Bryn Mawr Classical Review

BMCR 2014.03.61 on the BMCR blog

Bryn Mawr Classical Review 2014.03.61

Guillaume Bonnet, Abrégé de la grammaire de Saint Augustin. Collections des universités de France. Série latine, 405.   Paris:  Les Belles Lettres, 2013.  Pp. lii, 96.  ISBN 9782251014654.  €45.00 (pb).  

Reviewed by Vincent Hunink, Radboud University Nijmegen (


A Latin grammar by none other than Saint Augustine? Few readers and scholars will be aware that such a text actually exists. A splendid new critical edition with French translation and notes by Guillaume Bonnet will justly raise new interest in this forgotten testimony of Augustinian erudition.

It requires some space to explain what text we are dealing with. In his Retractationes 1.6, Augustine records a lost early work of his on Latin grammar. The treatise was meant to be part of an ambitious work on disciplinae, which he planned to write in the years before his baptism in Milan (387). The project remained unfinished: just the extant De musica and the book on grammar were actually composed. For centuries, scholars attempted to identify the lost grammar by Augustine. A grammatical work called Regulae, transmitted under Augustine's name, was an obvious candidate, but this had been labeled apocryphal as early as the 17th century. In 1852 however, cardinal Mai published a little known text from a French manuscript as ars sancti Augustini episcopi ad Petrum Mediolanensem, and this was argued to be the lost Augustinian book on grammar.

The new Latin text, however, failed to meet the high expectations of the age. Its subject matter and contents were generally thought to be far below the elevated standards of the church father, and doubts were raised on it authenticity and integrity. For many decades, the text was considered a summary or abstract made by Cassiodorus or later, and scholarly attention accordingly grew weaker.

It was only in 1984 that an Augustinian scholar, Vivian Law, seriously readdressed the issue and came to new conclusions (her paper was published in Recherches augustiniennes 19, 1984, 155-183). According to Law there are actually some important ideas and concrete examples in the text that can be taken as clues for Augustinian authorship. The evident shortening to which the text has been subjected at several places can be attributed to a second author working at some later stage, but before the Carolingian age, to which the three extant manuscripts can be dated.

Guillaume Bonnet has now taken up and extended Law's arguments. He has established a new critical text of the treatise, with Augustine's name clearly appearing in the title. The main merit of Bonnet's edition is surely to have brought into the open again an early text largely to be attributed to Augustine.

In his sound introduction, Bonnet discusses the history of the text, and analyses all relevant arguments that seem to point to Augustine. These include some ideas about language, remarkable parallels with clearly authentic writings of Augustine, and, somewhat more intriguing, details and examples that appear to reflect a North African context and readership. For example, in c.14 the author uses infector as one of his paradigmata for the declension of nouns (see also below); this is a rare noun referring to African manufacturers of purple dye. In another, rather more curious instance, he uses an example sentence about creatures 'jumping higher than the trees' (c.96: super arborem saliunt, which may be taken as a reference to apes. Many other exemplary words and phrases evoke an atmosphere of school, rhetoric, community life, and positive Christian values.

To be sure, none of this can count as definite proof, but the cumulative evidence is convincing. A reader familiar with Augustine will also have the general intuition of dealing with a text that is essentially Augustinian.

On the other hand, it is also obvious that the text has been abridged in a number of parts. The sections on nouns and verbs look rather extended and show many detailed considerations and thoughts, which suggests that these are more or less authentic. But other sections, such as the ones on adverbs, participles, conjunctions, and interjections are so short or even deficient that they can hardly be attributed to Augustine himself and must be the result of a subsequent shortening of his texts. As Bonnet argues, unlike some earlier scholars, a 'dialogue form' of the original text by Augustine may also have been stripped in the process. Such a dialogue form would have mainly consisted in formal questions like 'What is the declension of Tullius?', 'How many kinds of attributes can the verb have?' and could fairly easily have been removed for reasons of space. Although this must remain uncertain, Bonnet's arguments to date the revision and shortening of Augustine's text to a period shortly after the early 7th century seem sound.

So here is what we seem to have: a seventh-century, shortened version of an authentic, early text by Augustine (ca. 375), whose hand may be still seen in large sections where no major later editing seems to have taken place.

Bonnet's introduction also contains informative paragraphs on the manuscripts, with a new proposal for a stemma (p.XXXVI), on the medieval and later references to the treatise, and on the printed editions. A list of some 35 textual changes adopted here (p. XLVII), mostly involving relatively unimportant matters, is useful for those working with earlier editions.

The Latin text is faced by a French translation that leaves little to be desired. Wherever necessary, Latin words and phrases are reproduced in Latin with French translation added between square brackets (e.g. "mulier [femme]"). This is a useful and reader-friendly procedure, which makes the French translation accessible to linguists whose grasp of Latin is less than perfect.

As to the interpretation conveyed by the translation, I found hardly any points of debate or shortcomings, with the exception of the very first sentence: Latinitas est obseruatio incorrupte loquendi secundum Romanam linguam, which is rendered as follows: 'Le bon usage est l'observance d'un langage sans défaut conforme à la langue de Rome.' Here, the rather essential first word Latinitas has been given too general a sense: the sentence does not discuss language in general, but the proper use of Latin. It would certainly be better to render as e.g. 'Good Latin is...'

Bonnet's extensive explanatory notes (pp.51-90) are a great help in reading and studying the text. He discusses numerous points of historical linguistics, notably by comparing the treatise in question with other texts by late-Latin grammarians, while also raising various interpretative points. It is a pleasure to study Augustine's approach of the Latin language with Bonnet's notes as a helpful guide. Succinct indices round up the modest volume of some 250 pages (as usual in Budé editions, the number of '96' in the bibliographical descriptions refers to 'the text' and must be doubled to account for both Latin text and French translation).

Given his testimony in Retr. 1,6, Augustine seems to have composed his treatise on grammar as a simple, firm basis on corporalia that would allow subsequent study of incorporalia, that is, as a stepping stone for spiritual studies. Accordingly, spiritual matters are largely absent, with the minor exception of general Christian notions in example phrases like propter salutem (c.94) or examples referring to three bishops (c.7).

It provides interesting reading, however, also to those who look for subject matters which may been seen as more worthy of Augustine than grammar. The church father evidently writes for an audience of local non-native users of Latin, who need some first help and backup for their own writing in that language, be it as schoolmasters or clerks. For example, he leaves unexplained what must have been obvious to these users, such as the main functions of the subjunctive, and he deals at some length with matters that seem rather theoretical or irrelevant to daily practice. Nearly always, he appears to be eager to provide morally sound instruction, selecting positive words such as uirtus, oboedire, and above all, scribere, which is, perhaps not surprisingly, his main paradigm of the Latin verb, fully spelled out in all its forms.

To non-specialist readers interested in Latin grammar, it can be a nice little game to look for minor and major differences between Augustine's outlines of the system and modern grammatical descriptions, as presented in text books. Many patterns, examples, and subdivisions, appear familiar, but there are also notable differences. For example, the division of nouns is not according to declensions, the defining criterion rather being grammatical gender, which comes in six forms: male, female, neuter, male/female, male/neuter, or male/female/neuter (with homo, infector and prudens as the paradigms of the last three categories).

It is the combination of the familiar (or perhaps all too familiar...) with the unexpected which makes interesting reading here.

The treatise is probably not among Augustine's masterpieces, and not even Bonnet's fine edition will have the effect to make it count as such. But if this exceedingly high aim is left aside, the new Budé has the merit of bringing one of Augustine's early experimental works back to where it belongs: in the hands of readers.

Read comments on this review or add a comment on the BMCR blog

Read Latest
About BMCR
Review for BMCR
Support BMCR

BMCR, Bryn Mawr College, 101 N. Merion Ave., Bryn Mawr, PA 19010