Most of us are probably aware that the original manuscripts and early copies of the books of the Bible did not have chapter and verse numbers. These were added centuries later for convenient reference. However, some might not be aware that there are actually many competing reference schemes for dividing the Bible into books, chapters and verses. Take this snapshot by way of example:
In the LXX (the Greek version of the Old Testament) Esther 5:1a is Esther 15:2-15:4 in the KJV. The use of letters here (as in 1a) indicates that this material is in the Greek, but not the Hebrew, edition of Esther, which added material the Latin translators moved to the end of the book. The English tradition of versification more closely follows the Latin in Esther, thus accounting for the radically different chapter number. But it gets more complicated than that: there are differences between the Latin numbering and the English, so Esther 15:2-15:4 in the KJV is Esther 15:5-15:7 in the Latin Vulgate. But after Vatican II there was a concerted effort to make the Vulgate follow the older Greek and Hebrew traditions more closely, so the Nova Vulgata, or New Vulgate, numbers that section as Esther 5:2a-5:2c. But to make matters worse, the LXX numbering we use today comes from the Stuttgart edition, but the numbering in the older Cambridge edition edited by Swete and several other, older editions follow a different LXX numbering. They designate Esther 5:1a as Esther D 2 – D 4, introducing the use of letters as chapter indicators for the Greek additions.
If one were to make an xml file that mapped out these relationships, one node might look like so (taking the LXX as the starting point):
- <verse id=”Esther.5.1a”>
- <map scheme=”KJV” t=”Esther.15.2-15.4″ />
- <map scheme=”NRSV” t=”Greek_Esther.15.2-15.4″ />
- <map scheme=”LXX2″ t=”Esther.D.2-D.4″ />
- <map scheme=”NVUL” t=”Esther.5.2a-5.2c” />
- <map scheme=”VUL” t=”Esther.15.5-15.7″ />
Of course, one XML file would be insufficient to resolve all the possible transformations, because as you can see, these types of transformations are not one-to-one, but rather many-to-many, so to really nail these sorts of conversions, you’d need an xml file for each system of versification.
But the problems are more complicated than just with chapter and verse divisions – they operate all the way up to the book level. The English and Greek Bibles have a book called Bel and the Dragon which has 42 verses, but this is Daniel 13:65-14:41 in the Vulgate and Daniel 14:1-14:42 in the New Vulgate. The English and Greek Bibles have a book called the Epistle of Jeremiah, but this is the 6th chapter of Baruch in the Latin tradition.
Ready to really make your head spin? The LXX has a book called 2 Esdras (or Esdras B) that is a combination of Ezra and Nehemiah. The English bibles have a book called 2 Esdras that is not at all related to Ezra or Nehemiah, but is a non-canonical book not found in the Greek tradition at all – it is found in the Latin Bibles, where it is called 4 Esdras (or Esdrae or Ezrae) – 2 Esdras in the Vulgate refers to Nehemiah. But modern editions separate the core Jewish material of 4 Esdras from the Christian additions, and break 4 Esdras into EITHER 2, 4 and 5 Esdras OR 4, 5 and 6 Esdras. And to really add insult to injury, in the Slavonic and Russian bibles, 2 Esdras is the name for the English and Greek 1 Esdras, which is called 3 Esdras in the Latin because 1 Esdras in Latin is the book of Ezra. Except when it isn’t. So when a reference book, or a user of Bible software, writes ‘2 Esdras’ there are a great many places that reference could refer to.
Early versions of Logos Bible Software tried to impose some order on this tangled mess of reference schemes. You can link your Hebrew Bible with an English translation and Malachi 3.19 in the Hebrew will sync with 4.1 in the English and so on, and these mundane transformations work rather well. But where book boundaries and names are radically different, or the material is massively rearranged, a preference for the English system was enforced or no versemap alignment rules were defined.
We’ve just finished a massive data project that defines the various structures of the different Bibles in all their glorious, gory detail, and allows for effortless transformations from one versification system to another. As it stands now, the project consists of 56.6 megabytes of xml files that look a lot like the snip pasted above. The new Bible data type generated from this data will be available to you, our dear customers, in the next major version of the software, which means the next version will have a higher degree of precision when it comes to Bible navigation, comparing Bible versions and viewing them in parallel, and Bible reference tagging. However, if any versification schemes don’t line up perfectly in the next generation tools, please, blame my horse.
[Ed.–If topics like this interest you, be sure to check out BibleTech 2008, a conference on Bible and technology to be held January 25-26, 2008, in Seattle.]