Final Report Summary - GREEK INTO ARABIC (Greek into Arabic: Philosophical Concepts and Linguistic Bridges)
"Greek into Arabic" features among the Success Stories of the ERC funded grants (See URL: http://erc.europa.eu/succes-stories/bridging-greek-and-arabic-philosophical-and-linguistic-heritage).
The core of the project Greek into Arabic (www.greekintoarabic.eu) was a foundational text of Arabic-Islamic philosophy: the so-called pseudo-Theology of Aristotle, namely the Arabic adaptation of parts of Plotinus’ Enneads. This work inspired al-Farabi and was commented upon by Avicenna. When “Greek into Arabic” was launched, it was available in two non-critical editions, based on few manuscripts. Missions have been organized in the countries whose libraries were expected to preserve little known or unknown manuscripts of this work; the reproductions of the manuscripts have been collected and their contents classified; sample collations have been conducted by team members and the P.I. resulting in a stemma codicum whose details can be refined, but whose basic structure has been outlined on the grounds of a systematic inspection of the manuscript tradition; a provisional text has been established, all this being accompanied by historical and doctrinal studies, disseminated via the two main activities of the team: the international workshops and the journal of the project, Studia graeco-arabica. The missions conducted by members of “Greek into Arabic” have raised the number of the known manuscripts of the pseudo-Theology of Aristotle to more than 100 (See UNIPI_Fig. 1). The critical edition of this work, a major desideratum in the field, is on its way; the philological, doctrinal and historical studies to be published together with the critical edition will change the state of the art in the field. Studia graeco-arabica, the journal of the project, is now established as a respected publication in the field. To this activity in the field of philology and history of philosophy, two ground-breaking research activities have been carried on within the context of "Greek into Arabic": a research in computational linguistics as applied to ancient texts, and a lexicographical enterprise in the field of Graeco-Arabic studies.
Computational linguistics as applied to ancient texts.
The research carried on by ILC-CNR under the guidance of Dr. Andrea Bozzi in the field of textual scholarship and computational philology followed a development strategy that guarantees both the high specialization requirements of the research done by "Greek into Arabic", and elements of compatibility with other existing data, especially the lexicographical terminology developed at the Seminar fuer Orientalistik, RUB Bochum.
a) The system G2A Web App (http://g2a.ilc.cnr.it) has been created: a research infrastructure for scholars interested in texts and translations. Tools were designed to be flexible and implementable in different projects with structural similarity and common purpose. Therefore, the computational work done for "Greek into Arabic" should be considered as a successful pilot initiative, useful also for further adaptation.
b) The technological system and data have been designed to be compatible with the principles at the basis of a RI in the field of humanities, so that the processed data and information produced are interoperable with others already available in internet. Standards have been met for technological component development and, albeit indirectly, for the insertion (automatic or manual) of specialized annotations on digital documents: delimitation of corresponding Greek-Arabic pericopes, comments, hermeneutical notes, linguistic annotations, vocabulary, etc. Based on these principles, a “library” of sofware components has been built whose main modules are the following:
1) Parallel contexts or parallel pericopes module. It is used to align two Greek and Arabic texts, so that the interpretation could be performed under the best possible conditions.
2) Module for parallel reading of the two texts. The Greek text and its Arabic version can be read according to their original sequence as well as according to the order of the Arabic version: the graphical interface allows the user to position himself on the Greek and see the text of the corresponding Arabic translation or, vice versa, to position himself on the Arabic text and scroll the corresponding Greek text.
3) “Glossary” type annotations module. It is possible to select the entire pericope (or of one of its parts) in the graphical interface, so that it can be associated with an annotation. The information is recorded in a database and represents important material, that is shared with the editors of the Lexicon at Bochum.
4) Synthetic evaluation of the translation. G2A Web App has also a further possibility of navigating within the textual data, using a classification of the annotations as an access key. Offering the scholars the possibility of organizing their comments and annotations according to a typology represented by key words (a type of subject catalogue) is a considerable advantage especially in the phase of information retrieval.
5) Linguistic analysis. Both the Greek text and its Arabic version have been semiautomatically analyzed, using the tools developed within the project: each Greek and Arabic word has been labelled with its morphological information and the lemma.
6) Indexing and querying engine. A specific indexing engine has been developed able to list in alphabetical order all the Greek wordforms and lemmas. G2A Web App provides both basic and combined search. The basic search is intended to be performed on one text at a time, while the combined search is performed on both texts simultaneously.
The application architecture provides the use of components some of which are active for Greek into Arabic, others are ready to be used for future research projects. The value of the system is, therefore, its great flexibility and interoperability.
Graeco-Arabic Lexicography
The Greek and Arabic Lexicon (http://telota.bbaw.de/glossga/) in the form of a reference dictionary of the Arabic translations of Greek philosophical and scientific source texts has been the starting point and material basis of research activities at Ruhr-University of Bochum, in the workgroup of Prof. Gerhard Endress, assisted by Dr. Rüdiger Arnzen (mainly responsible for the Greek and Arabic Dictionary), and Dr. Yury N. Arzhanov (mainly responsible for the Glossarium Graeco-Arabicum database). With the support of the ERC Advanced Grant, and working as an affiliate of the main editorial and philological project of the Principal Investigator at UniPisa, this has been continued following the procedures established in the published parts, up to the end of the Arabic letter bâ', i. e. including all lexical materials lemmatized before the Glossarium Graeco-Arabicum online database was being set up. The Glossarium Graeco-Arabicum database has not only taken over where the printed publication left off, but is complementing the material presented here, and enhancing accessibility to a wider range of lexical, grammatical and stylistic features of scientific language. In the first instance, the database was created for presenting the basic data compiled in our card-files and additional material collected by the Bochum team. For this purpose, the full card file was digitalized, and for each entry, searchable data were entered from the file-cards, after being checked and supplemented on the basis of the original source texts. The software platform was developed as an online mySQL database, and continues to be hosted, by the TELOTA centre of the Berlin-Brandenburg Academy of Sciences, URL: http://telota.bbaw.de/glossga. Since the start of our ERC Advanced Grant project, the principal development work — setting up the database, programming the user interface and creating and enhancing sophisticated query, input and output facilities — was achieved by Torsten Roeder of the Berlin Academy, while the data input was managed by the Bochum team. The database is meant to open up the full range of the terminology, and more generally, the grammar and style of scientific Arabic: registering the entire vocabulary of altogether 70 scientific and philosophical texts dating from the 8th to the 10th centuries, a total of 95.500 records, and thus documenting the language of scientific categorization and demonstrative discourse created by the translators and their Arabic readers. The task we set ourselves for this project of making our materials, compiled in the earlier as well as in the more recent phase of our project, accessible in a searchable database, has been completed under the ægis of the ERC grant. In addition, (a) further texts, edited recently and parsed in view of the Greek and Arabic correspondences, have been included, and (b) for many of the sources, the basic lexemes were supplemented by the relevant context and by comments regarding the translators’ techniques and transposition. All fields of the datasets are searchable through a database search engine. During the final period, our work has focused on the development of appropriate tools and techniques for visualization of the results of the search queries and on the creation of a user friendly export function which provides the possibility to download or transfer via e-mail the results of any search query, and the output of large subsets of data in the shape of lexical entries to be used for the continuation of the Greek and Arabic Lexicon.
Successful cooperation with the informaticians of Associate Participant ILC-CNR, using the G2A Web application developed at the ILC-CNR, Pisa (providing parallel Greek and Arabic texts of the Theology of Aristotle and other Arabic Plotiniana in structured pericopes, and searchable for Greek and Arabic word pairs), and our partners at the Berlin-Brandenburg Academy of Sciences, have provided first results in creating import facilities into the TELOTA database, in view of including lexical datasets of the Theologia Aristotelis (and further texts, more of which are becoming available on the G2A platform).
The core of the project Greek into Arabic (www.greekintoarabic.eu) was a foundational text of Arabic-Islamic philosophy: the so-called pseudo-Theology of Aristotle, namely the Arabic adaptation of parts of Plotinus’ Enneads. This work inspired al-Farabi and was commented upon by Avicenna. When “Greek into Arabic” was launched, it was available in two non-critical editions, based on few manuscripts. Missions have been organized in the countries whose libraries were expected to preserve little known or unknown manuscripts of this work; the reproductions of the manuscripts have been collected and their contents classified; sample collations have been conducted by team members and the P.I. resulting in a stemma codicum whose details can be refined, but whose basic structure has been outlined on the grounds of a systematic inspection of the manuscript tradition; a provisional text has been established, all this being accompanied by historical and doctrinal studies, disseminated via the two main activities of the team: the international workshops and the journal of the project, Studia graeco-arabica. The missions conducted by members of “Greek into Arabic” have raised the number of the known manuscripts of the pseudo-Theology of Aristotle to more than 100 (See UNIPI_Fig. 1). The critical edition of this work, a major desideratum in the field, is on its way; the philological, doctrinal and historical studies to be published together with the critical edition will change the state of the art in the field. Studia graeco-arabica, the journal of the project, is now established as a respected publication in the field. To this activity in the field of philology and history of philosophy, two ground-breaking research activities have been carried on within the context of "Greek into Arabic": a research in computational linguistics as applied to ancient texts, and a lexicographical enterprise in the field of Graeco-Arabic studies.
Computational linguistics as applied to ancient texts.
The research carried on by ILC-CNR under the guidance of Dr. Andrea Bozzi in the field of textual scholarship and computational philology followed a development strategy that guarantees both the high specialization requirements of the research done by "Greek into Arabic", and elements of compatibility with other existing data, especially the lexicographical terminology developed at the Seminar fuer Orientalistik, RUB Bochum.
a) The system G2A Web App (http://g2a.ilc.cnr.it) has been created: a research infrastructure for scholars interested in texts and translations. Tools were designed to be flexible and implementable in different projects with structural similarity and common purpose. Therefore, the computational work done for "Greek into Arabic" should be considered as a successful pilot initiative, useful also for further adaptation.
b) The technological system and data have been designed to be compatible with the principles at the basis of a RI in the field of humanities, so that the processed data and information produced are interoperable with others already available in internet. Standards have been met for technological component development and, albeit indirectly, for the insertion (automatic or manual) of specialized annotations on digital documents: delimitation of corresponding Greek-Arabic pericopes, comments, hermeneutical notes, linguistic annotations, vocabulary, etc. Based on these principles, a “library” of sofware components has been built whose main modules are the following:
1) Parallel contexts or parallel pericopes module. It is used to align two Greek and Arabic texts, so that the interpretation could be performed under the best possible conditions.
2) Module for parallel reading of the two texts. The Greek text and its Arabic version can be read according to their original sequence as well as according to the order of the Arabic version: the graphical interface allows the user to position himself on the Greek and see the text of the corresponding Arabic translation or, vice versa, to position himself on the Arabic text and scroll the corresponding Greek text.
3) “Glossary” type annotations module. It is possible to select the entire pericope (or of one of its parts) in the graphical interface, so that it can be associated with an annotation. The information is recorded in a database and represents important material, that is shared with the editors of the Lexicon at Bochum.
4) Synthetic evaluation of the translation. G2A Web App has also a further possibility of navigating within the textual data, using a classification of the annotations as an access key. Offering the scholars the possibility of organizing their comments and annotations according to a typology represented by key words (a type of subject catalogue) is a considerable advantage especially in the phase of information retrieval.
5) Linguistic analysis. Both the Greek text and its Arabic version have been semiautomatically analyzed, using the tools developed within the project: each Greek and Arabic word has been labelled with its morphological information and the lemma.
6) Indexing and querying engine. A specific indexing engine has been developed able to list in alphabetical order all the Greek wordforms and lemmas. G2A Web App provides both basic and combined search. The basic search is intended to be performed on one text at a time, while the combined search is performed on both texts simultaneously.
The application architecture provides the use of components some of which are active for Greek into Arabic, others are ready to be used for future research projects. The value of the system is, therefore, its great flexibility and interoperability.
Graeco-Arabic Lexicography
The Greek and Arabic Lexicon (http://telota.bbaw.de/glossga/) in the form of a reference dictionary of the Arabic translations of Greek philosophical and scientific source texts has been the starting point and material basis of research activities at Ruhr-University of Bochum, in the workgroup of Prof. Gerhard Endress, assisted by Dr. Rüdiger Arnzen (mainly responsible for the Greek and Arabic Dictionary), and Dr. Yury N. Arzhanov (mainly responsible for the Glossarium Graeco-Arabicum database). With the support of the ERC Advanced Grant, and working as an affiliate of the main editorial and philological project of the Principal Investigator at UniPisa, this has been continued following the procedures established in the published parts, up to the end of the Arabic letter bâ', i. e. including all lexical materials lemmatized before the Glossarium Graeco-Arabicum online database was being set up. The Glossarium Graeco-Arabicum database has not only taken over where the printed publication left off, but is complementing the material presented here, and enhancing accessibility to a wider range of lexical, grammatical and stylistic features of scientific language. In the first instance, the database was created for presenting the basic data compiled in our card-files and additional material collected by the Bochum team. For this purpose, the full card file was digitalized, and for each entry, searchable data were entered from the file-cards, after being checked and supplemented on the basis of the original source texts. The software platform was developed as an online mySQL database, and continues to be hosted, by the TELOTA centre of the Berlin-Brandenburg Academy of Sciences, URL: http://telota.bbaw.de/glossga. Since the start of our ERC Advanced Grant project, the principal development work — setting up the database, programming the user interface and creating and enhancing sophisticated query, input and output facilities — was achieved by Torsten Roeder of the Berlin Academy, while the data input was managed by the Bochum team. The database is meant to open up the full range of the terminology, and more generally, the grammar and style of scientific Arabic: registering the entire vocabulary of altogether 70 scientific and philosophical texts dating from the 8th to the 10th centuries, a total of 95.500 records, and thus documenting the language of scientific categorization and demonstrative discourse created by the translators and their Arabic readers. The task we set ourselves for this project of making our materials, compiled in the earlier as well as in the more recent phase of our project, accessible in a searchable database, has been completed under the ægis of the ERC grant. In addition, (a) further texts, edited recently and parsed in view of the Greek and Arabic correspondences, have been included, and (b) for many of the sources, the basic lexemes were supplemented by the relevant context and by comments regarding the translators’ techniques and transposition. All fields of the datasets are searchable through a database search engine. During the final period, our work has focused on the development of appropriate tools and techniques for visualization of the results of the search queries and on the creation of a user friendly export function which provides the possibility to download or transfer via e-mail the results of any search query, and the output of large subsets of data in the shape of lexical entries to be used for the continuation of the Greek and Arabic Lexicon.
Successful cooperation with the informaticians of Associate Participant ILC-CNR, using the G2A Web application developed at the ILC-CNR, Pisa (providing parallel Greek and Arabic texts of the Theology of Aristotle and other Arabic Plotiniana in structured pericopes, and searchable for Greek and Arabic word pairs), and our partners at the Berlin-Brandenburg Academy of Sciences, have provided first results in creating import facilities into the TELOTA database, in view of including lexical datasets of the Theologia Aristotelis (and further texts, more of which are becoming available on the G2A platform).