TranscriboQuest2025 Medieval Vernacular Religious Texts

Team Medieval Vernacular Religious Texts

**Description**

This dataset was created by a collaborative working group with the aim of transcribing medieval vernacular religious texts across a range of European languages. To reflect the linguistic expertise of the group members, the project included Old and Middle French, Old and Middle Irish, Old Castilian, Old Swedish, and Early New High German (Bavarian). Religious texts were chosen as a common thread because of their wide diffusion in the European vernacular tradition, their high survival rate in manuscripts, and their relevance for the study of medieval cultural and textual practices.

The dataset is based on manuscripts preserved in France, Spain, Sweden, Germany, and Ireland, dating from the 11th to the 15th centuries, with a particular concentration in the 15th century. All manuscripts belong to the category of medium to highly decorated literary manuscripts. They are written in clearly identifiable scripts, predominantly in one or two columns; two manuscripts also include marginal texts.

**Text**

The transcribed texts represent a wide spectrum of medieval vernacular religious writing across Europe. They include hagiographic works such as the *Legend om S. Barbara* in Old Swedish and the *Legenda aurea* in Old Castilian, devotional poetry like the *Amra Colum Cille* in Old/Middle Irish, as well as historiographical and moralizing texts such as Simon de Hesdin’s and Nicolas de Gonesse’s French translation of Valerius Maximus (*Les Faits et Dits Mémorables*). Other contributions comprise Marian devotional literature (*La conception Nostre Dame*), vernacular adaptations of Latin translations (*Décadas de Tito Livio* in Old Castilian), and the widespread compendium *Der Heiligen Leben* in Early New High German (Bavarian) again based on the Legenda Aurea.

**Script**

The manuscripts reflect the diversity of medieval book hands used in different European regions between the 11th and 15th centuries. They include early Insular Carolingian minuscule, as well as various Gothic books and cursive hands characteristic of high and late medieval manuscript culture.

Our dataset is composed of following texts :

| Manuscript | Date | Language | Text | Scripta | Number of transcribed lines | Name of the transcriptor |

| :---- | :---- | :---- | :---- | :---- | :---- | :---- |

| Paris, BnF, fr. 282 | 1401 | Middle French | Les Faits et Dits Mémorables de Valère le Grant, by Simon de Hesdin and Nicolas de Gonesse | Gothic cursiva | 192 | Roxane Bougrelle |

| Paris, BnF fr. 818 | 1251-1300 | Old French | La conception Nostre Dame | Gothic | 143 | Catherine Dehaut |

| Dublin, RIA 23 E 25, pp. 5-8 | 11-12th c (MS). 8/9th c (Text) | Old/Middle Irish and Latin | Amra Colum Cille | Insular Carolingian miniscule | 19 | Ciaran McDonough |

| Stockholm, KB A 110, ff. [250r] | 1385-1400 | Old Swedish | Legend om S. Barbara, from Codex Oxenstiernianus (Järtteckensboken) | Cursiva gothica recentior | 110 | Pontus Henningsson |

| Escorial, ms. h-I-14 | 13-14th c. | Old Castillan | Legenda aurea | Gothic cursiva | 141 | María Florencia Vieira |

| Munich, University Library, 2° Cod. ms. 314 | 15th c. | Bavarian, Early Modern | Der Heiligen Leben (Sommerteil) | Bastarda | 287 | Barbara Denicolò |

| BNE ms. 12732 | 1433 | Old Castilian | Décadas de Tito-Livio (Pero López de Ayala from Bersuire’s translation) | Gothic Cursiva | 141 | Irene Salvo García |

The following metrics displayed below our corpus consisting of the manuscripts listed in the table above: in terms of its total number of documents (18); number of regions (74); total number of lines (1279); total number of words (11365); total number of characters (59696):

**Dataset structure**

The data folder contains each manuscript and its transcriptions. For each manuscript, the transcription is provided in an ALTO XML, along with a METS XML file, along with images of the manuscript pages that were transcribed. These are provided in one folder per manuscript as a subfolder to the data folder.

**Guidelines**

We followed the guidelines of the project CatMus (1.6) :

Pinche, A., Clérice, T., Vlachou-Efstathiou, M., Chagué, A., Camps, J.-B., Gille Levenson, M., Brisville-Fertin, O., Boschetti, F., Fischer, F., Gervers, M., Boutreux, A., Manton, A., Gabay, S., & Ferrante, G. (2025). CATMuS Medieval (1.6.0). Zenodo. [https://doi.org/10.5281/zenodo.15030337](https://doi.org/10.5281/zenodo.15030337)

And used the SegmOnto controlled vocabulary :

Simon Gabay, Ariane Pinche, Kelly Christensen, Jean-Baptiste Camps, Nicola Carboni, *SegmOnto, A Controlled Vocabulary to Describe the Layout of Pages*, version 0.9, Genève/Lyon/Paris, 2023, [https://segmonto.github.io/](https://segmonto.github.io/)

For our collective project, we decided to follow some common guidelines :

**Segmentation guidelines**

**MainZone** : see SegmOnto. MainZone contains principle text, as a single block, without any paratext. If there are many columns, we made different blocks. Interlinear glosses are part of the MainZone.

**NumberingZone** : see SegmOnto. Contains page numbers or folio numbers.

**GraphicZone** : images or decorations.

**MarginTextZone** : marginal glosses or additions.

**DropCapitalZone** : see SegmOnto. Any type of capital letters, without any subtype. Do not transcribe it.

**RunningTitleZone** : see SegmOnto. Zone containing a running title.

**DefaultLine / default (Kraken Result)** : standard textline.

**HeadingLine** : Any heading, without consideration for the level.

**InterlinearLine** : Any line between standard baseline.

Verse initials (appearing at the start of each line of a versified text, similar in size to the remaining text of the verse) should be included in the line of the corresponding verse. They should thus be segmented in the MainZone as normal text.

Some initial letters (not illuminated), which hang slightly below the line are not included in the mask, meaning that they have to be later added to the transcription.

**Transcription guidelines**

We also followed the CATMus guidelines when transcribing our documents. Given the diversity of our corpus, we chose a graphemic transcription system. Allographic variants (‘u/v’ and ‘i’/’j’) were normalised. When it comes to ‘z’/’s’, every allograph of ‘s’ was transcribed by ‘s’, and we made a difference between ‘s’ and ‘z’. We didn’t separate agglutinated words and verse initials were to be kept separated (using a space) from the rest of the word. When it comes to the choice of signs, we used four main signs: “.” for single dots, “:” for more than single dots, “/” for diastoles and signs that look like virgulas and “¶” for section markers. We didn’t normalize capital letters. We also transcribed hyphenations and diastoles. Also, if a character does not exist in MUFI but carries linguistic importance and is a common occurring character in a manuscript, a new character can be created by combining MUFI characters and symbols. Abbreviations were not developed and we transcribed them adhering to MUFI. In other words, we transcribe what we see, so that the dataset works for every vernacular language of our corpus. Every character is described in the CatMus guidelines:

[https://catmus-guidelines.github.io/html/guidelines/en/character\_table.html](https://catmus-guidelines.github.io/html/guidelines/en/character_table.html)

For the transcription of the manuscript in Old Swedish, specifically, a new combination of characters, “a̶ ” (small antiphon), was created since it is a common occurrence in Old Swedish manuscripts and carries linguistic importance. The character “a̶ ” has similar pronunciation as “ä” and “æ”, but since these letters appear in other languages with other pronunciation/meaning and look different than the small antiphon, a new character was created. A large antiphon already exists in MUFI,

[https://mufi.info/q.php?p=mufi/chars/unichar/59610](https://mufi.info/q.php?p=mufi/chars/unichar/59610), but the small one does not. In the future, hopefully the small antihpon, “a̶ “, will be added to MUFI.

**Creators of the dataset**

Ciaran McDonough, Aarhus University 0000-0002-5198-9205

Pontus Henningsson, Linneaus University, 0009-0002-8956-7312

Barbara Denicolò, Paris Lodron University of Salzburg, 0000-0001-7155-9790

Roxane Bougrelle, Université Lumière Lyon 2, 0009-0002-6343-2305

Catherine Dehaut, Université de Montréal, 0009-0000-5161-7906

María Florencia Vieira, ENS de Lyon, 0009-0001-8222-1178

Irene Salvo García, Universidad Autónoma de Madrid 0000-0003-0155-1886

### 中世纪俗语宗教文本数据集（Team Medieval Vernacular Religious Texts）

**数据集概述**
本数据集由协作工作组创建，旨在转录多语种欧洲中世纪俗语宗教文本。为体现工作组团队成员的语言学专长，本项目涵盖古法语与中古法语、古爱尔兰语与中古爱尔兰语、古卡斯蒂利亚语、古瑞典语以及早期新高地德语（巴伐利亚方言）。之所以选择宗教文本作为共同主题，是因为其在欧洲俗语传统中传播广泛、手稿留存率高，且对中世纪文化与文本实践研究具有重要价值。

本数据集基于保存在法国、西班牙、瑞典、德国与爱尔兰的手稿，年代跨度为11至15世纪，其中15世纪的手稿占比较高。所有手稿均属于中等至高度装饰性的文学手稿范畴，采用清晰可辨的书体书写，大多为单栏或双栏排版；其中两份手稿还包含页边文本。

**文本内容**
转录文本涵盖欧洲各地丰富多样的中世纪俗语宗教写作，包括：圣徒传记作品，如古瑞典语《圣芭芭拉传奇（Legend om S. Barbara）》与古卡斯蒂利亚语《黄金传奇（Legenda aurea）》；灵修诗歌，如古/中古爱尔兰语《阿姆拉·科隆·基尔（Amra Colum Cille）》；以及史学与教化文本，如西蒙·德·埃斯丹与尼古拉·德·贡塞合译的瓦莱里乌斯·马克西姆斯作品译本《难忘事迹与名言（Les Faits et Dits Mémorables）》。其他文本还包括圣母灵修文学《圣母领报（La conception Nostre Dame）》、拉丁语译本的俗语改编本（古卡斯蒂利亚语《提图斯·李维史论（Décadas de Tito Livio）》），以及基于《黄金传奇（Legenda aurea）》编纂的广泛流传的汇编文集《圣徒传（Der Heiligen Leben）》（早期新高地德语巴伐利亚方言版）。

**书体特征**
这些手稿展现了11至15世纪欧洲不同地区使用的中世纪书体多样性，涵盖早期海岛加洛林小写体，以及中世纪盛期与晚期手稿文化中典型的各类哥特体与草书手书。

本数据集收录的手稿详情如下表所示：

| 手稿编号 | 年代 | 语言 | 文本内容 | 书体 | 转录行数 | 转录者姓名 |
| :---- | :---- | :---- | :---- | :---- | :---- | :---- |
| 巴黎，法国国家图书馆（BnF）fr. 282 | 1401 | 中古法语 | 西蒙·德·埃斯丹与尼古拉·德·贡塞合译《瓦莱里乌斯·马克西姆斯难忘事迹与名言（Les Faits et Dits Mémorables）》 | 哥特草书体 | 192 | Roxane Bougrelle |
| 巴黎，法国国家图书馆（BnF）fr. 818 | 1251-1300 | 古法语 | 《圣母领报（La conception Nostre Dame）》 | 哥特体 | 143 | Catherine Dehaut |
| 都柏林，爱尔兰皇家科学院（RIA）23 E 25，第5-8页 | 手稿年代：11-12世纪；文本年代：8-9世纪 | 古/中古爱尔兰语与拉丁语 | 《阿姆拉·科隆·基尔（Amra Colum Cille）》 | 海岛加洛林小写体 | 19 | Ciaran McDonough |
| 斯德哥尔摩，瑞典皇家图书馆（KB）A 110，第[250r]叶 | 1385-1400 | 古瑞典语 | 源自Oxenstierna抄本（Järtteckensboken）的《圣芭芭拉传奇（Legend om S. Barbara）》 | 晚期哥特草书体 | 110 | Pontus Henningsson |
| 埃斯科里亚尔修道院图书馆，ms. h-I-14 | 13-14世纪 | 古卡斯蒂利亚语 | 《黄金传奇（Legenda aurea）》 | 哥特草书体 | 141 | María Florencia Vieira |
| 慕尼黑大学图书馆，2° Cod. ms. 314 | 15世纪 | 巴伐利亚早期新高地德语 | 《圣徒传（Der Heiligen Leben，夏季卷）》 | 巴斯塔德体 | 287 | Barbara Denicolò |
| 西班牙国家图书馆（BNE）ms. 12732 | 1433 | 古卡斯蒂利亚语 | 佩罗·洛佩斯·德·阿亚拉根据贝尔叙尔译本改编的《提图斯·李维史论（Décadas de Tito-Livio）》 | 哥特草书体 | 141 | Irene Salvo García |

本数据集所收录的上述手稿共计18份，覆盖74个区域，总转录行数为1279行，总词数为11365，总字符数为59696。

**数据集结构**
数据文件夹包含每份手稿及其转录文件。每份手稿的转录内容以ALTO XML格式提供，同时附带METS XML文件与对应转录手稿页面的图像文件。所有文件将按手稿分为独立子文件夹，存放于数据文件夹下。

**标注规范**
本研究遵循CatMus（1.6）项目的标注规范：
Pinche, A., Clérice, T., Vlachou-Efstathiou, M., Chagué, A., Camps, J.-B., Gille Levenson, M., Brisville-Fertin, O., Boschetti, F., Fischer, F., Gervers, M., Boutreux, A., Manton, A., Gabay, S., & Ferrante, G. (2025). CATMuS Medieval (1.6.0). Zenodo. https://doi.org/10.5281/zenodo.15030337

并采用SegmOnto受控词汇表（SegmOnto）：
Simon Gabay, Ariane Pinche, Kelly Christensen, Jean-Baptiste Camps, Nicola Carboni, *SegmOnto, A Controlled Vocabulary to Describe the Layout of Pages*, version 0.9, Genève/Lyon/Paris, 2023, https://segmonto.github.io/

**分割标注规范**
- **主文本区（MainZone）**：参见SegmOnto规范。主文本区包含核心文本，以单一区块形式呈现，不含副文本。若存在多栏排版，则分为多个独立区块。行间注释属于主文本区范畴。
- **编号区（NumberingZone）**：参见SegmOnto规范。包含页码或叶码。
- **图像区（GraphicZone）**：图像或装饰元素。
- **页边文本区（MarginTextZone）**：页边注释或附加内容。
- **首字母大写区（DropCapitalZone）**：参见SegmOnto规范。所有类型的首字母大写形式，无细分亚型，无需转录。
- **页头标题区（RunningTitleZone）**：参见SegmOnto规范。包含页头标题的区域。
- **标准行（DefaultLine / default (Kraken Result)）**：标准文本行。
- **标题行（HeadingLine）**：任意层级的标题行。
- **行间行（InterlinearLine）**：位于标准基线之间的任意行。

诗行首字母（出现在每行诗行开头，尺寸与诗行其余文本一致）应归入对应诗行的区块，作为正常文本在主文本区中进行分割。

部分非装饰性首字母，轻微悬挂于行下方，未被包含在掩码范围内，需在后续转录步骤中补充。

**转录规范**
本研究在转录过程中同样遵循CatMus项目的标注规范。考虑到语料库的多样性，我们采用音位转录体系。对异形变体（如‘u/v’与‘i’/‘j’）进行了归一化处理；针对‘z’/‘s’，所有‘s’的异形变体均转录为‘s’，并区分‘s’与‘z’。未对黏合词进行拆分，诗行首字母需与其余文本以空格分隔。在符号选用方面，我们使用四类核心符号："."表示单点符号，":"表示多点符号，"/"表示音节分隔符与类virgula符号，"¶"表示章节标记。未对大写字母进行归一化处理，同时转录连字符与音节分隔符。若某字符未在中世纪Unicode字体倡议（Medieval Unicode Font Initiative, MUFI）中收录，但具备语言学重要性且在手稿中频繁出现，可通过组合MUFI字符与符号创建新字符。未对缩写进行扩展，严格按照MUFI规范进行转录。简言之，我们如实转录所见内容，以确保本数据集适配语料库中的所有俗语语言。所有字符的详细说明可参见CatMus规范：https://catmus-guidelines.github.io/html/guidelines/en/character_table.html

针对古瑞典语手稿的转录，我们特别创建了“a̶ ”（小型antiphon符号）的组合字符，该符号在古瑞典语手稿中频繁出现且具备语言学重要性。“a̶ ”的发音与“ä”和“æ”类似，但由于这两个字母在其他语言中发音与含义不同，且外形与小型antiphon符号存在差异，因此创建了专属字符。大型antiphon符号已收录于MUFI（https://mufi.info/q.php?p=mufi/chars/unichar/59610），但小型antiphon符号尚未收录。未来我们期待小型antiphon符号“a̶ ”能够被加入MUFI字符集。

**数据集创作者**
Ciaran McDonough，奥胡斯大学 0000-0002-5198-9205
Pontus Henningsson，林奈大学 0009-0002-8956-7312
Barbara Denicolò，萨尔茨堡巴黎洛东大学 0000-0001-7155-9790
Roxane Bougrelle，里昂第二大学 0009-0002-6343-2305
Catherine Dehaut，蒙特利尔大学 0009-0000-5161-7906
María Florencia Vieira，里昂高等师范学院 0009-0001-8222-1178
Irene Salvo García，马德里自治大学 0000-0003-0155-1886