70,000 Prayers Sorted!

In September of 2021, I put up a short video explaining the task of sorting through the 50,000 prayers we had amassed from about 50 sources. Since that time we expanded our scope by another 25 dioceses from the Holy Roman Empire, the Baltics, Scandinavia, and a few other sources we deemed important. This brought the total number of collects up to more than 70,000.

Thirteen months later, this huge task is finally complete (it took almost two years in total). Unlike the Scripture readings, which come from the relatively stable text of the Vulgate, the text of a given prayer is much more liable to vary from one missal to the next. For example, the collect Pasti cibo spiritalis…consequamur, chiefly for St. Mark, was found with 13 variations across 35 sources (see the Incipit column below). Though the variations begin (and also end) with three different words, each belongs to the same collect (see the Text column).

We used the ten-volume Corpus Orationum (hereafter, CO) and its catalog of over 7,600 collects of the Western Church as the standard for our table of unique collects. Of the 4,089 individual collects that we identified, 2,500 aligned with the CO. (Missals contain a great many collects not found in the prayerbooks used by the compilers of the CO.) Our colleagues in Hungary plan to submit the remaining 1,500 for publication as an appendix to the CO.

In order to use the CO as the basis for identifying the collects, it was necessary to scan and perform OCR on all ten volumes. I am indebted to Mr. Eric Fattig for his tireless assistance during this phase of the project.

Once the CO had been converted to a searchable table, the sorting of collects began. We tackled them according to starting letter (collects that begin with A, etc…). Seminarian Conner Walts, then a senior at Indiana State University, had arranged with the university to serve as an official intern for The Lutheran Missal. He continued in this capacity unofficially after the semester’s end, working his way through half of the alphabet. Pastor Andrew Harris and Mr. Noah Vancina also made significant contributions. (Don’t be fooled by the letter D—one of every five collects begins with the word Deus.)

Now that all the Latin collects have been matched up, our software can present the data to our editors in a visual way, as shown below.

In this example from Trinity 10, there is a 2 to 1 majority for Deus qui omnipotentiam, which is the older, more established tradition for this day. The other collect, Pateant aures, more properly belongs to Trinity 11. (The minority sources follow a Roman innovation that shifted some, but not all, of the Trinitytide propers back a week. Perhaps Fr. Stefan will write something on the Trinity Schism in a future post.)

What’s next for the collects? Now that we can easily identify the historic tradition for each occasion, we will compare existing translations of the collects, such as TLH and the 1549 Book of Common Prayer, with the original Latin, paying special attention to any evangelical corrections made by the Reformers. Thankfully, we won’t need to translate all 4,089 unique collects—many of which are heretical. It was necessary to sort through all of these collects, but now we need only pay attention to those that will appear in the occasions of The Lutheran Missal.