Collective Intelligence Applied to Biblical Studies

Collective intelligence is a broad term covering many cases where intelligence or novel information result from the collaborative activities of many individuals. Recent and well-known examples include sites like

  • Wikipedia, where people work together to create encyclopedia-like content
  • i label (or ‘tag’) web page content, and others can look at my tags, or lots of people’s tags, to find things of interest.
  • slashdot, digg, reddit, and similar sites that collect votes on the interest of web pages and then ranked the pages by popularity

Though more popular perhaps in the last few years, these kinds of approaches have been around for some time. Google’s dominance of web search, arguably the current “killer app” on the internet (along with email), comes from a kind of collective intelligence. Their PageRank algorithm uses the number of links to a page from other web sites to estimate how important the page is, and assign its rank in the results you get back from a web search.

The interesting question to me is how collective intelligence might be usefully applied to Biblical studies. There have been a few projects in this area, though i think it’s fair to say they haven’t yielded too much yet. I’ve written a few posts (here, and almost 2 years ago here) about applying “Web 2.0” ideas to Bible study. YouVersion is perhaps the most promising of that bunch, but it still doesn’t collect nearly enough intelligence to really be different (meaning that the scale is too small, not that the comments are too stupid 🙂 ).

Another interesting set of data come from the ESV Bible Blog, where they analyzed their web searches to identify the most popular verses in the Bible. This provides some well-grounded analysis of people’s actual behavior (which is always better than guessing what they do). But as such it’s still just data, not information or knowledge (more about that in this rather conceptual post about the difference between data, information, and knowledge). In other words, how do we apply this data to do something new and different when it comes to Bible study?

Here’s one example collective intelligence project i’ve pondered (though i haven’t yet found time to actually construct it): identifying parables in the Gospels. We have numerous sayings of Jesus throughout the Gospels that use stories, allegories, or other metaphorical language to make a point. Some of these are explicitly described in their context as parables: for example, Mark 4:2 tells us

And he was teaching them many things in parables, and in his teaching he said to them …

We conventionally refer to the story that follows in Mark 4:3-8 as “the Parable of the Soils” (or, perhaps less appropriately given the focus of the story, the Parable of the Sower). However, other stories with the same character aren’t explicitly called parables in the text, or are labeled as parables in one gospel but not another. In fact, the Greek word parabolÄ“ (from which our word parable is a straightforward transliteration) doesn’t occur in the Gospel of John at all, though several of the teachings recorded there have a similar style as parables from the Synoptic Gospels.

If you consult the various Bible reference works, many of which contain lists of the parables of Jesus, you find a great deal of disagreement as to which passages are and are not parables. Not surprisingly, this also reflects divergence of opinion as to what ought to be considered a parable: only those instances where the term parabolÄ“ is used? Those as well as parallel stories? Any kind of figurative language? Wilmington’s Book of Bible Lists lists 38 parables of Jesus (several of which occur in multiple Gospels): the Baker Encyclopedia of the Bible lists 40; Harper’s Bible Dictionary has only 26 (plus a few others found only in the Gospel of Thomas).

Here’s a good candidate for applying collective intelligence to a real issue in Biblical studies: what should we list as a parable? You could approach it like this:

  • Identify the entire set of candidate passages that anybody anywhere has considered, or might consider, a parable (and maybe throw in a few others as a control group)
  • Create a web site where people could log in and simply vote up or down on each passage: Parable or Not?
  • Along with their votes, each participant should record their criteria for voting
  • Participants could also log in as proxies for existing reference lists or scholarly authorities and enter (as votes) what Wilmington, Dodd or Jeremias called a parable.

I’d think at least 100 participants would be required to make this exercise in distributed Biblical scholarship meaningful, and some might turn their noses up at the thought of letting unwashed masses have an equal say with the scholars. But wouldn’t this be an interesting exercise? In particular, rather than “the list” of parables, it would give us the basis for a distribution of opinions: for example, 95% might agree that Mark 4:3-8 is a parable, while perhaps only 10% would label Jesus’ saying about the vine and vinedresser (John 15:1-17) that way. And the criteria might provide some interesting clusters of votes. I’d love to add this kind of data to the Composite Gospel. In fact, that’s what started the idea: i sat down to label the parables, and quickly realized this wasn’t a straightforward task.

Additional resource: The Horizon Project is one product of the New Media Consortium that “charts the landscape of emerging technologies for teaching, learning and creative expression”. In my view, seminary education as well as pastoral preaching and teaching belong among this target audience. The Horizon Project produces an annual report on what’s here now, coming soon in the mid-term, and on the far-term horizon (3-5 years). Collective intelligence is one of their far-term horizon technologies: you can read more about in the Horizon Report.

3 thoughts on “Collective Intelligence Applied to Biblical Studies”

  1. It’s a good idea. I’ve thought about this before also, but in the context of how to interpret and apply various passages.


  2. I am not sold on collective intelligence being as “wise” as some would make out. Unquestionably, collective intelligence has solved or arrived at more correct answers than most individuals on their own, but consider that we know we are fallen and imperfect, with an inclination towards sin and falling away from God (Romans 1). When it comes to biblical study, I find I have a hard time thinking this is a good approach.

    Just look at the number of cults and false teachers in abundance today, leading people astray and, possibly worse, confusing people into thinking they know “Christianity” when most practicing Christ-followers would not recognize such a “Christianity.” For proof of my hesitation, see Paul to Timothy (2 Timothy 4:1-5). Not that this should stop us, but how do we ensure that sound teaching is understood and taught?

    Don’t get me wrong, I love the web and what it can bring us, but I’m just not sure how well this fits Bible study. Having tried eBible, I’m less than impressed so far. Maybe a Twine would be better… 😉

Comments are closed.