Evidence from the past and future that suggests religious revival is not happening in the United States

I appreciate this part of Ryan Burge’s approach to examining whether a religious resurgence is happening in the US: he looks at past patterns and he considers possible changes in the future.

Photo by Arina Krasnikova on Pexels.com

First, the present data:

The General Social Survey, for instance, reported a steady rise in the “nones” between the early 1990s and 2020. In 2018, the figure was 23%, rising to 28% in 2021. The two most recent estimates are slightly lower — 27% in 2022 and 25% in 2024. Similarly, the “headline finding” from Pew’s Religious Landscape Survey was that both the decline of Christianity and the rise of the unaffiliated have paused in recent years.

As Burge and others have noted, recent data seems to show a stopping/slowing/plateauing of two trends: fewer people affiliating with Christian traditions and more Americans identified as “religious nones.”

But these patterns are also related to aging and whose activity researchers can examine. Burge next turns to the future:

When you compare generations, the pattern is obvious. The youngest members of the Silent Generation were born in the early 1940s, and just 7% report no religious affiliation. In less than a decade, they — and a growing share of Baby Boomers (18% unaffiliated) — will disappear from survey samples.

Meanwhile, millennials are moving solidly into middle age, and 36% of them say they have no religion. Generation Z, all of whom will soon be adults, are even less religious: 43% are nones. That’s 25 points higher than the Boomers they’re replacing. So if the overall share of nones sits around 28% now, it will inevitably rise as generational turnover continues.

Could millennials and Gen Z find God in the years ahead? Possibly — but it would require a transformation unlike anything seen in modern times. Roughly 10 million millennials would have to reaffiliate with religion, followed by another 18 million Gen Zers. There’s no sign of that happening in any dataset.

In other words, for the percent of people in the United States to identify as Christian in the future at the same rates as now would require more young people to become Christian. For the percentage of Christians to grow, even more religious change would need to take place.

By looking at past, present, and future possibilities, Burge concludes: “I can say without equivocation that there’s no clear or compelling evidence that younger Americans are more religious than their parents or grandparents.”

When trying to understand what is happening in a social group or society, one data point or set of evidence is often not enough to fully understand what is happening. Patterns can change over time or the way we understand the world can change over time but a compelling case needs to be made. Seizing on new evidence that does not fit what we know about something might hint at significant change – or it could be a sampling outlier. Good steady research can help reveal these patterns even if there are multiple actors wishing that we could have identifiable patterns more quickly.

When a revival overwhelms a small town of 6,000 people

The recent religious revival at Asbury University brought a lot of people to Wilmore, Kentucky:

Photo by Aayush Srivastava on Pexels.com

By the time university leaders concluded the gathering, an estimated 50,000 students and visitors had come to the campus to pray, said Kevin Brown, Asbury’s president. The outpouring attracted students from more than 260 colleges and universities, many drawn by social media livestreams and posts. Similar prayer services cropped up at other Christian universities, including Lee University in Tennessee, Cedarville University in Ohio and Samford University in Alabama…

The surge of worshippers overwhelmed the campus and the sleepy town of Wilmore, which is home to roughly 6,000 residents, according to U.S. Census Bureau data. Cars streamed into the city, backing up traffic and filling the town’s parking spaces.

“We have two stoplights, to give you an idea of how large our town is,” Brown said. Suddenly having to figure out how to accommodate thousands of visitors “on the fly” was “unnerving and unsettling.”

“Our town and our institutions are just not equipped to absorb such a large influx of people,” he said. “On the other hand, it was really, really sweet and really beautiful to see so many different people, so many different ages, representing so many different geographies … just to see everyone in one space, united and experiencing something together.”

The juxtaposition of religious activity and visitors in a small town is worth considering. Three questions come to mind:

1. How many communities would be prepared for a large influx of visitors there for religious purposes? What exactly would they need to respond and what would mark the interactions and activity outside of the clearly marked religious spaces?

2. Would Christians in the United States be more or less surprised to find religious revival happening in a small town or in a major city (thinking of Billy Graham’s long meetings in 1957 in New York City or the Azusa Street Revival in Los Angeles in 1909)?

3. Are the conditions of small town life more or less conducive to religious fervor? Americans often have romantic notions of small towns yet big cities are denser and have more people coming and going.