Two “Communities” demographics: Don’t mention the Others

Roy Fisher does a deep dive into the census data, showing how it is almost impossible to opt out of the two communities model…

“When I told the people of Northern Ireland that I was an atheist, a woman in the audience stood up and said, ‘Yes, but is it the God of the Catholics or the God of the Protestants in whom you don’t believe?’”

– from The Wit and Wisdom of Quentin Crisp (1984)

When the quotation above was first published we lived in a different society. Aside from the more frequently noted differences, we were less diverse by country of origin, and self-identifying atheism was uncommon. Indeed, ‘no religion’ only became an option on the NI census seven years later in 1991. Yet, in 2018, despite increasing secularism and migration, we can still hear the echoes of Quentin Crisp’s audience member.

Last week, in a widely published news story, we heard academic Dr Paul Nolan predict that the 2021 census will find a “Catholic majority”. Media coverage all included the same 2011 census ‘religion’ data, without mention of ‘other’ or ‘no religion’. However, in order to achieve the reported ‘48% Protestant, 45% Catholic’ headcount, 16.9% of our census religion responses, and non-responses, went through a process of second-questioning and best-guessing which wasn’t even hinted at in most of the coverage. (There is an alternative analysis of the ‘religion’ statistics below.)

Most census respondents will only be asked to answer one religion question, “what religion, religious denomination or body do you belong to?” However, since 2001 the census, like Crisp’s audience member, has asked atheists to disclose which religion we no longer identify with.

The Northern Ireland Statistics and Research Agency (NISRA) provides a rationale for adding this ‘background’ question: “In response to the increasing proportion of the population who indicated that they had no religion or did not reply to the question, and to aid equality monitoring, it was proposed that an additional question be asked in the 2001 Census – about religion brought up in – but asked only of those who indicated that they did not belong to any current religion.”

No religion / religion not stated

Firstly, it must be questioned how state statisticians responded to the growth in atheism. But before we get to that, the claim itself of “the increasing proportion” can be contested. During ‘the Troubles’ there was serious contention related to the census – including the killing of a census taker in 1981.

The proportions of ‘not stated’ in 1971 and 1981 were 9.4% and 18.5% respectively. In 1991 this refusal decreased greatly to 7.3%, and as mentioned above it was the first time the census measured ‘no religion’, recording 3.7%. The combined ‘no religion’ and ‘not stated’ count in 1991 was 11%, significantly lower than the 18.5% ‘not stated’ in 1981.

Combining these groups is another issue. Why would statisticians cloud the proportions of both the non-religious and non-respondents? Atheism, agnosticism, humanism etc. and non-compliance are two quite different things. 

The 2001 census, nonetheless, reported a combined figure of 13.9%. There was no separate recording of ‘no religion’ and ‘not stated’. Any religion was given a recordable value, while ‘no religion’ had the same value as no answer at all.

In 2011 NISRA did record ’no religion’ and ‘not stated’ separately, 10.1% and 6.8% respectively, but in the Census results press briefing and for the ‘NISRA Short Story – Religion’ they presented a combined figure of 17%. This doesn’t inform the receiver what proportion of the population identified as having no religion, just that it’s somewhere between 0% and 17%.

A possible reason for conflating these groups is because of what we are not. We haven’t selected any religion. Or perhaps more importantly, we haven’t identified ourselves as part of the ‘two communities’. We don’t fit the narrative. We can’t be categorised in traditional binary terms.

For other censuses on these islands the ‘not stated’ category is presented separately, and there is no process of imputing a religious background value, as is the case for the NI census.

Equality Monitoring

The current equality monitoring form makes an assumption about how people ‘see’ each other, stating “Regardless of whether we practice religion, most of us in Northern Ireland are seen as either Catholic or Protestant. We are therefore asking you to indicate your community background by ticking the appropriate box below.”

The Equality Commission states, “The community background of an individual refers to whether that individual has been treated as belonging to the Protestant community or the Roman Catholic community in Northern Ireland.” If we don’t fill this box in, ‘equality monitoring officers’ are advised to try to impute a religious identity based on known information, like name, address or country of birth. While for the census there is Edit and Imputation software used to re-categorise us based on these various cues. This is called ‘the residuary method’.

For the purposes of equality monitoring we are looked at in this way to try to promote and achieve equality of opportunity between these two “communities”, and it has been largely successful. And is understandable in this light.

However, to use these statistics to define society, and to herd anyone who tries to opt out back into their perceived identity, promotes the narrative of two monolithic or competing ‘communities’.

Alternatively, the Equality Commission and NISRA could record ‘stated religion’ only, and allow atheists, agnostics and humanists to self-identify without trying to shoe-horn them into the binary. After all, the Fair Employment Treatment Order 1998 and The Northern Ireland Act 1998 both refer to discrimination “on the ground of religious belief or political opinion”.

The state ‘treating’ or ‘seeing’ me as a member of the Protestant community disguises my non-belief, and tells them nothing of my political opinions. People within these so-called communities have differing political opinions on almost every political question. There are correlations on some issues certainly, but as we are witnessing, religious background doesn’t determine someone’s opinion on Brexit, or abortion, or LGBT rights.

We know the Protestant faith has many denominations, some more religiously fundamental than others, but even the ‘religion or religion brought up in: Catholic’ census category contains a vast array of identities. Not only does it include thousands of atheists, but it includes many migrant Catholics. Whether they are devout, nominal, or perceived Catholics, whether they are from here, England, Poland, Lithuania, the Philippines, or anywhere, they are all treated as having the same community background.

Expounding the 2011 statistics and the ‘Catholic Majority in 2021’

The 2011 census reported that 40.8% of the population ticked the “Roman Catholic” box, while 41.6% ticked the boxes for one of the three main Protestant churches or answered with a denomination which is categorised as ‘Protestant or other Christian (including Christian related)’. 10.1% identified as being of no religion, 0.8% said they were of an ‘other [non-Christian] religion’, with the remaining 6.8% choosing lawfully not to answer the question. (0.1% rounding error)

As Catholicism is certainly not going to exceed 50% by 2021, I conclude that Dr Nolan means a Catholic majority in relation to Protestants only, rather than an overall majority within the state.

In the statistics above you can see the gap between Catholics and Protestants in 2011 stood at 0.8%. However, the Protestant/Christian grouping includes many different write-in answers – some of which are not, or may not, be Protestant.

Vague answers like ‘Christian’ and ‘Church’ add to this conflated group, as do the answers ‘non-denominational’, ‘Jehovah’s Witness’, ‘Orthodox Church’, ‘Mormon’, ‘mixed Catholic/Protestant’, ‘independent’, and more. The most frequent of these examples is ‘Christian’, which was the answer supplied by, or for, 14,630 people. They alone equate to the 0.8% gap.

Instead of analyzing the “religion” data to conclude that there was actually a ‘square brackets’ Catholic majority in 2011 when compared to definite Protestants, the media outlets reported the ‘background religion’ statistics (as usual). BBC NI’s Newsline and The View, the Irish News, Belfast Telegraph, iNews, the Scotsman, amongst others, all reported the 2011 census findings as 48% Protestant and 45% Catholic.

But omissions were made in the presentation of these statistics, and in the analysis of the story:

·       Predominantly receivers were not advised that these statistics are for “religion or religion brought up in” – this led to 81,995 self-identifying atheists (4.5% of the entire population) being interpreted as Protestant or Catholic.

·       ‘48% Protestant’ actually contains all of the other non-Catholic and mixed Catholic/Protestant Christians.

·       This set of statistics is completed with 5.6% ‘no religion’ and 0.9 ‘other religion’, but these did not feature in any of the media coverage. 

·       People who did not answer the religion question are not made aware that they have been subject to the census ‘residuary method’, and thus will most likely be represented as a census Protestant or Catholic.

·       11% of the 2011 NI population was born elsewhere. Part of the reason ‘Catholic’ numbers increased is due to migration, particularly from countries that became member states of the EU in 2004. 4.6% of the entire NI population in 2011 were Catholics who were born outside NI. The corresponding statistics for Protestants/Christians are 2.9% from outside NI. The census data accessed gives no indication as to how many of these Christians may not self-identify as Protestant.

 The second religion output is often referred to as religious or community background, but this misses another minority of people: religious converts. They are not second-questioned by the census, or thus regressed to a childhood or former church. If someone was brought to a Church of Ireland as a child, as I was, but prior to census day 2011 had converted to Buddhism, they were reported as ‘other religion’ in both outputs. This means religious converts have more agency in the census, and its results and interpretations, than first generation atheists.

The second output has no ‘not stated’ category. The 6.8% who chose not to answer the optional religion questions, along with any atheists, agnostics, Jedi Knights etc. who didn’t self-identify a ‘religion brought up in’, have a religious background imputed by NISRA.

This post-census process alters the “religion” results as follows:

 Protestant/Christian: +6.8%
Catholic: +4.3%
Other religions: +0.1%
No religion: -4.5%
Not stated: -6.8%
(0.1% rounding error)

Evolving identities and attitudes

Identities evolve. In the eyes of the state a ‘single’ person may get ‘married’, to later become a ‘widow’ or ‘divorcee’. Many women and an increasing number of men change their names after marriage, and we can all do it by deed poll. Transgender citizens can change their legal gender. Our physical and mental health identities change throughout our lives.

People who actively identify as Protestant or Catholic represent a huge range of views, from very religious to very secular, highly political to apolitical. However, they choose that category because it meant something to them.

However, when it comes to the evolution of an irreligious identity, the NI census responds ‘but what did your parents raise you as?’, or alternatively imputes an identity without consultation. There is no way to opt out.

The OSCE’s Ljubljana Guidelines on the Integration of Diverse Societies, a document obviously applicable to consociational democracies, promotes the “primacy of voluntary self-identification”, advising that “assimilation against one’s will by the State or third parties is prohibited”.

The NISRA / NI Neighbourhood Information Service (NINIS) website allows citizens, including employers and media, to enter a postcode to access census data relating to a specific Local Government District (LGD), Assembly Area etc. Under the heading “Ethnicity, Identity, Language and Religion”, NINIS provides the “religion or religion brought up in” statistics, but not the “religion” ones. While the website makes the user aware that they are the ‘background’ statistics, they do not advise the user of levels of religion in 2011, as the headline suggests they might. Similarly, the NI Assembly document “Census 2011: Key Statistics at NI and LGD level” contains “religion or religion brought up in” but not “religion”.

In short, as far as state statistics are concerned, it is barely possible to opt out of the binary.


When I asked NISRA about why the no religions were conflated into past religious identities, they replied that this is what the ‘census data users’ want to know. But who exactly wants to know which church I attended (without choice) decades ago? Why do they want to know? Why does NISRA want to report this to them? Other than workplace equality monitors, is it politicians, academics and/or the media?

If so, I question NISRA’s response to the many, many inaccuracies made by these users when they disseminate the data – ‘Religion or religion brought up in’ statistics are often interpreted and reported as representative of current proportions of Protestants and Catholics, and despite the points raised above there is a suggestion that these identities indicate an opinion on issues such as flags, parades, bonfires, or the constitution.

Yes, caveats are often added about conflating religious identities and political opinion – yet we keep hearing the ‘background religion’ statistics in relation to political or constitutional issues.

So this is a suggestion for NISRA and to those that report on the census data. You’re forcing the two community binary on people who don’t want it.

Please, make it stop.


Roy Fisher is an artist and trader at St George’s Market. Roy holds an MRes, in which he focused on the NI Census and Equality Monitoring.

Discover more from Slugger O'Toole

Subscribe to get the latest posts to your email.

We are reader supported. Donate to keep Slugger lit!

For over 20 years, Slugger has been an independent place for debate and new ideas. We have published over 40,000 posts and over one and a half million comments on the site. Each month we have over 70,000 readers. All this we have accomplished with only volunteers we have never had any paid staff.

Slugger does not receive any funding, and we respect our readers, so we will never run intrusive ads or sponsored posts. Instead, we are reader-supported. Help us keep Slugger independent by becoming a friend of Slugger. While we run a tight ship and no one gets paid to write, we need money to help us cover our costs.

If you like what we do, we are asking you to consider giving a monthly donation of any amount, or you can give a one-off donation. Any amount is appreciated.