LINGUIST List 30.2754

Mon Jul 15 2019

Review: Norse, Old; Language Documentation; Writing Systems: Leeuw van Weenen, de (2019)

Editor for this issue: Jeremy Coburn <>

Date: 13-May-2019
From: Bev Thurber <>
Subject: A Grammar of Möðruvallabók
E-mail this message to a friend

Discuss this message

Book announced at

AUTHOR: Andrea Leeuw van Weenen, de
TITLE: A Grammar of Möðruvallabók
YEAR: 2019

REVIEWER: Bev Thurber


Part of the Arnamagnæan manuscript collection currently housed in Reykjavík, Möðruvallabók is a fourteenth-century collection of eleven prose narratives: Njáls saga, Egils saga, Finnboga saga, Bandamanna saga, Kormáks saga, Víga-Glúms saga, Droplaugarsona saga, Ölkofra þáttr, Hallfreðar saga, Laxdœla saga ,and Fóstbrœðra saga (2). It may have been written at the monastery in Möðruvellir in northern Iceland and was certainly at Möðruvellir (though there is a question of which Möðruvellir) in 1628 (5–6). In this study, Andrea de Leeuw van Weenen describes the language of the manuscript in detail through a statistical analysis of the spelling. The main text of the book consists of four chapters: an introduction, a description of Möðruvallabók, and chapters on orthography and morphology. The first two chapters describe the project and the manuscript, and the second two present the results of the analysis.

The introduction recounts the history of the project, which began as early as 1976. De Leeuw van Weenen’s goal was to produce “a thorough description of a manuscript” dating to the middle of the Old Icelandic period (8). As a reasonably sized collection of prose originally composed in Icelandic and mostly written by a single scribe in the fourteenth century, Möðruvallabók proved a suitable candidate. Once she had selected the manuscript, de Leeuw van Weenen created a database that included seven pieces of information for each word: the form occurring in the manuscript, the reference, any punctuation, the normalized form, the associated lemma, the class the word belonged to, and additional grammatical information (13). This database is the foundation for everything presented in the book.

Chapter 2 provides a detailed physical description of Möðruvallabók from codicological and palaeographic perspectives, though most of the space is devoted to palaeography. At least four people participated in writing the main text (marginal notes are not discussed), but one fourteenth-century scribe did most of the writing. The sections written in the lone seventeenth-century hand are excluded from the analysis (22). Much of the chapter is spent enumerating the letters of each type by size. Large letters are combined, whether minuscule or not, as are small letters. Superscripts, subscripts, diacritics, ligatures, and abbreviations are also enumerated.

Chapter 3 summarizes the manuscript’s orthography. It includes a short individual description of how each phoneme is typically written, including capitalization and abbreviation, along with a list of variations. The descriptions are quite detailed and include counts of each occurrence and references where applicable. For example, “/m/ is spelled <m> 27 275 times, <M> 513 times initially (or following a capital), …, <f> in fikill 122rb33, and <n> in nargt 45rb27” (90). The long section of descriptions is topped off by a section containing a short summary for each letter and longer sections on abbreviations, compounds, numbers, hyphenation, and scribal errors. De Leeuw van Weenen concluded that the manuscript features regular spelling (130) and called the scribe “a scrupulous copyist who tended to keep not only the exact wording of his exemplar, but also its orthography and even some palaeographic features” (131) because the orthography is consistent within each individual text but varies between texts.

At 134 pages, Chapter 4 represents approximately 40% of the book. It is a full account of Icelandic morphology as expressed in the manuscript. The chapter includes sections on nouns, adjectives, pronouns, numerals, adverbs, and verbs; prepositions, conjunctions, particles, and interjections are lumped together under “Other word classes,” which lists the forms (261). De Leeuw van Weenen assembled complete paradigms of individual words to explain this phase of the language. As may be expected, the sections on nouns and verbs are the longest.

Möðruvallabók includes complete paradigms for 74 different nouns (when forms in compounds are included) and nearly complete paradigms for numerous others (133). The discussion of individual nouns is organized first by gender, then by stem and type. Each part of the section on nouns starts with sample paradigms and a list of the words in the manuscript that follow each sample pattern. This is followed by an enumeration of the forms present in the manuscript and descriptions of all the sound changes and spelling variations. De Leeuw van Weenan assigned each noun’s gender based on the forms present in the manuscript. When there was not enough information, as happened not infrequently, de Leeuw van Weenen followed standard dictionaries (135–136).

The sections on adjectives and pronouns are structured similarly. The adjectives are grouped by whether they are positive, comparative, or superlative. Unfortunately, no complete comparative or superlative paradigms are present in the manuscript (191). The section on pronouns mainly consists of tables of forms, and the section on adverbs is primarily a list of how many times each occurs in the manuscript.

As for verbs, there are 67,848 forms belonging to 955 lemmata in Möðruvallabók (224). Readers of the sagas will be unsurprised to find that approximately 60% of the forms are third person indicative --- 40% preterite and 20% present. The section on verbs is organized first by class, with assignments based on the forms appearing in the manuscript. This means that some classifications differ from those found in standard grammars and dictionaries. In particular, weak verbs of classes 3 and 4 are distinguished by their past participles, which results in some verbs that are usually placed in class 4 being placed in class 3 (224). Some verbs, including weak verbs of classes 3 and 4 whose past participles do not occur in the manuscript, could not be precisely classified, and de Leeuw van Weenen noted all the possibilities (see, e.g., 236, 241). The next part of the section on verbs is organized by conjugation, with a short discussion of each part of the paradigm that includes notes on scribal errors and unusual spellings of the ending.

The book concludes with two appendices (one listing the 22,247 compound words in the manuscript and the other listing all the corrections made to the database over the years), a bibliography, and an index (by paragraph number, not page number). The index has two parts, “Lemmata” and “Persons etc.”; the latter is a relatively short list mainly consisting of personal names and institutions.


Originally published in 2000, this book was meant as the third volume of Möðruvallabók, AM 132 fol. (1987), which was published by Brill under the author’s married name, van Arkel-de Leeuw van Weenen. How it became a separate book is a tale told in the preface. Essentially, the book was delayed due to various circumstances and other projects, most notably the author’s edition of the Icelandic Homily book, which involved the typesetting nightmare detailed in de Leeuw van Weenen (1997). A Grammar of Möðruvallabók ended up being moved to a different publisher (xiii), but has now returned to Brill in this reprint. This review treats it as a standalone volume rather than part of a set.

De Leeuw van Weenan’s work on Möðruvallabók spans nearly a quarter-century. This long timeline highlights the importance of this volume, which provides a record of the manuscript as it was by the time the transcription was finished in 1981 (9). When de Leeuw van Weenen revisited the manuscript in 1997, its condition was noticeably worse. Therefore, she gave more weight to earlier readings, including those of other editors, changing the transcription in 17 places (xiii). In particular, one of the exceptions to the rule that first-person pronouns follow verbs without the -m ending (slu, 11rb7) is no longer readable (253n200).

Such small details are important because one of the book’s themes is reliance on the manuscript as the authority. This high level of trust in the manuscript means that all the variant spellings are noted to ensure that no phonological or morphological changes are overlooked. The downside is that this trust may have allowed scribal errors to propagate. De Leeuw van Weenan notes that errors and important spelling variations can be indistinguishable (59). By trusting the manuscript, she allows scribal errors to pass undetected.

The analysis combines computational and manual methods. Although the database is the foundation, much of the work was done by hand. De Leeuw van Weenen’s descriptions of her methods and interpretations of the results are concise and remain close to the text. They help readers understand what the numbers represent, but do not delve deeply into interpretations. The observational nature of this volume leaves many openings for readers to pursue interpretations. The vast amount of data contained in the database and summarized in this book provides a firm foundation for future studies. The database could be made more readily available; de Leeuw van Weenen promised to provide it in a format that conforms to the Text Encoding Initiative Guidelines (13), but I was unable to find it online.

The book has been edited quite thoroughly, as is appropriate for its level of detail. I noted only a couple of typographical errors. Overall, it is a complete and precise description of Möðruvallabók and a valuable contribution to the literature.


Van Arkel-de Leeuw van Weenen, Andrea. 1987. Möðruvallabók, AM 132 fol. 2 vols. Leiden: Brill.

De Leeuw van Weenen, Andrea. 1997. A Medieval Icelandic manuscript: The making of a diplomatic edition. TUGboat 18.1:30–36.

Noreen, Adolf. 1970. Altnordische Grammatik I, Altisländische und altnorwegische Grammatik, Laut- und Flexionslehre, unter Berücksichtigung des Urnordischen. Sammlung kurzer Grammatiken germanischer Dialekte A4, 5th ed. Tübingen: Niemeyer.

Ordbog over det norrøne prosasprog. A Dictionary of Old Norse Prose. 1995. Vol 1: a-bam. Copenhagen: The Arnamagnæan Commission.


Bev Thurber is an independent scholar who is interested in historical and computational linguistics and the history of ice skating.

Page Updated: 15-Jul-2019