Over at fMh yesterday, Sara Katherine Staheli Hanks introduced a new series, “When the Temple Hurts.” I was particularly interested in a point she made in the post about how often we discuss the temple in lessons and talks at church:
The temple is a regular focus of meetings, lessons, talks, and discussions in church settings. I’d estimate that, in my experience, 1 out of every 4-5 Sundays in my adult life as a church member has included a talk or lesson where the temple was a primary focus.
I’ve been teaching primary now for a couple of years, and my memory of adult classes is maybe suffering from a bit of haziness, but Sara’s numbers sound good to me. This would mean an average of at least one temple-related lesson or talk per month, with of course lots of variation where there is a cluster of them, and then maybe no mentions for a longer period of time.
I would be interested to know how well that matches up with other people’s experience. It seems like this would be a difficult thing to measure well. Sure, we have correlated lesson manuals, but we also have locally chosen topics for things like sacrament meetings, first Sunday meetings in RS and priesthood quorums, and Teachings for Our Time lessons. And for that matter, even how correlated lessons are taught varies a lot from ward to ward and from teacher to teacher (much to the frustration of the Correlation people, I’m sure).
So I thought I would look at a related question that’s easier for me to answer, which is whether talk about the temple at the general level of the Church has been increasing or decreasing over time. If there’s a noticeable change over time, this is probably felt by people in their local experiences, as general-level material like Conference talks are not only used directly in the preparation of lessons and talks, but they also likely help drive local leaders’ perceptions of what topics are important at the moment.
I used the ever-wonderful Corpus of General Conference Talks to look at how often the word temple has been used in Conference since 1900. This graph shows the result. The dark purple line is the 10-year moving average, and the faded purple line shows the year-to-year rates.
It looks like mentions of the temple are as high now as they’ve been in the last century. (Actually, although I left the older years out of the graph, they’re as high as they have ever been, at least since 1851, which is as far back as the corpus goes.) There has been a steady increase since 1970 or so, although it looks like references might have flattened out since about 2000. This is kind of a surprising pattern to me. Thinking back to my last post on what’s been reported in the Church’s annual statistical report, there was a period of over a decade when temple building wasn’t even included. Of course, this ended in the late 1990s with the beginning of the temple-building boom. But it looks from the Conference data like GAs were talking more and more about the temple through the time when very few new temples were being built, but then they’ve leveled out in how much they talk about the temple even as more temples were built in the late 1990s and early 2000s.
I also looked at how often temples were talked about in Church magazines. To find this, I used Google to search for temple, restricting my search to lds.org, and to the magazine and year I was looking at. For example, to search for mentions of the temple in the Ensign in 2010, I used this search string: “temple site:lds.org/ensign/2010” (without the quotation marks). I wanted to know what percentage of articles were mentioning the temple rather than just the count (since the Church magazines have probably varied in size across time), so to estimate the total article count, I searched for a very common word that was likely to occur in most articles and not vary across time. I chose the word he, so for example to estimate the total article count in the Ensign for 2010, I just substituted he for temple in my search string: “he site:lds.org/ensign/2010” (again, without the quotation marks). One other thing to note on the off chance you try to replicate my work is that Google typically gives a really high estimate of the number of hits on the first results page, but then if you click through to the last page of results, it tells you how many there actually were in the end. I always clicked through to the end to find how many actual results were found, rather than how many were estimated on the first page.
Here’s the result:
The results for the Ensign are really surprising here, given what’s in the previous graph. The Ensign includes Conference reports, so the data for the graph above and the data for this graph actually overlap. But there’s no sign of the increase in mentions of the temple in this one. Of course, what’s being measured isn’t exactly the same. The first graph above is looking at what fraction of all words in Conference talks are temple. This graph is looking at what fraction of documents (articles) have at least one use of the word temple. Still, though, I would expect them to be more related.
The New Era and Friend lines look like they’re showing an increase since maybe the early 1990s until now, pushing the frequency with which articles mention the temple in these magazines to as high as it is in the Ensign. I’m not sure what was important about that time, but I think it’s interesting that both started to increase at about the same time.
So hit me with your hypotheses. What do you think is going on with these data? What’s behind the New Era and Friend trends, or for that matter, the Ensign lack of trend? Or what’s behind the Conference trend? Or why might these data be biased or otherwise untrustworthy?