Online index of biblical quotations in the literature of Late Antiquity and the Middle Ages – BIBLINDEX

BIBLINDEX must keep together two requirements: making available, as soon as possible, an enlarged corpus with the capacity for further extension which makes possible innovative research and the building of fruitful international partnerships with teams offering digitalized textual corpora.
We begin with the manual entry of references on the basis of handwritten analyses compiled in thousands of notebooks: by the middle of 2013, about 250,000 references have been put into spreadsheets.
At the same time, an information system has been developed through an iterative process; the database model and the web interfaces designed for the work based on quotations from the texts themselves have to be consistent with the numbered references.
The new database includes eleven biblical texts written in different ancient languages (Hebrew, Greek, Latin, and soon Syriac) and modern languages (French, English). A multilingual concordance between those Bibles has been created, allowing the user to visualize the biblical text in order to compare it with the quotations found.
Patristic researchers have also worked hard to define exactly what a “work” may be, in order to establish proper links between biblical and patristic texts. They established precise guidelines, giving a typology of quotations. On this basis, the analysis of texts from scratch was organized according to the corpus of authors, with each linguistic area requiring different skills.
In addition, prospective research about lemmatisation and semantic enrichment has been conducted on limited corpus samples. Two tools have been developed: a lemmatizer adapted to the Greek language used during Late Antiquity; a «pre tagging« software, which helps the analyst identifying biblical quotations and thus makes the enlargement of the corpus easier.
The integration of statistical-semantic tools in the information system is our main current work in 2013.

Adding references in the database available online for free is the first deliverable of this project. The input of 250,000 existing records is currently underway.
In addition, a research seminar takes place monthly, whose papers are published in a special series of the Centre for Patristics Analysis and Documentation in Strasbourg or in the hypotheses.org research notebook. It deals with the history of the reception of the Scriptures by the Fathers. Thanks to this, contacts with other institutions are tied to build future partnerships and graduate students can also contribute to the project by this means. Several scientific papers on both IT and patristic aspects of the text-tagging methodology have already been or are shortly to be published.
Besides, the fact that the site is already online generates results.
The hundreds of thousands of biblical references which were put together in the database already allow statistical approaches which were not possible before; the attached map is an illustration of this. Furthermore, the community of more than 7000 users of the site, coming from sixty different countries, is not the least of the achievements: on this basis, the further collaborative work become possible.
Finally, IT tools and software developed will be made available on the web platform for wider use: the ancient Greek lemmatizer, the Quotation Toolkit.

Various partnerships are being set up, in order to prepare a request for funding from the European Community. They involve following partners:
- other French teams working on the reception of the Bible;
- other biblical institutes understanding the high value of a tool useful at the same time for critical editions and patristic exegetical research;
- teams offering digitized corpora of ancient texts;
- projects working on text-reuse in other corpus;
- computer science labs dealing with natural language processing, particularly with search for plagiarism.

Two particularly innovative features of BIBLINDEX are its spatio-temporal breadth and its multilingual corpus. By providing access in a single database to the quotations of the Bible in Jewish and Christian literature, in any ancient language including those of the East, BIBLINDEX offers a new prospect for biblical studies, paying attention to the reception of the text through the study of its different versions and subsequent interpretations. In addition, because of its modular nature it could in future be extended to other corpora. This will facilitate intercultural dialogue without the distortions of fundamentalism.

Research dedicated to the semi-automatic identification of quotations and allusions will have multiple applications. It will be initially applied to the biblical corpus itself to track down all intra-biblical parallels, which are specified in highly variable way from a modern edition to another, and never identified in a completely multilingual way. Then it will be used for the identification of biblical quotations in patristic texts, but may also, as well as the visualization tools using synchronized parallel text windows connected according to multiple criteria, be used in other corpora.
The research which has already taken place has opened up a few areas that were not necessarily planned at first. A generic proposal for encoding quotes following the standards of the Text Encoding Initiative is being prepared, based on the methodological guidelines written for the specific use of BIBLINDEX.
The creation of a specific lemmatizer for late Ancient Greek will have applications wider than the biblical quotations, and meets a real need for the scientific community.
Moreover, a very promising phase is just beginning: the creation of innovative interfaces dedicated to spatio-temporal data visualization. The application of geovisualisation to the textual corpus of BIBLINDEX, a discipline which usually involves territorial sciences, is a real challenge.

The bilingual research notebook biblindex.hypotheses.org, created in 2011, hosts many methodological contributions to the project, including presentations made at the monthly seminar.
A first book has just been published, fruit of the collaboration between ITSEE in Birmingham and the Institut des Sources Chrétiennes : M. Vinzent, L. Mellerin, H.A.G. Houghton (eds), Biblical Quotations in Patristic Texts, Studia Patristica LIV, vol. 2 (Papers presented at the Sixteenth International Conference on Patristic Studies held in Oxford 2011), Leuven-Paris-Walpole, MA 2013. This book contains four contributions directly related to the analysis of corpora in BIBLINDEX, including the definition of criteria for distinguishing between quotations. Another publication is in press: L. Mellerin, «New Ways of Searching with Biblindex, the online Index of Biblical Quotations in Early Christian Literature«, chapter 11 in Claire Clivaz, Andrew Gregory, David Hamidovic (eds.), Digital Humanities in Biblical, Early Jewish and Early Christian Studies, Brill, Leiden, 2013, p. 175-192 [excerpts are readable on the research notebook]. It gives first examples of statistical visualization, foretaste of other results which each user will be able to customize, according to his or her own criteria, on the next available web site.
Finally, on the occasion of the publication of a collective book about Bible readings (I-XV c.) in the three monotheistic religions, coordinated by Sources Chrétiennes, a conference will take place in Lyon on 17 and 19 October 2013 devoted to the reception of the book of Ecclesiastes, which will also give rise to publications.

BIBLINDEX aims to create an online index of biblical references found in Jewish and Christian literature, as well as in Western and Eastern texts of Late Antiquity and the early Middle Ages, in order to renew the study of the interpretation and history of biblical texts.
It seeks to link:
- A corpus of Bibles, collections of scriptural books which were originally written in various languages of the Ancient East and translated early on. Quotations in ancient authors show these texts in the process of development;
- A corpus of ancient and medieval authors, who refer to the Bible as a fixed entity yet at the same time contribute through their quotations to the form and concept of 'the text'.
In reuniting these currently separated elements of cultural and literary heritage, BIBLINDEX will provide researchers and the general public with a tool for understanding how the Bible was woven into the intellectual fabric of society, becoming an interpretative matrix for Western civilizations, and supply a larger audience with the keys to a better knowledge of religious texts.
BIBLINDEX project is supported by the Institute of Christian Sources (Sources Chrétiennes), a CNRS research team in patristics, currently enjoying a high level of international recognition through its series of published editions. The project is based on the resumption of Biblia Patristica, printed indexes of biblical citations in the writings of Greek and Latin Church Fathers from the first centuries published by a team in Strasbourg between 1965 and 2000. Sources Chrétiennes is ideally placed to handle these valuable data, as it brings together, in addition to its own researchers and university lecturers, a wide-ranging network of French and foreign contributors.
With funding from the Rhone-Alps Regional Council from 2005 to 2009, the site BIBLINDEX-ß was developed and put online in December 2008, making 400 000 biblical references freely available: these consisted of both published and unpublished data from Biblia Patristica. While there remains room for improvement, this site already offers an interface with high 'added value'.
The current ANR project aims at considerable enlargement of the corpus, but not only: BIBLINDEX cannot remain a simple index of references, but must become a collaborative platform, offering direct and interactive access to a broad and brand new database, including direct or indirect access to biblical and patristic texts themselves. This is why the heart of the project is also two important IT research centres, LIRIS in Lyon and LIG in Grenoble, which will work on lemmatization and spatio-temporal visualization.
BIBLINDEX, without equivalent either in France or abroad, will be the first global, multilingual tool on biblical citations and their interpretation in Late Antiquity.
From the outset, the purpose of this site is international: what could be more polyglot than a biblical index? Via a site using e-working public software, researchers worldwide will have the opportunity to contribute to its improvement without costly investment. This will both optimize their own research and develop the data through interdisciplinary exchanges. Prestigious research institutes working on the biblical text, such as the INTF in Münster and the Peshitta Institute of Leiden University, have already given their support.
Two particularly innovative features of BIBLINDEX are its spatio-temporal breadth and its multilingual corpus. By providing access in a single database to the quotations of the Bible in Jewish and Christian literature, in any ancient language including those of the East, BIBLINDEX offers a new prospect for biblical studies, paying attention to the reception of the text through the study of its different versions and subsequent interpretations. In addition, because of its modular nature it could in future be extended to other corpora. This will facilitate intercultural dialogue without the distortions of fundamentalism.