Monthly Archives: July 2011

An Israeli algorithm sheds light on the Bible

Software developed by an Israeli team is giving intriguing new hints about what researchers believe to be the multiple hands that wrote the Bible.

The new software analyzes style and word choices to distinguish parts of a single text written by different authors, and when applied to the Bible its algorithm teased out distinct writerly voices in the holy book.

The program, part of a sub-field of artificial intelligence studies known as authorship attribution, has a range of potential applications _ from helping law enforcement to developing new computer programs for writers. But the Bible provided a tempting test case for the algorithm’s creators….
“We have thus been able to largely recapitulate several centuries of
painstaking manual labor with our automated method,” the Israeli team
announced in a paper presented last week in Portland, Oregon, at the annual
conference of the Association for Computational Linguistics. The team
includes a computer science doctoral student, Navot Akiva, and a father-son
duo: Nachum Dershowitz, a Tel Aviv University computer scientist, and his
son, Idan Dershowitz, a Bible scholar at Hebrew University in Jerusalem.

Read more :

The conference paper in question is available here:


CosmicRaysdit :

Similar work published in 2008:

Workshop at the 16th International Conference on Patristic Studies (Oxford): Biblical quotations in Patristic Texts

FRIDAY AUGUST 12, East School, Room 6, 4.00 – 6.30.

This workshop aims to enable research teams from different countries, whose work involves the identification and analysis of biblical quotations in Patristic texts, to share experiences and allow the audience to have an open discussion about methodological problems. Both patristic and biblical scholars will take part in this workshop, to consider how common tools could be developed to serve different purposes, such as the history of exegesis or a critical edition of biblical texts.

Laurence Mellerin (Sources Chrétiennes, Lyon): “Biblindex, Online Index of Biblical quotations in
Early Christian literature”

Numerous methodological issues have to be solved during the Biblindex project development, so that an effective XML-TEI schema for both biblical and patristic texts can be built: e.g., how to define the difference between allusion and quotation? How to distinguish and typify the Church Fathers’ ways of introducing and changing the biblical texts? How to decide where a quotation begins and where it ends? Through a few case studies, we’ll present Biblindex solutions and submit them to discussion.

Hugh Houghton (ITSEE, Birmingham): “Patristic Evidence in the New Edition of the Vetus
Latina Iohannes”
The paper is structured into two main parts:

a) the procedures followed
b) questions and textual insights arising from the patristic material for John

Bas ter Haar Romeny (Peshitta Institute, Leiden): “Biblical Quotations in Syriac commentaries”

This paper will first discuss the evidence for various versions of the Old Testament in Syriac exegetical works. In addition to quotations from the Peshitta, which is a second-century translation based on a Hebrew text, one finds references to the recentiores and to “the Greek” and “the Hebrew”. A complicating issue is the fact that references to “the Greek” may refer either to the Syro-Hexapla or to ad hoc renderings of the Greek biblical text of Greek commentaries that have been translated or excerpted.

Second, the significance of the material will be discussed. Quotations from the Fathers should play a role in the constitution of the critical text of the Peshitta, Syro-Hexapla, and Hexapla. This might seem to be self evident, but practical problems have often prevented editors from using the evidence. There are also more reasons why the quotations are important. For instance, they may help us to assign a place and date to a certain text form. In exegetical collections which do not state their sources, they may help to find out the provenance of the material. And last but not least, the role given to the quotations is a major aspect of the exegetical method that needs to be analysed.

The third issue that will be brought up has to do with the practical problems mentioned before: the Peshitta Institute will cooperate with the Biblindex Project and the Hexapla Institute in order to make at least a selection of the material available.

Claudia Wick (Thesaurus linguae latinae, München): “The Language of the Bible in the Thesaurus Linguae

The paper will discuss the procedures followed in ThLL to identify sources and establish the relevance of “VET.LAT” label for a biblical quotation, in a lexicographical way of thinking.

Another short paper:

Dominique Gonnet – The use of the Bible within Athanasius of Alexandria’s Letters to Serapion

My short communication has to be seen as a following of James D. Ernest’s The Bible in Athanasius of Alexandria (Boston-Leiden 2004). In fact, Athanasius of Alexandria’s Letters to Serapion did not belong to the corpus studied in that book. I collaborate to the Biblindex Project  HYPERLINK “” http :// ?lang=en which will be presented as workshop on “Biblical quotations in Patristic Texts” at this Conference. In my communication, I will summarize some conclusions about the use of the Bible in Letters to Serapion : how far do Athanasius knowledge of the Bible go? Does he use τρόποι, following the example of his adversaries? With how many verses did Athanasius answer the few verses quoted by Athanasius adversaries? How does he balance Old and New Testaments? Which place does he give to St. Paul’s writings?

Background: From Biblia Patristica to Biblindex

1. Biblia Patristica (1965-2000)

Hitherto, the only available tool for researching early Christian interpretation of the Bible was Biblia Patristica, a project held in Strasbourg from 1965 to 2000 by a CNRS team, CADP, Centre for Patristics Analysis and Documentation. The objective of the Centre was the production of a comprehensive inventory of all biblical quotations and allusions in ancient Christian literature. Unfortunately, due to lack of a sufficient team, Biblia Patristica had to stop its activity.

Its 7 published volumes already offer printed indexes of biblical quotations and allusions in the  Greek and Latin writings of the Church Fathers in the first three (and part of the fourth) centuries.

Seven volumes have been published, along with a supplementary volume for biblical references in Philo of Alexandria, who served as an exegetical model for many patristic authors.

Volume 1, which appeared in 1975, covers the beginnings of extracanonical Christian literature up to Clement of Alexandria and Tertullian.

Volume 2 indexes the literature of the third century, apart from Origen, to whom volume 3 is devoted exclusively.

Coverage of the fourth century begins with volume 4, which includes the writings of Eusebius of Caesarea, Cyril of Jerusalem, and Epiphanius of Salamis. Volume 5 covers Basil of Caesarea, Gregory of Nazianzus, Gregory of Nyssa, and Amphilochius of Iconium, and volume 6 turns to the Latin writers Hilary of Poitiers, Ambrose of Milan, and the Ambrosiaster.

Volume 7 on Didymus the Blind appeared in 2000.

The first step of Biblindex was to resume the  work of Biblia Patristica.

These volumes published make an inventory of 270.000 references, which have all been scientifically verified. A CD ROM had been prepared, funded by CNRS Editions, which was to have been released with the publication of the 6th paper-format volume. Thanks to the work of Jean Allenbach and André Pautler, former members of the CADP team, Sources Chrétiennes inherited this database in 2006.

2. Archive retrieval by Christian Sources, Biblindex “in the pipeline” (1999-2008)

In November 1999, CADP entrusted Sources Chrétiennes with the first part of its unpublished data, corresponding to 5 boxes of works analysis processed at least partially. These are around 100,000 verified references, which have been computerized. In 2006, again thanks to the work of Jean Allenbach and André Pautler, Sources Chrétiennes inherited the BIPALYON database, containing half of these unpublished data computerized : the 7th volume on Didymus the Blind, the verified and unverified data for the planned 8th volume, never published. It shoul have contained :

  • Athanasius
  •  John Chrysostom
  • Theodoret of Cyr
  • Procopius of Gaza
  • Jerome, and others

+ The spuria and dubia of authors of the other published volumes (Gregory of Nyssa, Epiphanius, Amphiloque, and Basil, Gregory of Nazianzus, Hilary of Poitiers, Ambrose, Justin).

In 2000, André Pautler entrusted Sources Chrétiennes with the second part of CADP’s data, this time 8 boxes containing printed or handwritten lists of references, untested and not computerized, for which there are no record in any database and even without an inventory list. These new boxes represent around 300,000 references.

As an example, Cyril of Alexandria’s complete verified data are on paper listings in the boxes, but whitout any record in the database. Now we have an inventory list of these archives : very different texts, from different space eras and time areas, are to be found in them. They were analyzed by members of the clergy in the three decades during which the CADP worked, without close coordination.

These data are difficult to operate, because the lists of references are often very difficult to read : pale listings, handwritten notebooks or sheets  which are difficult to decipher. You can see here (nice) examples of these materials.

One example : Jerome

In Biblindex, teams of specialist researchers have been organized for each important author. They are going to resume and complete these data. For example, none of Jerome’s works are included in Biblia Patristica, but most of them have been analysed in Biblindex.

The missing works include:
– Treatises on the Psalms
–  Comments on the Prophets (except Jeremy)
–  Commentary on Ecclesiastes
–  Comments on Matthew
–  Comments on the Pauline Epistles
–  Prefaces and Prologues
–  Lives of Malchus and Hilarion

For works already covered by Biblindex, are to be done data verifications, using a more recent version of the text where applicable. For the others, are to be inventoried in details the works presented in boxes. Works absent from Biblindex have to be entirely treated, according to our “Guidelines”.

Another example : John Chrysostom

There are 43,701 references to John Chrysostom in BIBLINDEX, i.e. more than one tenth of the data transmitted in electronic form by the CADP.

The corpus concerned is primarily the authentic corpus, with exceptions. This does not mean that the pseudo-chrysostomica are illegitimate : they should naturally have their place in Biblindex in due time.

Very few works are missing : the two Letters to Innocent, the Commentary on Job, and of course the still unpublished writings. These gaps will be filled gradually. An important work have already been done at Sources Chrétiennes to precise the titles of these works, which are very difficult to identify precisely.

The data are very reliable, according to a test done on a few works. They have been checked at least once by the CADP. In accordance with the rules of the CADP sticking to references to specific texts, some very slight or very loose allusions have not been reported.

The data were reported mainly from the text of Migne’s Patrologia. However, we must hope that results based on more modern editions would not have been fundamentally different, because in this specific case, to undertake a new comprehensive check would take nearly 30,000 hours of work…

3. Biblindex beta version (2006-2009)

The context in Sources Chrétiennes

From 2003 was developed in Sources Chrétienes an electronic database to inventory ancient authors and their works published or to be published in the series. Parallel to this initiative, 2006, an online scriptural index of Bernard of Clairvaux, on the basis of 12 volumes SC or 13000 Biblical quotations, was made. The completion of this project has helped to identify technical issues and define the means necessary for a larger-scale operation.

From 2006 to 2009, we applied successfully for funding at the Rhône-Alpes Region. We obtained € 19,000, ie 56 days for development and  a few equipment. The House of the Orient and the Mediterranean, Sources Chrétiennes Association, and a few donators on the web site helped us too, so that we could find 4,000 euros more.

Development of Biblindex-beta

We encountered technical and practical difficulties. We had to keep the database structure of the CADP, but we already had our own database. We had no IT engineer in our team, so we worked with vacations, not always precisely at the right moment. We had to let the IT development in abeyance, due to the lack of funding. The result form has not been tested and corrected by Patristicians after the engineer had finished his work. So there remains room for improvement.

Using the web site

First of all, you have to register. Then, you can access to the search form. Here, you have to read the instructions to figure out how things work, because it doesn’t work like you might think.
This form allows you to investigate the corpus by entering one or more biblical references, and to sort the results by ancient authors, works, dates, geographical areas (poles) or Clavis numbers. After selecting your criteria (biblical passage and patristic author/area), you have to hit the plus sign before you go on. As long as you can’t see a dimmed table with the asked choice in it, with a “-” button on its left, allowing to remove the choice, that means it wasn’t taken into account. The results are automatically calculated as you confirm your choices, but the page does not refresh (AJAX technology): the number that appears on the left of the light shows you the number of occurrences. You should therefore expect until the form is available to enter the following informations.

Ex. of biblical settings :

1 Samuel 1, 19-28 + 2,11-3,21 +4,1

= references to Samuel’s birth and life in Silo, without quotations of his mother Anna’s prayer.

After you hit the plus sign, the traffic light turns to green, and the search button is not dimmed anymore. You can load your query.

If the choices were properly validated, and that the “start search” button is staying dimmed, that doesn’t mean it doesn’t work, that means there are no matches. You can check this point with the counter top right next to the light.

It yields the results. Then you can see the references returned for your query and print them. On the top of the results form, there is a reminder of your search criteria. You have to select “expand all” and see details. The search results are not that easy to scroll through, but it gives you the critical edition along with the passage citation. There are not direct links to the works. Clicking on the an author cited provides a link to additional information and bibliographical help at Sources Chrétiennes

Our IT engineer is currently working to revise and upgrade search functionality and to improve this results form.


–         95% of Biblia Patristica data, published and archived by computer, are available

–         An interface with high « added value »

–         A great users community was created. Users feedbacks have helped us a lot to obtain new fundings : we have chosen to put the site online, although it still had to be deep improved, because we did not know if there really was a need for Biblindex. One of the conclusions we could draw from these feedbacks and registrations was the necessity of this tool

.. and deficiences

However the site developed is only a small step forward along a long path.

–         Gaps in the data:

  • Series of verses are taken into account only by their first and last verses
  • References to unnumbered verses do not appear
  • Septuagint verses with a letter do not appear

–         Search form is not intuitive enough

–         Difficult to sort, export, print results

e.g., the display of results when searching in a single work has to be simplified, to avoid repetition of the work title; full scriptural apparatus for a chosen work must be easy to obtain, the relative proximity of two verses in a work must be seen, etc.

–         It is still impossible to really work together with users, as we don’t have a sufficient collaborative interface.

4. The outlook of Biblindex (2011-…)

Our third applying for funds at the French Research Agency, ANR was at last successful : we obtained € 330,000 to fund 4 years. This time, we work together with four other institutions : two French labs working in data processing, lemmatisation, geovisualisation ; two biblical institutes, the INTF in Münster and the Peshitta Institute in Leiden.

We could recruit two people : an IT engineer, because first and foremost, computer development must be emphazised; and a patristic researcher who has biblical analyses in charge. They started working in February this year.

We will try to change the perspective of Biblia Patristica according to the problems we have pointed out. In Biblindex, Bible parts should be referred to works parts. Biblindex should not only be a tool to produce biblical apparatus, but a tool for matching biblical excerpts with patristical excerpts.

The access to the texts themselves

Until now, as stated above, only numerical references can be searched in Biblindex. The mere fact of knowing that a biblical text exists in a specific area at a specific time, or in another specific text, is already crucial information. Though researchers also need to know what sort of biblical manuscripts or compilation manuscripts the ancient writer was reading, what type of biblical text he quotes, and so forth: all this information can only be found in the details of the text quoted.

An experimental development, on a few texts written in each language we aim to explore, will be led to establish relationships between the index and the text itself.