Kent Universiy Cristian
Kent Universiy Cristian
Stephen Bullivant
Miguel Farias
Jonathan Lanman
Lois Lee
© Understanding Unbelief (2019)
Cover photos of atheists and agnostics in Brazil, Japan, and the United Kingdom
© Aubrey Wade (2019)
This publication was made possible through the support of a grant from the John Templeton Foundation (JTF grant ID#
60624) managed by the University of Kent. The opinions expressed in this publication are those of the authors and do
not necessarily reflect the views of the John Templeton Foundation.
1. Atheists (i.e., people who ‘don’t believe in God’) and 5. Unbelief in God doesn’t necessarily entail unbelief in
agnostics (i.e., people who ‘don’t know whether there is other supernatural phenomena. Atheists and (less so)
a God or not, and don’t believe there is a way to find out’) agnostics exhibit lower levels of supernatural belief than
exhibit significant diversity both within, and between, do the wider populations. However, only minorities of
different countries. Accordingly, there are very many ways atheists or agnostics in each of our countries appear to
of being an unbeliever (i.e., atheists/agnostics combined). be thoroughgoing naturalists. (2.2, 2.3)
(All)
6. Another common supposition – that of the purposeless
2. In all six of our countries, majorities of unbelievers unbeliever, lacking anything to ascribe ultimate meaning
identify as having ‘no religion’. Nevertheless, in Denmark to the universe – also does not bear scrutiny. While
fully 28% of atheists and agnostics identify as Christians; atheists and agnostics are disproportionately likely to
in Brazil the figure is 18%. 8% of Japan’s unbelievers say affirm that the universe is ‘ultimately meaningless’ in five
they are Buddhists. Conversely, in Brazil (79%), the USA of our countries, it still remains a minority view among
(63%), Denmark (60%), and the UK (52%), a majority of unbelievers in all six countries. (2.4)
unbelievers were brought up as Christians. (1.1, 1.2)
7. Also perhaps challenging common suppositions: with
3. Relatively few unbelievers select ‘Atheist’ or ‘Agnostic’ only a few exceptions, atheists and agnostics endorse
as their preferred (non)religious or secular identity. 38% of the realities of objective moral values, human dignity and
American atheists opt for ‘Atheist’, compared to just 19% attendant rights, and the ‘deep value’ of nature, at similar
of Danish atheists. Other well-known labels – ‘humanist’, rates to the general populations in their countries. (3.1)
‘free thinker’, ‘sceptic’, ‘secular’ – are the go-to identity for
only small proportions in each country. (1.3) 8. There is remarkably high agreement between
unbelievers and general populations concerning the
4. Popular assumptions about ‘convinced, dogmatic values most important for ‘finding meaning in the world
atheists’ do not stand up to scrutiny. Atheists and and your own life’. ‘Family’ and ‘Freedom’ ranked highly
agnostics in Brazil and China are less confident that for all. Also popular – albeit less unanimously so – were
their beliefs about God are correct than are Brazilians ‘Compassion’, ‘Truth’, ‘Nature’, and ‘Science’. (3.2)
and Chinese as a whole. Although American atheists
are typically fairly confident in their views about God,
importantly, so too are Americans in general. (2.1)
Prof. Stephen Bullivant (co-Leader, Understanding Unbelief) is Professor of Theology and the Sociology of
Religion at St Mary’s University, where he also directs the Benedict XVI Centre for Religion and Society. His
latest books are Mass Exodus: Catholic Disaffiliation in Britain and America since Vatican II (OUP, 2019), and
Why Catholics Leave, What They Miss, and How They Might Return (Paulist, 2019; co-authored with
C. Knowles, H. Vaughan-Spruce, and B. Durcan).
Dr Miguel Farias (co-Leader, Understanding Unbelief) leads the Brain, Belief and Behaviour group at
Coventry University. He has previously been a lecturer at Oxford University where he also did his doctorate in
experimental psychology. His major research interests are the psychobiological roots of beliefs and the effects
of spiritual practices.
Dr Jonathan Lanman (co-leader, Understanding Unbelief) is Assistant Director of the Institute of Cognition
& Culture, Senior Lecturer in Anthropology at Queen’s University Belfast, and Secretary General of the
International Association for the Cognitive Science of Religion. He previously served as Departmental and
College Lecturer at Oxford University, where he completed his doctorate on the cognitive anthropology of
atheism. His research aims to utilise the tools of both cognitive and social anthropology to examine religion,
atheism, morality and group dynamics.
Dr Lois Lee is Senior Research Fellow in Religious Studies at the University of Kent and principal investigator
on the £2.3m Understanding Unbelief programme. Her books include Recognizing the Non-religious:
Reimagining the Secular (OUP, 2015), The Oxford Dictionary of Atheism (with Stephen Bullivant; OUP, 2016),
and the co-edited volumes Secularity and Non-religion (Routledge, 2013) and Negotiating Religion: Cross-
disciplinary Perspectives (Routledge, 2017). She is founding director of the Nonreligion and Secularity Research
Network (NSRN), and co-editor of Secularism and Nonreligion and of the De Gruyter-NSRN book series,
Religion and Its Others: Studies in Religion, Nonreligion and Secularity.
Introduction .......................................................................................................... 6
Note on terminology ............................................................................................ 7
1. Unbelief, identity and religion .......................................................................... 8
2. (Un)Belief, science and naturalism ............................................................... 12
3. The Values of Unbelief.................................................................................. 18
Endnotes ........................................................................................................... 20
Appendix: Survey methodology........................................................................ 21
This report presents emerging findings from the core • To what extent does not believing in God correlate
research project of the Understanding Unbelief programme with not believing in other supernatural phenomena?
(2017-2020). Understanding Unbelief: Across Disciplines • What ranges of worldviews do such people have?
and Across Cultures (ADAC) seeks to map the nature and
• How does unbelief relate to religious and/or secular
diversity of the varied phenomena traditionally – albeit
identities, meaning, and values?
problematically and contestedly (see below) – labelled as
‘unbelief’, across different national settings. • What does unbelief look like outside of western (and
disproportionately Anglophone) countries, and/or
This multi-year research programme is motivated by within minority groups within those countries?
the growing public, scholarly, and media interest in • How do people from different countries, religious
atheism, nonreligion, and secularity, fueled by the growing backgrounds and socio-demographic groups
proportions of religious ‘nones’ and ‘unbelievers’ in understand such central terms as ‘God’, ‘religion’ and
many countries, the flourishing of secularist activism and ‘atheism’ — terms frequently used in international
nonreligious cultures such as ‘New Atheism’, and urgent surveys?
policy debates around the status and rights of atheists,
agnostics, humanists, and related groups. While the The Understanding Unbelief programme – led by an
last decade has seen a rapid expansion in research on interdisciplinary team from the disciplines of sociology,
these topics – see, most notably, the flourishing of the anthropology, and psychology – will address all of these
international Nonreligion and Secularity Research Network questions, among many others, in a more systematic
(NSRN) – understandings of these topics have typically and comprehensive way than has hitherto been
focused on narrow (and arguably unrepresentative) groups attempted. Our overarching methodology combines
within the ‘nonreligious milieu’, and with a very heavy focus in-depth, face-to-face interviews (n = 30 per country,
on the North Atlantic world. This leaves us with many across three separate regions in each) and conceptually
unanswered questions: linked, nationally representative surveys (n = 1100
Note on terminology
‘Unbelief’, ‘atheist’, ‘agnostic’
Our use of the term ‘unbelief’ follows that provided in The Oxford Dictionary of Atheism (Bullivant and Lee
2016): ‘The state of lacking (especially religious) faith or belief… unbelief is often used in a wide sense, implying
a generalized lack of belief in a God or gods.’ It is chiefly employed, along with its cognates (unbelievers,
unbelieving) as a convenient shorthand term, incorporating much of what is commonly termed atheism and/or
agnosticism.
More specifically, for the purposes of this report – and the methodology of the research underlying it – we
operationalise ‘unbelievers’ as those giving either the first (‘atheists’) or second (‘agnostics’) response to the
following widely used social survey question:
Which statement comes closest to expressing what you believe about God? (ISSP 2008)
‘God’
While ‘God’ is a common concept in Brazil, Denmark, the United Kingdom and the United States, it is less
common and less relevant in China and Japan. For the ADAC survey methodology, we have followed the
World Values Survey in using the following terms for ‘God’ in China and Japan, for the purposes of determining
atheism and agnosticism.
China
佛祖、真主, 上帝 或者神明
Fózǔ, zhēnzhǔ, shàngdì huòzhě shénmíng
Buddha, Allah, God, or Spiritual phenomena
Japan
神
Kami
Traditional Japanese term designating deity or spirit
Shown in percentages
85
81
79
77
73
63
28
18
15
1 12
7 8 7
5 5
3 3
1 5 2 1 00 1 2 0 1 000 0 1 2
0 1 11 0 0 1 1
It has long been known that by no means all of those who identify as ‘nones’ or as having ‘no religion’ on surveys
are, in reality, atheists or agnostics.1 As we see above, this also works both ways: not all atheists or agnostics would
describe themselves as having no religion.
Across our six countries, between 63% (Denmark) and 85% (China) of unbelievers answer our widely used religious
affiliation question with ‘no religion’. In all countries, a significant minority opted for a religious identity, with Christian
labels being most popular in Brazil (18%), China (7%), Denmark (28%), the United Kingdom (15%), and the United
States (12%), and Buddhist ones most popular in Japan (8%).
With the clear exceptions of China and Japan, a majority of atheists and agnostics in our countries say that they were
brought up within a religious tradition. Four in every five Brazilian unbelievers, three in five Danish and American
unbelievers, and half of all UK unbelievers, say that they were brought up as some kind of Christian.
Other religions are also well-represented in unbelievers’ backgrounds. Most notably, 8% in America were raised Jewish,
13% in Japan were raised Buddhist, and 4% in China were raised Muslim.
Shown in percentages
82
79
70
63
60
52
42
32
26
14 15
13
8 7
5 4
2 22 3 2 3 2
1 1 0 11 0 0 1 0 0 0 0 11 0 1 0 00
Conversely, a nonreligious upbringing predominates among Chinese (82%) and Japanese (70%) unbelievers, but not
among those in the United Kingdom (42%), Denmark (32%), the United States (26%), and Brazil (15%).
Shown in percentages
Agnostic
Non-religious Agnostic Spiritual but not religious Free thinker
16 27 13 6
16 3 6 17
17 9 14 9
34 2 9 13
27 13 14 11
21 26 18 11
As explained earlier in this report (see ‘Note on terminology’), this report uses i) atheist and ii) agnostic to describe
those respectively answering i) ‘I don’t believe in God’ and ii) ‘I don’t know whether there is a God, and I don’t believe
there is any way to find out’ on the commonly used seven-option survey question: ‘Which statement comes closest to
expressing what you believe about God?’. We are using unbelievers to refer to atheists and agnostics together.
As previous studies have shown, many people who are de facto atheists or agnostics do not choose to identify
themselves, either primarily or at all, by these terms. (There are many reasons for this, which are being explored in the
qualitative research undertaken within ADAC.)
Our data strongly reinforces this. In none of our countries is ‘Atheist’ or ‘Agnostic’ the preferred identity of atheists or
agnostics. The country with the highest percentage of ‘Atheist’-identifying atheists is the USA (39%), in Denmark, it is
only 19%. Meanwhile in Brazil, 27% of actual agnostics identify as such, compared to just 2% in Japan.
There is a great diversity of preferred self-descriptors both within and across the six countries. Among all the possible
options, only ‘Atheist’ (among Brazilian, UK, and US atheists), ‘Non-religious’ (among Danish, Japanese, and UK atheists,
and Japanese and UK agnostics), and ‘Agnostic’ (among UK and US agnostics) attract more than a quarter of unbelievers
from any one country.
Other, historically important designations are markedly less popular. With the single exception of Chinese atheists (17%),
‘Humanist’ was the preferred identity of fewer than one in ten atheists or agnostics in each of our countries. (See also
‘sceptic’ and ‘secular’.)
Small but non-trivial numbers of atheists and (especially) agnostics in all our countries primarily identified with a religious
designation. This includes 12% of Brazilian atheists and fully 17% of Danish agnostics.
All of this is important for two main reasons. Firstly, it means that social-scientific studies of atheists or agnostics which
focus only on those who self-describe in this way - e.g., typically as a write-in option on a survey’s ‘Other (please specify)’
category – are limited to only a very small, and likely atypical, subset of de facto atheists and agnostics. Secondly, it
underscores the point that there is no single term – or discrete set of terms – for referring to these people which a majority
of them would actively choose for themselves. We acknowledged in the introduction the contested nature of the terms
unbelief and unbelievers: there does not, however, appear to be any self-evident candidate to replace it.
-0
-1
-2
On a scale from -2 (strongly disagree) to +2 (strongly agree), we asked both our samples to rate their agreement with
a statement expressing confidence ‘that my beliefs about God’s existence are the right ones’. This produced several
interesting results.
First, agnostics tend to have the least confidence in their views (from -0.21 in Japan to +0.13 in the UK). This
is perhaps not wholly surprising, given that not knowing is, etymologically speaking, agnosticism’s defining
characteristic.
Second, and contrary to some popular suppositions, being an atheist does not necessarily entail a high level of
confidence or certainty in one’s views. All six of our countries’ atheists express overall levels of confidence in their
beliefs about God’s existence either notably lower than (Brazil, China), or broadly comparable to (Denmark, Japan,
UK, US), the general population’s. For instance, the comparatively high level of confidence exhibited by America’s
atheists matches more-or-less exactly the high ‘religious confidence’ of Americans-in-general. Overall, among
American atheists, 45% ‘strongly agreed’ and 20% ‘somewhat agreed’ with our question’s statement; among our
general population sample, these proportions were 44% and 24% respectively.
i) Atheists
50
40
30
20
10
ii) Agnostics
60
50
40
30
20
10
70
60
50
40
30
20
10
While ‘belief’ and ‘unbelief’ are normally used in relation to God (or gods), there are many other supernatural beings and
phenomena that substantial percentages of the general population believe to exist. Logically, unbelief in God need not
entail unbelief in these other things. Accordingly, our survey probed the extent to which unbelievers agree that various
supernatural phenomena either exist (e.g., ‘objects with mystical powers to heal or harm’) or are true (e.g., astrology).2
On the whole, and across all our countries, atheists show the lowest levels of supernatural belief; agnostics show slightly
higher levels; and within the general population such beliefs are typically rather prevalent.
While there is substantial variation both within and between countries on unbelievers’ most/least believed-in phenomena,
a few points stand out. First, the beliefs that there are ‘underlying forces’ of good and evil, that ‘there exists a universal
spirit or life force,’ and that ‘most significant life events are meant to be and happen for a reason’ are the most endorsed
among unbelievers globally.
Second, among our atheists, Japanese are the least, and Brazilian and Chinese the most, ‘supernaturally inclined’.
Chinese agnostics, meanwhile, are strikingly more likely to believe in supernatural phenomena than those from other
countries.
Shown in percentages
27 12 8 2 26 11 27 7 29 8 35 10
Atheist Agnostic
While Figure 2.2 reveals the percentages of unbelievers (in God) who nonetheless believe in other specific supernatural
phenomena, there is also the question of the proportion who don’t believe in any supernatural phenomena at all. Here, we
present the percentage of unbelievers who are thoroughgoing naturalists.
For the purposes of this report, we are using a consistently negative answer to each and every supernatural phenomenon
as a proxy for naturalism.
As can be seen above, in none of our six countries surveyed does the percentage of unbelievers who qualify as naturalists
approach 50%. Even among American atheists, the most naturalistic group across our surveyed countries, only a third
seem to have a wholly naturalistic world view. Among Chinese atheists meanwhile, fewer than one in ten does.
As one might predict from the previous graphs, agnostics are consistently less likely to be naturalists than atheists. In no
country does more than one in ten agnostics qualify; in China, it is more like one in fifty.
47
36
32
30 30
25 25 25
24
23
17
14
For many centuries, there has been substantial scholarly and public discussion regarding the wider implications of not
believing in God – debates which continue to the present day. For example, it has often been argued that for those
without a belief in the existence of a God, the universe can possess no ultimate meaning or purpose.
While this is normally discussed at the level of abstract argument, our data allow us explore whether or not, as a matter
of empirical fact, unbelievers do regard the universe as ‘ultimately meaningless’.
While there is some variation across countries, two points stand out. First, in no country surveyed does the proportion
of either unbelievers, or the general population, affirming a meaningless universe reach 50%.
With the exception of Brazil, where 47% endorse such a view, only around a third of unbelievers in each country regard
the universe to be ultimately meaningless. While unbelievers are admittedly more likely to take this view than the
population at large, curiously this is not so in Japan.
Shown in percentages
87
80
74 74
71 70 69
68 69 68
66 66
63 63
59
52
50 49 49
46
43
33
28
26
‘The scientific method is the only ‘Humans have developed over time from
reliable path to knowledge.’ simpler, non-human life forms.’
Discussions of unbelief often involve discourse on ‘science’ vs. ‘religion’. Consequently, we aimed to examine the extent
to which atheists and agnostics would endorse a scientific account of human origins, as well as a strong statement of
science as the only reliable path to knowledge.
The data appear to show a divergence between East Asian (China, Japan) and other countries (Brazil, Denmark,
the United Kingdom, the United States). In both China and Japan, the percentages of unbelievers and the general
population who regard science as the only reliable path to knowledge are nearly identical. By contrast, the percentages
of unbelievers in Brazil, Denmark, the United Kingdom and the United States endorsing this statement are substantially
higher than the general population. Further, in both China and Japan, members of the general population are more likely
to endorse an evolutionary account of human origins than the unbelieving subgroup. This relationship is reversed in
Brazil, Denmark, the United Kingdom and the United States.
While those figures may suggest a united East Asian approach to science and religion, it is also important to note the
substantial difference between China and Japan in relation to the status of science. Seven-in-ten of Chinese participants
agree that science is the only reliable path to knowledge, compared to only three-in-ten of Japanese participants.
86
83
77
73 75 75
73 72 73
70 69
68
64
53
45 46
40
36 36
33 33
30
21
13
‘All human beings, regardless of where ‘In the long-run, society becomes
they are born, are born with dignity and better over time.’
a special set of rights.’
93
89
82
78 79
77 76 77 78
70 71
60
55 55 56
50 50 51
47 46
45 44 44
41
‘Nature has deep value beyond its ‘What is right and wrong is up
usefulness for human beings.’ to each person to decide.’
Our findings suggest no consistent difference between unbelievers and the general population. Percentages agreeing
with this statement ranged from 41% among the general population of China, to 60% among the general population
of Brazil. While unbelievers are more likely to endorse this statement in China and the United States, members of the
general population are more likely to endorse it in Brazil, Denmark, Japan, and the United Kingdom.
As to human dignity and the objective existence of human rights, unbelievers are typically less likely to affirm these than
are the general populations – although note Denmark and the United States, where the proportions are near-exactly
the same, and the fact that affirmation of dignity and rights is the majority position among unbelievers in all surveyed
countries.
In relation to the ‘deep’ value of the natural world, regardless of its usefulness to humans, in half of the countries
surveyed (Denmark, the United Kingdom, the United States) our unbelievers and general samples endorsed it at near-
identical levels. In the other half (Brazil, China, and Japan), unbelievers were less likely to endorse the statement (82%
vs. 89% in Brazil, 77% vs. 93% in China, and 55% to 79% in Japan). Across all countries and across both samples,
however, the majority of participants endorse the claim of the inherent value of the natural world.
By contrast, a belief in progress – that ‘in the long-run, society becomes better over time’ – shows a huge amount of
cross-national variation. In Japan, just 13% of unbelievers and 21% of the population as a whole affirmed this statement.
The Chinese figures are vastly different: 69% of unbelievers, and 83% of the general population.
3.2 Five top-ranked ‘values’ choices by atheists/agnostics and general population (from 43 options)
‘When it comes to finding meaning in the world and your own life, which of the above are most important to you?
Please select the five items most important to you.’
To address these questions, we presented participants with a list of 43 words and short phrases (e.g. ‘Beauty’, ‘Art’,
‘Family’, ‘Romantic love’, ‘Nature’, ‘Justice’, ‘The connectedness of everything’, ‘A higher power’, plus ‘Other (please
specify)’) and asked them to select the five items most important to them for finding meaning in the world and in their
own lives. Despite all items being chosen by at least one person in every country, we found very high levels of agreement
across our samples. Out of the possible 43 options, only 14 made it into the overall Top Fives for any of our six countries’
unbeliever or general population samples.
Figure 3.2 shows a remarkable level of similarity between unbelievers and the general population across our countries in
what makes the world and life meaningful. ‘Family’ was the most frequently chosen item in all general population samples,
and in four of the six unbeliever samples. Further, where it was not the most frequently chosen item, it came either second
(Brazil) or third (China). ‘Freedom’ was also a frequently chosen item across all samples. It was the most frequently chosen
item by Brazilian and Chinese unbelievers, ranked second behind ‘Family’ in half of our samples, and never fell out of the
Top Five chosen items in any sample.
Other items frequently appearing in the Top Five across both samples are ‘Compassion’ (6), ‘Truth’ (5), ‘Nature’ (4),
‘Science’ (3), ‘Friendship’ (3), and ‘Equality’ (3).
Endnotes
1
For example, Pew data from 2017 shows that 17% of US nones ‘believe in God as described in the Bible’, and a further 53% believe in a
‘God, higher power, or spiritual force’. See: https://www.pewforum.org/2018/04/25/when-americans-say-they-believe-in-god-what-do-they-
mean/04-25-18_beliefingod-00-04/
2
For brevity’s sake, our graphs give shorthand descriptions of the phenomena in question. The actual wording used on the (English language) survey
were as follows: Life after death: ‘There is some sort of life after death.’; Reincarnation: ‘Sometime after I die, I expect that I’ll be born again in
another body.’; Astrology: ‘The positions of the stars and planets affect people’s lives.’; Objects with mystical powers: ‘Some objects have mystical
powers (e.g. to heal, harm or bring good luck).’; People with mystical powers: ‘Some people have mystical powers (e.g. to heal, harm or bring
good luck).’; Significant events ‘meant to be’: ‘Most significant life events are meant to be and happen for a reason.’; Supernatural beings: ‘There
exist supernatural beings, who might be good, evil or neither, such as angels, demons, ghosts or spirits.’ [NB: culturally appropriate entities were
suggested for each country]; Underlying forces of good and evil: ‘There are underlying forces of good and evil in this world.’; Universal spirit or life
force: ‘There exists a universal spirit or life force.’; Karma: ‘There is a power in the universe that causes good things to happen to people who behave
morally and bad things to happen to people who behave immorally.’
For each country (Brazil, China, Denmark Japan, the United Kingdom, the United States), we gathered two samples. The
first is a sample composed only of unbelievers (n = 900), which is made to be representative of the unbelieving population
of each country in relation to Age, Sex, and Region. The other is a General Population sample (n = 200), made to be
representative of the general population in relation to Age, Sex, and Region, which we use for the purpose of comparison.
Our total global sample for ADAC will be n = 6600. Please note, however, that these interim findings are based on a
sample of n = 5285, broken down like so: Brazil (n = 846), China (n = 903), Denmark (n = 593), Japan (n = 1016),
UK (n = 1109), and USA (n = 818).
In order to develop representative samples of unbelievers in our chosen countries, we utilised the most recently available
data from two largescale, highly respected survey programmes: the World Values Survey (WVS; 2010-2014) and the
International Social Survey Programme (ISSP; 2008) to establish a demographic picture of unbelievers according to three
variables (Age, Sex, and Region). We then set demographic quotas for our own survey based on this picture. For example,
if our data from the WVS indicates that 40% of the unbelievers in a particular country are female, we then set quotas of
40% female and 60% male for our own sample of said country.
Based on our analysis of existing WVS and ISSP data regarding belief in the existence of God, we established the
following quotas for our unbelieving samples.
Continued overleaf