Bild

ALMA (see below)

The main objective of the project is to investigate the interaction between language and knowledgein the Romance-speaking cultural sphere of the High and Late Middle Ages, where, between approximately 1100 and 1500, new vernacular knowledge networks developed, increasingly supplanting the previously dominant Latin ones. The project traces this process in the medieval varieties of Gallo-Romanic—French, Occitan, and Gascon—and of Italo-Romanic.

The increasing use of writing in numerous areas of life and the emergence of new educational institutions in the growing medieval cities, for example, fostered the emergence of new groups of experts, new ways of transmitting and generating knowledge, and thus the expansion of the vernaculars as a medium of written communication: While writing in Western Europe was almost exclusively in Latin until around 1150, there was now an increasing shift toward the vernacular languages. This linguistic shift was not complete, as Latin would continue to dominate the vernacular varieties as the language of administration, education, and scholarship for a very long time. The new vernacular forms of knowledge communication are more regional, more fragmented, and in some cases only loosely interconnected. They do not always contribute to the development of independent, innovative bodies of knowledge. They often remain dependent on Latin models and are therefore frequently imitative. Nevertheless, vernacular knowledge communication is an important factor in the process that will further develop—or even erode—the traditional, canonized, and authoritatively administered bodies of knowledge of the Latin tradition.
The language shift is not abrupt either, but takes place in several stages, and individual regions begin to write in the vernacular at different times and with varying intensity. At first glance, the transition from (Middle) Latin to the vernaculars may appear to be a superficial phenomenon—a simple shift between mutually interchangeable linguistic forms. From a socio-historical, cultural, and genre-historical perspective, however, such an assessment is incorrect, since Latin in the Middle Ages was linked to specific user groups and specific contexts of use that do not correspond to those of the vernaculars. When the vernacular varieties also become means of conveying knowledge in writing, the networks of knowledge undergo a transformation.

The conceptual and associated terminological expansion of the vernacular languages—two central levels of analysis in the project—serve both as an indication of and a catalyst for social change, in which a renewal of the intellectual landscape on the path to the Renaissance manifests itself. The languages of knowledge (and scholarship) not only bring an important component of Europe’s cultural heritage into focus. The Romance languages are also significant vehicles of cultural exchange, which in the Middle Ages established European identity as a knowledge society.

Methodology

The project deliberately interweaves methods from lexicology and lexicography, textual philology, and the history of knowledge and science with approaches from the digital humanities, the Semantic Web, and ontology engineering.

1. Text Editions and Corpus: Corpus Linguistic Analysis

Using digital methods, the rich Romance tradition in the field of knowledge communication is being cataloged and digitally processed into two exemplary text corpora in the domains of medicine and law for historically oriented research. These cross-linguistic text corpora cover a significant cultural area of medieval Romance-speaking Europe and serve as the basis for reconstructing the central concepts and conceptual networks of these knowledge domains. The cross-linguistic approach to creating domain-specific parallel corpora represents a long-standing need in previous research.

The corpora consist of digital text editions of previously unprocessed manuscripts newly created by ALMA, as well as retro-digitized editions originally published in book form. All texts are converted into a uniform XML/TEI format, tokenized, annotated, and lemmatized. The corpus texts are analyzed using quantitative-empirical methods of corpus linguistics (absolute and relative frequency analyses, co-occurrence analysis) to generate initial insights into relevant technical terms and key concepts.

“Medicine” Domain

The field of medicine plays an important role, as it reflects contact with the Greco-Arabic tradition, which introduces new groups of experts to the written communication of knowledge.
An example of one of the most significant texts in the medical corpus is the vernacular translations of Chirurgia magna [according to the text, 1363; oldest manuscript with a French translation, 2nd third of the 15th century] of Gui de Chauliac (ca. 1300–1368), a physician active in 14th-century France who served as personal physician to three popes. Contrary to what its name might suggest, this work’s scientific significance is not limited to the field of surgery.
Instead, the Chirurgia magna represents a high point of medieval medicine based on ancient knowledge, heralding the transition to modern medicine: By intertwining surgery—which at that time was regarded as a field separate from medicine—with medicine, Gui de Chauliac stands at the dawn of a more comprehensive concept of medicine characteristic of the modern era. At the same time, the Chirurgia magna is a compilation representing the complete medical and surgical knowledge of its time. Also significant here is the discussion of the dissection of human cadavers as a medical teaching method, which many of Guy de Chauliac’s contemporaries did not employ. Thanks to other documented medical innovations—such as the classification of pneumonic and bubonic plague, or the refinement of existing methods for healing and treating bone fractures—the Chirurgia magna remained a popular textbook and handbook well into the 17th century, serving as a direct bridge between medieval and modern medicine.

“Law” Domain

The “Law” domain is particularly relevant due to its centuries-long tradition of conveying legal terms and concepts in both vernacular and Latin.
Two of the key texts here are the Assises de Jérusalem [before 1266] by Jean d’Ibelin and the anonymous Grand coutumier de Normandie [late 13th century]. The Assises de Jérusalem is a compilation of customary law in the Latin-Christian Kingdom of Jerusalem from the mid-thirteenth century. Its author was the politically highly influential Jean d’Ibelin, Count of Jaffa and Ashkelon, one of the most prominent nobles of the Latin Middle East of his time. The work is of exceptional importance for several reasons:
First, it is by far the most comprehensive of several contemporary legal texts from its cultural sphere. Second, the Assises de Jérusalem are considered one of the most comprehensive—if not the most comprehensive—medieval treatises on Christian-European feudal law. Furthermore, it addresses formal and substantive issues concerning the royal court of justice, the constitution of the kingdom, and its governing structure. The text is also particularly relevant to the ALMA project because it was written in French without a direct Latin source and, moreover, provides insight into the impressive reach of new vernacular knowledge networks extending all the way to the Middle East.

2. Linguistic Analysis

The results of the corpus-linguistic, statistical analysis form the basis for the competence-linguistic elaboration of the vocabulary using the qualitative-hermeneutic approach of historical linguistics. Based on the relevant concepts and terms identified in the texts for analysis, lexical-semantic studies are developed from an onomasiological, concept-based perspective. These studies analyze the history of concepts (histoire du concept) as well as their connection to the history of words (histoire du mot), across languages and in the context of the emergence of new knowledge networks.

These studies also incorporate findings from related disciplines such as astronomy, botany, and pharmacology. Furthermore, they compare vernacular texts with the original Medieval Latin texts as well as with previously analyzed text corpora, particularly for Medieval Spanish and Catalan. The lexicological work is further validated by drawing on the lexicographical materials compiled in historical dictionaries: specifically, the lexicography of Gallo- and Italo-Romance languages, particularly the dictionaries Dictionnaire onomasiologique de l’ancien gascon - DAG (HAdW), Dictionnaire étymologique de l’ancien français - DEAF (HAdW), Dictionnaire de l’occitan medieval - DOM (BAdW) and Lessico etimologico italiano - LEI (AdW Mainz), French Etymological Dictionary - FEW and Dictionnaire du moyen francais - DMF, as well as Iberoromania and (Medieval) Latin.

3. Cross-disciplinary Analyses of the Development of Knowledge Networks

Building on these findings from lexical-semantic studies, separate, cross-disciplinary investigations examine and describe the development of vernacular knowledge networks and language expansion. Potential examples include contexts such as “plague,” “anatomy,” “donation practices,” etc.

4. Digital Humanities, the Semantic Web, and Ontology Engineering

One of the project’s innovative aspects lies in combining a language-data-based and cross-linguistic approach with the tools of ontology engineering: The primary goal is to translate the historical-philological research findings identified in the project into domain-specific, historicized ontologies (in Web Ontology Language – OWL). These further develop the conceptual framework of the examined knowledge domains into structured knowledge. As historicized ontologies, they do justice to the specific nature of medieval explanatory patterns: They make it possible to reveal the breaks in continuity between medieval and modern systems of knowledge and scholarship, rather than obscuring them through anachronistic classifications. Their cross-linguistic and subject-oriented approach makes them hubs for the content-driven linking of research data. This expands the ontologies’ user base far beyond the specific field of Romance studies to all disciplines engaged in historical research.

Linked Open Data

The corpus texts, the lexical-semantic studies, and a critical research bibliography (ALMABibl) are converted into the Resource Description Framework (RDF) and are thus available for ontology-based access. To achieve interoperability, ALMA employs ontologies, vocabularies, and repositories established as (de facto) standards for the discipline and its historical linguistic resources when modeling as Linked Open Data:

• OntoLex-Lemon, Ontologies of Linguistic Annotation – OLiA for the description of linguistic terminology, Model for Language Annotation - MoLA for language description, etc.,
• as well as those for place names (Geonames, Getty Thesaurus of Geographic Names – TGN), personal names (Virtual International Authority File – VIAF; Gemeinsame Normdatei – GND) etc.

The project will also utilize Linked Open Data modeling and historical domain ontologies for the innovative reuse of relevant lexicographic resources from the dictionaries of the DEAF, DOM, and LEI Academies (and, where applicable, DAG).

Publications

The findings generated by ALMA will be made available to the public through various types of publications: syntheses of results from specific research areas will be published in the form of monographs, conference proceedings, or articles in scholarly journals. A large portion of the research results is published on the project’s open-access online portal: the corpus texts, the lexical-semantic studies (closely linked to the online versions of the DAG, DEAF, DOM, and LEI, as well as other dictionaries), the ALMABibl research bibliography, and the domain ontologies.
The Linked Open Data resources are made queryable via a triplestore.

Copyright:
1. Visualization of the TextFactory pipeline, ALMA.
2. “Trotula,” holding a urine bottle, manuscript from the Wellcome Collection, London, MS. 544 [early 14th century]. Miscellanea medica XVIII. CC BY 2.0
3. “Trotula de Salerno, De passionibus mulierum,” manuscript from Salerno [1544].” Public domain.
4. Illustration from a text in eScriptorium.
5. RDF data.