The Monumenta Germaniae Historica (MGH)

In 1890 a German Imperial court put an end to a dispute which had erupted a few years previously between the city of Lübeck and the Grand Duchies of Mecklenburg-Schwerin and Mecklenburg-Strelitz – a question of property rights over a nearby river. Determinant in reaching a verdict was the interpretation of a diploma issued by Emperor Frederick I in 1188, about 700 years earlier. It was one of the last cases in Continental Europe in which a judicial proceeding was settled through recourse to a medieval document. Since the 1810s, in the aftermath of the French Revolution and under the impulse of Napoleonic reforms, the old judiciary system had given way throughout the Continent to the new vogue for legal codification enshrined in the Code Napoléon. Medieval documents were no longer a subject of interpretation for jurists; from then on, they became the preserve of historians.

Jurists had not been the only group interested for practical reasons in the preservation of documents. Charters and diplomas had been kept for centuries, carefully preserved in the archives of monasteries, and not infrequently carefully forged. Many of these foundations were swept away by the turmoil blowing through the Continent in the nineteenth century: between 1802 and 1867, the number of suppressed monasteries in Italy and Germany reached an impressive total. With the suppression of monasteries, their archives and libraries were confiscated: ancient manuscripts and charters found their way into city-based state archives and university libraries. On the one hand, legal codification was depriving documents of their practical validity, and on the other, their migration into the fold of the bureaucratic modern state secularized them.

Through the combined action of these two factors – the loss of currency of documents and their acquisition by secular institutions striving for historical self-affirmation and prestige – we can trace the birth of modern historical science in the nineteenth century, and, with it, of the Monumenta Germaniae Historica.

The diplomatic career of Baron Karl vom Stein, minister of the Prussian State 1804–7 and a prominent reformer, was affected by Napoleon just as much as legal systems and monasteries. In 1808, suspected of conspiratorial activity, he was declared an enemy of France by the Emperor for his staunch nationalism. A year later he was forced into exile. His career marks a decisive turn in the history of German nationalism. Belief in the cultural, linguistic and ethnic unity of the German nation and in its ancient origins had been widespread among writers and thinkers connected to the Romantic movement and indeed earlier. Key political figures in Germany now began appreciating the power of these ideas as political rallying cries during the French occupation and the Napoleonic wars.

As part of the struggle against France, Stein was instrumental in promoting a state-sponsored nationalism founded on widespread education among the bourgeois sectors of society. Germany came to be defined as a territorial political entity, striving towards unification and seeking expression in the modern nation-state. After the Congress of Vienna in 1814–15 Stein retired to private life and set to work on the idea that the establishment of a German nation had to rest upon the foundations of its medieval past. In 1819 he constituted a society, Die Gesellschaft für Deutschlands ältere Geschichtskunde, explicitly devoted to the task. By 1823 he had secured the collaboration of two young scholars, Georg Heinrich Pertz and Johann Friedrich Böhmer, who were to remain Editor and Secretary respectively of the society for half a century.

In 1824 Leopold von Ranke, a history teacher in a provincial secondary school, published a history of the Italian and German peoples in the early modern period (Geschichten der romanischen und germanischen Völker von 1494 bis 1535, whose preface explained how and why the narrative had been built only from original sources. It appeared jointly with a methodological treatise titled Zur Kritik neuerer Geschichtsschreiber, which concluded with a chapter on ‘What is still to be done’. In it Ranke called for digging into archives and unpublished primary sources, documents which were the only guide that enabled the researcher to penetrate the real intentions of historical actors. The medieval and early modern history of Germany, he argued, had to be refashioned upon a documentary foundation. Finally, he sketched a profile of the prospective historian for the years to come. He wrote:

'What we need is a man equipped with reasonable knowledge, lavish letters of recommendation and good health, who would traverse Germany in all directions in order to hunt down the remains of this world, which is half sunken and yet so close to us. We pursue unknown grasses into the deserts of Libya: how can the life of our forefathers, in our own country, not deserve the same zeal?' (Quoted by A. Grafton, The Footnote: A Curious History, Cambridge MA, 1997, p. 49.)

This was a kind of cover letter for the career Ranke had in mind for himself. It was also inspired by the first successes of Pertz. Not far from the stacks where the IHR keeps its entire collection of the MGH, the library also holds seven of the original twelve volumes of the Archiv der Gesellschaft für Ältere Deutsche Geschichtskunde (1819–74). In it are documented, in the words of Walter Pohl, the efforts of ‘dozens of German scholars [who] criss-crossed Europe in a quest for medieval manuscripts; their archival networks evolved parallel to, sometimes even anterior to the expansion of railways on the continent’ (W. Pohl, ‘History in Fragments: Montecassino’s Politics of Memory’, Early Medieval Europe, 10, 2001, pp. 343–74 at 343). After Stein’s death (1831), and the securing of the society’s financial position, Pertz was in a position, from the 1830s and 1840s, to enrol a group of young researchers and dispatch them throughout the Continent in search of sources – or, as Stein would have put it, memorials of the German past. Although Ranke never had any direct involvement in the MGH project, ‘Pertz’s boys’, the Monumentisten, perfectly matched his ideal of ‘what was still to be done’.

The ideological background framing the MGH project at its inception and throughout its initial development, therefore, rested on two pillars: political nationalism and the modern bureaucratic state. They are the key to understanding two essential aspects of the MGH as it still appears today.

In the first place, this ideological underpinning shaped the textual canon which was the prime objective of the editorial enterprise. From the beginning, the MGH editors’ purported task was to publish texts pertaining to German history from the disappearance of the Roman Empire to the invention of printing – that is, across the entire millennium from 500 to 1500. The problem was to define what counted as German history. The ‘Germany’ of which these texts were supposed to be monuments had to be invented (to borrow from Benedict Anderson) – and the MGH itself proved a formidable element in this national mythopoesis. The mentality and methodology of its promoters created monuments and, in so doing, contributed to the creation of ‘Germany’. As Patrick Geary has put it:

'These editors claimed as these monuments all texts written in or about regions in which Germanic-speaking peoples had settled or ruled. First, the Monumenta editors claimed all those regions that had been part of the “Holy Roman Empire of the German Nation,” from the south of Italy to the Baltic. In addition, they annexed the whole of Frankish history, including the chronicles and acts of Merovingian and Carolingian kings in the regions of Gaul that are today France and Belgium. They absorbed the laws of the Visigoths, the Burgundians, and the Lombards, Germanic-speaking groups that had settled into what is today Italy and the Rhone valley. They appropriated the county of Flanders and all of the Netherlands east of the Schelde because these areas were settled by the Germanic-speaking Frisians. … By defining the corpus of what was German history, the Monumenta set the parameters within which Germany would search for its past. The Goths, the Franks, the Burgundians, the Vandals, and other early “peoples” were identified by an uninterrupted history, which preceded the establishment of the medieval Holy Roman Empire and which reached through the nineteenth century.' (P. J. Geary, The Myth of Nations: The Medieval Origins of Europe, Princeton and Oxford, 2002, pp. 28–9.)

In the second place, the model of the bureaucratic state informed the classification of sources – a typological classification that, with some additions, still forms the essential backbone of the MGH. Pertz himself, in 1824, envisioned the present system of source classification by establishing five canonical divisions:

• Scriptores: historiography, including Mommsen’s edition of the Liber pontificalis (a collection of papal biographies) and a complete subseries of works linked to the Investiture Contest of the eleventh century
• Leges: law codes, also including church councils (Concilia), a complete subseries of episcopal capitularies (Capitula episcoporum), and Merovingian and Carolingian formularies (Formulae Merowingici et Karolini aevi)
• Diplomata: charters, chiefly of the Carolingian period
• Epistolae: letters, including a selection of papal missives from the thirteenth century
• Antiquitates: poetry and necrologies

Over time, these five major sections have been subdivided into thirty-three subseries. In more recent years, the scheme has been expanded to include three further major sections:

• Quellen zur Geistesgeschichte: sources for intellectual history, from 1955
• Hebräische texte: devoted to Hebrew text editions, started in 2005
• Reiseberichte des Mittelalters: devoted to medieval travel literature, inaugurated in 2020 with the publication of Hieronymus Münzer’s Itinerarium

Initially sketched by Pertz, the canonical system came to full fruition only in the succeeding period, the so-called Waitz era, when the different sections were entrusted to individual directors responsible for each division as heads of department. It was also in this period that the influence of Ranke became most prominent. Ranke’s school produced most of the key figures of the Waitz era, starting with Waitz himself, a pupil of Ranke and the author of a massive German Constitutional History

While practical and intellectually powerful, the MGH’s structure has significant drawbacks. One example will suffice to clarify the point: in MGH SRL pp. 1–6 Waitz published a short mid-seventh-century text titled Origo gentis Langobardorum – a king-list with commentary, preceded by an account of how the Lombards had migrated from Scandinavia and acquired their ethnonym in battle through the intervention of the gods. In all three manuscripts which transmit it, the Origo is copied alongside the Lombard laws, notably the near-contemporary Edict of Rothari. In the early ninth century Paul the Deacon considered the Origo and the Edict interdependent and relied on both for his classic history of the Lombards.  Rothari’s Edict, however, was published in a separate volume of the MGH series, in the Leges section, MGH LL 4 (ed. Bluhme).

Later interpreters have established four crucial points about these two texts. First, the Origo is not, as the MGH editors believed, an historical account of the Lombards’ migration drawn directly from their immutable ‘Germanic’ oral lore, but rather a foundation myth promoting their ethnic identity as envisaged at the later time of composition. Second, Rothari’s Edict is not a faithful representation of the spirit of ancestral ‘Germanic law’ as it was practiced and enacted by the Lombard ‘people’ from ancient times, but an act of legitimacy promoted by a king and military élite to foster the consolidation of their rule over former Roman provinces. Third, the interdependence of ‘law’ (lex) and ‘origins’ (origo) was instrumental in shaping and enacting the construction of Lombard identity as an ethnic group and a polity. An account of the people’s ethnic origin legitimized the ruling classes’ legislative power, while the origo narrative was canonized through the legal context framing it. Fourth, this very interdependence not only points to the fact that the producers and readers of these texts considered the boundaries between the two genres of lex and origo as fuzzy; it is also the key to their practical value and social significance. The MGH editors’ decision to separate the two texts and allocate them to autonomous realms of discourse – historiography and law – was intended to provide evidence, respectively, for the history of the Lombards and the prescriptive functioning of their legal system. It not only distorts how historical actors perceived those two domains, but also potentially misinterprets their function and meaning in the society which produced them.

The volumes of the Archiv der Gesellschaft mentioned above are an eloquent testimony to the methodological principles that drove the first MGH editors. Research reports from close enquiry into the sources (Quellenforschung) were entrusted, and confined, to monographs and articles rather than the editions. Each edition was bent on providing a critical text as close as possible to ‘the original’, whose value in turn depended on its chronological proximity to the events it purportedly described. The result was often to obscure the process of transmission, and the roles and motivations of the social actors who preserved the texts, constantly modifying and improving upon them over time.

The scholarly methodology of the early MGH combined the textual criticism perfected by Classical philologists in the late eighteenth century with the historical source criticism pioneered by Ranke. Ranke’s method, which became increasingly sophisticated under his pupils, provided subtle means to detect forgeries and interpolations, to date texts, and to edit them by establishing the best possible approximation to the hypothetical Urtext from which the surviving copies had stemmed. Scholars rarely questioned the transparency of texts as accurate reflections of social reality, or the motives which stood behind their production and preservation.

The canon established by the MGH editors therefore projected back onto the Middle Ages a modern and ideologically charged vision of Germany. Hence the structure of society and its institutions was presupposed by the scholarly model of source-typology and by the methodology of the editors in handling the texts. The whole enterprise ran the risk of imposing its own terms of reference on the medieval past: the idea of a socio-political field partitioned into institutional spheres whose function was regulated in autonomous written domains (histories, laws, charters, and so on) endowed with an intrinsic ‘rationality’. To quote Pohl once again: ‘The MGH … established an archive for early medieval studies in the double sense of the word arché: starting point and authority. It lent credibility to a far-reaching reconstruction of early medieval history.’ (Pohl, ‘History in Fragments’, p. 344.)

This is a brief account of the consequences for the MGH and its users of its insertion into a digital environment. This takes the form of a brief guide to the digital materials currently available, while also providing some scattered notes on the possible impact of these developments in changing the function of the MGH’s ‘archive’ two centuries after its inception.

The digital output of the MGH can be divided into four core branches which respond to different demands and functions, and might therefore appeal to different kinds of scholars and students.

Elektronische Monumenta (eMGH) was the first digitization project of the MGH corpus: it started in 1994 in partnership with Brepols Publishers. From 1996 onwards, eMGH published a series of 6 CD-ROMS, the last of which was released in 2006. The product, based on the CETEDOC software, and intended chiefly as a philological tool, is now accessible through Brepolis, the digital platform of Brepols Publishers. Within the IHR, it can be accessed via the library catalogue. The eMGH is updated annually, with precedence given to narrative texts.

Texts can be approached in three main ways:

1. Two Search Screens (Quick and Advanced) permit searches based on word-forms or groups of word-forms. Search-queries can be conducted by using wildcards as well as proximity and Boolean operators. Advanced searches can be filtered according to five criteria: Collection, Author, Text, Century, and Period. The last comprises three sub-divisions: the Patristic Age (Aetas Patrum c. 200–735; the Middle Ages (Medii Aeui Scriptores, 736–1500); the Modern (or Neo-Latin) Age (Recentior Latinitas, 1500–1965).
2. The Table of Contents allows researchers to access and display specified passages from individual texts starting from explicit bibliographic references.
3. Distribution of Word-Forms. The tool allows users to explore the distribution of occurrences and the position of single word-forms (or lists of word-forms corresponding to queries containing wildcards) in increasingly more specific portions of the database: across the entire database; within the three MGH Periods described above; in the works of a single author; and in a single work. Results pertaining to single works are displayed in the form of a traditional concordance: the keyword is embedded in its context, which never exceeds a sentence.

The eMGH database is integrated in the Brepolis Latin database, which includes the Library of Latin Texts, the Aristoteles Latinus Database, and the Archive of Celtic-Latin Literature — all employing the same principles and approach for text-analysis and search interface. The Cross Database Search Tool allows users to carry out searches simultaneously in the four full-text corpora (over 9,000 texts in all).

The Digital MGH (dMGH) gives unrestricted access to all the MGH volumes with a moving-wall embargo of two years. A previous version of dMGH made all these editions available as scanned images and electronic full-texts in HTML, allowing the user to switch between the two visualization options, as well as to perform additional search functions. As at November 2022, the current dMGH Beta version only allows searches of the entire MGH corpus and access to scanned digital images in PDF format. The images can be downloaded as single or multiple pages, or even an entire volume.

The canonical status accorded to the printed editions is evident: the digital version provides an image of the printed page; the website’s main menu presents the series and volumes of the printed version; the web interface is chiefly a tool allowing users to flip through the pages, browse a table of contents or by page number, and zoom in and out of the images; the downloadable PDFs contain only images, rather than full texts. In sum, rather than a digital edition proper, dMGH is a digitized version of the MGH — or, as Patrick Sahle has put it, ‘a scholarly digital library’.

One of the most relevant features of dMGH is its permalink strategy. To enable citations of online reproductions, links to exact pages are available, as explained in the guide to Linking and Citing dMGH. The digital image viewer of dMGH displays the digitized pages of the original, and the readily understandable permanent URL reproduces closely the standard abbreviations in which MGH editions are cited. The only relevant limitation is that the URL’s syntax cannot be used to address page ranges. For example, the established abbreviated reference MGH SS 32 p. 98, referring to MGH Scriptores (in folio), vol. 32, p. 98 — Holder-Egger’s 1905–13 classic edition of Salimbene de Adam’s Cronica — can be written as follows: https://www.mgh.de/dmgh/resolving/SS_32_S._98. In that form, the fixed address which all URL links display (https://www.mgh.de/dmgh/resolving)  is followed by the series abbreviation (SS), volume number (32), and page (‘S.’ stands for Seite (page) number).

The recent development of openMGH will prove particularly valuable to linguists and philologists. It provides a more integrated approach to the possibilities offered by digital technologies, especially in the field of text-mining for historical semantics. This initiative, achieved through the joint collaboration of the MGH and Munich’s Bayerische Staatsbibliothek, aims to provide lemmatized texts of all MGH volumes, encoded as TEI-XML files. The data currently available — in particular the lemmatization and morphological annotations — were initially generated under the aegis of Frankfurt University’s Computational Historical Semantics project (the original website is no longer maintained).

The project’s chief aim was to apply to medieval texts the corpus-linguistic methods developed in modern historical research, in order to explore the transmission of political ideas and concepts in the Middle Ages. A qualitative approach to the history of ideas as constituted through the communicative activities of social actors was integrated through the quantitative analysis of data made possible by applying computational analysis to large textual corpora. As the project focused on the process though which meanings are generated by and in communicative practices, specific tools for the analysis of word frequency, patterns of occurrence and co-occurrence of relevant terms, expressions, syntactic correlations and argumentative interrelations were developed for the purpose, notably the corpus management system infrastructure Historical Semantics Corpus Management (HSCM). The generated data are now stored and managed in the Latin Text Archive (LTA) platform (access the beta version).

The LTA is, in the first place, a vast historical reference corpus, offering open-access digital editions of texts in their most trusted versions. The editions are not meant to replace critical editions as they only offer the main reading of the texts, without critical apparatus. In its current form, the LTA features a sizeable repository of lemmatized texts covering roughly the same timeframe as the MGH (400–1500), provided by three core institutions: the MGH, the Institut de recherche et d’histoire des texts (IRHT) and the Corpus Corporum. Users can approach the data in two distinctive ways.

The first is from a corpus-based perspective. Texts can be analysed in predefined reference corpora, either genre-based (narrative texts, legal texts, charters, letters, and theological texts) or diachronic (where texts are allocated to time slots of quarter-centuries). Additionally, users can constitute their own corpora, browsing texts by author or according to a model of classification operating on two levels: functional (called ‘text-type class’, e.g. administrative, educational; at present, twenty classes have been identified) and typological (called ‘text type’, following their actual form, e.g. ‘tax list’, ‘encyclopedia’). Corpus analysis can be carried out either using the historical reference corpora directly integrated in the platform or external computational tools such as Voyant Tools. Individual texts and corpora can be visualized as plain text, downloaded as TEI-XML, or visualized and/or downloaded as lemmatized (tagged) texts.

Second, the LTA hosts the Frankfurt Latin Lexicon (FLL) a morphological lexicon of Medieval Latin and the world’s largest free Latin word-form lexicon. FLL has been developed since 2009 to support the automatic lemmatization of medieval Latin texts. Entries are structured according to the so-called Wiktionary model, a four-tiered model comprising superlemmata (lemmata with normalized spellings); lemmata (preserving standard and non-standard spellings); syntactic words (inflected forms under these lemmata); and word-forms. The introduction of superlemmata, which makes possible the preservation of non-normalized spellings at the lemma level, is particularly important for the analysis of sociolinguistic variation in medieval Latin, and a key feature in measuring its diachronic development. As new texts are uploaded into the database, tags are generated by automatic lemmatization with the help of the Text-Technology Lab Latin Tagger (TTLab Tagger); after lemmatization they are checked by Latin philologists: a crowdsourcing approach but limited to a selected body of experts. The results of are incorporated in the tagging of the LTA text repositories, where lemmatized texts are presented in a colour code differentiating levels of lemmatization.

One of the most interesting developments which the digital turn has opened up for the MGH is the research materials gathered as Digital Resources. Organized under the usual MGH Sections, they comprise chiefly pre-print versions of projected editions and auxiliary materials. On the one hand, these allow researchers to access the data without having to wait for the printed edition. On the other, by publishing working texts, the editors can take advantage of ‘social editing’ by sharing their work and collecting feedback from the wider scholarly community.

One notable example among many is an edition of Hugo of Flavigny’s Chronicle in the Scriptores section. It reproduces Georg Pertz’s original edition of 1848 (MGH Scriptores in folio 8, S. 288–503), integrated with the corrections proposed by Mathias Lawo, Studien zu Hugo von Flavigny, Schriften der MGH 61, 2010 (see esp. the Appendix). In his study, Lawo has established that the manuscripts which preserve the chronicle were Hugo’s autograph working copies, which therefore represent an invaluable source for reconstructing Hugo’s own working habits. Pertz’s editorial methods — apparently due to lack of time and resources, and in line with the then dominant paradigm of philological reconstruction — ironed out the many additions, corrections, interruptions, and second-thought modifications which characterize Hugo’s original work. While a new edition is currently in preparation, the MGH’s Digital Resources allows researchers to freely download a PDF file in which Lawo’s highlighted suggestions for correction are incorporated into Pertz’s original text. The latter is reproduced page by page and line by line, in order to maintain citability.

Databases with search functions are available for the following

Leges

• the Clavis Canonum a selection of Canon Law collections prior to 1140, MGH Hilfsmittel 21)

Diplomatica

• a database with addenda to older volumes of the Diplomata

Epistolae

• Epistolae variorum 877–923 (MGH Epistolae 9)
• Epistolarium of Pseudo-Marinus of Eboli
• The Lombard Letter Collection

Finally, in 2019 the MGH published its first digital scholarly edition: Ulrich Richental’s Die Chronik des Konzils von Konstanz, ed. by Th. M. Buck. Buck is the editor of the last published edition, from 2010, which was the first to take into account the entire manuscript tradition, but which still presented a single version as the main text. Since the Chronik had soon become a fluid repository of urban collective memory on the Council of Konstanz, manuscripts and prints transmit the text in significantly different textual versions. The digital format has allowed Buck to present three parallel textual versions, each provided with a historical commentary. As explained by the editors: 'MGH digital editions are integrated in the MGH edition programme as fully equivalent to the printed publications. In cases where digital format offers clear advantages compared with printed publication, digital editions will be preferred.'