Background | Aims and objectives | Transcription policy | Additional features | Navigation and collation | Other witnesses | Title | Future developments | Feedback |



What is Partonopeus de Blois?

Partonopeus de Blois is a medieval French verse romance, composed in the last third of the twelfth century. It tells of how the young hero, Partonopeus, is transported to a mysterious city where he encounters the heroine Melior. He becomes her lover without ever seeing her, but eventually breaks the taboo on bringing a light into her bedchamber and is banished by her. After various adventures, he is forgiven, wins her hand in marriage at a three-day tournament, and becomes emperor of Byzantium. A continuation then tells the story of the hero’s former squire, Anselot, and of an unsuccessful invasion by Partonopeus’s former love-rival, the sultan of Persia.

Partonopeus seems to have been something of a medieval best-seller: it is preserved in seven complete or near-complete manuscripts (hereafter abbreviated as MSS), and fragments from a further three copies (there is evidence that at least another two MSS were also in circulation). It had a significant influence on other twelfth- and early thirteenth-century French writers. There are clear allusions to it in Hue de Rotelande’s Ipomedon, in Aymon de Varennes’s Florimont and Renaut de Bâgé’s Le Bel Inconnu, amongst others. The story also proved popular across medieval Europe: adaptations appeared in Middle High German, Old Norse, Middle Dutch, medieval Spanish and Catalan, and in medieval Italian. The Partonopeus story was revived in the nineteenth century in the form of Massenet’s opera Esclarmonde.

Why an electronic edition?

For all its popularity, Partonopeus has not proved an easy text for modern critics to study, largely because of the complicated relationships between the existing MSS. In two MSS, the story ends with the marriage of Partonopeus and Melior, but these two versions are far from identical: the events that take place between the end of the three-day tournament and the wedding differ significantly from one to the other. One of these two MSS also has a rather different account of the tournament itself, but we cannot be sure just how far it diverged from the version found in all the other MSS, because some folios (pages) are missing. The other five MSS contain part or all of the continuation of the story, but each of them stops at a different point. In the continuation and the main body of the poem, there is evidence that some scribes were abbreviating the original text, while others were expanding it (and at least one was apparently doing both, in different places).

As a result, there are a number of important questions relating to Partonopeus that have still not been settled. What was the original version of the poem like, and which of the extant MSS is closest to it? Was the continuation written by the original poet, or by another writer (or writers)? When and why was the story continued? Why did scribes decide to add material to the poem or to leave sections of it out? Which version was known to each of the subsequent poets who drew on Partonopeus? There is controversy over when the original version was composed: one view is that it was first written around 1170, while another is that it dates from the mid 1180s. This question of dating is bound up with another, namely whether Partonopeus influenced the work of the most famous writer of Old French romance, Chrétien de Troyes, or the other way around.

It has been very difficult to address these questions using conventional printed editions of the poem, which present the (corrected) text of one ‘base’ MS, with the variants from the others listed at the bottom of each page, or in appendices, in the case of lengthy passages. From printed editions, it is hard to get a clear sense of how the story progresses in each MS, and of exactly how and how much the MSS differ from one another. The only other way to do a detailed comparative study up till now has been to work with printouts from microfilms of the original MSS, which poses all sorts of practical problems. Partonopeus is such a key work in the history of Old French literature, and raises so many important questions about how medieval romances were composed, altered and adapted, that it seemed to call out for an electronic edition, that would give scholars access to the full text of each of the MSS, and allow them to compare MSS easily and efficiently.

This is the background that led to the setting-up in 2001 of the Partonopeus de Blois Electronic Edition Project, with funding from the then AHRB (now AHRC) Resource Enhancement Scheme. The project team spent nearly four years transcribing and checking all the MSS, researching their history and characteristics, designing the user interface, and preparing the additional documentation. We were assisted in this by a wide variety of individuals and organisations, whose help we gratefully acknowledge (see Acknowledgements for details). We hope that the research tool that we have created will allow medievalists to explore Partonopeus in new ways and come up with answers to the questions it poses.

Top of page


Aims and objectives

The Partonopeus de Blois Electronic Edition project was set up with the aim of creating a research tool that would facilitate comparative literary and linguistic studies of the different extant versions of this important twelfth-century text. Its purpose was not to produce a critical edition, but to provide researchers with full transcriptions of all the manuscript witnesses, together with software that would allow for effective searching and comparing of text. A key objective was to present the text of each MS without the type of editorial intervention that might involve value judgements as to its reliability or its place within a conventional stemma or any other schema of MS relationships. For this reason, we have been careful to avoid designating any of the MSS as a base MS for purposes of comparison or collation, and have not made any textual emendations, however obvious they might seem.

Top of page


Transcription policy

At the same time, the project aimed to create a resource that would be easy to use, and would not be off-putting for users who had little or no experience of reading manuscripts. For this reason, it was decided not to produce a fully diplomatic transcription of each MS, complete with all the original abbreviations. Instead, the transcriptions are semi-diplomatic in nature:

Wherever uncertainty as to a reading remained, this is recorded in a note. An initial transcription of each MS was made from microfilm; this was then checked and corrected where necessary by at least one other member of the project team; uncertain readings were then checked against the original MS (except in the case of fragment C, where the expense of a visit to the Beinecke library could not be justified). We recognise that no transcription is perfect, however careful the checking process, and would welcome corrections from users. These will be incorporated into subsequent versions of the edition. Corrections can be notified to us at

Top of page


Additional features

We also recognise the value to users of certain types of information contained in MSS, which would not normally be included in a semi-diplomatic transcription. The majority of the Partonopeus MSS are too long, and some libraries' charges for digital images too high, to make inclusion of images of every folio a realistic option. It was therefore decided to mark up scribal features such as erasures and insertions, and material features such as stitching, holes and creases that affect the text, and to enable users to view them via an alternative display mode. Features of mise en page, decoration and hands are described in manuscript information pages, and can be viewed on images of a selection of representative folios.

Top of page


Navigation and collation

Various navigational aids have been included in the edition (see the User's Guide for details). Amongst these is an episode map which allows users who are familiar with Partonopeus to locate a passage of text by means of its narrative content. We are keen to stress that the episode map is a navigational device, not a definitive guide to the narrative structure of the romance. Its form was dictated by the need to divide the story into appropriately-sized subsections, rather than by a methodologically rigorous analysis of narrative content. To avoid possible confusion, we would therefore advise against using the labels of subsections in the episode map as a way of identifying parts of the narrative in publications prepared with the help of this edition. Users are encouraged to use the metalineation numbers as the principal form of reference.

Each line in each MS is allocated two ID numbers. One is a standard line number representing the location of the line within the sequence of lines in that MS. The other ID number is a metalineation number, corresponding to the location of the line within the sequence made up by aggregating all the MSS. The metalineation thus corresponds to a "virtual MS" that contains not only the text common to all the MSS, but also every "additional" passage that appears in one or more of the individual witnesses. This makes it possible to refer unambiguously to any line in the whole MS tradition without having to resort to the paraphrases necessary with printed editions based on one MS.

Collation is also based on the metalineation. The metaline number for each line reflects its relative position within the aggregate sequence rather than its content. If the lines of a couplet appear in reverse order in one MS in comparison to the others, the line that appears first in that MS has the same metaline number as the first line of the couplet elsewhere, even though its content matches the second line of the couplet in other MSS. The metaline sequence is defined on a mathematical basis : it reflects the order of lines in a simple majority of the MSS. If four MSS have lines in the order abcdef and three have the equivalent passage in the order abefcd, the metaline numbering follows the sequence abcdef. Within that sequence, the line identified as d may be worded quite differently in one MS from all the others, but this does not alter its metalineation number. That line may not appear at all in another MS; this MS will therefore have a blank line at the metaline in question. When two or more MSS are collated, the transcriptions are displayed in metaline order, with blanks where lines, couplets, or longer passages do not feature in certain MSS. This makes it easy for the user to identify and analyse potential lacunae, additions and abridgements. (Transcriptions can also be viewed without blanks, reflecting the arrangement of each individual MS).

As with any collation system, establishing the metalineation has involved a number of compromises. We have been guided principally by the anticipated needs of users and the purposes for which the edition is likely to be consulted. Defining "equivalent passages" has proved problematic in some instances. In the case of the Arsenal MS, 60 lines of Ernoul's speech that come immediately after the major lacuna differ in terms of detailed content from the central section of Ernoul's speech in all the other MSS. However, this passage has been treated as equivalent for the purposes of metalineation. It is quite short, occurs in exactly the same context and has the same narrative function : users will need to be able to compare it line by line with the other MSS. Likewise, the ending of the Sornegur episode in the Berne MS has been treated as equivalent to the ending of the same episode in the other MSS, despite differences in the events it narrates. However, the long final section of the Arsenal MS differs so radically from the ending of the main narrative in all the other MSS that line-by-line collation would serve little purpose. This passage has therefore been metalined separately. Users can still compare the MSS by viewing them simultaneously without invoking the metalineation or collation function.

Top of page


Other witnesses

This first version of the edition contains transcriptions of the seven complete or near-complete MSS of Partonopeus, plus the two known fragments of the text and one extract copied as a free-standing item in a collection of vernacular texts. It does not contain transcriptions of the passages from Partonopeus incorporated into the 13th-century romance Cristal et Clarie or into the version of the vernacular adaptation of the Disciplina clericalis preserved in BnF nouv. acq. fr. 7517. None of these passages appears to have been copied from any of the extant Partonopeus MSS, so there may be a case for including them in a subsequent version. However, it is not always clear, particularly in the case of the Disciplina, how far the extracts have been adapted to their new context; further work is needed to establish whether, and how, they could be successfully incorporated. In the meantime, users can find the full text of the extracts from Cristal et Clarie in Penny Eley, Catherine Hanley, Mario Longtin and Penny Simons, 'Cristal et Clarie and a lost manuscript of Partonopeus de Blois', Romania, 121 (2003), 329-347.

Top of page



It is normal practice for modern editors of Old French texts to use the oblique form of characters' names in the titles of eponymous romances. This is usually uncontroversial. In the case of our romance, however, this is an area where a choice has to be made, and where a value judgement is inescapable. The first printed edition of the romance, by Crapelet, had the form Partonopeus in its title, because this is by far the most common form of the oblique case in the Arsenal MS on which the edition was based (54 occurrences – including one written as Partonopexas against 9 of Partonopeu). The second editor, Gildea, used the form Partonopeu, on the basis of an analysis of nine instances of the hero's name appearing in the oblique case at the rhyme, which showed that the oblique form without an s predominated in this position in the seven complete or near-complete MSS. While Gildea's analysis may have provided some evidence as to the pronunciation of the name, it does not resolve the question of the graphical form to be adopted here. For an edition of all the MSS, a different approach is needed. One possibility would be to adopt the oblique form that is most common across all witnesses, irrespective of their dating. Alternatively, dating can be taken into account, on the premise that the sooner a MS was produced after the date of composition of the romance, the greater the probability of its preserving the graphical forms of proper names used by the original author. We have opted for this latter approach. Since the Arsenal MS is probably the oldest of the extant MSS, we have used the predominant graphy from that witness. (Users familiar with previous work by Simons and Eley will be aware that we do not subscribe to the view that the Arsenal MS represents an anomalous variant form of the text.) It is also worth noting that Partonopeus / Partonopex occurs as an oblique form in some of the other MSS, and that the rubric in BnF 91512 reads Apres vient de partonopex de bloys.

Top of page


Future developments

The remit of the project covered the preparation of the transcriptions and XML mark-up, plus development of a user interface allowing for consultation, collation and searching, and provision of information about the MSS. In the course of preparing the edition, we have identified additional features not included in the initial proposal, which we would like to incorporate at a later stage. These include a fully searchable index of proper names and the option of displaying transcriptions (or parts of transcriptions) with the original, non-regularised word division, for the benefit of users with a particular interest in this aspect of scribal practice.

Top of page



The editorial team would welcome feedback from users that will help us to improve content and functionality in future versions of the edition. We would also appreciate comments on what works well in this first version, so that we are not tempted to change features that are valued by our users. Comments can be sent to

Top of page