Those who still hold to some more nuanced form of the theory take great pains to clarify that of course they don't think religion is in an inexorable, inevitable decline and will, like an evaporating puddle, slowly disappear from the earth. To think that's secularization, secularization theorists say, is a misunderstanding, a misrepresentation of the idea.
In his book Secularization: In Defense of an Unfashionable Theory, for example, Steve Bruce very avidly argues secularization has happened and is happening, and yet he says this idea of religion's inevitable disappearance is critics' straw-man misconstrual of his position. He says no modern sociologist has taken secularization to be inevitable.
Rather, Bruce says, secularization will happen and has happened in the sense that religion is increasingly displaced from the center of human life. Or, more boldly, that "the social power of religion, the number of people who take it seriously, and how seriously they take it" are all decreasing.
Secularization doesn’t mean that religion will disappear, but that it will change. And more, that there will be increasing variety in what it means to be religious, and how people are religious, and what people take “religious” to mean. It's just not the case that religion will decline indefinitely: that’s not what secularization theorist are saying, and not what the numbers actually show. There's no data and no good interpretation of the data that supports that.
Which is not to say there are not plenty of bad interpretations out there. Take the United Kingdom's Office for National Statistics in the report "Religion in Great Britain" (pdf), which was widely, uncritically reported.
E.g., "Atheists likely to outnumber Christians in England in 20 Years,” the Religion News Service's report on the study, which was picked up and run as-is by Christian Century, Crosswalk, the Washington Post and the Richard Dawkins Foundation for Science and Reason.
The big news: "If these populations continue to shrink and grow by the same number of people each year," the study said, "the number of people with no religion will overtake the number of Christians in Great Britain in 20 years."
It's stupid to think that a decline in self-reported identifications with Christianity will continue indefinitely into the future without any change. Speculating about what would happen in 20 years if everything continued on exactly as it is right now is not helpful. Why should we suppose that everything will just keep happening exactly as it has happened, without any change?
Self-described adherence to Christianity declined by 7.6 percent in the UK between 2004 and 2010. If you break that down by year, it's a continuing but irregular trickle. The number of self-identified Christians declined by:
- 1.6 percent from 2004 to 2005
- .7 percent from 2005 to 2006
- 2.2 percent from 2006 to 2007
- 1.4 percent from 2007 to 2008
- 1 percent from 2008 to 2009
- 2.3 percent from 2009 to 2010
It may actually be the case, as the study suggests, that the religious adherence numbers will "continue to shrink and grow by the same number of people each year" for the next 20 years. Maybe in the next six years, adherence will decline another 7.6 percent, and then another 7.6 percent in the six years after that.
But that speculation is premised on the assumption all adherence is equal, that there will always be another .7, or 1, or 1.4, or 1.6, or 2.2, or 2.3 percent of Christians with a weak and weakening sense of connection to Christianity.
It seems much more plausible to suppose there's been some percentage of Christians in the UK who were nominally associated with a church and had rather weak attachments to the dominant religious disposition, attachments which were re-enforced by Christianity's privileged position in the culture. As that privilege weakens, so does that kind of adherence. But not all adherence. What we’re seeing is more likely the end of a specific sort of religious adherence, which was the way in which a segment of the population affiliated with Christianity, and which can be distinguished from other sorts of adherence that are not affected by Christianity’s loss of privilege.
It may be the case that this percentage of the population actually lost their faith, and that something has happened in the culture so that the Christian faith is just less plausible, just crumbling away, with a little more lost faith every day. It could, however, alternatively be the case that some considered themselves Christians not out of a deeply held, individual belief, or, say, out of personal affirmation of the hope of the resurrection, but because there were tangible, social benefits to nominal confessions of adherence to Christianity.
That has changed, and so they have "lost their religion." Except that they didn't themselves change at all.
This is just to say that what "being a Christian" means can be a lot of different things.
That’s what happens with secularization, when faith is an individual thing, a private choice, and there’s no coercive enforcement of orthodoxy. There are a lot of competing understandings of Christianity, and a lot of different ways to be a Christian. There’s no reason to assume being a Christian means the same thing to the 7.6 percent that used to consider themselves Christians but don't anymore and all self-identified Christians in the UK.
The same study shows, in fact, that depending on how you ask the question about religious identification, there can be as much as a 31 percent increase or decrease in identifications with Christianity.
Between 2010 and 2011, the question was changed slightly in the UK study, and the order of possible answers was changed, and the percentage of self-identified Christians dropped by 4.5 percent, a larger decrease than any single year in the previous decade. But it wasn't likely a decrease at all, but just a demonstration of the real wiggle room in these numbers. If you clarify that one can be a Christian "even if you're not currently practising," the number of people who choose to identify as Christian goes up. If you imply a stricter definition of Christianity, the numbers go down.
The same is true with non-religious: A lot depends on the definition.
If non-religious means "atheist," as the wire service headline would have it, the percentage who self-identify that way is actually pretty small. If it's taken to mean a much broader, looser, general description of one's orientation towards the world, more people in the UK will say they're that. Even as many as the 22.4 percent who chose to identify as non-religious in the 2010 survey, or the 27.1 percent who chose that answer when it was listed first among the choices at the beginning of 2011, when "Christianity" was given a stricter definition.
Bruce argues the numbers themselves don't matter as much as "the overall pattern" of declines in religious adherence, and the fact it doesn't seem to easily reverse. This is fine, but won't support the sensational headlines.
The repletion of the stupid version of secularization isn't the study's fault, really. It was one ill-thought-out line that the wire service turned into the crux of a story. There are problems with the study, though, when you dig into it.
For one thing, it bases the stats it has on self-reports, which are problematically unreliable, exactly because secularization means there are various, conflicting and competing understandings of what it means to be religious or not religious.
What does it mean when someone says they are a Christian? It can mean a lot of different things. What does it mean when someone says they're non-religious? It could mean they're atheists, but a lot of non-religious people aren't thinking about God even enough to formulate an opinion on whether there is one, and really aren't atheists but just not religious. Whatever people say in self-reports has to be contextualized, and not simply taken as true. It's not a statement of simple fact, but rather a presentation of themselves to a stranger. Most of the time, whatever one says to the pollster, the real message is "I'm a good person," and people say what they say because they think their answer will communicate that. Measurements of self-identification are largely measurements of what people think the right answer is.
For another thing, the survey allowed heads of households to answer for their households. That should give one pause. If people are inaccurate in their own self-reports -- information they presumably have access to -- are they really to be trusted in reporting their spouses' states of belief of unbelief? Their children's? Do wives actually accurately report the religious identifications of their husbands, husbands accurately report the religion of their wives? At least some of the time, it would seem likely the answer is no.
Do non-believing adults know if their teens are dabbling in faith, and would they say if they did? Do religious parents know when their children are losing their religion, and, more, would they tell a stranger on the phone that that was the case? It seems unlikely. The accuracy of this data has to be called into question.
This gets at one of the weirdest factoids of this survey, a number which kind of throws the easy, bold conclusions into question. By age bracket, the least religious group in the UK today is not teens, according to this report, and not young adults. The least religious Britons are babies.
Between ages 0 and 4, non-religion spikes to 35 percent, and identification with Christianity is at an all-time low of slightly more than 50 percent. The second least-religious and least-Christian age group is 25 to 29 and 30 to 34, where slightly more than 30 percent say they have no religion and about 55 percent say they're Christians. It's this older group that people tend to think of when they think of declining adherence to Christianity -- headlines about a coming age of atheism and the abandoning of religion aren't being understood as the non-religiousness of toddlers.
The fact that 35 percent of those aged 0 to 4 aren't religious doesn't even mean they're not being raised in a religion, either. It may just mean a lot of their parents aren't in churches where babies are considered Christians, i.e., they're in churches that believe one must choose one's religion, making an adult decision. The implication of the fact seems to me to be that fewer children are being raised Catholic and Anglican, more raised Baptist or Brethren, and maybe that, among the Anglicans and others who traditionally believe children are Christian by virtue of being the children of Christians, anabaptist sensibilities have increasingly taken hold.
That's a guess, though, because the point of these studies is that there are lots of ambiguities, lots of nuances, lots of vagueness and variety in what's understood by belief, and unbelief, Christianity, atheism, religion and religious adherence.
That state of overlapping ambiguities and multiple, competing definitions of everything is, it turns out, exactly what scholars of secularization are talking about when they say secularization has happened. This is the secularity brought about by the process called secularization.
The only solid, supportable conclusion one can actually draw from a study like this is a need for caution in interpreting numbers, and a healthy respect for the real complications of the secular condition.