Jon Morgan

Often when using BPBible I needed to find the particular verse I was thinking of (exactly what search is designed for). Unfortunately, though, I didn’t always remember the exact wording of the verse, and even when I did the words I used might have been from a different version from the one I was searching. Nowadays, I’m much more likely to turn to Google to help me, since it turns out that with the help of the Internet they have solved the problem pretty well.

Trying to find a particular verse in BPBible

Take for example a well-known concept: avoiding even the appearance of evil. In BPBible, the ESV is my default version, and the search doesn’t turn up what I’m looking for:

As it happens, the verse is from the KJV, and if I did that search in the KJV it would find it. But it can get frustrating trying to figure out what are the exact words to search for, and which version is the right version to look in. For many years we talked about adding a multi-version search to BPBible. It would have done a lot to address these problems. But the question was how to fit this into our existing Bible search functionality.

The search we had was powerful, flexible, and relatively fast searching a single Bible on a modern computer. But to use it on multiple versions would have required the search to be run for each version separately and then combined. Each Bible would need to be indexed in advance, and building indexes wasn’t particularly fast. The user would probably have wanted to choose which versions were going to be searched, and in which order. And how would we display the results? Show a bunch of matches in different versions with duplicates removed? Try to map all the matches back to the original version? Something else?

In short, I had questions, questions, and more questions. I’m sure all of these issues could have been addressed, but it was never quite important enough to work on.

Google to the rescue

Like my BPBible search, Google allows a very simple search:

But look how much more useful the results are. The very first result is the verse I’m looking for, in the version which has that wording (the KJV). But it doesn’t just stop there. It also has a list of relevant articles and blog posts discussing the meaning of the verse. I would usually then go back to BPBible to find out more about the verse, but that could just as easily be done through the online Bible programs Google links to.

What I would usually then do is go back to BPBible to look at the context of that verse, see if it actually means what I thought it did, and find any relevant cross-references. I use BPBible because it’s what I’m used to, but these things could easily be done through the online Bible programs Google links to.

When interpreting a verse like this, context is important: it’s easy to remember a Bible verse as a throwaway line without really understanding how it fits or considering other related verses. In this case, the “obvious” interpretation of the verse - that believers should avoid doing anything which others might think are wrong - is not backed up by modern versions like the ESV. A contrast seems to be being drawn between holding on to things that are actually good, and avoiding things that are actually bad. In fact, the simple message would have condemned the Jesus of the gospels, who viewed his job as reaching the sinners, not as toeing the party line of the Pharisees.

And I’m sure the various articles Google suggests say the same thing.

Google as Bible software

In this case, Google isn’t just a search tool. It addresses another debate we had: whether we should have a magic reference bar which allows the user to either go to a specific reference or to search the current Bible. In BPBible, we opted to keep a distinction between looking up a reference and searching for one or more words. It was just so much easier that way. How were we to determine whether a search like “Luke” was looking for a book of the Bible or a character in the Bible? And why should a mistyped reference become a useless search?

Google takes that magic search box and actually makes it work. If I want to look up or find any verse or passage on any internet enabled computer, the verse is just a Google search away. No installation required. It’s also not limited to the standard Protestant English canon. I can easily find a verse in 2 Maccabees or Sirach. If I want to look for concepts rather than specific verses, I’ll get plenty of matches for searches like “What does the Bible say about love?” And the results in those posts will have been curated by a human, rather than a list of the 744 times the word ‘love’ occurs in the ESV. That’s pretty powerful. And yes, Google relies on the many specialised Bible reading tools online to actually provide the different texts. But as a user I don’t need to think about it: I just ask Google, and it’s a genuine case of “seek and you will find” (which Google tells me is from Matthew 7:7).

I’ve talked about the “Logos philosophy” before: that you should build up as large a theological library as possible, since the different resources in your library are more valuable when integrated together with a global search. Google can provide a similar service. However, where Logos indexes an individual’s (expensive) theological library, Google indexes the world’s (freely available) Bibles and theological musings. It’s not just a search tool: Google provides a lot of the user interface elements required to access the Bible texts stored in different software around the web.

Of course, Logos is a much more specialised and powerful tool, and its carefully prepared scholarly resources may be more accurate than some random article on the web. It also allows searches that Google can’t, such as realising that “he” in a particular verse is actually Peter, or John, or Jesus. But for the casual user Google has a lot going for it. And BPBible is not really in the running.

General purpose tools vs specialist tools

Frequently, we assume that a specialised tool is guaranteed to get better results than a general purpose tool. But I’m not sure that’s always true. This is related to the Inventor’s Paradox: Sometimes, it is easier to produce a general solution than a very specific solution. It also relates to data analysis: Sometimes, having more data to work with is more useful than searching for a better algorithm or building a better tool. And here, I don’t know about Logos, but Google has a lot more data available to it than BPBible ever did.

In this case, I don’t know how much of Google’s functionality is general purpose. From playing with it, I’m pretty certain there is support in Google search specifically to identify Bible reference searches. Probably a number of different people have tweaked it over the years to work better with Bible verses, the same as they have done for many other search domains. And maybe now their wonderful AI is able to detect which searches look like Biblical searches and react accordingly.

But to get the features of a Logos, BPBible would have required specific code and UI design. We would have required reference lookup. Multi-version search. Integrated search of commentaries, dictionaries, and other Bible resources. Intelligent ranking of different search results. Each of these things we would have had to try and guess exactly what the user was trying to do, and we would probably have guessed wrong on many of them. Even if we got them all right, it would still have required each and every relevant Bible version and commentary to be prepared for use by the application and then installed by the user.

Google, on the other hand, has a very powerful generalised search and ranking tool. Through the wonder of the web it is able to access the wisdom of the crowds. Bible versions, commentaries, and blog posts are available to it without anyone having to specifically prepare them for use with Google. Some of these resources are resources that we could not even access: For example, we could not even offer the NIV as a pay-for Bible, while Google enables it to be accessed and searched freely.

Conclusion

The Google Bible tool is a powerful tool, allowing the user to look up Bible references anywhere, search multiple Bible versions, analyse a wealth of commentary on Biblical passages, and then present a list of results ranked for relevance. I doubt there is much specialised code required to do this. It is just a part of Google’s general mission to organize the world’s information and make it universally accessible and useful.

Obviously, it will not appeal to everyone. Specialist tools still allow specialised searches that are difficult with Google. And anyone qualified to study the Bible in the original languages or wanting access to scholarly resources will probably be better served by a specialist tool. Even online there are tools which do a much more thorough job of Bible study. But Google provides a valuable starting point before linking through to these tools.