.
00:16:01.948 --> 00:16:05.029
Here's more evidence
for a pro-religious stereotype,
00:16:05.029 --> 00:16:08.559
that we can find by looking at
similar studies that are conducted
00:16:08.559 --> 00:16:10.849
in different cultures than our own.
00:16:10.850 --> 00:16:15.199
For example some report that happiness,
life satisfaction
00:16:15.198 --> 00:16:17.019
and personality measures
like agreeableness
00:16:17.019 --> 00:16:20.539
are more closely associated
with religion in the United States
00:16:20.539 --> 00:16:25.068
than in the United Kingdom or Northern
Europe where religion is less dominant.
00:16:25.068 --> 00:16:28.750
To give you one specific example:
this particular study here
00:16:28.750 --> 00:16:34.789
asks people to rate their impressions
of people just from looking at photographs
00:16:34.789 --> 00:16:36.726
of faces and smiling faces were judged
00:16:36.726 --> 00:16:40.633
to be more religious
than non smiling faces.
00:16:40.633 --> 00:16:44.350
That was in the United States.
In the United Kingdom
00:16:44.350 --> 00:16:47.068
the exact opposite was true and
00:16:47.068 --> 00:16:50.889
tends to be a general relationship
00:16:50.889 --> 00:16:54.039
between religion and self-control,
mental well-being,
00:16:54.039 --> 00:16:57.099
psychological adjustment,
social support.
00:16:57.100 --> 00:17:00.870
In general in societies where the
non-religious are the majority
00:17:00.870 --> 00:17:04.588
the non-religious are rated more favorably
00:17:04.588 --> 00:17:08.848
on all those particular measures.
So, again evidence that there
00:17:08.848 --> 00:17:10.809
is a cultural bias at work here.
00:17:10.809 --> 00:17:13.598
I think by now it should be clear that
self-report data doesn't provide
00:17:13.598 --> 00:17:17.969
reliable evidence for the religious
pro-social reality hypothesis.
00:17:17.970 --> 00:17:21.920
Self-evaluations on religious subjects
are vulnerable to self-enhancement,
00:17:21.920 --> 00:17:26.348
impression management, distortions and
others and they are contaminated
00:17:26.348 --> 00:17:30.639
quite possibly by a widely-held
pro-religious cultural stereotype.
00:17:30.640 --> 00:17:34.890
It might be better
instead of trusting self-reports
00:17:34.890 --> 00:17:38.720
to look at experimentally
controlled measures of behavior
00:17:38.720 --> 00:17:43.199
or, if you prefer the way Jesus might say
it, we should judge believers
00:17:43.199 --> 00:17:46.809
by their fruit not by their words.
Let's look at
00:17:46.809 --> 00:17:50.329
what their actual deeds tell us.
Well, for one religious individuals
00:17:50.329 --> 00:17:53.649
claim to value forgiveness
more than others
00:17:53.650 --> 00:17:57.430
but actually any effect of their religiosity
on actual forgiveness
00:17:57.430 --> 00:18:01.600
has been found to be negligible.
This is just one study
00:18:01.599 --> 00:18:06.049
that shows that the comparison
of self-reports with controlled experiments
00:18:06.049 --> 00:18:09.269
on behavior reveal how often believers
00:18:09.269 --> 00:18:11.916
fail to live up to their high opinion
of themselves.
00:18:11.916 --> 00:18:17.023
Those in high in intrinsic religiosity again
- this is a measure of belief -
00:18:17.023 --> 00:18:19.710
reported a more grateful disposition
but don't do not perform
00:18:19.710 --> 00:18:21.920
better than anyone else in studies
00:18:21.920 --> 00:18:25.380
measuring reciprocal behavioral
gratitude: "Will they give back?"
00:18:25.380 --> 00:18:29.460
High intrinsic religiosity doesn't seem
to reduce aggression
00:18:29.460 --> 00:18:32.700
here is the scary detail:
it tends to make people think
00:18:32.700 --> 00:18:34.869
they're less aggressive
than they really are.
00:18:34.869 --> 00:18:38.238
Fundamentalists in particular report
00:18:38.238 --> 00:18:41.009
higher levels of altruism
towards everybody
00:18:41.009 --> 00:18:44.599
but in reality they are
more willing to help friends
00:18:44.599 --> 00:18:48.279
or like-minded individuals.
They're not as likely to help strangers
00:18:48.279 --> 00:18:51.519
or what is sometimes called
'value violators',
00:18:51.519 --> 00:18:54.710
perhaps like a homosexual or
something like that,
00:18:54.710 --> 00:18:59.000
that is clearly on the wrong side
of the divide on some culture war issue.
00:18:59.000 --> 00:19:02.369
We have to remember again
that everyone tends
00:19:02.369 --> 00:19:05.389
to overestimate
how moral they actually are.
00:19:05.390 --> 00:19:09.560
So, this is really common to find a gap
between how individuals predict
00:19:09.559 --> 00:19:11.720
they'll behave
and how they actually behave.
00:19:11.720 --> 00:19:14.500
If we were to take everybody in this
room we would probably
00:19:14.500 --> 00:19:16.509
see the exact same phenomena.
00:19:16.509 --> 00:19:20.190
We are no different.
It's just that the fact
00:19:20.190 --> 00:19:23.370
seems to be more pronounced
in religious populations.
00:19:23.369 --> 00:19:28.260
As we said before, highly religious do
show a greater tendency
00:19:28.280 --> 00:19:32.020
towards self-enhancement
on questionnaires and this disjunction
00:19:32.020 --> 00:19:34.200
between self-reported measured behavior
00:19:34.200 --> 00:19:38.019
is actually wider in the religious
than in the rest of the population.
00:19:38.019 --> 00:19:40.750
In fact the greatest gap
we can see
00:19:40.750 --> 00:19:44.650
between altruistic beliefs and
altruistic behaviors
00:19:44.650 --> 00:19:47.629
is actually found
in those who rate religion
00:19:47.629 --> 00:19:49.799
as more important to them personally.
00:19:49.799 --> 00:19:53.779
What's funny is that sometimes
the experimental evidence
00:19:53.779 --> 00:19:57.730
is so contrary
to our stereotypes about religion
00:19:57.730 --> 00:20:01.400
that some researchers
put a spin on this conclusion.
00:20:01.400 --> 00:20:05.950
What they're publishing
is showing no positive effect for religion
00:20:05.950 --> 00:20:09.169
but their abstracts or the way
they interpret the data
00:20:09.169 --> 00:20:13.230
speaks in glowing terms. For example
this particular study here.
00:20:13.230 --> 00:20:16.920
McCullough & Worthington in 1999 said that
00:20:16.920 --> 00:20:20.830
"Even if religious people
are no more facile
00:20:20.829 --> 00:20:24.869
at forgiving in real life situations
than are less religious people,
00:20:24.869 --> 00:20:29.699
they do you desire to be forgiving and
go on about how great it is that they
00:20:29.700 --> 00:20:32.370
want to be good forgiving people."
00:20:32.369 --> 00:20:36.219
I'd like to argue to you that praising
the leaders for their moral intentions
00:20:36.220 --> 00:20:37.950
kind of misses the point.
00:20:37.950 --> 00:20:41.390
It's not that we just desire to be
better - and that's good -
00:20:41.390 --> 00:20:44.690
in some cases they already
think they're superior.
00:20:44.690 --> 00:20:49.360
And there are major dangers in having an unrealistic assessment
00:20:49.359 --> 00:20:51.049
of one's own character and limits
00:20:51.049 --> 00:20:54.409
Let's move on to a different type of study:
00:20:54.410 --> 00:20:58.340
religious priming studies.
Although religious individuals do not seem
00:20:58.339 --> 00:21:00.359
to behave as morally as they report,
00:21:00.359 --> 00:21:03.599
it would still be very odd.
I personally would find it strange
00:21:03.599 --> 00:21:07.089
if religion didn't have some impact
on moral behavior.
00:21:07.089 --> 00:21:11.439
After all, scriptures and sermons abound
with exhortations to love thy neighbor,
00:21:11.440 --> 00:21:14.820
to do unto others
as you would have them do onto you
00:21:14.819 --> 00:21:17.830
and I think frequent exposure
to these messages
00:21:17.830 --> 00:21:20.120
would result in pro-social behaviors.
00:21:20.119 --> 00:21:24.239
So priming studies are a useful way
of seeving this out.
00:21:24.240 --> 00:21:28.420
Again participants are primed somehow.
Maybe they have to unscramble
00:21:28.420 --> 00:21:30.359
a word bank
and it has religious words in it.
00:21:30.359 --> 00:21:32.208
Or perhaps they have to read
00:21:32.208 --> 00:21:35.259
a portion of Scripture
and write a response to it.
00:21:35.259 --> 00:21:38.410
Sometimes it can be really subtle:
symbols in the room,
00:21:38.410 --> 00:21:40.039
a crucifix in the room,
00:21:40.039 --> 00:21:44.279
jewelry or clothing.
Sometimes it's just the context.
00:21:44.279 --> 00:21:48.250
Conducting the experiment in a
church instead of,
00:21:48.250 --> 00:21:51.440
say, a high school gymnasium
or something like that.
00:21:51.440 --> 00:21:55.210
Well the good news of priming studies
is that this is the best evidence
00:21:55.210 --> 00:21:58.539
we can find for the religious
pro-sociality hypothesis.
00:21:58.539 --> 00:22:00.612
There's a lot of data supporting it.
00:22:00.612 --> 00:22:01.955
There are good studies
00:22:01.955 --> 00:22:04.759
showing greater honesty and generosity
00:22:04.759 --> 00:22:07.819
amongst the religious,
increased sharing,
00:22:07.819 --> 00:22:11.000
increased cooperation,
better self-control in distressing situations
00:22:11.000 --> 00:22:15.670
and greater resistance to temptation.
00:22:15.670 --> 00:22:20.100
So why are religious concepts so good at
priming these kinds of behavior?
00:22:20.099 --> 00:22:22.719
Several studies cited a possible
mechanism here.
00:22:22.719 --> 00:22:25.279
'Supernatural surveillance' they called it.
00:22:25.279 --> 00:22:29.250
The belief that one's actions
are constantly and inescapably
00:22:29.250 --> 00:22:31.549
being observed by a divine being.
00:22:31.549 --> 00:22:34.559
Thinking that this is a strong
reminder to us
00:22:34.559 --> 00:22:37.829
to be aware of our actions
and perhaps that's why
00:22:37.829 --> 00:22:41.729
religious concepts prime
these pro-social behaviors.
00:22:41.729 --> 00:22:44.995
God might be watching after all.
00:22:44.995 --> 00:22:46.861
But I wont to share
some curious details
00:22:46.861 --> 00:22:50.849
that aren't as often shared
in these priming studies.
00:22:50.849 --> 00:22:54.139
Neutral religious works
like Bible, the Cycle ????
00:22:54.140 --> 00:22:58.049
or Chapel don't seem to promote
any helping behavior.
00:22:58.048 --> 00:23:01.259
It seems to be only positive words
like heaven, miracle or bless
00:23:01.259 --> 00:23:05.019
that have that effect on people.
Even more interesting
00:23:05.019 --> 00:23:08.019
the positive effects
don't seem to be dependent
00:23:08.019 --> 00:23:12.139
on the participants level of religiosity.
You can be just kind of religious
00:23:12.140 --> 00:23:14.050
or you could be a hardcore fundamentalist
00:23:14.050 --> 00:23:17.280
and the priming affects you the same way.
00:23:17.280 --> 00:23:21.750
Also, non-religious people respond
positively to religious primes
00:23:21.750 --> 00:23:24.760
and to the exact same degree
as their religious counterparts.
00:23:24.760 --> 00:23:30.839
If you were to look at all those symbols,
you would act more morally too.
00:23:30.839 --> 00:23:34.369
Even more interesting:
priming secular concepts, like civil
00:23:34.369 --> 00:23:37.699
or court, seem to have
the same power to promote
00:23:37.700 --> 00:23:41.600
honesty or lower hypocrisy
as religious primes do.
00:23:41.599 --> 00:23:44.849
And religious destructive atheists:
00:23:44.849 --> 00:23:48.469
the distrust they have for us
goes down
00:23:48.470 --> 00:23:52.179
when the religious are primed
with concepts of secular authority.
00:23:52.179 --> 00:23:55.409
That's really interesting.
Why would that possibly be?
00:23:55.409 --> 00:24:00.010
Well, one idea, not exactly sure,
but one idea is that the leaders know
00:24:00.009 --> 00:24:03.679
that atheists do not live their lives
as if God is watching them.
00:24:03.679 --> 00:24:06.240
So without the supernatural monitoring
they may wonder
00:24:06.240 --> 00:24:09.579
what reason we have
for behaving well.
00:24:09.579 --> 00:24:14.198
But this distrust can be ameliorated
when we are reminded
00:24:14.198 --> 00:24:16.869
that morality can be
monitored in different ways.
00:24:16.869 --> 00:24:18.928
So these kind of pacific primes
remind everybody;
00:24:18.928 --> 00:24:22.650
"Oh wait, there is a social order,
there is something keeping these
00:24:22.650 --> 00:24:27.059
evil atheists in check." And so their negative impressions go down.
00:24:27.058 --> 00:24:30.658
Amazingly even the presence of a mirror
00:24:30.659 --> 00:24:36.130
or just pictures of eyes in the laboratory
will actually have these same effects
00:24:36.130 --> 00:24:41.109
which actually really boost
that notion of supernatural surveillance.
00:24:41.109 --> 00:24:46.110
Obviously this has implications
for the religious pro-sociality hypothesis.
00:24:46.109 --> 00:24:49.479
Religious concepts do not seem
00:24:49.479 --> 00:24:54.309
to prime pro-social behavior like honesty
because they're religious.
00:24:54.309 --> 00:24:58.888
It may be that any concepts that are
associated with morality in a particular
00:24:58.888 --> 00:25:03.269
culture trigger greater concern
for protecting your reputation.
00:25:03.269 --> 00:25:05.899
Again since there's a widespread
cultural stereotype
00:25:05.899 --> 00:25:08.389
that religion is linked to morality here,
00:25:08.390 --> 00:25:12.250
religious concepts will activate moral
behavior, but as we pointed out
00:25:12.250 --> 00:25:14.596
secular primes do just as well.
00:25:14.596 --> 00:25:17.002
While the positive effects
of religious priming
00:25:17.002 --> 00:25:21.058
are the stuff of headlines, what you don't
usually hear about is the dark side
00:25:21.058 --> 00:25:22.545
of religious primes.
00:25:22.545 --> 00:25:24.592
Numerous studies demonstrate
00:25:24.592 --> 00:25:26.359
that socially undesirable behaviors
00:25:26.359 --> 00:25:30.689
also manifest when subjects
are exposed to religious messages.
00:25:30.689 --> 00:25:34.679
So for example, participants
who read passages from the Bible
00:25:34.679 --> 00:25:38.990
depicting God sanctioned violence,
administer more electrical shocks
00:25:38.990 --> 00:25:40.558
than the control group
in studies of aggression.
00:25:40.558 --> 00:25:45.329
We should note this works
on non-believers as well.
00:25:45.329 --> 00:25:49.308
Even a non-believer reading
those passages from the Bible
00:25:49.308 --> 00:25:53.678
will also become more vicious
in their behaviors toward somebody.
00:25:53.679 --> 00:25:57.929
It's just that the effect seems to
be more pronounced for believers.
00:25:57.929 --> 00:26:01.159
Especially disturbing is this subgroup
00:26:01.159 --> 00:26:05.099
of religious believers
high in intrinsic religiosity
00:26:05.099 --> 00:26:07.759
and also high in levels of submissiveness.
00:26:07.759 --> 00:26:12.149
This group was very disturbing
because they became the most vengeful
00:26:12.149 --> 00:26:16.178
after being primed with religious words.
They really seemed to go off the rails.
00:26:16.178 --> 00:26:19.489
So I guess what I'm saying is:
it doesn't affect all people equally.
00:26:19.489 --> 00:26:23.528
Certain personality characteristics
come into play here too
00:26:23.528 --> 00:26:26.950
to either aggravate
or kind of mute these responses.
00:26:26.950 --> 00:26:31.330
I'd like to share this study real quick.
Experiments where people were assigned
00:26:31.330 --> 00:26:34.320
to read the biblical version
of the golden rule
00:26:34.319 --> 00:26:38.070
actually had no effect on diminishing
Christians’ homophobia.
00:26:38.070 --> 00:26:40.342
So, negative attitudes
towards homosexuals
00:26:40.342 --> 00:26:41.714
were not at all diminished
00:26:41.714 --> 00:26:45.308
by reading what we think
is a very positive prime,
00:26:45.308 --> 00:26:49.558
right, the golden rule. Strangely enough
reading the Buddhist version
00:26:49.558 --> 00:26:53.980
of the golden rule actually increased
their homophobic responses.
00:26:53.980 --> 00:26:57.750
If they read another religious text
telling them to be merciful
00:26:57.750 --> 00:27:01.429
and do onto others as you would have
them to do want to them,
00:27:01.429 --> 00:27:05.415
they wanted to do that even less.
This is perhaps
00:27:05.415 --> 00:27:08.241
because the moral imperative
00:27:08.241 --> 00:27:11.649
was coming from this
distrusted out-group source.
00:27:11.720 --> 00:27:15.870
Likewise unscrambling words associated
with Christianity increased
00:27:15.869 --> 00:27:17.779
racial prejudice
towards african-americans
00:27:17.779 --> 00:27:21.949
that was found by Johnson,
lead author Johnson in 2010.
00:27:21.949 --> 00:27:26.250
And attitudes toward all out-group
members became more negative
00:27:26.250 --> 00:27:29.278
when experiments were conducted
in a church setting rather than
00:27:29.278 --> 00:27:32.359
than in a civic context.
00:27:32.359 --> 00:27:34.999
This is a strange paradox
we're looking at here.
00:27:34.999 --> 00:27:37.109
Religious priming seems to increase
00:27:37.109 --> 00:27:42.158
both pro-social behaviors like honesty and
sharing, and non pro-social behaviors
00:27:42.159 --> 00:27:47.270
like aggression and prejudice. This will
make more sense I think to us when we
00:27:47.269 --> 00:27:49.079
consider another curious,
00:27:49.079 --> 00:27:52.528
but consistent finding in this literature
00:27:52.528 --> 00:27:55.740
that the kindness of religious individuals
is typically not
00:27:55.740 --> 00:27:58.769
extended universally to everyone.
00:27:58.769 --> 00:28:02.849
Instead the primary beneficiaries
of a religious pro-sociality
00:28:02.849 --> 00:28:09.110
are usually other believers. This can be
most clearly seen in economic games.
00:28:09.110 --> 00:28:12.359
So, to save a little bit of time
I'm not going to go into how all of
00:28:12.359 --> 00:28:17.658
these games work, but they basically
start with people trading or exchanging money.
00:28:17.679 --> 00:28:22.218
Those games are designed
to encourage cooperation and trust.
00:28:22.218 --> 00:28:26.349
So basically
if the players work together,
00:28:26.349 --> 00:28:29.869
they will both get further along,
but one player
00:28:29.869 --> 00:28:34.918
might have the opportunity to make off
with more money if they deceive or lie
00:28:34.919 --> 00:28:37.628
or cheat the other players.
So this is all trying
00:28:37.628 --> 00:28:39.829
to assess cooperation, trust,
00:28:39.829 --> 00:28:43.678
giving, that sort of thing.
The economic games shown
00:28:43.679 --> 00:28:48.769
in behavioral economic studies where
the religiosity of the participants is none.
00:28:48.769 --> 00:28:52.220
- so we actually know what they are -
a general trend emerges:
00:28:52.220 --> 00:28:55.950
religious individuals cooperate more
and give more money
00:28:55.950 --> 00:28:59.149
than non-religious participants.
So they do that overall.
00:28:59.148 --> 00:29:02.808
They give more and they trust more
than the non-religious.
00:29:02.808 --> 00:29:06.139
The pro-sociality hypothesis is true.
00:29:06.140 --> 00:29:09.196
It's just has that twist:
they only give it to those
00:29:09.196 --> 00:29:10.972
who share their religious identity.
00:29:10.972 --> 00:29:15.308
For example this study, Ahmed, 2009
found the clergy students
00:29:15.308 --> 00:29:19.339
exchanged greater money offers
than non clergy students,
00:29:19.339 --> 00:29:23.928
but only to those from their own
religious group.
00:29:23.929 --> 00:29:25.788
These findings are almost,
well, they are most likely due
00:29:25.788 --> 00:29:29.679
to that previous phenomenon
we mentioned of in-group favoritism.
00:29:29.679 --> 00:29:33.659
But there also might be something else
going on here. This might be that
00:29:33.659 --> 00:29:35.600
pro-religious cultural stereotype
00:29:35.599 --> 00:29:39.449
happening again, because notice:
non-religious participants
00:29:39.450 --> 00:29:42.610
did not show the same in-group favoritism
00:29:42.609 --> 00:29:46.098
in those economic games.
They also trusted
00:29:46.098 --> 00:29:50.778
religious participants more
than their non-religious peers
00:29:50.778 --> 00:29:54.609
and allocated more money to them overall,
even though that money
00:29:54.609 --> 00:29:56.889
would not be reciprocated.
00:29:56.889 --> 00:30:01.509
Yeah, it's amazing
how ingrained that stereotype is.
00:30:01.509 --> 00:30:03.408
This pattern of preferential treatment
00:30:03.408 --> 00:30:06.027
is not limited
to behavioral economic studies.
00:30:06.027 --> 00:30:10.268
It constitutes a general trend
across the entire literature.
00:30:10.269 --> 00:30:13.620
In fact a new word had to be coined
just to explain it.
00:30:13.619 --> 00:30:17.908
One researcher who is very popular in
this by the name of Saroglou
00:30:17.909 --> 00:30:21.580
coined the term "minimal prosociality",
00:30:21.579 --> 00:30:25.089
meaning the greater helping on the part of
the religious that extended to friends
00:30:25.089 --> 00:30:27.849
an in-group members
but not too out-group members
00:30:27.849 --> 00:30:30.879
who threatened religious values.
00:30:30.880 --> 00:30:35.179
So, I guess the correct way to say it
or was consistent with most of the evidence
00:30:35.179 --> 00:30:36.330
in these economic games
00:30:36.329 --> 00:30:39.558
are that religious people
are ‘minimally pro-social’.
00:30:39.558 --> 00:30:43.629
And actually if we take this idea
of limited pro-sociality seriously
00:30:43.630 --> 00:30:46.760
it explains a lot of other trends
that we see in the data.
00:30:46.759 --> 00:30:50.119
For example across different cultures
we see that religiosity
00:30:50.119 --> 00:30:56.119
is weakly but still positively correlated
with the value of benevolence,
00:30:56.119 --> 00:31:01.009
charity, helping people out
and yet at the same time is negatively
00:31:01.009 --> 00:31:03.398
related with the value of universalism,
00:31:03.398 --> 00:31:06.449
helping out, you know, your neighbor,
your stranger,
00:31:06.450 --> 00:31:10.600
the Good Samaritan, that type of thing.
Again it seems like a contradiction,
00:31:10.599 --> 00:31:13.600
but when you take the idea of limited
00:31:13.600 --> 00:31:15.860
or minimal pro-sociality seriously,
00:31:15.859 --> 00:31:19.509
it tends to make more sense.
It's that in-group favoritism again.
00:31:19.509 --> 00:31:23.129
Also it might explain things
like why religious primes
00:31:23.129 --> 00:31:24.308
increase ethnic prejudice
00:31:24.308 --> 00:31:27.308
and derogation of out-group members,
00:31:27.308 --> 00:31:31.500
because religious concepts activated
in-group bias in people's minds.
00:31:31.500 --> 00:31:34.548
This also plays through
religious research on giving.
00:31:34.548 --> 00:31:37.710
This one conclusion
I'm not as sure about,
00:31:37.710 --> 00:31:42.470
but it is very clear that religious
organizations themselves
00:31:42.470 --> 00:31:46.589
are the largest source of charitable giving.
Religious people give way more to charity
00:31:46.589 --> 00:31:51.418
than the non-religious and that finding
has held up across the board.
00:31:51.419 --> 00:31:54.960
But as other studies note,
many of the recipients of these,
00:31:54.960 --> 00:31:58.340
even ones that are labeled secular,
00:31:58.339 --> 00:32:02.589
tend to be religious
or some religious organization.
00:32:02.589 --> 00:32:05.230
So all this money is exchanging hands
within the in-group.
00:32:05.230 --> 00:32:07.379
This would be really
interesting one to test
00:32:07.379 --> 00:32:09.408
if we can tease out that in-group favoritism
00:32:09.409 --> 00:32:12.880
would we still see a charity gap
between the non-religious
00:32:12.880 --> 00:32:17.090
and the religious? We might, actually I
suspect, we probably would
00:32:17.089 --> 00:32:21.418
and for this reason
there's another aspect
00:32:21.419 --> 00:32:25.210
to religious charitable giving,
and that is generosity
00:32:25.210 --> 00:32:29.048
measured as a function
of religious importance
00:32:29.048 --> 00:32:33.139
was smaller than those measured as a
variation in religious attendance.
00:32:33.140 --> 00:32:37.390
That is church attendance
seems to be the key factor
00:32:37.390 --> 00:32:40.690
in how much a religious person will give.
00:32:40.690 --> 00:32:44.869
If you actually
measure religiosity by belief,
00:32:44.869 --> 00:32:48.889
how much conviction do you have
that God exists
00:32:48.890 --> 00:32:52.049
we'll see that
that predicts giving to a lesser degree
00:32:52.048 --> 00:32:55.048
then church attendance.
I think what's going on here is
00:32:55.048 --> 00:32:58.519
when you're actually in the building,
you're given an opportunity
00:32:58.519 --> 00:33:01.919
to give, right?
The plate is passed around
00:33:01.919 --> 00:33:05.820
and there's social pressure for you
to put something in that plate.
00:33:05.819 --> 00:33:09.480
I still think the religious should get
credit for this, but they get credit for
00:33:09.480 --> 00:33:12.528
building institutions
that support charitable giving.
00:33:12.528 --> 00:33:14.950
It may not be the belief,
the religious belief,
00:33:14.950 --> 00:33:17.130
that's really motivating this behavior.
00:33:17.130 --> 00:33:19.470
So I guess that kind of brings up
an interesting question here.
00:33:19.470 --> 00:33:22.839
How actually are we measuring religiosity
00:33:22.839 --> 00:33:25.878
because, as we just saw,
depending on how we measured it,
00:33:25.878 --> 00:33:28.259
we might get different effects.
00:33:28.259 --> 00:33:32.470
Typically the methodology
that's employed here is to compare
00:33:32.470 --> 00:33:35.980
a general population of people
to highly religious people
00:33:35.980 --> 00:33:39.950
and weekly religious people.
And then the atheists agnostics
00:33:39.950 --> 00:33:43.588
or all the nones, we call them,
those who declare no religious affiliation,
00:33:43.588 --> 00:33:45.648
are mixed into that sample as well.
00:33:45.648 --> 00:33:48.970
There are different ways
again of measuring
00:33:48.970 --> 00:33:51.370
intrinsically religiosity as I
mentioned is a measure
00:33:51.370 --> 00:33:54.190
of metaphysical belief or commitment.
00:33:54.190 --> 00:33:58.538
Extrinsic religiosity, as I call it,
is often a measure of behavior,
00:33:58.538 --> 00:34:01.638
how often do you pray,
engage in rituals.
00:34:01.638 --> 00:34:04.959
That sometimes includes another way
that is measured
00:34:04.960 --> 00:34:08.990
is measuring religiosity purely
through church attendance alone.
00:34:08.990 --> 00:34:13.179
So whenever you see a study
that says religious people are better
00:34:13.179 --> 00:34:16.570
at XYZ, the next question you should ask is;
00:34:16.570 --> 00:34:19.809
"Better compared to whom?"
And the reason is:
00:34:19.809 --> 00:34:24.139
how one measures religiosity
has a major impact on your findings.
00:34:24.139 --> 00:34:27.199
For example, frequent church attendance
has been linked
00:34:27.199 --> 00:34:31.340
to modestly lower rates of mental
illness such as depression,
00:34:31.340 --> 00:34:34.519
but the effect is negligible
when you measure
00:34:34.519 --> 00:34:37.098
religiosity as strength of belief.
00:34:37.098 --> 00:34:40.268
Again, people have better mental health
because they're
00:34:40.268 --> 00:34:44.079
in a congregation of people, they have a
support social support network,
00:34:44.079 --> 00:34:49.359
like-minded people to talk to. The belief
doesn't seem to be as important.
00:34:49.359 --> 00:34:51.900
Studies that control
for purely social factors
00:34:51.900 --> 00:34:54.440
find a greatly diminished
or non-existent effect
00:34:54.440 --> 00:34:57.493
of religious beliefs
on pro-social measures.
00:34:57.493 --> 00:34:59.336
So you can see how we measure religion
00:34:59.336 --> 00:35:03.599
and who we compare our groups to
is very important in this debate.
00:35:03.599 --> 00:35:06.069
Most frequently
the strongest pro-social effects
00:35:06.069 --> 00:35:08.369
are associated with church attendance
00:35:08.369 --> 00:35:12.170
and social contacts
rather than just metaphysical belief.
00:35:12.170 --> 00:35:15.930
So it appears that group affiliation
drives many of these behaviors.
00:35:15.929 --> 00:35:19.949
Could a committed secular group
- like this one right here -
00:35:19.949 --> 00:35:23.569
have effect on its membership
similar to that of a church?
00:35:23.570 --> 00:35:27.920
In this book that I mentioned earlier -
unfortunately it's buried on page 472 -
00:35:27.920 --> 00:35:32.769
you have to get
through all the good stuff
00:35:32.769 --> 00:35:36.060
to finally see this qualification,
00:35:36.060 --> 00:35:40.220
but Robert Putnam mentions
"even an atheist
00:35:40.219 --> 00:35:43.269
who happens to become
involved in the social life
00:35:43.269 --> 00:35:47.710
of a congregation is much more likely
to volunteer at a soup kitchen
00:35:47.710 --> 00:35:50.356
then the most fervent believer who prays alone."
00:35:50.356 --> 00:35:54.242
And then it goes on to say
- or slightly before that on page 465 -
00:35:54.242 --> 00:35:59.450
he says: "Religious belief turns out to be
utterly irrelevant to explaining the religious
00:35:59.450 --> 00:36:04.080
as in good neighbourliness."
That should've been on page 1.
00:36:04.079 --> 00:36:09.590
But both reviewers in that book
didn't get that far.
00:36:09.590 --> 00:36:14.160
You can guess how it was depicted
in the popular press.
00:36:14.159 --> 00:36:18.129
In fact that's a major problem.
The problem with most studies is
00:36:18.130 --> 00:36:22.119
that they are lumping all nonbelievers
together, without considering how
00:36:22.119 --> 00:36:24.130
confident they are in their non-belief,
00:36:24.130 --> 00:36:27.110
whether or not they attend groups
like you do right here,
00:36:27.110 --> 00:36:29.300
how involved they are
with the community overall.
00:36:29.300 --> 00:36:32.890
They're just all dumped
into one pool: the non-religious.
00:36:32.889 --> 00:36:36.789
And then they're compared with weekly
religious and highly religious,
00:36:36.789 --> 00:36:40.610
typically highly religious people
who are in a church context.
00:36:40.610 --> 00:36:44.900
When you do that, you do get
what's called a linear effect.
00:36:44.900 --> 00:36:49.539
If pro-social, being happy, healthy
and more helpful is all on this axis,
00:36:49.539 --> 00:36:54.210
and religiosity on this one,
we would see as religiosity rises
00:36:54.210 --> 00:36:57.339
the more religious you get,
the more happy, helpful
00:36:57.339 --> 00:37:00.070
and honest you are as an individual.
00:37:00.070 --> 00:37:03.380
But what we're kind of doing is
we're cutting off half of our sample.
00:37:03.380 --> 00:37:06.559
The few studies that compare
highly religious people
00:37:06.559 --> 00:37:09.570
with the confidently non-religious
actually show
00:37:09.570 --> 00:37:11.880
what's called a curvilinear effect
00:37:11.880 --> 00:37:14.949
between religiosity and pro-sociality.
00:37:14.949 --> 00:37:19.459
To explain what's going on
with this curvilinear effect,
00:37:19.460 --> 00:37:23.690
- I should have had noticed, but I didn't -
00:37:23.690 --> 00:37:26.900
Essentially what we do, what we've
done is we've expanded our sample.
00:37:26.900 --> 00:37:32.340
So before the atheists and agnostics and
humanists were getting lost in this side of
00:37:32.340 --> 00:37:34.560
the curve now we brought it out
00:37:34.559 --> 00:37:38.690
and we actually see that it's the
less confident, the weekly religious,
00:37:38.690 --> 00:37:40.449
the weekly secular in the middle
00:37:40.449 --> 00:37:44.539
that tend to have poor
ratings on pro-social measures.
00:37:44.539 --> 00:37:48.289
Oh, here's what I was looking for.
Nominal believers,
00:37:48.289 --> 00:37:51.529
not atheists, show the highest levels
of depression actually,
00:37:51.530 --> 00:37:53.180
the poorest mental health
00:37:53.180 --> 00:37:56.309
and they generally report
less satisfaction with life.
00:37:56.309 --> 00:38:00.650
And fact is, this is true of the
cross-cultural data on this too.
00:38:00.670 --> 00:38:04.659
The world value survey found that both
those who claim religion is very important
00:38:04.659 --> 00:38:06.910
and those who claim
that it wasn't important at all,
00:38:06.910 --> 00:38:08.720
tended to be the happiest.
00:38:08.719 --> 00:38:13.509
So curvilinear effects
are also found in the moral realm,
00:38:13.510 --> 00:38:18.620
for example physicians, Doctors
Without Borders and that sort of thing
00:38:18.619 --> 00:38:23.309
highest membership is going to be
highly religious and totally atheist.
00:38:23.309 --> 00:38:24.880
This is true when
00:38:24.880 --> 00:38:29.099
Milgrams famous obedience trials
- if you're familiar with those studies -
00:38:29.099 --> 00:38:31.779
where we get to see
just how much will somebody
00:38:31.780 --> 00:38:34.546
obey the experimenter.
When those were replicated,
00:38:34.546 --> 00:38:37.730
it was the extreme believers
and the extreme non-believers
00:38:37.730 --> 00:38:42.170
that were most likely to disobey the
researchers unethical orders.
00:38:42.170 --> 00:38:46.779
So actually being highly
religious or highly non-religious
00:38:46.779 --> 00:38:50.309
seems to give you a little
bit more moral integrity.
00:38:50.309 --> 00:38:54.039
Part of the hypothesis
why this might be is because
00:38:54.039 --> 00:38:58.350
these pools of individuals, they're so
certain of their world view
00:38:58.350 --> 00:39:03.729
that they're not as kicked around
by the pressure of social conformity as others.
00:39:03.729 --> 00:39:06.460
So it appears that confidence in one's worldview
00:39:06.460 --> 00:39:09.129
and regular affiliation with like minded people
00:39:09.129 --> 00:39:12.630
are far more important to well-being
and moral integrity
00:39:12.630 --> 00:39:17.190
than your particular beliefs
about metaphysics. Sorry guys,
00:39:17.190 --> 00:39:22.050
even some non-believers are sad to hear
that sometimes, they want to believe that
00:39:22.050 --> 00:39:24.990
believing the right thing, having the
right grasp on reality
00:39:24.990 --> 00:39:27.420
will make you a better person
00:39:27.420 --> 00:39:31.960
and it doesn't seem that metaphysical
beliefs are all that important.
00:39:31.960 --> 00:39:37.290
But sadly studies are not designed to
notice curvilinear effects a lot of times
00:39:37.290 --> 00:39:41.889
And when they aren't, they can give
the impression that atheists are in danger
00:39:41.889 --> 00:39:44.375
of poor physical or mental health
00:39:44.375 --> 00:39:50.170
and this is exactly what we see with the military's spiritual fitness scale, that they have.
00:39:50.170 --> 00:39:52.750
I don't know if anybody has
ever heard of that?
00:39:52.750 --> 00:39:56.540
The US military has a spiritual fitness
dimension in their instrument
00:39:56.540 --> 00:40:00.100
that they use to assess
a soldier's wellness and mental health.
00:40:00.100 --> 00:40:03.890
And they conclude that soldiers
have the greatest resiliency
00:40:03.890 --> 00:40:06.348
when they are spiritual,
when they are religious
00:40:06.349 --> 00:40:11.070
and this has prompted
some superior officers
00:40:11.070 --> 00:40:14.170
to go find their underlings
who are non-religious
00:40:14.170 --> 00:40:18.559
and to pressure them into prayer meetings
and other religious services, right,
00:40:18.559 --> 00:40:22.060
because it's bad for their health.
They might be in a suicide risk.
00:40:22.060 --> 00:40:29.110
However though an examination of the
actual question items on the spirituality scale
00:40:29.110 --> 00:40:31.948
shows a major flaw in the way
these concepts are measured.
00:40:31.948 --> 00:40:36.699
And it's going to be my last major point
about how this research is conducted.
00:40:36.699 --> 00:40:39.173
"Criterion contamination"
00:40:39.173 --> 00:40:42.180
this is where the pro-sociality literature
00:40:42.180 --> 00:40:46.119
defines spirituality in a way
that kind of begs the question.
00:40:46.119 --> 00:40:50.650
So for example, usually when we
make a prediction
00:40:50.650 --> 00:40:55.309
of some sort of criterion, you want the
items used in the prediction
00:40:55.309 --> 00:40:58.650
to not contain elements
of what is being predicted.
00:40:58.650 --> 00:41:03.278
If you flip the conclusion
and you put it in your premise,
00:41:03.278 --> 00:41:05.690
you're arguing in a circle, right?
00:41:05.690 --> 00:41:08.439
But yet we see
this happen all the time,
00:41:08.439 --> 00:41:10.629
we see the reverse
happening all the time.
00:41:10.630 --> 00:41:14.970
For example this right here.
Religiously engaged individuals
00:41:14.969 --> 00:41:19.009
have greater social networks,
but religious engagement
00:41:19.010 --> 00:41:23.000
was defined by having church social contacts.
00:41:23.000 --> 00:41:26.760
So really all this is saying
- I mean it sounds really good, right? -
00:41:26.760 --> 00:41:31.130
Doesn't it? Wow? Religious
engagement really benefits us.
00:41:31.130 --> 00:41:34.340
All this is saying, is;
"Socially engaged religious people
00:41:34.340 --> 00:41:36.318
are socially engaged religious people."
00:41:36.318 --> 00:41:41.349
That is all that is said .
Many spirituality scales measure concepts
00:41:41.349 --> 00:41:45.649
that do not necessarily refer
to supernatural believes either.
00:41:45.650 --> 00:41:49.800
For example, these are all the things
that will get you a high rating as a
00:41:49.800 --> 00:41:52.750
spiritual person on these fitness scales.
00:41:52.750 --> 00:41:55.960
"I believe there is
a larger meaning to life.
00:41:55.960 --> 00:41:59.559
It's important for me
to give something back to my community."
00:41:59.559 --> 00:42:03.378
If you answer yes to that,
you're labeled as religious on this scale.
00:42:03.378 --> 00:42:06.958
"I believe that humanity
as a whole is basically good."
00:42:06.958 --> 00:42:09.629
If you have a positive humanistic outlook,
00:42:09.640 --> 00:42:13.920
you might say you're going to score
on that spirituality scale too.
00:42:13.940 --> 00:42:18.809
"I'm concerned about those
who will come after me in life."
00:42:18.809 --> 00:42:22.929
So numerous studies including this
military spiritual fitness assessment
00:42:22.929 --> 00:42:27.318
claims to demonstrate that religiosity is
related to pro-social outcomes,
00:42:27.318 --> 00:42:30.780
but they are really
just criterion contamination effects.
00:42:30.780 --> 00:42:33.069
Having pro-social traits here
00:42:33.069 --> 00:42:35.650
is what defines being religious.
00:42:35.650 --> 00:42:38.930
Just begging the question.
And as we know many atheists
00:42:38.929 --> 00:42:42.789
with a broader sense of meaning
would score ‘spiritual’ on these same scales.
00:42:42.789 --> 00:42:47.440
This artificially inflates the apparent
relationship between religiosity
00:42:47.440 --> 00:42:51.019
or spirituality and these positive
pro-social outcomes.
00:42:51.019 --> 00:42:54.710
All right.
So, tying it all together.
00:42:54.710 --> 00:42:58.849
The question; “Does religion make us
better?” actually doesn't admit
00:42:58.849 --> 00:43:02.579
of a simple answer. You've already seen
evidence showing: "yes and no"
00:43:02.579 --> 00:43:07.189
or "yes in particular ways and no and other
particular ways".
00:43:07.190 --> 00:43:10.470
Unfortunately this stuff just doesn't
work in a sound bite
00:43:10.469 --> 00:43:12.489
and we live in a sound-bite culture.
00:43:12.489 --> 00:43:14.929
The conclusion one reaches depends
00:43:14.929 --> 00:43:19.799
on the measure of religiosity being used;
the way pro-sociality is defined.
00:43:19.800 --> 00:43:23.043
We have to be cognizant of a host of
00:43:23.043 --> 00:43:25.746
complicating factors if we're going to be accurate.
00:43:25.746 --> 00:43:30.509
Really this is like a minefield for a critical thinker.
00:43:30.509 --> 00:43:35.529
Even the most experienced critical thinker
is going to run into problems
00:43:35.530 --> 00:43:37.870
with how complex this data is.
00:43:37.870 --> 00:43:43.329
So we came up with 10 questions for
thinking critically about religious pro-sociality
00:43:43.329 --> 00:43:48.698
that will help people in the future
to think more carefully about these studies.
00:43:48.698 --> 00:43:53.030
Number 1: has the research controlled
for the possibility that stereotypes
00:43:53.030 --> 00:43:58.040
- such as the expectation that
religious individuals will be more pro-social -
00:43:58.040 --> 00:44:01.250
have those stereotypes affected
self-reports and ratings?
00:44:01.250 --> 00:44:04.699
2: Are the results based on evidence
that have been compromised
00:44:04.699 --> 00:44:07.139
by in-group favoritism or bias?
00:44:07.139 --> 00:44:10.909
3: When pro-social effects follow
the priming of religious concepts,
00:44:10.909 --> 00:44:13.540
will those same effects follow secular prime?
00:44:13.540 --> 00:44:16.010
That's a great one
for the priming study.
00:44:16.010 --> 00:44:20.460
Number 4: is the study also able
to detect potential negative
00:44:20.460 --> 00:44:23.490
as well as positive effects
for religious primes?
00:44:23.489 --> 00:44:29.228
5: Is the research based on self-reports
or does it also measures actual behaviors?
00:44:29.228 --> 00:44:33.559
If it doesn't measure actual behaviors,
it's worthless.
00:44:33.579 --> 00:44:38.339
6: could the context of this study have
an impact on the results? For example,
00:44:38.338 --> 00:44:42.670
would this study get the same results in
the United States as opposed to
00:44:42.670 --> 00:44:47.990
other nations in Northern Europe that
are predominately non-religious?
00:44:47.989 --> 00:44:52.169
7: are the results solely attributable to
religious belief itself
00:44:52.170 --> 00:44:54.419
or is there a group affiliation effect
going on?
00:44:54.419 --> 00:44:58.410
If church attending believers are compared to non church attenders,
00:44:58.410 --> 00:45:01.699
the sources of any differences
might be unclear.
00:45:01.699 --> 00:45:05.868
Number 8: does the study conflate non-believe with low religiosity
00:45:05.869 --> 00:45:09.480
or do we have a clear measure
of the non-believers?
00:45:09.480 --> 00:45:14.029
By the way, for we gonna fulfill number 8
we need more research on secularists.
00:45:14.039 --> 00:45:17.818
So we need more researchers willing
to study communities like this
00:45:17.818 --> 00:45:20.061
and answer surveys and that sort of things.
00:45:20.061 --> 00:45:22.284
If you ever see those things pop up in your inbox.
00:45:22.284 --> 00:45:25.509
Please take'm.
You will help us all.
00:45:25.510 --> 00:45:28.399
Number 9: do the religious groups
under comparison allow
00:45:28.399 --> 00:45:31.039
for an examination of curvilinear effects?
00:45:31.039 --> 00:45:33.630
That is, if you're comparing a church group,
00:45:33.630 --> 00:45:39.159
you got to compare it with an equal group like this.
00:45:39.159 --> 00:45:42.120
Number 10: has religion or spirituality
00:45:42.120 --> 00:45:45.818
been defined in a way that
would also include
00:45:45.818 --> 00:45:48.949
pro-social behavior
just from the definition?
00:45:48.949 --> 00:45:52.849
I think if you watch for those things
you're going to have a leg up
00:45:52.849 --> 00:45:58.060
on most other people who are paying attention
to this particular research.
00:45:58.060 --> 00:46:01.960
I hope you got something out of that.
I hope that brings a little more clarity
00:46:01.960 --> 00:46:04.400
to this often confusing debate
00:46:04.400 --> 00:46:09.459
and a last thing I just wont to put
in another plug for my podcast:
00:46:09.459 --> 00:46:16.010
if you happen to enjoy what you heart tonight,
found it enlightening at all,
00:46:16.010 --> 00:46:19.499
both I and the author of the
the Psych Review, Luke Galen,
00:46:19.499 --> 00:46:22.285
we both work
on this podcast "Reasonable Doubts",
00:46:22.285 --> 00:46:25.929
you can find it at doubtcast.org.
00:46:25.929 --> 00:46:28.362
It is one of the most informationally dense
podcasts you'll find
00:46:28.362 --> 00:46:31.425
that still manages to be funny from time to time.
00:46:31.425 --> 00:46:33.000
I thank you very much.
00:46:33.000 --> 00:46:40.840
(Applause)
00:46:41.560 --> 00:46:44.709
To catch up on past Reasonable Doubts episodes
00:46:44.709 --> 00:46:46.358
or to email your questions or comments,
00:46:46.358 --> 00:46:50.190
check out www.doubtcast.org
00:46:50.190 --> 00:46:53.712
Reasonable Doubt is a production
of WPRR Reality Radio.
00:46:53.712 --> 00:46:59.029
You can find out more about Reality
Radio at publicrealityradio.org
00:46:59.029 --> 00:47:04.008
Reasonable Doubt's theme music is performed
by Love Fossil and used with permission
00:47:04.334 --> 00:47:17.961
Subtitled by www.kritischdenken.info