#### LXIII

Erika Kihlman and Denis Searby eds Ars Edendi Lecture Series Vol. V

This is the fi fth and fi nal volume of lectures on textual criticism and classical philology – broadly understood – given within the framework of the Ars edendi

Two of the six papers in this volume stem from a 2015 workshop on editorial theory and method, the theme of which dealt with fragments and the writing of commentaries. As regards the former, S. Douglas Olson problematizes the creation and continuation of scholarly knowledge concerning texts that have only come down to us in a fragmentary state, emphazising the challenges and pitfalls that lay in wait for the editor. Benjamin Millis offers a nuanced homage and apology for the traditional text edition with a scholarly commentary, especially underscoring its importance as a connective pathway between text and reader as well as

The other four lectures were given at the concluding conference of the Ars edendi programme, held in August 2016. In a case study Cynthia Damon shares her refl ections on how to digitally edit Pliny's *Natural History* in a form that will provide this work's rich reception history and at the same time its extensive use of sources, many of which are now lost. The digital component is also prominent in Odd Einar Haugen's contribution in which he shows that digital mark-up is also an editorial enterprise and how it can be useful for the textual scholar. Dorothea Weber gives an insider's view of the *Corpus Scriptorum Ecclesiasticorum Latinorum*, an editorial project on-going since 1864, and especially how improved cataloguing has led to numerous discoveries of texts by St. Augustine. As a conclusion to the volume, David Greetham, one of the founders of the Society for Textual Scholarship, refl ects on three different methods for editing texts that have undergone various degrees of rescription, namely the *oeuvres* of Eriugena, Coleridge, and Eliot.

research programme (2008 – 2015).

the impetus it can give to scholarly research.

Volume V

Edited by Erika Kihlman and Denis Searby

# STUDIA LATINA STOCKHOLMIENSIA LXIII

# *Ars Edendi* LECTURE SERIES Volume V

Edited by Erika Kihlman and Denis Searby

2020

Published by Stockholm University Press Stockholm University SE-106 91 Stockholm Sweden www.stockholmuniversitypress.se

Text © The Author(s) 2020 License CC-BY

Supporting Agency (funding): Riksbankens Jubileumsfond

First published 2020 Cover Illustration: Miniature from Den Haag, Koninklijke Bibliotheek, ms 71 A 24, fol.2v, containing the legend of the monk Theophilus. Cover image copyright: CC-BY-NC-ND Cover design by Stockholm University Press

Studia Latina Stockholmiensia (Online) ISSN: 2002-472X

ISBN (Paperback): 978-91-7635-119-2 ISBN (PDF): 978-91-7635-116-1 ISBN (EPUB): 978-91-7635-117-8 ISBN (Kindle): 978-91-7635-118-5

DOI: https://doi.org/10.16993/bbd

This work is licensed under the Creative Commons Attribution 4.0 Unported License. To view a copy of this license, visit creativecommons.org/licenses/by/4.0/ or send a letter to Creative Commons, 444 Castro Street, Suite 900, Mountain View, California, 94041, USA. This license allows for copying any part of the work for personal and commercial use, providing author attribution is clearly stated.

Suggested citation: Kihlman, E. and Searby, D. (eds.) 2020. *Ars Edendi Lecture Series, vol. V*. Stockholm: Stockholm University Press. DOI: https://doi.org/10.16993/bbd. License: CC-BY

To read the free, open access version of this book online, visit https://doi.org/10.16993/bbd or scan this QR code with your mobile device.

# **Studia Latina Stockholmiensia**

Studia Latina Stockholmiensia (SLS) (ISSN 2002–472X) is a peerreviewed series of scholarly editions of Latin texts as well as monographs and anthologies of Latin studies. While continuing Stockholm University's tradition of Medieval studies, the series is also open to Latin from the Ancient and the Modern periods.

The textual editions include an introduction, a critical apparatus and usually a translation. The literary, linguistic and other philological studies of Latin feature a range of methodological approaches, both modern and traditional.

SLS includes Ars Edendi Lectures as a sub-series.

Studia Latina Stockholmiensia was previously included in the series Acta Universitatis Stockholmiensis (ISSN 0491–2764).

### **Editorial Board**

Maria Plaza, PhD, Associate Professor of Latin, Dept. of Romance Studies and Classics, Stockholm University, Sweden.

ORCID: https://orcid.org/0000-0003-1643-2660

Hans Aili, PhD, Professor emeritus of Latin, Dept. of Romance Studies and Classics, Stockholm University, Sweden.

Erika Kihlman, PhD, Associate Professor of Latin, Dept. of Romance Studies and Classics, Stockholm University, Sweden.

ORCID: https://orcid.org/0000-0001-8081-1740

#### **Titles in the series published by Stockholm University Press**


# **Peer Review Policies**

Stockholm University Press ensures that all book publications are peer-reviewed. Each book proposal submitted to the Press will be sent to a dedicated Editorial Board of experts in the subject area. The full manuscript will be reviewed by chapter or as a whole by two external and independent experts.

The editor Erika Kihlman, who is also a member of the Editorial Board of Studia Latina Stockholmiensia, did not take part in any recommendations or decisions for this book.

A full description of Stockholm University Press' peer-review policies can be found on the website: http://www.stockholmuniversitypress.se /site/peer-review-policies/

#### **Recognition for reviewers**

The Editorial Board of *Studia Latina Stockholmiensia* applies singleblind review during manuscript assessment. Many thanks to the reviewers involved in this process.

# **Contents**

Introduction vii *Erika Kihlman and Denis Searby*

Contributors xiii List of Illustrations xvii

Eupolis' Lost *Taxiarchoi* and the Problem of Academic Consensus 1 *S. Douglas Olson*

Commentaries and the Problem of Authority (with particular attention to editing fragments) 15 *Benjamin Millis*

On editing the reception of Pliny's *Natural History* 29 *Cynthia Damon*

Beyond the Edition: On the Linguistic Annotation of Vernacular Texts 67 *Odd Einar Haugen*

Challenges of Editing Latin Patristic Texts: A report from inside the *Corpus Scriptorum Ecclesiasticorum Latinorum* 89 *Dorothea Weber*

Editing in Three Movements: Eriugena, Coleridge, Eliot 109 *David Greetham*

# **Introduction**

*Erika Kihlman and Denis Searby*

The *Ars edendi* Research Programme at Stockholm University, financed by the Bank of Sweden Tercentenary Foundation, ran from 2007 to 2015. With an emphasis on the editing of texts that had often been overlooked in the methodological and theoretic literature on editing, it focused on unusual editorial problems as well as on contrasting methodological solutions, particularly those related to medieval Greek and Latin literature. It dealt, not least, with commentary and compilatory traditions in various genres, model sermons, biblical glosses, anthologies and both prose and poetry for use in the liturgy. During these same years, the *Ars edendi* lecture series provided a dynamic forum for leading textual scholars to discuss their editorial decisions and share both their practical experience of methodological aspects of textual criticism, the mise-en-page of edited texts as well as wider perspectives on textual philology. These volumes preserve to a large extent the style of the original oral lectures. All of this is thematically and stylistically reflected in the present volume as well.

The final conference of the research programme, entitled *The Arts of Editing: Past, Present, and Future*, was held at Stockholm University in August 2016. Exactly one year earlier, a final workshop on editorial methods and theory took place, the theme of which dealt with fragments and writing commentaries. Selected papers from these two events fill the pages of this, the fifth and final volume of the *Ars edendi Lecture Series*, for, alas, there is an end to all things, and that principle applies not least to research funding.

The volume begins with two papers related to classical Greek comedy. Using as his frame-work attempts at reconstructing the *Taxiarchoi* of Eupolis, S. Douglas Olson launches into a lively and thought-provoking discussion of 'how scholarly knowledge is created and maintained in

**How to cite this book chapter:**

Kihlman, E. and Searby, D. 2020. Introduction. In: Kihlman, E. and Searby, D. (eds.). *Ars Edendi Lecture Series, vol. V.* Pp. vii–xii. Stockholm: Stockholm University Press. DOI: https://doi.org/10.16993/bbd.a. License: CC-BY

regard to texts that are so emphatically 'lost' that one might better say that they simply do not exist any longer.' In other words, what can we really know and how can we make claims to knowledge of the plots and structures of lost plays based on the scant fragments that have chanced to survive to our day? Yet, modern and even contemporary scholars do make knowledge claims about no longer existent works that prove to be more products of imagination than of science, and Olson takes one of them in particular to task in order to prove his deeper point about critical methodology and the creation of illusory academic consensus by means of accumulation and ingenuity rather than through an unprejudiced approach to the evidence. His deeper point, in other words, has to do with our responsibilities as scholars and teachers, and that this has wider social implications than his obstensible topic of the reconstruction of a lost play of an obscure Greek comic poet.

Benjamin Millis's wide-ranging essay presents the edition and commentary as a form of scholarly engagement requiring much imagination and 'the sort of detailed engagement with the text that is seemingly less and less stressed and valued.' He argues that a shift in scholarly production occurred over the course of the nineteenth and twentieth centuries away from editions and commentaries and toward the monograph as the prime means of scholarly discourse. As part of this process, commentaries took on a more subsidiary role in the production of advanced scholarship. He describes the effect that Eduard Fraenkel's edition of Aeschylus, *Agamemnon* had in shaping subsequent scholarship which was 'to establish a template for what a scholarly commentary on an ancient text could be and perhaps even should be', and this was the idea of the 'monumental, definitive commentary'. His aim in tracing the development of scholarly commentaries over the course of the twentieth century is to encourage reflection on the writing of commentaries as well as on their mediating role between texts and so-called higher criticism. He sees editions and commentaries as 'themselves a dialogue with the material and a grappling with it much like any other type of criticism or analysis.' Millis argues for a re-legitimization of editions and commentaries, and sees a role here precisely in the editions of literary fragments. In this regard, he discusses the difficulties faced by editors of fragments with regard to deference to previous authorities, the great names of scholarship, as well as the more immediately practical problems of the presentation and ordering of fragments and of the evaluation of the trustworthiness of the sources.

Both Olson's and Millis's reflections are recommended reading to any young scholars about to embark on work with fragments of Greek and Latin authors. However, they also touch on the value, the difficulties, and the complex decisions involved in two pillars of philology, namely textual editing and commentary, as well as the transmission and reception of textual scholarship and how it can frame later scholarly discussions. They thus provide medievalists with food for thought. Olson and Millis presented their papers at the last workshop organized within *Ars edendi*. The remaining contributors to this volume presented theirs at the final conference in 2016.

Like Olson and Millis, Cynthia Damon is also primarily a classicist. Her paper begins with a reflection on the ways in which textual editing resemble pathways connecting not only the editor and the original text, but also different generations of readers and writers as well as different fields of scholarship. One such pathway, connecting classicists and medievalists alike, is the challenge to create critical editions in digital form in order to contribute to contemporary scholarly explorations of texts and literatures. With a conventional OCT edition of Caesar behind her, Damon moved on to studying the challenges of digitally editing Pliny's *Natural History*, a work with a rich reception history and itself an example of a text made up of material taken from other texts, most of which have not survived. Book 9, which deals with fish, serves as the case study here. Damon explores important parts of its reception history in *De piscibus*, Book 4 of *Hortus sanitatis*, printed in Mainz in 1491 and recently edited in both paper and digital form by Catherine Jacquemard and her colleagues. The most important source of this compendium was the *Speculum naturale* of Vincent of Beauvais, who in turn points back to Pliny. Damon asks how one can 'edit the reception' of Pliny's encyclopedia or 'edit the genesis' of Vincent's? Damon's paper is deliberately intended to stimulate imaginative solutions to providing durable infrastructure for the connections enabled by texts and editors, given our modern resources.

In his contribution, Odd Einar Haugen gets into the nitty-gritties of precisely that: how the digital mark-up or annotation of medieval texts is an editorial enterprise. He draws on his experience in working on medieval vernaculars in the Medieval Nordic Text Archive (MeNoTA) project and illustrates how annotation may usefully be seen as an integral part of the whole editorial process. Haugen discusses in particular three focal levels in the process: fascimile, diplomatic and normalised; thus he speaks of the multi-level rendering of manuscript texts, and exemplifies it with short extracts from the digital edition of the Old Norwegian Homily Book. Perhaps surprising for classicists or medievalists working in the classical languages is the amount of morphological and syntactical mark-up involved; Haugen reminds us, however, of the importance of vernacular texts as sources for language history where these details become cumulatively significant. Yet other categories of annotation can be chosen, of course, depending on the scholarly objectives of the editorial project and the nature of the text, for instance annotation can be an aid for better understanding obscure or ambiguous passages in historical works, thus making annotation a close cousin of the scholarly commentary. Haugen concludes his chapter with a cost-benefit analysis, comparing relatively small textual traditions with those that have large numbers of manuscripts, and discussing how the level of canonicity will be decisive in the final cost-analysis.

Like Haugen, Dorothea Weber provides a close-up look at an ongoing editorial project, although, in this case, the project is the *Corpus Scriptorum Ecclesiasticorum Latinorum* which started already in 1864. Weber provides an historical overview of CSEL, the foundation of which was directly connected to the planning of the *Thesaurus linguae Latinae*. She is thus also able to give us an interesting sidelight on the difference in the perceptions of both late antiquity itself and of the editing of late antique works that prevailed in the later nineteenth century and those of today. One result of the changing appreciation for late antiquity is the fact that CSEL has not limited itself to editing only Christian Latin texts from the period. Weber gives us short but detailed glimpses into the work currently being done at CSEL, beginning with the essential work of improved cataloguing, especially in the series devoted to works attributed to St. Augustine, which has yielded discoveries of new texts of Augustine such as the twenty-nine letters published in 1981 by Johannes Divjak, the recent discovery of the commentary on the Gospels by Fortunatianus, or Ars edendi's own Brian Jensen's discovery of a sermon. Weber offers some details of the kinds of editorial problems encountered and solutions proposed in recent editing projects. Of course, each and every edition has challenges and problems of its own, but a good edition must also make the structure of the text visible, which can be done in the *constitutio textus* as traditionally understood or by other means, which Weber illustrates with examples from the edition of Augustine's corpus of the *Enarrationes in psalmos*. In general, CSEL relies on tried and true stemmatic methods, and Weber notes that the kinds of texts edited within CSEL are generally not adequate to the methods of the so-called New Philology, because they 'in most instances are not texts that underwent systematic rescription, quite the contrary: they were held in esteem as authoritative and were thus not adapted freely.'

David Greetham, by contrast, examines three editions of works that did undergo rescription, although the rewriting in these cases was done by the authors themselves, namely John Scotus Eriugena, Samuel Coleridge and T. S. Eliot. Eriugena's *Periphyseon* was edited by É. Jeauneau between 1996–2003, the works of Coleridge by J.C.C. Mays in 2001, and those of Eliot by C. Ricks and J. McCue in 2015. Although the methods of the editors vary they are united in their 'aim to provide a comprehensive, indeed exhaustive, access to the extant documents of their authors' and one of the main editorial challenges lies in the treatment of the revisions, the versions thus created and the variants, the key word of 'New Philology'. In Jeauneau's edition, variance takes pride of place. The different versions of the texts are placed on par with the 'critical' text, all set in parallel columns on the page. Although this procedure, Greetham argues, places great demands not only on the editor but also on the reader, it also 'moves the readerly eye (or ear) away from the plainchant of a single utterance into a polyphony with multiple voices and variance as a normative condition – where a blank space is 'just as much a *presence* as is a positive textual variant'. Like Eriugena, Coleridge was also an avid revisioner of his writings but his editor Mays opted for a different solution than Jeauneau, creating a 'reading text' for each piece – not to be considered 'the' text or even a 'standard' text – against which a full documentary record of variants can be set, the 'variorum text'. Also, in the Eliot edition by Ricks and McCue, which they preferred to call an annotated rather than a critical text, a clear reading text is accompanied by the editors' commentary including a documentary history on the textual genesis. Despite Eliot's general reluctance to revise, variance still exists in the different impressions of Eliot's work, a fact lamented by Eliot himself. From his observations and remarks regarding the editorial principles and the realisations of the edited texts, Greetham opines that these new 'exhaustive' editions will not end but rather encourage further textual examination. The author, as Greetham points out, is not dead but very much alive and this is through 'the loyalties and devotions (and sheer hard work) of editors working in the long tradition extending from the Alexandrian librarians to the present day.' It was with great sadness we learned that David Greetham passed away just a few weeks before this volume was to be sent off to the Stockholm University Press.

The editors of this volume, who are the only remaining members of the *Ars edendi* project at Stockholm University, wish to extend our thanks to the contributors to the present volume and to previous volumes, as well as to our dear, now former colleagues in the *Ars edendi*, both fellow researchers and student assistants, to all our colleagues who have lent their support over the years. We dedicate this volume to the memory of Benkt-Erik Hedin, husband of Gunilla Iversen on whose initiative the *Ars edendi* was formed and who was its leader throughout the duration of the project. Benkt-Erik was ever a splendid host when the whole team assembled, which we often did, at their home; he passed away as we were doing the final revisions of the papers. We end by expressing our hope that the art of editing medieval and classical texts will long continue at Stockholm University, so that the legacy may be both transmitted and developed.

Erika Kihlman and Denis Searby

# **Contributors**

**Cynthia Damon** is a Professor of Classical Studies at the University of Pennsylvania, having received her undergraduate degree from Stanford University, her MA from Boston College and her PhD from Stanford University. She has written extensively on classical Latin literature and Roman historiography. In the field of textual studies, she has edited Caesar's *Commentariorum libri III de bello civili* (OCT 2015) and the Loeb volume of Caesar's *Civil War* (2016); she translated Tacitus' *Annals* for the Penguin Classics series (2012). She is currently working on the Loeb volume of Caesar's *Gallic War* (forthcoming 2021). Another current project is on Pliny's encyclopedic *Natural History* and its reception.

**David Greetham** (October 21, 1941–March 24, 2020) was an American literary scholar and critic. He was a Distinguished Professor at The Graduate Center, CUNY, having received his undergraduate degree from the University of Oxford and completed his PhD in English at CUNY. He was a founding member of the Society for Textual Scholarship and served as the Society's President from 1999 to 2001. Author of such books as *The Margins of the Text: Editorial Theory and Literary Criticism* (1997), *Theories of the Text* (1999) and *The Pleasures of Contamination* (2010), he was a towering figure in the field of textual studies. He is remembered by his graduate students as a generous and inspiring mentor, and will be much missed by his colleagues.

**Odd Einar Haugen** is a Professor of Old Norse Philology (norrøn filologi) at the University of Bergen, Norway, having received his undergraduate and graduate degrees at the same university. He has written extensively on Old Norse philology and linguistics. As a textual scholar, he has been active in editing medieval manuscripts and is the current head of the Medieval Nordic Text Archive. He has published Old Norse grammars in Norwegian and German, edited a handbook of Old Norse philology in the same languages, and co-edited a reader of Medieval Nordic in Italian, *Le lingue nordiche nel medioevo*. He is a member of the Royal Norwegian Society of Sciences and Letters, The Royal Gustavus Academy for Swedish Folk Culture, The Royal Society of Arts and Sciences in Gothenburg as well as the Comité International de Paléographie Latine (CIPL).

**Benjamin W. Millis** holds a PhD in classics from the University of Illinois, has been Research Associate in Greek Epigraphy at the University of Oxford for the ERC-funded research project the Social and Cultural Construction of Emotions, as well as collaborated in the Heidelberg Academy *Kommentierung der Fragmente der griechischen Komödie* Project. Recent publications include his edition of the fragments of *Anaxandrides* (*Fragmenta Comica* vol. 17 2015), *Inscriptional Records for the Dramatic Festivals in Athens: IG II2 2318–2325 and Related Texts* (co-authored with D. Olson 2012), *Greek Theatre in the Fourth Century BC* (contributing author, 2014) as well as the English translation of the Basel Homer Commentary.

**S. Douglas Olson** is Distinguished McKnight University Professor of Classical and Near Eastern Studies at the University of Minnesota. He is closely associated with the Heidelberg Academy *Kommentierung der Fragmente der griechischen Komödie* Project and is also the General Editor of the Basel Homer Commentary English Edition. He is particularly well known for his work on 5th- and 4th-century Athenian comedy; for his critical editions and commentaries on Greek poetic texts of all period; and for his general interest in ancient literary fragments. Among his over twenty books is a new critical edition of Athenaeus of Naucratis' *Learned Banqueters* for the Teubner series.

**Dorothea Weber** is Professor of Latin at the University of Salzburg and director of the Corpus Scriptorum Ecclesiasticorum Latinorum (CSEL), having received her PhD at the University of Vienna. She has worked intensively on the manuscript tradition and editing of works of Saint Augustine as well as on other works within Latin patristics and Neo-Latin drama. Among her publications within the field of textual scholarship are *Die handschriftliche Überlieferung der Werke des Heiligen Augustinus* (1993), her edition of Augustine, *De Genesi contra Manichaeos* (1998), and *Augustinus Conversus*. *Ein Drama von Jakob Gretser. Einleitung, Text, Übersetzung und Kommentar* (2000), and *Prosper, De vocatione omnium gentium* (2009).

## **Editors**

**Erika Kihlman** received her PhD from Stockholm University in 2006 and has since worked on textual cricism and critical editions of medieval commentaries and other teaching-related texts. She was one of the founding members of the Ars edendi research programme and acted as its co-director for the duration of the programme.

ORCID: https://orcid.org/0000-0001-8081-1740.

**Denis Searby** received his PhD from Uppsala University in 1998 after having first studied at Columbia University. He has worked on the Greek anecdotal tradition, the reception of Thomas Aquinas in Byzantium and the Revelationes of Saint Birgitta. He is a professor at Stockholm University and was also one of the founding members of the Ars edendi research programme and acted as its co-director for the duration of the programme.

ORCID: https://orcid.org/0000-0002-3983-9132.

# **List of Illustrations**

## **Cover Illustration**

Miniature from Den Haag, Koninklijke Bibliotheek, ms 71 A 24, fol.2v, containing the legend of the monk Theophilus. Cover image copyright: CC-BY-NC-ND.

## **Cynthia Damon**

Figure 1. Citations of NH books 2–37 in the Speculum naturale, using the formula "Plinius libro ... ." 35

Figure 2. *NH* 9.1–186, excerpted chapters (**bold**) and re-used chapters (**large bold numbers**). 36

## **Odd Einar Haugen**

Figure 1. An admonition in the Old Norwegian Homily Book. Copenhagen, The Arnamagnæan Collection, AM 619 4º, f. 8v, l. 23–26. Copyright: The Arnamagnæan Collection, License: CC BY-NC-ND. 68

Figure 2. Display of a multilevel transcription of an Old Norwegian law fragment. Oslo, National Archives, Norse Fragment 7, f. 5r, col. A (ca. 1320–1350). Copyright: Odd Einar Haugen, License: CC BY-NC-ND. 75

Figure 3. A morphological analysis of the last sentence in Fig. 1. Copyright: Odd Einar Haugen, License: CC BY-NC-ND. 77

Figure 4. A dependency analysis of the last, coordinated sentence in Fig. 1. Copyright: Odd Einar Haugen, License: CC BY-NC-ND. 80

Figure 5. Stanza 2.5–8 of Vǫluspá in Reykjavík, The Árni Magnússon Collection, GKS 2365 4º, fol. 1r, l. 4–5. Copyright: The Arnamagnæan Collection, License: CC BY-NC-ND. 82

Figure 6. The morphological annotation of Vǫluspá stanza 2.5–8. Copyright: Odd Einar Haugen, License: CC BY-NC-ND. 83

Figure 7. The syntactic annotation of Vǫluspá stanza 2.5–8. Copyright: Odd Einar Haugen, License: CC BY-NC-ND. 84

## **David Greetham**

Figure 1. Pages 254–255 in *Iohannis Scotti seu Eriugenae Periphyseon: editionem novam a suppositiciis quidem additamentis purgatam, ditatam vero appendice in qua vicissitudines operis synoptice exhibentur*, ed. by É. Jeauneau, vol. 5, CCCM 165 (Turnhout: Brepols, 2003), License: CC BY-NC-ND. 114

# **Eupolis' Lost** *Taxiarchoi* **and the Problem of Academic Consensus**

*S. Douglas Olson*

The nominal topic of this paper is Eupolis' *Taxiarchoi*, or "*Taxiarchs*", and in particular what, if anything, can be said of the action of the play: who the characters were, how they interacted, and what went on onstage. My real and deeper interest, however, is in how scholarly knowledge is created and maintained in regard to texts that are so emphatically "lost" that one might better say that they simply do not exist any longer. While I discuss *Taxiarchoi* in some detail, therefore, Eupolis' comedy is only an excuse, an opportunity to expose in a particularly clear fashion a process that goes on constantly in the field of classical studies, and indeed in any academic field devoted to making sense of the past. What I argue in what follows is that we do not and cannot really "know" anything about *Taxiarchoi*, at least in the way the verb "know" is conventionally and freely used; that once that point becomes clear, issues of critical methodology become more pressing than they might otherwise appear to be; and that these matters are of far greater significance than the essentially trivial question of what modern readers can agree might have gone on onstage in a lost comedy by an obscure poet over two millennia ago. But first the play itself.

Eupolis apparently began to stage dramas in 429 BCE (thus test. 2.6), and *Taxiarchoi* is traditionally dated to the next year, 428 BCE, on the ground that the Athenian general Phormio (*PA* 14958; *PAA* 963060), who seems to have been a central character in the action, died that summer. Dead men do not normally appear onstage in comedy, the argument goes, so we have something close to a fixed date for the play. This argument has been vigorously challenged by Ian Storey, who does his best to move Eupolis' comedy down to 415 BCE, on the eve of the

#### **How to cite this book chapter:**

This lecture was given at the *Ars edendi* workshop on Fragments and Commentaries 27–28 August, 2015.

Olson, S. D. 2020. Eupolis' Lost *Taxiarchoi* and the Problem of Academic Consensus. In: Kihlman, E. and Searby, D. (eds.). *Ars Edendi Lecture Series, vol. V*. Pp. 1–13. Stockholm: Stockholm University Press. DOI: https://doi.org/10.16993/bbd.b. License: CC-BY

Sicilian Expedition.1 As I discuss this question in detail elsewhere,2 and as it does not impinge directly on my argument here, I will say nothing more about it except to observe that I believe that Storey is demonstrably wrong and that he offers the same sort of problematic claims in support of his position regarding the date of the play as those I discuss in what follows in connection with its action.

Seventeen book-fragments of *Taxiarchoi* are preserved, along with broken bits and pieces of a first-century CE papyrus commentary on the play (= fr. 268) first published as *POxy*. 2740. The title of the play is not mentioned in the papyrus. But it does refer to Phormio at one point (fr. 268.33), while at another it overlaps with one of the book-fragments (= fr. 281), and it is universally accepted today that this is in fact a commentary on Eupolis' lost comedy. What is known of *Taxiarchoi* — and for most of the rest of this paper I will use the words "known" and "knowledge" within quotation marks, as it were, to describe what those of us who are interested in the text *think* we know — is approximately the following. According to a *scholion* on Aristophanes' *Peace* (= *Taxiarchoi* test. i), quoting fr. 274, in the course of the play "Dionysus learned the customs of generals and wars at Phormio's side." Many if not all of the book-fragments can be made to fit this general theme, as I illustrate below, including the verses of Eupolis' comedy quoted as lemmata for commentary in the papyrus. The text of the papyrus is readily accessible in Kassel–Austin's edition of the comic fragments and (even easier, at least for English-speakers) in translation in Storey's Loeb edition of the so-called Old Comic poets and in Rusten's *Birth of Comedy* volume.3 At 175 lines long and containing substantial portions of about a dozen verses of Eupolis, the papyrus apparently treats a scene from *Taxiarchoi* in which Phormio taught Dionysus what it meant, in practical terms, to be an Athenian soldier. In particular, A. M. Wilson has identified what looks to be part of a rowing scene (fr. 268.48–53), in which the god — unsurprisingly — did a very bad job of one of the tasks that were set him, splashing everyone around with water and being barked at by the old military commander

<sup>1</sup> Ian C. Storey, *Eupolis: Poet of Old Comedy* (Oxford, 2003) 247.

<sup>2</sup> S. Douglas Olson (ed.), *Eupolis* Heilotes — Chrysoun genos *(frr. 147–325). Translation and Commentary* (*FrC* 8.2: Heidelberg) 370–1.

<sup>3</sup> Ian C. Storey (ed. and trans.), *Fragments of Old Comedy* Vol. II *Diopeithes to Pherecrates.* Loeb Classical Library 514 (Cambridge Mass. and London: Harvard University Press, 2011); Jeffrey Rusten (ed.), *The Birth of Comedy: Texts, Documents, and Art from Athenian Comic Competitions, 486–280* (Baltimore, 2011).

Phormio.4 Intriguingly, the scene appears to anticipate the action at Aristophanes' *Frogs* 188–270, where Dionysus similarly does a miserable job of rowing, in this case under Charon's direction, suggesting that part of the inspiration for the latter play came from Eupolis. Finally, a vase-painting from a well-deposit in the Athenian Agora dating to around 400 BCE and published by Crosby shows two figures, perhaps comic actors, one of them clearly labeled "Dionysus", the other "Phor-", seemingly confirming the centrality of those characters in Eupolis' comedy.5 The above can reasonably be described as the general state of contemporary knowledge about *Taxiarchoi*, the sort of information taken more or less for granted by those interested in the play, encoded in various ways in Kassel–Austin's edition of the fragments, and used by other scholars as a basis on which to build further, hopefully more revealing hypotheses.

Any modern evaluation of *Taxiarchoi* must inevitably begin with the treatment of the play by Storey, who in his 2003 monograph on Eupolis notes that in the case of *Taxiarchoi* we are in a relatively privileged position, at least as far as "lost" fifth-century drama goes, in that we can say "a fair bit about this comedy".6 In particular, the *scholion* to *Peace* "tells us that Dionysus went to Phormio ... to learn the rules of generals and wars",7 on which basis Storey notes: "I am assuming that Dionysus had a major role in *Taxiarchoi*, that his scene with Phormion was not limited to a brief encounter in an episode."8 As Storey himself observes, this is by no means a radical interpretative step, but instead represents what everyone working on the play has always done and believed, *inter alia* because the thesis appears to be supported by the evidence of the fragments. Storey then makes two further assumptions: first, that the Dionysus of *Taxiarchoi* was similar to the effeminate, clownish impostor seen in *Frogs* — a thesis that receives some initial, provisional support from the connections between the rowing scenes in the two comedies noted by Wilson, and that suggests that we should look for further parallels between the two plays wherever possible9 — and second, that Eupolis' chorus was firmly on Phormio's side throughout

<sup>9</sup> Storey, *Eupolis*, p. 251–252.

<sup>4</sup> A. M. Wilson, 'A Eupolidean Precedent for the Rowing Scene in Aristophanes' *Frogs*?', *CQ* NS 26 (1974), 250–252.

<sup>5</sup> Margaret Crosby, 'Five Comic Scenes from Athens', *Hesperia* 24 (1955), 81–82 with pl. 34c.

<sup>6</sup> Storey, *Eupolis*, p. 250.

<sup>7</sup> Storey, *Eupolis*, p. 246.

<sup>8</sup> Storey, *Eupolis*, p. 252.

the comedy, just as the chorus of knights is firmly on the side of the Sausage-seller in Aristophanes' play of 424 BCE.10

With this broad framework in place, Storey reconstructs an opening scene, beginning with fr. 272, which has long been thought to represent a reaction to Dionysus' arrival in camp carrying an enormous load of equipment that has no place in a soldier's life:

ὅστις πύελον ἥκεις ἔχων καὶ χαλκίον ὥσπερ λεχὼ στρατιῶτις ἐξ Ἰωνίας

whoever you are, who have come with a bathtub and a bronze cauldron, just like a new mother from Ionia joining the ranks.

Storey observes that this "should be directed at the newly arrived Dionysus", as Kaibel for example thought,11 adding "I have no problem with the attribution of these lines to Phormio himself".12 Elaborating on this hypothesis, Storey brings in fr. 285 σκευοφοριώτης, an odd comic word for a baggage-bearer perhaps modeled on εἰραφιώτης, a cult name of Dionysus himself (*hHom*. 1.2, 17, 20), and fr. 279 ὄνος ἀκροᾷ σάλπιγγος ("a donkey heeds a trumpet"), all of which evidence taken together, he suggests, may hint that "Perhaps *Taxiarchoi* began, like *Frogs*, with a comic scene involving Dionysus, his baggage, and an ass that reacted badly to the sound of the military trumpet".13 As for what Dionysus is doing in Phormio's camp in the first place, Storey notes the traditional theme of Hera's wrath on the one hand, and fr. 274 ὡς οὐκέτ' ἂν φάγοιμι † στιβάδας ἐξ ὅτου ᾽φυγον ("that I could no longer eat † since I/they fled camp-beds" — admittedly corrupt — on the other, and writes: "I wonder if in *Taxiarchoi* ... Dionysus entered fleeing from Hera to hide himself by joining the Athenian forces".14

Storey's next step toward reconstructing the action of *Taxiarchoi* begins with what are universally agreed to be fragments from the portion of the action in which Phormio teaches Dionysus what military life involves: fr. 276

οὔκ, ἢν φυλάττῃ γ' ὧδ' ἔχων τὴν ἀσπίδα

<sup>12</sup> Storey, *Eupolis*, p. 254.

<sup>10</sup> Storey, *Eupolis*, p. 250.

<sup>11</sup> ap. Kassel–Austin.

<sup>13</sup> Storey, *Eupolis*, p. 254.

<sup>14</sup> Storey, *Eupolis*, p. 252.

(They) won't (*sc*. kill you *vel sim*.) if you stay on guard, holding your shield like this,

where Storey notes that the line "should come from a scene of instruction in the proper use of a shield";15 and fr. 269

(Φο.) οὔκουν περιγράψεις ὅσον ἐναριστᾶν κύκλον; (Β.) τί δ' ἔστιν; εἰς ὤμιλλαν ἀριστήσομεν; ἢ κόψομεν τὴν μᾶζαν ὥσπερ ὄρτυγα;

(Phormio) Draw a circle big enough to have lunch in, won't you? (B.) What's going on? Are we going to play *eis ômillan* for lunch? Or are we going to smack our barley-cake like a quail?,

where Storey, in this case following Kassel–Austin (whose version of the text is printed above; further discussion of this point below), observes: "The first speaker is Phormio ... and the other must be Dionysus".16 Storey then turns to the papyrus, first discussing a handful of verses so badly damaged that I will not treat them here, except to note that they appear to involve a discussion of military matters such as passwords (fr. 268.26–7) and soldier's pay (fr. 268.18–20). After this, Storey considers Wilson's rowing scene, fr. 268.48–53 (a combination of text and ancient commentary):

γ̣ὰρ οὐκ ἐπίσ̣τ̣αμ̣α̣[ι *παρὰ*] *τὸ πεζῇ βαδίζω*, [*νεῖν*] *γὰρ οὐκ ἐπίσταμα*[*ι*] παύσει ῥαίνων ἡμ[ᾶς, οὑκ πρῴρας; *εἰώθασι λ*[*έγειν·* "*ὁ ἐκ πρώρας*, *μὴ ῥ*[*αῖνε*".

for I don't know how: alluding to the saying "I go by foot, for I don't know how to swim". Stop sprinkling us, you toward the bow!17 They are accustomed to say: "You toward the bow, don't get (us) wet!"

The final line of the commentary sparks a mocking comment from Storey: "One does not need to look far for an incompetent oarsman",18

<sup>15</sup> Storey, *Eupolis*, p. 253.

<sup>16</sup> Storey, *Eupolis*, p. 253.

<sup>17</sup> Thus Storey; see further discussion of the sense of ἐκ πρῴρας below.

<sup>18</sup> Storey, *Eupolis*, p. 257.

meaning that the awkward, unwarlike Dionysus is patently the incompetent rower who cannot swim.

With much of the basic plot of *Taxiarchoi* established, Storey goes on to reconstruct what one might call the logical next step in the dramatic action, in which Dionysus realizes that the soldier's life is not for him: fr. 271

Offer me Naxian almonds to chew on and wine from Naxian vines to drink!,

on which Storey comments "Meineke is surely right to attribute these lines to the god";19 fr. 275

ἐπιφαγεῖν μηδὲν ἄλλ' ἢ κρόμμυον λέποντα καὶ τρεῖς ἁλμάδας

to eat nothing else, except an onion one peels and three brined olives,

in regard to which Storey notes that this "could be either Phormio describing the lot of a soldier ... or Dionysus' own complaint about the poor quality of the food. The latter seems preferable in my judgment";20 fr. 280

ἀντὶ ποικίλου πιναρὸν ἔχοντ' ἀλουσίᾳ κάρα τε καὶ τρίβωνα

in place of an embroidered robe having a head filthy from lack of washing and a peasant's robe as well,

on which Storey comments: "The subject is clearly Dionysus ... I suspect Dionysus himself is speaking, complaining about his physical squalor and the state of his clothing";21 and fr. 270

(A.) ὅτ' ἦν μέντοι νεώτερος, κρόκης πέντε στατῆρας εἶχε. (B.) ναὶ μὰ τὸν Δία, νῦν δὲ ῥύπου γε δύο τάλαντα ῥᾳδίως

(A.) When it was newer, however, it contained five *statêres* of woof-thread. (B.) Yes, by Zeus; whereas now (it contains) two talents of dirt at least,

<sup>19</sup> Storey, *Eupolis*, p. 255.

<sup>20</sup> Storey, *Eupolis*, p. 255.

<sup>21</sup> Storey, *Eupolis*, p. 255.

in regard to which Storey says: "In view of the clear association of the *krokoton* with Dionysus, it is likely that Dionysus is complaining about what has happened to his garments in the course of his military training".22 In addition, Storey notes the mysterious fr. 273:

οὐ θᾶττον αὐτὴν δεῦρό μοι τῶν τοξοτῶν ἀγαγὼν ἀποκηρύξει τις, ὅ τι ἂν ἀλφάνῃ;

One of the bowmen bring her here quickly and auction her off for whatever price she might fetch!

on which he comments: "I wonder if this fragment does not come from a scene early in the play, where an Athenian official encounters Dionysus and his entourage and reacts accordingly",23 the point being that the god in his effeminate clothing has been mistaken for a foreign woman (to which one might compare Pentheus' threats in Euripides' *Bacchae*). Finally, to bring the play to a close, Storey cites a now-lost *phlyax* vase that depicts a male figure, perhaps equipped with a dangling comic phallus, and riding a huge fish, that comes from the same well-deposit as the Dionysus-Phormio vase and which Crosby in the original publication of both pots compared to it. Storey asks rhetorically: "I wonder if the oinochoe is showing another scene from *Taxiarchoi*?", and goes on to spell out his idea, which is that the pot represents the very end of Eupolis' comedy, as Dionysus — who has apparently learned how to handle a boat or the rough equivalent in the meantime — returns in triumph on the back of a giant sea-creature reminiscent of the dolphins into which the Tyrrhenian pirates are transformed at the conclusion of the *Homeric Hymn* in his honor. <sup>24</sup>

Storey's book has been respectfully reviewed, which is not to say that everyone who has read it has agreed with everything he has to say.25 But the treatment of the individual comedies of Eupolis has rarely been challenged directly, and the initial impression produced by Storey's handling of *Taxiarchoi* in particular, I suspect, is likely to be not skepticism but an astonished admiration at his ability to reassemble the plot of the play so neatly and effectively. What I argue in the rest of this paper is that Storey has in fact led us badly wrong and that the problem is not just the individual arguments and the philology on which they are

<sup>22</sup> Storey, *Eupolis*, p. 256.

<sup>23</sup> Storey, *Eupolis*, p. 254.

<sup>24</sup> Storey, *Eupolis*, p. 260.

<sup>25</sup> See e.g. Nesselrath, review of Storey, *Eupolis* in *Bryn Mawr Classical Review* 2005.02.44.

based, but his general methodological orientation, including the notion that it makes sense to talk about reconstructing lost Greek comedies as if this were a task that sufficient historical imagination and critical brilliance might allow us to accomplish.

I begin with the greatest improbability of all, which is Storey's presupposition — never stated, but clear throughout — that all the preserved fragments of *Taxiarchoi* can be made to fit coherently together. The problem with this assumption is that the fragments were not chosen by ancient scholars with 21st-century goals in mind, which is to say that this is not a selection of material from representative and highly significant points in the action of the comedy designed to facilitate the process of reconstruction. Instead these are random bits and pieces of Eupolis' play — perhaps 2% of the original text — almost all of them chosen for quotation only because they include a rare vocabulary item, a reference to an interesting historical person or the like. The notion fundamental to Storey's project and the way he carries it out — that we can nonetheless find a likely place in the original structure of the play for almost every tiny piece of *Taxiarchoi* quoted for us is thus untenable, and because it is untenable, there is no point in undertaking the process — although I will qualify this argument modestly below, in what I hope is an interesting fashion. But the simple fact of the matter is that we have no hope of fully understanding the action of *Taxiarchoi*, and that — and this is the crucial point — the more synthetic and complete the explanation produced, the less likely it is to be correct. Storey's ingenious ability to "make everything fit" is thus the clearest sign that his reconstructions must be wrong.

Second, Storey's most basic tool for reconstructing lost comedies generally, and *Taxiarchoi* in particular, is analogy, combined with a readiness to fold into the argument what he takes to be related material of various sorts. Thus in the case of *Taxiarchoi*, if fr. 272 might be taken to represent Dionysus' arrival at Phormio's camp, and if frr. 279 and 285 can be made to fit into such a scene, and particularly if we seem to have a similar scene in Aristophanes, then on Storey's handling of the evidence we abruptly have three fragments of an initial arrival scene featuring a donkey and a porter and reminiscent of *Frogs*. But all of this is merely a series of unfounded guesses, each dependent on the one preceding it and lacking any solid basis in the evidence. Nor does Storey content himself with the known fragments of *Taxiarchoi* itself, but reaches ever further afield, by bringing in the Wrath of Hera theme, for example, to explain Dionysus' initial arrival onstage, and details from the *Homeric Hymn* to justify his vision of the end of the action, even though there is no concrete hint anywhere that any of this played a part in Eupolis' play, and above all else by whimsically appealing to the lost *phlyax* pot, which has no organic historical, literary or archaeological connection whatsoever to *Taxiarchoi*, to argue in favor of an invented thesis regarding the content of the (in fact utterly obscure) end of the play.

But perhaps the most unfortunate — although arguably also the most interesting — aspect of Storey's methodology in his handling of Eupolis' play is the combination of a readiness to guess and assert with a quiet assertion of academic and social authority. In the quotations offered above, Storey routinely informs the reader that these are merely his opinions, but seemingly not in order to make us doubt what he is saying. Instead, the rhetorical function of the language is to insist that this is what the reader too should believe, because this is what Storey believes — as I have now repeatedly pointed out, for generally insufficient reasons.

Storey's *Taxiarchoi* is thus a complex and unstable argumentative house of cards, which stands no chance of being an accurate account of the content of Eupolis' play, and which Storey himself acknowledges consists merely of a long string of guesses, assertions and intuitions, but which he nonetheless expects his readers to assent to and indeed build upon. Nor is this a unique case, for all Storey has done is to expand on arguments that others like Meineke and Kock made before him, pushing forward in already well-established critical directions. To illustrate concretely how this process of illusory academic consensus building works and how problematic its effects can be, I return to what we know of the *Taxiarchoi* papyrus, with its 176 lines of Greek text and its rowing scene.

When Lobel first published *POxy*. 2740 in 1968, he printed it as two main fragments with separate numbering, accompanied by a handful of tiny scraps that seemed to be in the same hand but had been found separately, and which were therefore relegated to an appendix at the end of the volume. Whether fr. 1 or fr. 2 came first in the text of the commentary on *Taxiarchoi* was — and remains — unclear, as was and is — the size of the gap between the fragments, however arranged; perhaps it is hundreds of lines. When Austin republished the papyrus in 1973, however, he brought together all the fragments — including the stray and dubious bits in Lobel's appendix — into a single document with a single numbering system, in the arbitrary order in which Lobel had placed them on the page.26That artificial creature has turned into Kassel–Austin fr. 268 of Eupolis' play and has been translated in that form — a form it never had before, certainly not in Roman Egypt, where the commentary was copied — in Storey's Loeb and the Rusten volume, which is where all but the most sophisticated modern readers will encounter it.

As for the supposed rowing scene, Storey has made Dionysus more of a land-lubber than he is, by reading more into the text than exists in lines 48–50. We know that Eupolis wrote "for I don't know how" (which is the lemma), and we know that the commentator claimed, rightly or wrongly, that this was an allusion to the saying "I go by foot, since I don't know how to swim". But we do not know that Eupolis' character said anything about swimming, for he may just as well have said e.g. "I go by ship, since I don't know how to walk". Even more to the point, a study of the uses of ἐκ + genitive in fifth-century comedy makes clear that ὁ ἐκ πρᾠρας cannot mean "you toward the prow", but must mean instead "you on the prow", which is to say that, however water is being scattered here, figuratively or in fact, rowing is not in question, because one does not row from the prow but from the sides of a ship. To put all this more concretely: there is no 175-line long section of papyrus commentary on *Taxiarchoi*, and there also appears to be no rowing-scene, meaning that there is no evidence that Eupolis' play exercised an influence on *Frogs* — which ought to have been regarded as a dubious argument in any case, given that *Frogs* is securely dated about 25 years later than *Taxiarchoi*. To all this one can add that we should not have expected *Taxiarchoi* to include a rowing scene in any case, for taxiarchs are tribal hoplite commanders rather than naval commanders, and the other fragments of the play having to do with military life all seem to be concerned with hoplite fighting. Finally and most tellingly, we do not even know that Dionysus played a central part in Eupolis' comedy. We know that he played *some* role in the play, which is to say that he appeared onstage at one point, learning about war from Phormio. But perhaps he merely arrived as an exemplary figure in a short scene in the second half of the action, like those that are common in Aristophanic comedies (including with divine visitors), and the more interesting point in any case is that the scholarly community has seized on the one isolated fact we have about the action of

<sup>26</sup> C. Austin (ed.), *Comicorum Graecorum fragmenta in papyris reperta* (Berlin and New York: De Gruyter, 1973), 113–118.

*Taxiarchoi* and has used it as a basis on which to reconstruct scenes, assign speakers to fragments and the like. To cite only the most obvious example of this tendency: in fr. 269 as printed above from Kassel– Austin and as translated by Storey and Rusten, the second speaker is identified as Dionysus. But we have no idea who the second speaker is, and Dionysus is merely a guess by Meineke that has made its way into the body of so-called "knowledge" about Eupolis' comedy.

My central point should by now be clear: Almost everything that, by general scholarly consensus today, is "known" about *Taxiarchoi* is not true. Better put, our "knowledge" of the play consists of a network of weakly grounded hypotheses the scholarly community has chosen to believe, and "progress" in understanding the play consists in practical terms of producing further such hypotheses — most of them better described as wild guesses or simple errors backed by rhetorical devices such as "In my opinion" — and asking others to accept them. My larger concern in this paper accordingly has less to do with *Taxiarchoi* than with critical methodology and the nature and significance of our enterprise. The problem with reconstructing lost comedies is not that this is a difficult business that requires ever greater ingenuity allowing for the discovery and integration of new evidence. That, at bottom, is the idea behind Storey's approach to the plays, and it is misguided as can be seen concretely from the fact that such arguments lead to consistently misleading conclusions. What we are engaged in is not a fundamentally scientific process, like e.g. discovering as much as we can about the moons of Jupiter, which are very far away and very difficult to see and understand, but about which we can gradually, provided we are clever and industrious enough, learn more and more. The problem is that Eupolis' *Taxiarchoi* is unlike the moons of Jupiter because *Taxiarchoi* does not exist. Once upon a time, it did exist. But it does not exist any longer; that is what "lost" means. This is not to say that it is pointless to discuss such texts. But the realization that we can never test our hypotheses against their object changes the nature of the enterprise entirely, and puts the focus where, I have quietly attempted to argue throughout this paper, it belongs: on critical methodology. The problem with Storey's hypotheses, for example, is in most cases not precisely that they are incorrect, for no one can tell. Perhaps Dionysus is the second speaker of fr. 269; but the question is unanswerable. The problem is instead that Storey uses what I will now freely call wrong-headed methods to reach his conclusions, and that, as part of that process, he invokes authority — his own authority and the authority of the scientific process — in an attempt to compel his readers into accepting his conclusions. Put another way, because we can never establish the "truth" about lost comedies (since the comedies do not exist, meaning that there is no "truth" to discover), the terms on which the scholarly debate about what remains of them proceeds can only be *methodological*: Are we willing to accept certain ways of handling evidence and of arguing, and to treat them as normal and appropriate?27To do so is to render them not just normal but normative, a model for how we and our students and colleagues can and should proceed. The fundamental point of this paper is that in this case that would be a mistake, an abdication of our responsibilities as scholars and teachers.

There is a sense in which what went on Eupolis' *Taxiarchoi* which is to say, what we today are willing to say went on in Eupolis' *Taxiarchoi* — is a matter of almost complete indifference. This is a lost comedy by an obscure poet who died over 2400 years ago, and aside from the handful of classicists who will read this paper, and a few other colleagues scattered about the world, no one cares and no one needs to care. As I have tried to show, however, there is another sense in which fundamental academic and political questions are at issue in how we choose to understand this emphatically lost play, and those questions deserve our close attention and concern.

## **Bibliography**


<sup>27</sup> See in general Han Baltussen and S. Douglas Olson, 'Epilogue: A Conversation on Fragments,' in Jan Kwapisz (ed.), *Fragments, Holes and Wholes: Reconstructing the Ancient World in Theory and Practice.* Journal of Juristic Papyrology Supplement, 30 (Fundacja im. Rafała Taubenschlaga: Warsaw, 2017), pp. 393–406.


# **Commentaries and the Problem of Authority (with particular attention to editing fragments)**

*Benjamin Millis*

Editing and commenting on fragments and fragmentary texts is an often difficult endeavour that has its own problems and concerns intrinsic to the nature of the material, but many of the basic issues are essentially the same as those faced when dealing with any sort of text. Editing texts, and equipping these texts with commentaries of various sorts and levels of complexity, is a very old process that has its roots in antiquity. However much this process may have evolved over the past two millennia or so, the essential activity – producing a text in accord with certain aims (usually increased readability or accuracy) and explicating this text in accord with the needs of a certain imagined readership – has remained much the same. Adherence to a long and successful tradition has doubtless played no small part in the continued vitality of editions and commentaries, but they no longer occupy the same central role in scholarship that they did until well into the modern period. Over the course of the 19th and, particularly, the 20th centuries, the edition and commentary was eclipsed by the monograph as the prime means of scholarly discourse. As part of this process of a shift in the mode of scholarly expression, commentaries have become viewed much more as an aid to producing advanced scholarship than as advanced scholarship itself.

As commentaries have moved to a more subsidiary role over the past century or so, perception of some fundamental differences between different sorts of commentaries and editions has likewise changed. A distinction between different levels of commentaries and editions, namely between 'shorter' commentaries on the one hand and 'longer' or 'comprehensive' (a debatable term to be discussed below) ones on the other

#### **How to cite this book chapter:**

This lecture was given at the *Ars edendi* workshop on Fragments and Commentaries 27–28 August, 2015.

Millis, B. 2020. Commentaries and the Problem of Authority (with particular attention to editing fragments). In: Kihlman, E. and Searby, D. (eds.). *Ars Edendi Lecture Series. vol. V*. Pp. 15–28. Stockholm: Stockholm University Press. DOI: https://doi .org/10.16993/bbd.c. License: CC-BY

hand,1 is still commonly recognized, although the different uses and aims of the two types, aside from a general feeling that the one is more appropriate for 'advanced' readers or suitable for 'experts' in a given text, appears to be less clearly understood. Similarly fading from common appreciation is the distinction that has aptly been described as 'insular' vs. 'continental'.2 As commentaries and editions have become less important in their own right as a vehicle for pushing the bounds of scholarship, differences between the various sorts have become elided, resulting in a much more monolithic conception of what a commentary or an edition is and what it is trying to do. The greater the extent to which commentaries become viewed as all doing essentially the same thing, namely supplying answers to a discrete and limited set of questions, the more the pattern repeats and the less commentaries are seen as interpretative works.

The long-term shift of scholarly production away from editions and commentaries and toward monographs is not necessarily a bad thing in itself; aside from anything else, it is indicative of a laudatory broadening interests toward topics that are perhaps better suited to the flexibility of the monograph. But one clear drawback is an ever narrower conception of what editions and commentaries are meant to

<sup>1</sup> The distinction here is the one often described as between 'student' and 'scholarly' commentaries. It is true that 'shorter' commentaries are far more frequently used in the classroom and that 'longer' commentaries are largely the preserve of scholars and, to a lesser extent, advanced students, but the line between the two is not always easy to draw. Elementary commentaries aside, 'shorter' commentaries are regularly works of real scholarship despite any aiming at an ostensibly less advanced audience, and the differences tend to lie more in level of detail and style of presentation than in content. The difference between the two sorts of commentaries is brought out well by the comparison on pp. 348–353 in Roy Gibson, 'Fifty Shades of Orange: *Cambridge Classical Texts and Commentaries*', in *Classical Commentaries: Explorations in a Scholarly Genre*, ed. by Christina S. Kraus and Christopher Stray (Oxford: Oxford University Press, 2016), pp. 346–375.

<sup>2</sup> See M. L. West, 'Forward into the Past', in *Hesperos: Studies in Ancient Greek Poetry Presented to M. L. West on his Seventieth Birthday*, ed. P. J. Finglass, C. Collard and N. J. Richardson (Oxford: Oxford University Press, 2007), pp. xx–xxviii, where West relates on p. xxiv an anecdote in which he was asked by Stefan Weinstock which of the two sorts he (West) was writing. West goes on to define the distinction as 'the sort of commentary that seeks only to elucidate the particular work which is its object, or the sort that reaches out in all directions and is full of material relevant to other authors in which related things occur.' The formulation 'insular' vs. 'continental', particularly in the mouth of a refugee German scholar, might suggest a dig at British scholarship, but West does not suggest any such implication in his account.

accomplish, to whom they are directed, and how they fit with other scholarship in the field.3 Seemingly one of the most common misconceptions about commentaries is that there is an essentially finite series of questions that a commentary on a given text attempts to answer and that the fundamental difference between commentaries is the greater or lesser level of nuance and detail given in providing these answers. In accord with this view, the basic questions about a text are already largely known and once these have been addressed as best as possible by restoring the text as closely as possible to what the author wrote, or is believed to have written, and by explicating the content (linguistic, stylistic, literary, etc.) to an appropriate degree, the editor's job is done. Of course this view of the texts themselves, and of what editors do, is overly simplistic and notably stands in stark contrast with how texts are understood by scholars working, for example, on literary analysis; it is, nonetheless, a perception seemingly widely held by the field at large, including by many editors and commentators themselves. One indication of this is the idea of a 'standard' edition, by which is too often meant 'definitive', or at least 'definitive for our time', and the pernicious effect that this concept can have on scholarship. Once a commentary sufficiently detailed has been written on a given work, the general consensus commonly arises that that work has been 'done', leaving no room or point to producing another; further work along these lines is not just discouraged but usually never even considered, at least until enough time has passed.4 In contrast, monographs that provide a literary analysis of a particular work, for example, seem not to be held to the same strictures, i.e. monographs seem somehow distinct from one another in a way that commentaries are not. The same is also true for translations; they also apparently seem individualistic enough

<sup>3</sup> For brief comments about discrepancies between the intentions of commentary writers and the expectations of commentary readers, see Gibson, pp. 366–367.

<sup>4</sup> For much the same observation, see Gibson, p. 365. He goes on to note, à propos of F. R. D. Goodyear, *The Annals of Tacitus*, vol. I: *Annals 1.1–54* (Cambridge: Cambridge University Press, 1972) and *The Annals of Tacitus*, vol. II: *Annals 1.55–81 and Annals 2* (Cambridge: Cambridge University Press, 1981), 'But who – at least in English – would take on *Annals* 1 again at such length, despite the widely perceived deficiencies in Goodyear's editions? (And what press would publish such a commentary?)' Although 'competing' commentaries do occasionally appear, and Gibson cites one such example, this is very much the exception; overwhelmingly more common is the avoidance of 'duplication', even in cases like that of Goodyear's Tacitus which is both nearly half a century old and not unproblematic.

that 'competing' translations are generally seen as a positive rather than as something to be avoided.

One factor seldom remarked on, yet among the most important, at least for Anglo-American scholarship, is the effect that Eduard Fraenkel's edition of Aeschylus, *Agamemnon* has had in shaping subsequent scholarship.5 Fraenkel's immense personal authority combined with the utterly impressive philological achievement of his commentary, largely unparalleled in effect,6 to establish a template for what a scholarly commentary on an ancient text could be and perhaps even should be.7 There is no particular reason to believe that Fraenkel had any doubts whatsoever about the excellence of his commentary, but there is also no reason to believe that he anticipated the status his commentary

<sup>5</sup> Eduard Fraenkel, *Aeschylus, Agamemnon* (Oxford: Clarendon Press, 1950).

<sup>6</sup> Fraenkel's commentary is not entirely without parallel, in that a number of the older variorum commentaries can compare with it in terms in scope if not in learning, but almost the only really comparable works in terms of both breadth and erudition are a handful of the commentaries produced by the leading scholars of the 17th and 18th centuries. There are also a number of works that are comparable in scope and erudition and very closely contemporary, i.e. Rudolf Pfeiffer, *Callimachus* (Oxford: Clarendon Press, 1949–1953), Felix Jacoby's work on the Atthidographers (*Atthis: The Local Chronicles of Ancient Athens* [Oxford: Clarendon Press, 1949], *Die Fragmente der griechischen Historiker* III B [Leiden: E. J. Brill, 1950] and *Die Fragmente der griechischen Historiker* III b (Supplement) [Leiden: E. J. Brill, 1954]) and A. S. F. Gow, *Theocritus* (Cambridge: Cambridge University Press, 1950); note that the one work not produced by a German exile was also the only one not produced at Oxford. But for all the virtues, tremendous learning and influence of these other works, their dominance has not been quite as long lasting nor have they occupied quite the same place in the imagination as a scholarly exemplar. The obvious comparison is of course Ulrich von Wilamowitz-Moellendorff, *Euripides, Herakles* (Berlin: Weidmann, 1889 [2nd ed. 1895]) (cf. p. 44 n. 12 in Christopher Stray, 'A Teutonic Monster in Oxford: The Making of Fraenkel's *Agamemnon*', in *Classical Commentaries: Explorations in a Scholarly Genre*, ed. by Christina S. Kraus and Christopher Stray (Oxford: Oxford University Press, 2016), pp. 39–57), but for all its tremendous influence and learning, Wilamowitz' commentary is a very different work and never became the exemplar that I am suggesting Fraenkel's work did.

<sup>7</sup> For a recent account of Fraenkel's *Agamemnon*, see Stray, 'Teutonic Monster'; cf. Jaś Elsner, 'Pfeiffer, Fraenkel, and Refugee Scholarship in Oxford during and after the Second World War', in *Ark of Civilization: Refugee Scholars and Oxford University 1930–1945*, ed. by Sally Crawford, Katharina Ulmschneider and Jaś Elsner (Oxford: Oxford University Press, 2017), pp. 25–49. For Fraenkel the man, the basic account is Gordon Williams, 'Eduard Fraenkel 1888–1970', *Proceedings of the British Academy* 56 (1970) 415–442; most recently, see Christopher Stray, 'Eduard Fraenkel (1888–1970)', in *Ark of Civilization: Refugee Scholars and Oxford University 1930–1945*, ed. by Sally Crawford, Katharina Ulmschneider and Jaś Elsner (Oxford: Oxford University Press, 2017), pp. 180–197 (with all essential bibliography).

would eventually achieve, and indeed he is fairly explicit that he saw his task largely as clearing away masses of accumulated error and highlighting real insights and important work in order to promote further scholarship, not to forestall it.8 Regardless of Fraenkel's own intentions and despite some discomfort from others at the time that Fraenkel's views might prove overly dominating,9 the lesson that many seem to have taken away from Fraenkel's work was that aside from details here and there he had said the last word on the play, at least for the foreseeable future. Contemporaries influenced by Fraenkel, but to a much greater extent his students, and the students of his students, absorbed this idea of a monumental, definitive commentary and took this as a model for their own work.10 The essential idea seems to be that if one is diligent enough and does one's work properly, there would be little left to say, thus rendering other attempts largely superfluous. In any case, the concept of the standard commentary, so prevalent by the end of the twentieth century, is completely foreign to the model that prevailed at the outset of the century, when for example in the twenty-five years spanning the turn of the nineteenth to twentieth centuries no less than six or eight new editions of *Acharnians* were produced, and this number could be increased still further if revised editions of older work were also taken into account.11

<sup>8</sup> See Fraenkel, pp. vii–ix.

<sup>9</sup> See, for example, E. R. Dodds as quoted at Stray 'Teutonic Monster', p. 54: 'it is not good that any book, however outstanding, should acquire the status of an oracle.'

<sup>10</sup> For the claim that Fraenkel 'decisively influenced the whole approach to the study of Classical antiquity in Britain' and for a general overview of his influence, see Williams pp. 422–423. Contrast John Dewar Denniston and Denys Page, *Aeschylus, Agamemnon* (Oxford: Clarendon Press, 1957), p. iv, where Page assumes that the commentary by Denniston and himself and that by Fraenkel are merely two points on a long continuum that stretches into the past but also the future; Page's implication is that a commentary on a scale similar to his own would appear not so far in the future, but sixty years later that has yet to happen.

<sup>11</sup> Editions of *Acharnians* published in this period include the following: Frederick H. M. Blaydes, *Aristophanis Acharnenses* (Halle an der Saale: Waisenhaus, 1887); J. van Leeuwen, *Aristophanis Acharnenses* (Leiden: A. W. Sijthoff, 1901); W. Rennie, *The Acharnians of Aristophanes* (London: Edward Arnold, 1909); W. J. M. Starkie, *The Acharnians of Aristophanes* (London: Macmillan, 1909); Benjamin Bickley Rogers, *The Acharnians of Aristophanes* (London: George Bell & Sons, 1910); Richard Thomas Elliott, *The Acharnians of Aristophanes* (Oxford: Clarendon Press, 1914). The above list includes only substantial editions that contain both text and commentary and that were first published within this narrow window; the list could be expanded by including school editions, editions of texts without commentary and subsequent editions of works published earlier. Lest the example of *Acharnians* give the appearance of being an extreme case, the situation is not far

The reason for going through all this at such length is not to criticize any particular commentary or scholar but to try to encourage greater reflection on what writing a commentary is meant to accomplish and on the place of editions and commentaries in modern scholarship more generally. There is of course no single answer to this question, and that fact is indeed part of the answer. Editions and commentaries form an important, in fact crucial, mediating role between texts and so-called higher criticism and in making texts accessible, in the widest sense of the word, to scholars of every sort. But beyond this, and what is often forgotten, editions and commentaries are not solely a sort of middle point in this way but are also themselves a dialogue with the material and a grappling with it much like any other type of criticism or analysis. As such, it is perhaps worthwhile that they be re-legitimized as an end in themselves, not simply as a means to an end. What this means in practical terms is that editions and commentaries should be viewed as interpretative works like any other, rather than solely as reference works answering a circumscribed set of questions, and that any notion of a standard or definitive edition be largely abandoned.

Turning to editing and commenting on fragments in particular, editions of fragments tend, on the one hand, to reinforce this notion of the one authoritative edition in that such editions appear seldom, and only rarely more than once in a generation or so. While this is natural enough, given that fragments are often on the margins of mainstream work, editions of fragments can, on the other hand, work against this trend and help to re-legitimize editions and commentaries as a means of engaging with texts as valid as any other, largely because the nature of fragments is such that often they are best handled via commentaries.

In his commentary on the *Agamemnon*, Fraenkel was explicit that his commentary and its structure was informed by the need to sift through vast amounts of bibliography in order to root out entrenched, but mistaken, ideas and to resurrect illuminating insights that had been forgotten or overlooked. This process is of course a large part of writing any scholarly commentary, although perhaps relatively few have plumbed the depths of previous scholarship to the extent that Fraenkel did, even if complaints in prefaces about mountains of bibliography are something of a trope. But in the case of fragments, the situation is reversed, and it is very often possible to control the bibliography in its

different for several other plays of Aristophanes, numerous tragedies and a variety of other popular texts both Greek and Latin.

entirety if one so desires. The danger in this possibility lies in the fact that an editor is not forced to be ruthlessly selective and so the exercise of critical judgement can easily be side-lined as a range of views are presented as if they are all equally valid. Since the amount of previous work is usually so limited, the temptation to discuss it all and to take it all seriously is strong. In reality, simply because a suggestion is one of a handful instead of one of many does not mean it is necessarily more worthy of discussion. A related issue involves the interpretation of difficult passages. In extant works, particularly those that have been studied repeatedly and in detail for centuries, the basic interpretational possibilities of difficult passages have often been long known, and thus in practice, interpretation can frequently consist of picking from one of a number of opposing viewpoints. In the case of fragments, following this same procedure is a trap that is easy to fall into, but in fact it not infrequently transpires that the best interpretation of a particular passage is one that has not yet been suggested.

Obedience to authority and the great name is a problem in all walks of life, but particularly for the editor of texts, and among editors of tests especially for the editor of fragments. The material is difficult, in most cases only a handful of scholars have worked on it, and this handful often includes some of the greatest names in the history of the field. The temptation to follow them uncritically is easy to give in to, especially when few alternatives have been suggested, but must be avoided. Scholars like Casaubon or Bentley may well be right more often than they are wrong, but that does mean they are incapable of blunders or even stupidity; scholars like van Herwerden or Blaydes may overwhelm their shrewd suggestions with oceans of wild conjectures, but that does not mean their views should necessarily be treated with contempt. Every suggestion should be judged on its own merits regardless of its author; as a piece of advice, this is an old chestnut of seemingly obvious truth, but practice has repeatedly shown that its application is not as easy as it sounds.12 A case in point is the dating of a number of minor comic poets. The chronology of the Greek comic poets generally, as we understand it today, is largely the work of August Meineke together with some minor modifications resulting from epigraphical discoveries

<sup>12</sup> See R. D. Dawe, *Repertory of Conjectures on Aeschylus* (Leiden: E. J. Brill, 1965), pp. 4–7; cf. N. G. Wilson, *Aristophanes Fabulae*, vol. I (Oxford: Clarendon Press, 2007), p. viii.

of the late nineteenth century.13 Meineke judged that a number of poets could not be dated, and a number of others he dated to late in the Hellenistic period. With very few exceptions, this same chronology has been followed uncritically by all subsequent editors, including the justly lauded Kassel and Austin, even in cases where investigation and attention to detail shows that the received dates are almost certainly wrong.14 One strongly suspects that had these dates been suggested not by Meineke, but instead by Kock or Edmonds, the evidence for these dates would have been examined far more carefully and the dates themselves not just argued against but actively ridiculed.

At least as insidious and liable to unthinking obeisance to authority is the issue of presentation and ordering of fragments. Most scholars tend, reasonably enough, to favour inertia and the retention of inherited ordering in the absence of a compelling reason for change. But the ordering of fragments, even when superficially innocuous, has a very real effect on how fragments are understood both individually and in relation to one another and can drive interpretation in directions that are unwarranted.15 A good example is the three fragments of the play *Agroikoi* (*Rustics*) by the comic poet Anaxandrides. These three fragments, all preserved by Athenaeus, were arranged by Meineke and earlier scholars, and thus also by all subsequent editors, in an order that does not reflect their occurrence in Athenaeus but that does seem to form a narrative. The fragments refer respectively to participation in a symposium, a description a previous feast or symposium, and a recollection of heavy drinking. This sequence has often been taken as informing our understanding of the structure of the play: a rustic

<sup>13</sup> Meineke presented his chronological conclusions, together with much of the evidence, as *Fragmenta Comicorum Graecorum*, vol. I: *Historia Critica Comicorum Graecorum* (Berlin: Reimer, 1839); these results of course provided the overarching structure for the remainder of his edition of the comic fragments (as for all subsequent editions aside from that by Kassel-Austin who sensibly used alphabetic order).

<sup>14</sup> For some examples, see Benjamin Millis, 'Post-Menandrian Comic Poets: An Overview of the Evidence and a Checklist', in *The Oxford Handbook of Greek and Roman Comedy*, ed. by Michael Fontaine and Adele Scafuro (New York: Oxford University Press, 2014), pp. 871–884.

<sup>15</sup> Jackie Elliott, 'Commenting on Fragments: The case of Ennius' *Annales*', in *Classical Commentaries: Explorations in a Scholarly Genre*, ed. by Christina S. Kraus and Christopher Stray (Oxford: Oxford University Press, 2016), pp. 136–156 has a good discussion of some of the issues and interpretive problems that can result even from an entirely reasonable organization of fragments as well as thoughtful consideration of a number of other issues that arise when editing fragments.

participates in a symposium, later in the play he looks back to it and describes it to someone and then in the same or a later scene also discusses the effect the drinking had on him, possibly realizing that city life is not for him. This interpretative sequence appears to make sense and is superficially attractive. But one main stumbling block is that the same series of events is thus both acted out on stage (fr. 1) and related, apparently in some detail, in a narrative description (fr. 2). Since this is inherently unlikely, the standard interpretation that links the three fragments in a clear narrative sequence is almost certainly incorrect and obviously so; ordering of the fragments so as to suggest this interpretation, and then retaining this order in obedience to tradition, has worked mainly to reinforce for centuries an unlikely interpretation.16

These three fragments also incidentally exemplify a related pitfall to be guarded against constantly, namely the compulsion to take the meagre snippets of information that survive and try to combine them into a coherent narrative. Attempting to reconstruct the plots of lost plays, a path that is easy to be tempted onto and difficult to withstand, has traditionally formed a large part of work on fragments yet has served in many ways more to obscure difficulties than to elucidate real problems. While it is true that a group of fragments that seems to form a coherent narrative sequence could reflect in some way the overall structure of a work with reasonable accuracy, it is at least as likely that either such a sequence comes from a single scene that has no larger structural significance or the sequence is in fact illusory and is composed of fragments that belong to widely divergent parts of a work and thus have no real relation to one another in terms of plot. The fact of the matter is that being able to shoe-horn most, if not all, fragments of a given play into an intelligible sequence with structural significance for the plot is no guarantee that such an interpretation is correct and, somewhat counter-intuitively, the more loose ends that can be absorbed, the less likely the over-all interpretation is likely to be true.17

<sup>16</sup> For discussion of the relationship between the three fragments, see Benjamin Millis, *Anaxandrides* (Heidelberg: Verlag Antike, 2015), p. 35 (pp. 36–48 for the fragments themselves in greater detail); the traditional order of the fragments was retained there also, although a better decision would have been to print them in the order in which they appear in Athenaeus.

<sup>17</sup> This issue has come much more to fore recently, particularly in the work of S. Douglas Olson, e.g. 'Athenaeus' Aristophanes and the Problem of Reconstructing Lost Comedies', in *Fragmente einer Geschichte der griechischen Komödie / Fragmentary History of Greek Comedy*, ed. by Stylianos Chronopoulos and Christian Orth (Heidelberg: Verlag Antike, 2015), pp. 35–65 (see esp. pp. 47–48);

A third issue, perhaps the most difficult of all, that the editor of fragments faces is the trustworthiness of his sources. We normally have such little information that any pointer given by an ancient source is grasped with the alacrity of a drowning man seizing a raft. The problem is that sources can be misleading, are only as good as their sources, and can even actively misrepresent what little knowledge they actually have. A good example is the quotation of several comic fragments in the twenty-eighth oration of Aelius Aristides. Aristides quotes two lines from a comic poet, noting that they came from the beginning of the play, presumably meaning the prologue. He then goes on to quote another line that he says came from the end of the poet's play *Cheirons*. 18 Aristides' phrasing implies that the two quotations probably came from different plays, although that is perhaps not quite certain, but clearly seems to indicate that both quotations belong to the same poet. We thus have here what appears to be specific and exceedingly valuable information concerning the placement of several quotations within a play or plays by a learned, well-connected, major cultural figure who lived at a time when Greek theatre was probably still a living phenomenon and when copies of plays by various authors should have still been available, at least for someone with the right connections. Unfortunately, this seemingly invaluable information begins to fall apart as soon as it is examined at all closely. First, both the metre and the content of Aristides' first quotation dictate that it must belong not to a prologue or elsewhere near the beginning of a play, but to the parabasis; any other conclusion is contrary to everything we know about such things. Perhaps even more seriously, while the second quotation seems to be convincingly attributed to Cratinus' *Cheirons* on the basis of the title, a marginal note informs us that the first quotation actually comes from Eupolis' *Marikas*. The most plausible explanation for at least part of Aristides' confusion is that he had no first-hand knowledge of the plays at all, despite his implication to that effect, but instead was working from an anthology of some sort from which a lemma had fallen out, thus enabling the conflation of quotations from two different plays by two different authors. Also worth noting is that Aristides, and thus presumably his audience, had difficulty distinguishing a prologue from a parabasis. In this particular case, we are fortunate in having information that allows us to check Aristides'

cf. 'On the Fragments of Eupolis' *Taxiarchoi*', in *Studi sulla commedia attica*, ed. by Matteo Taufer (Freiburg im Breisgau: Rombach, 2015), pp. 201–213 (see esp. pp. 207–209).

<sup>18</sup> The two fragments are in fact Eup. fr. 205 and Cratin. fr. 255 respectively.

assertions; in the vast majority of cases where we have no such information, similar errors must lurk, but we have no means of identifying them. Perhaps even more worrying is that bad information is presented with absolute assurance, and derivative knowledge is presented as if it were the result of first-hand acquaintance with ancient texts.

A common theme linking the first part of my paper, dealing with commentaries more generally, and the second part, looking a few specific problems more closely aligned with editing fragments, is the problem of authority, both ancient and modern, and the pressures one faces from it. It is very difficult to break away from these pressures, which tend to drive interpretation, define the scope of the work, and determine the approach taken. As a result, commentaries can often be reactive, that is responding largely to a set of traditional questions or adhering to a normalized approach. For example, the language of Sophocles or Euripides is not necessarily more interesting or worthy of study than that of Thucydides or Plato, but the traditions of commenting that have developed, particularly over the twentieth century, mean that such investigations are largely side-lined in the case of historians and other prose authors.19 This in turn has meant a narrowing of approach to many authors. Although it is not often explicitly acknowledged, editions and commentaries wield enormous power not only in setting the agenda for what sorts of questions are asked of authors but even for what authors are studied at all seriously. It is thus incredibly important that commentators make every effort not to be led by their predecessors but also to move away from the idea of the 'standard' commentary, since this implies that there is a circumscribed number of questions to be dealt with and only one valid approach.

Good commentaries are in many ways about asking questions of the text and expanding the ways of looking at both individual problems and a text as a whole. In this regard, some of the most exciting commentary work being done today is on fragmentary texts because the model of how to write a commentary on such texts is still changing and being developed and so has not ossified into a set approach. The edition and commentary as a form of scholarly engagement with ancient texts has a very long history, has led to great advances in knowledge, and still has much potential. At the same time, it requires much imagination,

<sup>19</sup> For a sketch of the development and norms of commenting on historical texts, see John Davies, 'The Historical Commentary', in *Classical Commentaries: Explorations in a Scholarly Genre*, ed. by Christina S. Kraus and Christopher Stray (Oxford: Oxford University Press, 2016), pp. 233–249.

a wide range of skills, and the sort of detailed engagement with the text that is seemingly less and less stressed and valued. Commentators must in many ways strive to be all things to all people, but in doing so we must not lose sight of the real goal, which is furthering knowledge and breathing new life into ancient texts. For many of the problems we face, there is no right answer, or rather no single right answer, and that is why we must constantly examine what we are doing and what we hope to accomplish and, at the same time, allow for a multitude of approaches and aims.

## **Bibliography**


C. Collard and N. J. Richardson (Oxford: Oxford University Press, 2007), pp. xx–xxviii.


# **On editing the reception of Pliny's** *Natural History*

*Cynthia Damon*

## **1. Pathways**

Editing texts is an enterprise that may be compared to pathways connecting discrete and sometimes distant entities: the editor and the text of course, but also, I submit, scholarly generations and fields.1 As the exemplary projects undertaken in the course of the *Ars edendi* program have shown, editorial methods are extremely heterogeneous, appropriately so given the heterogeneous nature of our texts. But it is important to acknowledge as we go about our editorial work that the aims of that work are the same, *grosso modo*: to preserve, understand, and communicate the textual legacy of the past.

Unlike the classicist, the medievalist can say, to paraphrase Andrew Dunning, 'if I need an edition of a particular text to do my work, I just find a good manuscript and make one.'2 Classicists meanwhile are weighed down by the legacy of what was once a great strength of the field, the existence of critical editions, and usually more than one, for the majority of our texts. These editions can be still be improved upon in significant ways − Richard Tarrant gives a list of the some of them

#### **How to cite this book chapter:**

This lecture was given at the *Ars edendi* concluding conference 18 August, 2016.

<sup>1</sup> Cf. Elisabet Göransson, 'Networking the Case Studies: Editorial Methods and the Editorial Circle Model', in *The Arts of Editing Medieval Greek and Latin: A Casebook*, ed. by E. Göransson *et al*. Studies and Texts 203 (Toronto: PIMS, 2016), pp. 400−429, p. 403 ('the link'), p. 427 ('mediation'), and esp. p. 428 ('build a bridge between the complex material and the reader by means of an edition').

<sup>2</sup> 'Where a printed version of a medieval text does not allow me to answer my research questions, my solution has been to create a new edition of a significant manuscript, giving credit to the intellect of earlier editors by citing their corrections.' Quotation from his blog post of 7 May 2015, 'Networking Scholarly Editions,' https:// medium.com/@dunning/networking-scholarly-editions-ec38fdefef0f (accessed 17 April 2020).

Damon, C. 2020. On editing the reception of Pliny's *Natural History*. In: Kihlman, E. and Searby, D. (eds.). *Ars Edendi Lecture Series. vol. V*. Pp. 29–66. Stockholm: Stockholm University Press. DOI: https://doi.org/10.16993/bbd.d. License: CC-BY

in 'The future: Problems and prospects,' the final chapter of his recent book on Latin textual criticism3 − but the bigger challenge confronting us at the moment is to get those critical editions into digital form so that they can join the various modern scholarly conversations about texts and literature. And that brings us back to the pathways that enable such conversations. In the present paper I proceed along some still quite rickety pathways in the hope that eventually a proper infrastructure will be built.

During the past ten years or so I have been on a steep learning curve. As a graduate student I decided not to follow my dissertation supervisor, Ted Courtney, into the area of textual criticism. He had made a distinguished career as an editor of classical texts, but at that time − the 1980s − the job market for classicists did not look with favor on that sort of work.4 As the years passed, however, and I became increasingly aware of the new digital ecosystem into which classical texts were being migrated, I realized that my pragmatic graduate-student decision had left me woefully underequipped to contribute productively to this movement. So, I decided to learn something about textual criticism, and to look for allies in the quest for better texts.

Phase one of this assignment took a very traditional form: I did an old-fashioned edition for the old-fashioned Oxford Classical Texts series. It could hardly have been more retro: the text was a military narrative by Julius Caesar, and the type of edition that this tradition called for was classically Lachmannian, stemma and all. It was good for the elementary education I needed, but the further I got into the project the more I could see of the broader editorial landscape.

Phase two of the assignment accounts for a much steeper part of the aforementioned learning curve. This is the part concerned with finding allies who are themselves eager to obtain and if necessary, produce good digital editions of works with transmission histories comparable to those of the classical texts I study. As a practical matter I

<sup>3</sup> R. J. Tarrant, *Texts, editors, and readers: Methods and problems in Latin textual* 

*criticism* (Cambridge: Cambridge University Press, 2016), p. 145. 4 Twenty years later the picture remained much the same, see D. Shanzer, 'Editions and editing in the classroom: A report from the mines in America,' in *Vom Nutzen des Edierens: Akten des internationalen Kongresses zum 150-jährigen Bestehen des Instituts für Österreichische Geschichtsforschung, Wien, 3.−5. Juni 2004*, ed. by B. Merta *et al*. (Wien: Oldenbourg, 2005), pp. 355−368. It would be nice to think that the push for digital editions will revive the appreciation of editorial work in the field of Classics. For some hopeful signs see http://digitallatin.org/ (accessed 17 April 2020).

am concentrating on texts written in Latin, but that limitation is not inherent to the project itself. The text I have chosen as a focus for this phase is Pliny's *Natural History*, which seems to me both interesting in itself and, in view of its robust reception history, a promising source of alliances. The *Natural History* is also a text that challenges familiar ideas about what a text is, and such challenges have always been a stimulus to editorial innovation. Plus, in following the traces of Pliny's *Natural History* through the centuries I came upon an admirable recent edition of another text with a large debt to Pliny, so I decided to use that edition to explore the question of how best to reveal the echoes of and transformations to Pliny's text.

More precisely, I will look at what happens to the ninth book of Pliny's *Natural History*, the fish book, in the *Speculum naturale* of Vincent of Beauvais, where fish are the topic of the seventeenth book. This will illustrate, I hope, the value of networked and dynamic information for the study of reception and sources, since it addresses the challenge of coping editorially with the presence of texts within texts, the topic of a number of stimulating papers in the *Ars edendi Casebook*. 5 Pliny's *Natural History* is of course a prime example of a text that is built of material taken from earlier texts, but relatively few of his sources survive intact. The *Natural History*, by contrast, and despite a complicated transmission history, is basically here with us today, standing ready to shake hands across the centuries with texts that reproduce and repurpose its words. The question before us is, how can we make the contact between texts as informative as possible? What sorts of pathways do we need?

<sup>5</sup> *The Arts of Editing Medieval Greek and Latin: A Casebook*, ed. by E. Göransson *et al*. Studies and Texts 203 (Toronto: PIMS, 2016). The challenge is of course relevant well beyond the encyclopedic sphere, and a variety of approaches to satisfying it have already been devised. See, for example, for an encyclopedic text Ilse De Vos *et al*., eds, 'L'art de compiler à Byzance: La lettre Γ du Florilège Coislin', *Byzantion* 78 (2008): 159−223, and *eadem*, 'La lettre B du Florilège Coislin: editio princeps', *Byzantion* 80 (2010): 72−120); for an epistolary text, Jeffreys, Elizabeth, and Michael Jeffreys, eds, *Iacobi Monachi Epistulae*. Corpus Christianorum Series Graeca 68 (Turnhout: Brepols, 2009) with discussion by Elizabeth Jeffreys in 'Tapestries of quotation: The challenges of editing Byzantine texts', in *Ars Edendi Lecture Series, Volume II*, ed. by A. Bucossi and E. Kihlman. Acta Universitatis Stockholmiensis: Studia Latina Stockholmiensia LVIII (Stockholm: Stockholm University, 2012), pp. 35−61, and by Macé, C, 'Rules and guidelines in book series and their impact on scholarly editions', in E. Göransson *et al*., pp. 248−266. Other papers in *The Arts of Editing* that deal with the editorial challenge of embedded texts include those by Andrée, Bucossi, Crostini, and Hicks.

## **2.** *Hortus sanitatis ... Speculum naturale ... Naturalis historia*

The work that will furnish my point of departure is the 2013 edition of *De piscibus* by Catherine Jacquemard and a team based at Caen.6 The *De piscibus* is the fourth book of an 8-book compendium called the *Hortus sanitatis* printed in Mainz in 1491. The *Hortus* was originally given to the world without an author's name, and its modern editors describe it as a last belated flowering of the encyclopedic urge that peaked some two centuries earlier.7 In book form the edition is impressively learned and aesthetically pleasing. The Latin text is supported by notes on the text, notes on the content, and a facing-page French translation. The layout is clear and spacious, and the editors incorporate the woodcut illustrations that appeared on almost every page of the *Hortus* into the presentation and analysis. There is also a long and helpful introduction. I have found only one brief published review, but I will stick out my non-medievalist neck and say with that reviewer that it is an admirable edition: 'claire, complète, et prudente, un modèle du genre.'8 The print volume is only the half of it, for the material is also published online in a version that offers additional functionalities, including one that was very helpful for me in trying to track down Plinian material in the *Hortus.*<sup>9</sup>

The *Hortus* is basically a compilation of texts that were themselves compilations. For the fish of Book 4, the most important source was a compendium produced in the mid-thirteenth century, the *Speculum naturale* of Vincent of Beauvais. Both points are made with exemplary clarity in the 2013 edition. In the print edition of the chapter on snails, *cochleae*, for example, the Latin text and a woodcut illustration occupy the right-hand side of page 197, the references to Vincent and Pliny the

<sup>6</sup> Jacquemard, C. *et al*., eds, *L'Hortus sanitatis*, Livre IV: *Les poissons* (Caen: Presses Universitaires de Caen, 2013).

<sup>7</sup> Jacquemard, *L'Hortus sanitatis*, pp. 46–47.

<sup>8</sup> Van den Abeele, B., Review of Jacquemard, C. *et al*., eds, *L'Hortus sanitatis*, Livre IV: *Les poissons* (Caen: Presses Universitaires de Caen, 2013), *Scriptorium* 69 (2015), 182−183.

<sup>9</sup> https://www.unicaen.fr/puc/sources/depiscibus/accueil (accessed 17 April 2020). The planning and work that went into the website are well documented: Jacquemard, C., and P.-Y. Buard, 'Le projet Ichtya entre édition critique et constitution de corpus', *Schedae*, prépublication 14, fascicule 1 (2011), 139−152, Gauvin, B., *et al*., 'Emprunts, compilation, et réécriture dans l'*Hortus sanitatis*,' *Schedae* prépublication 1 (2011), 1−21; Jacquemard *et al.*, 'L'*Hortus sanitatis*: Transmission et réorganisation de la matière encyclopédique au XVe siècle', *Revue d'histoire des textes*, n.s. VII (2012): 353−369; and P.-Y. Buard, *Modélisation des sources anciennes et édition numérique* (PhD diss. Université Caen, Normandie, 2015).

left, and the text of Pliny excerpted by Vincent the footnotes. The online version of the page is a little different: the references to Vincent and Pliny are now hidden behind the siglum [.] and only appear if you hover over it.10 This version of the page looks less like a critical edition and more like the original *Hortus*, which copies but does not cite Vincent and cites but does not reproduce Pliny.11 The fish book of the *Hortus sanitatis* comprises 106 chapters about specific fish arranged in rough alphabetical order. To make a very long story very short, the compiler of the *Hortus* imported Vincent's chapters on specific fish wholesale. The *Hortus* gives the reader information about the source of each datum, but cuts out the middleman, so to speak. One of the many virtues of the 2013 edition of the *De piscibus* is to put Vincent back into the picture after 500 years of erasure.

Now as many a frustrated medievalist knows, it is not easy to work on Vincent of Beauvais.12 The edition that is still the standard for citation today was published in 1624. I will discuss its methodology briefly below, but for the moment it is enough to say that it lacks some crucial features of a modern critical edition, even if it is available online in searchable form.13 The 2013 *De piscibus* edition, as I said, puts Vincent back into the picture by providing references to the *Speculum naturale*. The edition was designed to interact with the digital version of the 1624 *Speculum* prepared by the Atelier de Vincent de Beauvais.14 This

<sup>10</sup> https://www.unicaen.fr/puc/sources/depiscibus/consult/hortus\_fr/FR.hs.4.23

<sup>(</sup>accessed 17 April 2020). 11 The online version of *De piscibus* also contains page images from two fifteenthcentury editions of the *Hortus*, including this page on the *cochleae*, citing 'Pli.li.ix': https://www.unicaen.fr/puc/sources/depiscibus/ui/images/Template.php

<sup>?</sup>edition=Pruss1&vue=12 (accessed 28 July 2017). 12 For a recent overview of the difficulties see M. Franklin-Brown, *Reading the World: Encyclopedic Writing in the Scholastic Age* (Chicago: Chicago University Press, 2012), pp. xvi−xxii.

<sup>13</sup> Website: http://sourcencyme.irht.cnrs.fr/encyclopedie/voir/133 (accessed 17 April 2020). One can also find manuscripts of the *Speculum* online, including this one from the fifteenth-century: http://daten.digitale-sammlungen.de/~db/0003 /bsb00035779/images/index.html?fip=193.174.98.30&seite=5&pdfseitex= (accessed 17 April 2020).

<sup>14</sup> Jacquemard *et al.*, 'Transmission et réorganisation', p. 353. Here is how the texts on the Sourcencyme site, the *Speculum naturale* among them, are described: 'Les textes ont été enregistrés suivant les éditions ou les manuscrits indiqués. Il s'agit donc de transcriptions, pas de nouvelles éditions du texts. Cependent, des corrections (signalées) ont été apportées, par exemple pour corriger de coquilles, rétablir la graphie médiévale en supprimant les diphtongues appliquées par les éditeurs de l'époque moderne, proposer une meilleure leçon en cas de leçon incompréhensible, etc. Ces aménagements ne disposent donc pas de retourner à l'édition critique ou,

is a very useful text, and especially so in Vincent's book about fish, Book 17, since the *De piscibus* team supplied precise references for Vincent's Pliny extracts. That is as far as the interaction between these two projects has gone to date, so in practice one has to use the two sites in tandem. But this particular pathway is well begun.

Back to tracking down Plinian material in the *Hortus*. With a little persistence one can assemble a chain of texts pertaining to each of about 100 fish. For each fish, the *Hortus* contains a copy, sometimes truncated and occasionally garbled, of the *Speculum* chapter or chapters on said fish, and many of the *Speculum* chapters contain extracts of one or more Pliny passages. Here is the 'chain' for snails, with the *Hortus* at the top, Vincent in the middle, and Pliny at the bottom.

*Hortus sanitatis* 4.23.1: Cochleae aquatiles terrestresque sunt exerentes se domicilio, binaque ceu cornua protendentes contrahentesque. Oculis carent, ideoque corniculis earum praetentant iter.

*Speculum naturale* 17.45.1: Cochleae aquatiles terrestresque sunt exerentes se domicilio, binaque ceu cornua protendentes contrahentesque, oculis carent. Ideoque corniculis earum pretentant iter.

*Naturalis historia* 9.101: In eodem genere cocleae aquatiles terrestresque, exerentes se domicilio binaque ceu cornua protendentes contrahentesque, oculis carent; itaque corniculis praetemptant iter.

In the description of the *Hortus* given above I have simplified the situation slightly, because some of its chapters are based on Albert the Great, not Vincent, and both medieval compendia cite more sources than Pliny, and some of the sources they cite, such as Isidore, also cite Pliny, and so on, but the basic picture of a chain of sources is sound. The details underlying the expression 'extracts of one or more Plinian passages' will occupy us in the fourth and fifth sections of this paper.

# **3. Compiling extracts**

Before I get to the philological details, however, here are some numbers to provide a broad sense of the relationship between the texts of the thirteenth-century Dominican friar Vincent and the first-century Roman administrator Pliny. The first numbers come from a chart published by Baudouin Van den Abeele in 1997 as an appendix to his paper on the

le cas échéant, à l'édition ancienne ou manuscrit.' (http://sourcencyme.irht.cnrs.fr /encyclopedie/liste [accessed 17 April 2020]).


**Figure 1.** Citations of *NH* books 2–37 in the Speculum naturale, using the formula "Plinius libro ... ."

sources of the animal lore in the *Speculum naturale*. 15 His chart shows the number of extracts per source in *SN* 16−22, which cover the animal kingdom, or the products of the fifth and sixth days of creation, to use Vincent's narrative framework. Pliny's 979 extracts are more than twice as numerous in these books as those of the next most important source, the *Liber de natura rerum* by Vincent's contemporary Thomas of Cantimpré, and nearly three times as numerous as the 376 citations of Aristotle in his Latin guise.16 Other classical authors are far down the list, including poor Cicero with exactly one. A comparable chart is given in the 2013 edition of *De piscibus*, showing that Pliny's predominance persists two centuries on.17 Equally impressive is the coverage of Vincent's Pliny extracts, which I summarize for the *Speculum naturale* in Figure 1.

Vincent cites the great majority of Pliny's thirty-six content books with markers of the form *Plinius libro secundo* etc. (Book 37 is the last book of the *Natural History*, and Book 1 contains only a preface and an index.) The only books Vincent ignores are those on geography, Books 3−6. The numbers themselves are not important − or even precise, since

<sup>15</sup> B. Van den Abeele, 'Vincent de Beauvais naturaliste: Les sources des livres d'animaux du *Speculum naturale*', in *Lector et compilator: Vincent de Beauvais, frère prêcheur: Un intellectuel et son milieu aux XIII*<sup>e</sup> siècle, ed. by M. Paulmier-Foucart and S. Lusignan (Grâne: Créaphis, 1997), pp. 127−151, pp. 144−145.

<sup>16</sup> And the real extent of Vincent's debt to Pliny is somewhat larger still, since the works of Thomas and the next most prominent source, Isidore, themselves contain a lot of Plinian material.

<sup>17</sup> Jacquemard *et al*., *L'Hortus sanitatis*, from 'Annexe 2: Sources des citations réferencées dan le *Tractatus de piscibus*, répartation quantitative par auteur': Pline l'Ancien 86, Thomas de Cantimpré 84, Aristote 26.



the actual count is much higher owing to citations introduced by *Plinius eodem* or *Plinius ubi supra*, etc.− but even this glimpse shows that the major fish books, 9 and 32, are well represented in Vincent's *Speculum,*  with 72 and 40 direct references respectively; *NH* 9 treats fish themselves, *NH* 32 discusses remedies derived from fish and other creatures.

The next number-based figure shows the volume and rough distribution of Vincent's extracts from Pliny's fish book. Each cell in Figure 2 represents a chapter of *NH* 9, and the bold-numbered cells represent those that were excerpted in Vincent's fish book, Book 17.

The distribution is somewhat uneven, the more so if one takes into consideration the fact that some chapters − those with larger numbers in bold font − are cited more than once; there is also a noticeable gap about two thirds of the way through the book (Ch. 107−126) to which we will return. Still, it is clear that Vincent's extracts are drawn from a wide range of paragraphs in *NH* 9. In short, Pliny was a major source for Vincent, and the pathway from one encyclopedist to the other seems worth investigating.

### **4. From text to text**

This is easier said than done. I alluded above to the difficulty of working on Vincent's massive *Mirror*, which runs to more than three million words in length and appeared in at least three different versions in the middle of the thirteenth century. Earlier scholars perforce limited themselves to broad-brush characterizations of Vincent's extracts, such as 'citations livrées telles quelles' or 'textes ... reproduits dans les mots mêmes de l'auteur.'18 Such an understanding of the excerption process underpins the online edition that I mentioned above, which at best gives a bare reference to Pliny by book and chapter, and more often than not by book alone. The 'dans les mots mêmes' view also justifies (sort of) the editorial practice of the 1624 edition, which proudly claims on its title page to have vetted Vincent's extracts against their originals and corrected the text where the two did not match, 'the exact wording of each author's statements having been restored' (*suis unicuique autori* 

<sup>18</sup> Van den Abeele, 'Vincent de Beauvais naturaliste', p. 135, and M. Paulmier-Foucart and M.-C. Duchenne, *Vincent de Beauvais et le Grand Miroir de monde* (Turnhout: Brepols, 2004), p. 35: 'la plupart des textes retenus sont reproduits dans les mots mêmes de l'auteur; il ne s'agit pas de résumer, sauf exception pour des textes narratifs, en particulier hagiographiques, et pour les extraits d'Aristote, mais il s'agit bien de faire des extraits, en abrégeant éventuellement selon les règles grammaticales courantes ... .'

*redditis exacte sententiis*).19 But in doing justice to Vincent's sources the editors have overruled Vincent's own editorial decisions. Recent scholars now point to the 1476 edition as being more faithful to Vincent's manuscripts.20 But leaving that can of worms aside, even Vincent's references to the excerpting process suggest that 'telles quelles' and 'dans les mots mêmes' misrepresent the relationship between our two texts.

The clearest evidence comes from the second version of Vincent's introductory *liber apologeticus*. 21 In the *apologia* Vincent speaks with some exasperation, *cum stomacho* he says, to critics who have accused him of adulterating the excerpted texts.22 Specifically, of changing the order of the words and the words themselves. As he explains the process, 'The words have generally been rearranged, and there have been occasional slight alterations to the form of the words themselves, but the authority's opinion is still present.'23 He argues that this approach was dictated by the need to reduce verbiage (*prolixitatis abbreviande*), unite disparate passages (*multitudinis in unum colligende*), and clarify obscurity (*obscuritatis explanande*). His policy, he says, like Jerome's, was not *verbum e verbo* but *ex sensu sensus*, and he is deterred from it

<sup>19</sup> The editor, Balthazar Bellerus, in introducing the book to readers (specifically, to the Benedictine Collegium Vedastinum in Douai), speaks of a *citationum fidelis recensio*, saying that he and the printer collated the *Speculum* text against manuscripts and two printed editions, and that they collated and emended the extracts themselves. How far their diligence − the title-page motto is *labore et perseverantia −* extended has yet to be determined. See also Franklin-Brown, pp. xx–xxii and C. Silvi, 'Citer Pline dans les encyclopédies médiévales: L'exemple des notices zoologiques chez Thomas de Cantimpré et Vincent de Beauvais', Archives internationales d'histoire des sciences, 61 (2011), 27–55, p. 39 n. 50.

<sup>20</sup> For a link to an online copy see note 13 above. *SN* 17 starts on 347v for the index, 348v for the text.

<sup>21</sup> This brief but important *apologia* is the only bit of the *Speculum* to have appeared in a modern edition, that of Serge Lusignan, *Préface au Speculum maius de Vincent de Beauvais: Réfraction et diffraction* (Montreal: Bellarmin, 1979), which presents its first and second versions.

<sup>22</sup> The chapter, entitled 'Apologia de modo excerpendi in quibusdam libris Aristotilis,' focuses on the extracts from Aristotle made by his assistants (*quos nequaquam ego ipse excerpseram, sed a quibusdam fratribus excerpta susceperam*). But that the method was applied more widely is suggested towards the end of the chapter, where he defines the relevant material thus: *de flosculis Aristotilis et de ceteris quos in hoc opere per diversa capitula inserui et propriis actorum nominibus annotavi*. On Vincent's use of assistants see also *lib. apol*. 3, which appears in all versions of the *liber*: *non omnia manu propria, sed pleraque per manus notariorum abbreviavi, ut potui*. For the *liber apologeticus* and its various versions see Lusignan, *Préface*; the text quoted here is taken thence.

<sup>23</sup> *Lib. apol.*10 *ordine plerumque transposito, nonnunquam mutata paululum ipsorum verborum forma, manente tamen actoris sententia.*

neither by pricks of conscience nor by attacks from contemporaries. In fact, he retorts that those who don't approve of his policy should check his extracts against the originals before complaining about them.24 'I'd ask them to read the passages first, and compare them with the originals, and then turn up their noses, if they please.' That is what I propose to do in the balance of this paper, not in order to turn up my nose, but with an eye to how editors might make it easier for scholars to take up Vincent's challenge, '*Legant, obsecro*.'

One last bit of overview, this one designed to convey the scope of the differences that you will find if you read Vincent's extracts against Pliny's original.25 The details are given in Appendix 1, which contains the direct extracts from Pliny's Book 9 in Vincent's *Speculum* that were carried over into the *Hortus*. Graphic variations in the Appendix indicate how much of Vincent's extract differs from Pliny's original and roughly how it differs, with different indicators for (1) variations in word order, (2) additions, usually of connectives and explanatory particles, (3) simplifications to vocabulary or lists, (4) clarifications, (5) alterations of meaning, and (6) readings in Vincent and the *Hortus* that appear in the apparatus but not the text of modern Pliny editions. How many of the eighty-four extracts transmit Pliny's original 'dans les

<sup>24</sup> *Lib. apol.*10 *Legant obsecro prius et cum originalibus suis conferant, ac postmodum si videtur eis despiciant*. Some intrepid scholars have recently done just that, including Christine Silvi, who concludes, apropos of Pliny, that 'l'*HN* n'est pas seulement abrégée, réorganisée, remise en ordre, en un mot réécrite pour être adaptée aux contraintes de la notice, elle est aussi dénaturée, corrompue, d'où des approximations dan les citations, des reformulations parfois maladroites, des fautes aussi' (Silvi, pp. 43–44). Similarly on page 37: 'Citer Pline, ce n'est pas recopier ce que Pline a écrit, c'est d'abord localiser et sélectionner, abréger, synthétiser, dénaturer et corrompre aussi, quand ce n'est pas mettre à distance et finalement dominer l'énorme masse d'informations contenues dans l'*HN* dans le but de les adapter aux contraintes de la notice, élément de base de l'encyclopédie médiévale.'

<sup>25</sup> The broad scope of Vincent's changes also comes across in the summary of Eva Albrecht's 2007 dissertation, *De ontstaansgeschiedenis en de compilatie van het Speculum Naturale van Vincent van Beauvais († 1264)*, published in the online *Vincent de Beauvais Newsletter*: 'Vincent a élaboré ses sources d'une manière tres intensive. Il ne s'est pas limité à simplement copier les citations. Alors qu'il citait régulièrement ses sources d'une manière textuelle, il n'a pas hésité à sauter des passages, à adapter le début et la fin d'un passage au contexte du chapitre concerné, ou à inverser l'order de la source. Si nécessaire il a résumé les sources ou les a paraphrasées. Aux passages un peu obscurs, il a ajouté ses propres explications didactiques. Une fois prêts, les traités pouvaient ensuite être copiés et reliés dans les manuscrits.' (http://www.vincentiusbelvacensis.eu/bibl/recent.html; accessed 17 April 2020).

mots mêmes'? Only four.26 Now let us take a closer look at these differences and how students of Pliny's reception might communicate them.

## **5. Legant, obsecro**

It might seem that a citation of the Plinian original would suffice to reveal the divergence between Pliny and Vincent. But things are rarely so simple. Apropos of the *echeneis*, or remora, for example, Vincent tells us that it is used as an amulet to prevent miscarriages and is therefore preserved in salt (presumably to reduce the nastiness of having a dead fish tied to one's body for months at a time).

*SN* 17.51.2 = *HS* 4.36.10 Plinius libro XXIIo: Echineis vel echinus gravidis adalligatus usque ad maturitatem continet lubricos partus ideoque asservatur sale.

Pliny in Book 22: The echeneis or echinus, when worn by pregnant women as an amulet, keeps fetuses in up to maturity—they are inclined to slip out—and is therefore preserved in salt.

What Pliny says, as the citation supplied by the editors of *De piscibus* tells us, is that the marvel of miscarriage-prevention is an idea purveyed by 'some Greeks,' and that he credits the fish-preserved-in-salt idea to some other sources, who maintain that it induces labor.27

*NH* 32.6 mirumque, e Graecis alii lubricos partus atque procidentes continere ad maturitatem adalligatum, ut diximus, prodiderunt, alii sale adseruatum adalligatumque grauidis partus soluere, ob id alio nomine odinolyten appellari.

<sup>26</sup> The relatively short bits that are adopted without change are: *HS* 4.26.3 = *SN* 17.114.3 = *NH* 9.82 *Draco marinus captus et immissus in arenam cavernam sibi rostro mira celeritate excavat*; *HS* 4.51.3 = *SN* 17.119.2 = *NH* 9.84 *Loligo etiam volitat, extra aquam se efferens*; *HS* 4.67.7 = *SN* 17.78.6 = *NH* 9.160 *Pectines sponte naturae arenosis proveniunt*; *HS* 4.92.3 = *SN* 17.96.2–3 = *NH* 9.143 *Novit torpedo vim suam, ipsa non torpens mersaque in limo se occultat*. The information in Appendix 1 pertains to extracts that appear in the alphabetic list of fishes in *De piscibus*. The amount of alteration here is perhaps higher than in the longer extracts in the earlier part of Vincent's Book 17, but I have not compared the texts there

systematically (see further notes 37, 42, and 47 below). 27 The *De piscibus* editors also supply notes that explicate the confusion between the *echeneis* and the *echinus*, or sea urchin (e.g., 236 n. 1, and 239 n. *d'*, and the relevant entry in the 'Catalogue des créatures aquatiques décrites dans le *Tractatus de piscibus'* (102). Like them, I cite Pliny from the Budé editions of de Saint-Denis (1955, 1966).

An amazing thing that, as I said, some Greeks reported, is that, worn as an amulet, it keeps fetuses in up to maturity when they are inclined to slip out and actually emerging. Other Greeks reported that, preserved in salt and worn by pregnant women as an amulet, it releases fetuses and for this reason is called by another name 'releaser of birth-pangs'.

Comparing the *Speculum* text and the Plinian original, as Vincent asks us to do, we learn that the book number he cited is wrong (32, not 2228), and that he has excised Pliny's Greeks and his amazement at the works of nature, expressed, as often, by the term *mirum*. We also see that Vincent has omitted the labor-inducing function of this fish amulet. Why? Is it uninteresting or otherwise objectionable, and if so, why?29 Or perhaps we should instead take this excision as a testament to the power of etymology, since the extract from Isidore that opens the discussion of the fish in question asserts that the *echeneis* was so called 'from the fact that it holds a ship back by attaching itself' (*SN* 17.49.1: *ex eo quod navem adhaerendo retineat*).30 It makes sense that a ship-delaying fish should delay fetuses, too, but it may have seemed counterintuitive that it should induce labor.31 Another possibility, however, is that Vincent echoes the language of this Pliny passage from Book 32 here (*ad maturitatem* and *lubricos partus* are particularly close) but reproduces the content of a passage to which Pliny gives a back-reference in *ut diximus*. For in his Book 9 discussion of the remora Pliny reported miscarriage prevention but not labor induction:

*NH* 9.79 est paruus admodum piscis adsuetus petris, echeneis appellatus ... fluxus grauidarum utero sistens partusque continens ad puerperium.

There is a fish, quite small and adapted for rocky areas, called *echeneis* ... stopping the vaginal fluxes of pregnant women and keeping fetuses in until delivery.

Vincent cited the sentence that follows this one earlier in his discussion of the *echeneis* (*SN* 17.49.4 = *HS* 4.36.3 ≈ *NH* 9.79), so in 17.51.2 he

<sup>28</sup> On errors in Vincent's book numbers see Silvi, p. 39 n. 51.

<sup>29</sup> It may (or may not) be relevant that Vincent had earlier reported that snails, taken as food, induce labor (*SN* 17.45.2).

<sup>30</sup> *HS* 4.36.1 = *SN* 17.49.1 *Isidorus: Echeneis parvus et semipedalis pisciculus, nomen sumpsit ex eo quod navem adherendo retineat. Ruant licet venti, seviant procelle. Navis tamen radicata in mari stare videtur nec moveri, non retinendo, sed tantummodo adherendo: hunc Latini remoram appellaverunt, eo quod cogat stare navigia ipsa*.

<sup>31</sup> For a discussion of an entry in the *Speculum* that embraces conflicting information see Franklin-Brown, pp. 223–232 on the *rana*.

may have blended two disparate bits of Pliny. In other words, the juxtaposition of two passages facilitated by the *De piscibus* edition answers some questions about the relationship between Vincent and Pliny, but raises others that require access to more of the source text, and possibly also to more of the target text if one is curious about Vincent's attitude to Greeks, or the marvels of nature, or abortifacients.

The point is even clearer in a passage from Vincent's chapter on the *squatina*, or angel-shark.

*SN* 17.94.3 = *HS* 4.84.3 Plinius libro XIo: Squatina est ex piscium genere qui pro spina cartilaginem habent.

Pliny in Book 11: The angel-shark belongs to the category of fish that have cartilage instead of a spine.

Vincent credits Pliny's Book 11 with this information, but the citation supplied by the editors of the *De piscibus* is from Pliny's Book 9.

*NH* 9.78 Planorum piscium alterum est genus quod pro spina cartilaginem habet, ut raiae, pastinacae, squatinae [...]

There is another category of flat fish that has cartilage instead of a spine, for example rays, prickly rays, angel-sharks, ...

The verbal alignment is close enough to make the identification of Vincent's source plausible, and the difference in number suggests that there is again an error in Vincent's marker.32 But the Pliny citation has more to tell us: we can see that Vincent has trimmed away references to similar fish at both the front and rear ends of his extract. Pliny's *alterum* links this passage about 'another category of flat fish' to a previous discussion of flat fish, at *NH* 9.72. And the citation ends with a list of *squatina*-like fish, which continues into the editors' ellipsis.33 In fact, if we had the full Plinian context of the citation supplied for Vincent's extract we would see that he has elided six species,34 a reference to

<sup>32</sup> It might be also useful to have easy access to *NH* 11, if only to assure oneself that

there is nothing relevant there; the book is in fact about insects. 33 For another example, illustrating Vincent's extraction of the mouse from a Plinian list of animals that are neither wild nor tame, see Silvi, pp. 42–43 on *SN* 19.126, adapting *NH* 8.22–21.

<sup>34</sup> A fuller version of the Pliny passage is given earlier in *SN* 17.19 on *de diversis figuris piscium*, which is part of the general discussion of marine creatures that precedes Vincent's alphabetic list of specific fish. Here too the reference to Aristotle and Pliny's coinage are missing, as they are in a third related passage, *SN* 17.109.3 on the dolphin.

a terminological innovation by Aristotle, and a Latin coinage offered rather coyly by Pliny, who seems unwilling to be outdone by the Greek naturalist.

*NH* 9.78 Planorum piscium alterum est genus, quod pro spina cartilaginem habet, ut raiae, pastinacae, squatinae, torpedo et quos bouis, lamiae, aquilae, ranae nominibus Graeci appellant. quo in numero sunt squali quoque, quamuis non plani. haec Graece in uniuersum selaches appellauit Aristoteles primus hoc nomine iis inposito. nos distinguere non possumus, nisi si cartilaginea appellare libeat.

There is another category of flat fish that has cartilage instead of a spine, for example rays, prickly rays, angel-sharks, the electric ray, and those that the Greeks call 'the ox,' 'the witch,' 'the eagle,' and 'the frog.' Sharks, too, are in this group, although they are not flat. In Greek Aristotle was the first to call these fish as a group the selachians, having imposed this term on them. We are unable to distinguish them—unless we want to call them cartilagineans.

That is, both the word *alterum* and the ellipsis symbol in the citation are tantalizing markers of a text that lives elsewhere.

The ghost of an absent presence also hangs over a textual divergence in the next sentence of the *Speculum*, which supplies a generalization about these cartilagenous fish.

*SN* 17.94.3 = *HS* 4.84.3 Omniaque talia carnosiora sunt, et supina vescuntur.

And all fish like this are rather fleshy, and they lie on their backs to feed.

*NH* 9.78 Omnia autem carniuora sunt talia et supina uescuntur.

All of them are carnivorous and lie on their backs to feed.

According to Vincent, all such fish are 'rather fleshy,' *carnosiora*, whereas for Pliny they are 'carnivorous,' *carniuora*. Vincent deploys this same extract in two other passages (17.19.1, 17.109.3), each time with the reading *carnosiora*. Where did 'rather fleshy' come from? *Carnosus* is a word that both he and Pliny use with some frequency, and Vincent uses it more than he uses *carniuorus*, but is it a deliberate substitution here, or is it the result of transmission problems? For this sentence the location of the 'text that lives elsewhere' is uncertain at present, but its lure is no less strong for that.

Answering the question of how best to effect the confrontation, and engineer the pathway, between passages of the *Speculum naturale* and its sources will be necessary if we want to figure out how to 'edit the reception' of Pliny's encyclopedia or 'edit the genesis' of Vincent's. So, my next examples are meant to suggest some further desiderata.

Here is Vincent on shellfish called *conchae*, including those that produce pearls:

*SN* 17.44.1 = *HS* 4.22.1 Plinius libro IXo: Concharum genera35 firmioris sunt teste, in quibus magna varietas apparet ludentis nature. Tot colorum differentie, tot figure, inter quas principatum tenent margarite.

Pliny in Book 9: Types of *conchae* are characterized by quite a firm shell, in which you can see the great variety belonging to nature at her games. So many different colors, so many shapes! Among the *conchae*, pearls hold first rank.

The concluding relative clause *inter quas principatum tenent margarite* is a bit elliptical, substituting the pearl for its fishy host, but the idea that the pearl-producing oyster holds first rank among these shellfish makes perfectly good sense. The *De piscibus* editors identify two Plinian passages underlying Vincent's *conchae*.

*NH* 9.102 Firmioris iam testae murices et concharum genera, in quibus magna ludentis naturae uarietas tot colorum differentiae, tot figurae.

Murexes and types of *conchae* are characterized by quite a firm shell. In these, [he exclaims] the great variety belonging to nature at her games, so many different colors, so many shapes!

*NH* 9.106 Principium ergo columenque omnium rerum pretii margaritae tenent.

The first and highest rank among all precious objects is thus held by pearls.

From the gap in numbers (102, 106) it is evident that Vincent has skipped three chapters' worth of material between these two sentences.36 What is not immediately obvious is that he has reframed the value-system in which pearls hold first rank. For Pliny, as the intervening context shows, the pearl is the most costly precious object, and he devotes 20 chapters to a diatribe inveighing against their popularity and retailing

<sup>35</sup> In the 1624 edition the text reads *generis*; *genera* is the reading of the 1476 printing of the *Speculum* and of the *Hortus*.

<sup>36</sup> Many of Vincent's 'Pliny' extracts are in fact made up of discontinuous chunks of Plinian text drawn from within a single book or from disparate books. See Silvi, p. 34; she mentions invisible amalgamations of up to eight distinct extracts.

shocking stories about pearl-related extravagance (*NH* 9.104–23).37 Vincent, however, does not find pearls of much interest, at least in the context of shellfish, where this brief extract suffices. But Pliny's pearl diatribe does find a home in the *Speculum naturale*: it is reassigned to the book on stones, Book 8, where a condensed and reworked form of it occupies three longish chapters (*SN* 8.81–83), the last of them entitled 'on the pearl-related extravagance of the ancients' (*de luxuria antiquorum in margaritis*). In the case of this extract on shellfish, then, the 'text that lives elsewhere' revealed by the citations supplied by the *De piscibus* editors lives both in Pliny's *Natural History* (9.103–105, the omitted chapters) and in Vincent's *Speculum* (8.81–83, the chapters in the books on stones). Its absence can perhaps be ascribed, broadly speaking, to the medieval author's relocation of a heatedly rhetorical section of the ancient text. The desideratum here, I would argue, is the ability to find where a particular bit of Pliny ends up. At present one can try to find this out with a word search in the online version of the *Speculum*, but the search only works if you happen to hit on a word that is present in both passages, spelled the same in both passages, and rare enough that you aren't buried by an avalanche of results. The *Speculum naturale* and the *Natural history* are very long texts, after all, roughly 1,200,000 words in the former, 400,000 in the latter.38

Another procedure revealed by juxtaposing Vincent and Pliny is the reuse of a single Plinian passage in multiple contexts, the cells containing large bold numbers in Figure 2 above. Pliny's *NH* 9.57, for example, is pressed into service five times in Vincent's Book 17, and Vincent has five different versions of it. The chapter is something of a grab-bag; its topics are announced in Pliny's index as 'fish that have a pebble in their heads, fish that hide during the winter, fish that are caught only on fixed

<sup>37</sup> Pliny makes a similar point with more circumspection at *NH* 37.204: *rerum autem* 

*ipsarum maximum est pretium in mari nascentium margaritis*. 38 The Zoomathia project, which is in partnership with the *De piscibus* team, is working on semantic annotation that would make such searching more efficient: M. Tounsi *et al*., 'Studying the History of Pre-modern Zoology by Extracting Linked Zoological Data from Mediaeval Texts and Reasoning on It', in *The Semantic Web: ESWC 2015 Satellite Events. ESWC 2015. Lecture Notes in Computer Science*, vol. 9341, ed. by F. Gandon *et al*. (Cham: Springer, 2015), pp. 406–415; Tounsi, M., *et al*., 'Studying the History of Pre-modern Zoology with Linked Data and Vocabularies', in *Proceedings of the First International Workshop Semantic Web for Scientific Heritage (SW4SH 2015), Portorož, Slovenia, June 1, 2015. CEUR Workshop Proceedings vol. 1364*, ed. by A. Zucker *et al*. (http://ceur-ws.org /Vol-1364/), pp. 7–14; and http://www.cepam.cnrs.fr/zoomathia/?lang=fr (accessed 17 April 2020).

dates' (*NH* 1.9 *qui calculum in capite habeant, qui lateant hieme, qui hieme non capiantur nisi statis diebus*). Here is the chapter in full:

*NH* 9.57 Praegelidam hiemem omnes sentiunt, sed maxime qui lapidem in capite habere existimantur, ut lupi, chromis, sciaena, phagri. cum asperae hiemes fuere, multi caeci capiuntur. itaque his mensibus iacent speluncis conditi, sicut in genere terrestrium retulimus, maxime hippurus et coracini, hieme non capti praeterquam statis diebus paucis et isdem semper, item murena et orphus, conger, percae et saxatiles omnes. terra quidem, hoc est uado maris excauato, condi per hiemes torpedinem, psettam, soleam tradunt.

A particularly cold winter is felt by all fish, but especially by those that are thought to have a stone in their heads, for example bass, the *chromis*, the grayling, and porgies. When the winters have been harsh, many are caught blind. Thus in these (i.e., winter) months they lie hidden in caves (as we reported concerning the class of land animals), especially the gilthead and blackfish, and are not caught except on a few fixed days, always the same days; likewise the moray eel and the *orphus*, the conger and perch and all rockfish. People say that the electric ray, the plaice, and the sole hide themselves in the ground—that is, in a depression hollowed out in the sea—through the winter.

The paragraph contains both general information about fish—'all fish (*omnes*) feel a particularly cold winter'—and lots of fish names: *lupi, chromis, sciaena, phagri, hippurus, coracini, murena, orphus, conger, percae, saxatiles omnes, torpedo, psetta, solea*. Vincent uses this extract in connection with three topics in his general discussion of fish at the beginning of Book 1739 and twice apropos of specific fish; most of the fish in Pliny's list do not get their own entries in Vincent's catalog.40 The specific fish are the *lupus*, or bass, and the *pagrus*, seabream or porgy, both of which fall into Pliny's category of 'fish that have a pebble in their heads.' Vincent's first two extracts from *NH* 9.57 basically split Pliny's chapter between them, but they double up on one sentence and garble the ending badly (17.7, 13). The third general extract condenses

<sup>39</sup> The general headings under which Vincent places the extract are 'on their sense perception' (17.7), 'on fishes' changing from one location in the water to another' (17.13), and 'on fishes' variability according to the seasons of the year' (17.14).

<sup>40</sup> Vincent has no *chromis* (but there seems to be a textual problem in Pliny here; see note below), no *sciaena*, no *orphus*. For the *perca* at *SN* 17.78.7 Vincent does not adapt Pliny but writes the text himself: *Actor: Perca est piscis fluvialis varii coloris, cursu velocissimus, squamis et pinnulis acutis et asperrimis armatus, quibus etiam se defendit contra pisces maiores, ne predantes eam invadant: pre ceteris piscibus fluvialibus et stagnensibus infirmis in cibo precipue convenit*.

Pliny's chapter and gets the basic information of the ending right (17.14). The two extracts that are attached to specific fish truncate the passage and are similar but not identical. Apropos of the *lupus* Vincent says:

*SN* 17.65.3 = *HS* 4.54.3 Idem in libro IXo: Pregelidam hyemem omnes sentiunt pisces, sed maxime, qui lapidem in capite existimantur habere, ut lupi et pagri.41 Cumque aspere fuerint hyemes, multi capiuntur ceci.

The same source in Book 9: A particularly cold winter is felt by all fish, but especially by those that are thought to have a stone in their heads, for example bass and porgies. And when the winters have been harsh, many are caught blind.

And apropos of the *pagrus*:

*SN* 17.78.3 = *HS* 4.67.3 Idem libro IXo: Omnes quidem pisces pregelidam hyemem sentiunt, sed maxime, qui lapidem42 habere in capite existimantur, ut lupi et pagri. Cumque aspere fuerint hiemes, multi ex eis ceci capiuntur, sicut de lupo iam dictum est superius.

The same source in Book 9: All fish feel a particularly cold winter, but especially those that are thought to have a stone in their heads, for example bass and porgies. And when winters have been harsh, many of these [*sc.* porgies] are blind when caught, as was already said above about the bass.

The word order in the second passage is different at the beginning, and it includes a back reference to the former passage at the end. As we saw earlier, Vincent tells us in the *Liber apologeticus* that he employed assistants in excerpting material from his sources, and some scholars attribute the sort of variations we see here to extracts by different hands, perhaps even from different manuscripts.43 But I would want to see more examples before drawing any conclusions, since the second passage here is clearly aware of the first.44 Anyway, the desideratum

<sup>41</sup> On variations in this phrase in particular: *SN* 17.7.4 *ut lupi, coracini, sciena, pagri*; 17.14.1 *ut lupi, sciena et pagri*; 17.65.3 *ut lupi et pagri*; 17.78.3 *ut lupi et pagri*. *Coracini* looks like an attempt to repair a problem in the manuscripts, which have, for *chromis*, various misspellings, including *choromis*, *choramis*, *chorami*. *Coracini* are mentioned later in Pliny's paragraph, beyond the end of the extract in *SN* 17.7.4. In *SN* 17.14.1 the problematic word is simply omitted.

<sup>42</sup> The online edition of Vincent has *lapide* here, but the 1624 edition has the correct reading, *lapidem*.

<sup>43</sup> For details see notes 24–26 above, with Silvi, p. 40. 44 One further example can be found in connection with the pearl extract discussed above. The sentence about the pearl's preeminence (*SN* 17.44.1 *inter quas principatum tenent margarite*) appears two more times and in two other forms: at the

suggested by this example is obviously the ability to find out where else a passage is cited.

Juxtapositions can also reveal the origin of errors. For instance, Vincent and the *Hortus* collapse Pliny's information about four fish that breach the ocean's surface into one.45 Is this a deliberate alteration or an unintentional slip? The Pliny passage comes in a series of fish characteristics: fish that change color (*mutant colorem*, 9.81), a fish that flies (*uolat*, 9.82), fish that lack blood (*sanguine carent*, 9.83), and so on. Here is the bit about the fish that fly, along with a typical Plinian addendum about similar behaviors.

*NH* 9.82 Volat sane perquam similis uolucri hirundo, item miluus, subit in summa maria piscis ex argumento appellatus lucerna, linguaque ignea per os exerta tranquillis noctibus relucet. attollit e mari sequipedanea fere cornua quae ab iis nomen traxit.

A fish in fact flies exceedingly like a bird, the swallow-fish, likewise the kitefish. A fish that comes up to the sea's surface is called the lantern-fish for good reason, and by sticking its fiery tongue out of its mouth it sheds light on calm nights. A fish raises foot-and-a-half-long horn-like appendages out of the ocean and takes its name from them.

First Pliny mentions two 'fish that fly,' the *hirundo* or 'swallow' and the *milvus* or 'kite.' Then he mentions a fish, the *lucerna* or 'lantern,' that comes to the surface and sticks its phosphorescent tongue up into the air. And then he mentions a fish named for the sesquipedalian horns that it raises like periscopes; it was presumably the *cornua* fish he mentions at *NH* 32.145. But in Vincent's alphabetical compilation the kite-fish has lost its capacity of flight and acquired a phosphorescent tongue and long horns:

*SN* 17.66.5 = *HS* 4.55.5 Milvus subit in summa maria, piscis ex argumento appellatus lucerna: linguaque ignea per os exerta tranquillis noctibus relucet. Attollit e mari sesquipedalia fere cornua ab hisque nomen traxit.

The kite comes up to the sea's surface, a fish called 'the lantern' for good reason, and by sticking its fiery tongue out of its mouth it sheds light on calm nights. It raises foot-and-a-half long horn-like appendages out of the ocean and takes it name from them.

beginning of 8.81.4 (*principium culmenque omnium rerum tenent margarite precia*), and some way into 8.83.1 (*precipue ergo rerum precium tenent, ut dictum est, margarite*). These earlier passages are closer to the sense of the Plinian original. The latter also contains a back-reference to the former.

<sup>45</sup> The *De piscibus* edition has helpful notes on the confusion of this chapter, which involves more than the *milvus*.

The excerptor chose the boundaries of his extract without regard for Pliny's verb-initial sentence structure, making 'kite' the subject of 'comes up' instead of attaching it to 'flies.' He also altered the syntax of the last clause to make 'kite' the subject of 'takes.' It is possible to grasp the logic of Vincent's alterations from the citation supplied by the editors of the *Hortus*—*NH* 9.82 above—but the point of Pliny's verb-initial sentence structure is much easier to see when the extract is read in its Plinian context, where 'a fish in fact flies' picks up on the preceding 'fish change color.' Furthermore, the comparison between the extracts and the original is unnecessarily difficult in the online version of *De piscibus*, since the extract of Plinian Latin, when revealed by a hovering mouse, obscures the text of the *Hortus* itself. There is a workaround on the page devoted to the 'Répertoire des citations,' where you can see all of the Pliny extracts next to their *Hortus* reworkings, but for those who want to compare the original text and its adaptation the print page is vastly superior.46 However, the print page is also immutable, and in a later example we will see how that can be a problem. But for now I just highlight the desirability of being able to work outwards from a given extract.

In the next example the quoted Pliny passage does not in fact suffice to explain the *Speculum* extract. The passage in question appears in Vincent's chapter on the *lepus*, or 'sea hare,' which shares a *Speculum* chapter with the sea lion. It begins as follows:

*SN* 17.61.1–4 = *HS* 4.48.1–2, 4.49.1–2 Isidorus: Leo marinus dictus est a similitudine terrestris. Ambrosius: Leo terribilis est in terris, sed dulcis in fluctibus. Lepus est animal timidum in terris, in mari formidabile, nam citam et que non facile possit auferri, corruptelam invehit. Isidorus: Lepus a similitudine leporini capitis nuncupatus est ...

Isidore: The sea lion is named from its likeness to the terrestrial lion. Ambrose: The lion is terrifying on land but gentle in the waves. The hare is a timid animal on land, a fearful one in the sea, for it brings a harmful substance that is quick-acting and not easily removed. Isidore: The hare is named from its likeness to a hare's head ...

Now comes the passage I want focus on. Here is Vincent's Pliny:

Plinius: Venena dira non cessant in lepore, qui in indico mari etiam tactu pestilens vomitum dissolutionemque stomachi protinus creat.

<sup>46</sup> The index to the Répertoire is at https://www.unicaen.fr/puc/sources/depiscibus /citations (accessed 17 April 2020).

Pliny: Terrible poisons do not abate in the sea hare, which in the Indian Ocean causes illness by contact alone and immediately induces vomiting and a loosening of the stomach.

And here is Pliny:

*NH* 9.155 Nec uenena cessant dira, ut in lepore, qui in Indico mari etiam tactu pestilens uomitum dissolutionemque stomachi protinus creat.

Not even poisons, terrible ones, are absent, for example in the sea hare, which in the Indian Ocean causes illness by contact alone and immediately induces vomiting and a loosening of the stomach.

Only three seemingly minor alterations have been made to the wording of the Plinian original: *nec* has been changed to *non*, *ut* has been omitted, and the word order has been simplified, replacing a nounadjective hyperbaton with a simple adjective-noun expression. These look like the sorts of changes made by an excerptor to create a standalone factoid about a named entity such as the sea hare. In Vincent's context *venena dira non cessant* means 'terrible poisons do not abate' in the sea hare; the dangerous character of the fish was mentioned in the citation from Ambrose, who said that the hare was a fearful animal in the sea, and Pliny apparently supplies the details: the poisons are persistent and cause vomiting and diarrhoea. But the Plinian context from which this factoid is extracted is centered on the proposition that everything that occurs on land is also found in the sea (9.154 *admodum nihil non gignitur in mari*), including annoying bugs and terrible poisons. That is, the point of Pliny's *nec uenena cessant dira* is that terrible poisons are no less present in the marine environment than in the terrestrial one. The sea hare is then brought on as an example of a terribly poisonous sea creature, *ut in lepore*; Pliny says nothing about the persistence or otherwise of its poisons. At a minimum, then, enough context needs to be supplied to show how the initial connective, *nec*, works in the original passage. Pliny's Latin is famously difficult, and little words like *nec* are essential sign-posts to the underlying train of thought.47 The editors of *De piscibus* do not supply a note about this divergence, but it would be easy enough to extend the citation in an updated version of the on-line edition.

<sup>47</sup> Pinkster, H., 'The language of Pliny the elder', in *Aspects of the language of Latin prose*, ed. by T. Reinhardt *et al*. Proceedings of the British Academy 129 (Oxford: Oxford University Press, 2005), pp. 239–256.

However, in my view links to a good online version of Pliny's text would be a more effective way to supply the context for these extracts. This is hypothetical of course, since there is no good online version of the *Natural History* at present. But naturally I would like there to be one. And the following example, my last, shows that it needs to contain both a text and a critical apparatus.48

Here is the beginning of Vincent's extract from Pliny on the *urtica*, or sea nettle, which is presumably a kind of jellyfish, juxtaposed with Pliny's version. I have underlined some significant textual divergences. (I only discuss the first divergence, but the textual analysis could be extended to the rest of the passage, which is given here for context.)49

*SN* 17.99.2 ≈ *HS* 4.102.3 (much abbreviated50) Idem in libro IXo: Urtica noctu vagatur noctuque mutatur. 51 Carnose frontis52 est: carne vescitur. Vis eius pruritu mordax eademque que terrestris urtice. Contrahit ergo se quam

<sup>48</sup> My example illustrates the value of having easy access to a critical edition of the *Natural History*. I would like to be able to show an example where it would be helpful to have a critical edition of the *Speculum naturale*, but as I said earlier, the standard edition was published in 1624 and its editors do not assert fidelity to Vincent's text as their highest priority.

<sup>49</sup> The first extract in *SN* 17.99 concerns a different creature, the *uranoscopus*, and it is an imperfect digest of information from *NH* 32.69. Here is Vincent's text: *Plinius libro XXXIIo: Uranoscopus vocatur ab oculo quem habet in capite, a quo subter et supra intendit. Alio nomine gallio nuncupatur. Huius sel cicatrices sanat et carnes oculorum supervacuas consumit, ut dicit Menander in comediis*. And here is Pliny's: *callionymi fel cicatrices sanat et carnes oculorum superuacuas consumit. nulli hoc piscium copiosius, ut existumauit Menander quoque in comoediis. idem piscis et uranoscopos uocatur ab oculo, quem in capite habet*. See the notes on *HS* 4.102.1 for the garbling involved in Vincent's *subter et supra* and *gallio*.

<sup>50</sup> The *Hortus* omits much of Vincent's report on the *urtica*, both the problematic phrase *carnosae ... vescitur* and everything after *urticae*: *HS* 4.102.3 *Urtica noctu vagatur noctuque mutatur. Vis eius pruritu mordax eadem quae terrestris urticae*. The two cuts are related, since what follows in Vincent pertains to the foliage-like

nature of the fronds of the *urtica* (cf. *frondem*, *algae vice*). 51 In order to highlight the lemma, *urtica*, Vincent moves the sentence with which Pliny introduces this material to the end of his discussion: *Equidem et iis inesse sensum arbitror, quae neque animalium neque fruticum, sed tertiam quandam ex* 

*utroque naturam habent, urticis dico et spongeis*. 52 *Frons, frontis* is probably not a variant spelling for *frons, frondis*. Vincent repeats a grammarian's rule about forming the genitive of nouns whose nominatives end with -*ns*: normally you take away -*s* and add -*tis*, but *frons, frondis* is an exception: *SD* 2.67.1 *Desinentia in ns, vel rs, vel ls, ablata s et addita tis, faciunt genitivum, ut mons, montis; ars, artis; puls, pultis; exceptis frons, frondis; lens, lendis; glans, glandis; libripens, libripendis*. Forms of both *frons, frontis* and *frons, frondis* occur throughout the *Speculum*.

maxime rigens ac prenatante pisciculo frontem suam spargit, complectens quem devoret. Alias marcenti similis: et iactari se passa fluctu alge vice, contractos pisces contrituque petre scalpentes pruritum invadit. Eademque noctu pectines et echinos perquirit. Cum autem admoveri sibi manum sentit, colorem mutat et contrahitur. Tacta uredinem mittit: paululumque si fuerit intervalli absconditur. Ora ei radices esse traduntur et excrementa per summa tenui fistula reddi.

The jellyfish moves at night and changes by night. It is a fish with a fleshy forehead, carnivorous. Its effect is an itchy sting, the same as that of a nettle on land. Accordingly, it draws itself together, becoming as stiff as possible, and when a small fish swims by it spreads its forehead, surrounding the fish it aims to devour. At other times it resembles something languid and, allowing itself to be tossed on the waves like seaweed, attacks fish that have been gathered in and are scraping away the itch by rubbing against a rock. The same fish seeks out scallops and urchins by night. When it perceives a hand approaching, it changes color and contracts. Once touched, it transmits a burning sensation and, if there a very small gap, hides itself. Their mouths are said to be 'roots' and their excrement to be emitted by a slender tube through their topmost parts.

*NH* 9.146–47 Vrticae noctu uagantur locumque mutant. Carnosae frondis hiis natura, et carne uescuntur. Vis pruritu mordax eademque quae terrestris urticae. contrahit ergo se quam maxime rigens ac praenatante pisciculo frondem suam spargit conplectensque deuorat. 147 alias marcenti similis et iactari se passa fluctu algae uice, contactos piscium attrituque petrae scalpentes pruritum inuadit. eadem noctu pectines et echinos perquirit. cum admoueri sibi manum sentit, colorem mutat et contrahitur. tacta uredinem mittit, paulumque si fuit interualli, absconditur. ora ei in radice esse traduntur, excrementa per summa tenui fistula reddi.

Jellyfish move at night and change their location. Their nature is that of fleshy foliage, and they are carnivorous. Their effect is an itchy sting, the same as that of a nettle on land. Accordingly, it draws itself together, becoming as stiff as possible, and when a small fish swims by it spreads its 'foliage,' surrounding it, and devours it. (147) At other times it resembles something languid and, allowing itself to be tossed on the waves like seaweed, it attacks fish that have come into contact with it as they scrape away the itch by rubbing against a rock. The same fish seeks out scallops and urchins by night. When it perceives a hand approaching, it changes color and contracts. Once touched, it transmits a burning sensation and, if there is a small gap, it hides itself. Its mouth is said to be in a 'root' and its excrement to be emitted by a slender tube through its topmost parts.

If one approaches the opening sentence of Vincent's extract with a modern text of Pliny in hand, it looks like the medieval author has prettied up a plain Plinian expression. In addition to switching the plural to the generalizing singular, an unremarkable change, he has replaced *locumque mutant*, 'they change their location,' with *noctuque mutatur,* 'it changes by night' or more literally 'is changed by night'; the anaphora is nice, but the change of voice makes the meaning objectionably vague. But before floating any theories about stylistic priorities one would be wise to check a critical apparatus for Pliny's text. Here are the notes for the lemmata *locumque* and *mutant* from the Budé and Teubner editions:53


In the surviving manuscripts of the *Natural History* the opening sentence of this extract in fact reads 'jellyfish move by night, by night they change' (*noctuque mutant*). Vincent has repaired Pliny's lacunose syntax—*mutant* wants a direct object—using an equally vague but syntactically complete passive. That is, the anaphora was in the text he inherited. The emendation we read in modern editions, *locumque mutant*, is based on Aristotle's Greek; before Harduin's 1685 edition editors changed the voice of the verb, as Vincent did. To make a long story short, citing the text of Pliny will lead you astray ... unless of course it leads you to a critical edition of Pliny. The editors of *De piscibus* are commendably attentive to variants in the text of Pliny, but they missed this one. As it happens, they are silent about the others underlined above as well, because the compiler of the *Hortus* excised most of Vincent's problem-filled extract. In short, the information in the critical apparatus of Pliny's text is necessary to understand the passage's medieval progeny.

# **6. Getting from here to there**

I have focused here on showing the value of the finest of fine-grained information about our texts. But that is me remaining in my philological

<sup>53</sup> I supply the notes from both editions here since the editors disagree on what to report. The Teubner edition has a negative apparatus, so lemmata are sometimes implicit, as in the entry on *mutant*.

comfort zone. If I look up from page and screen and stare off into space for a moment, I can glimpse the possibility that compilation methods might be one of the areas of 'distant reading' that could enlist texts from a wide range of periods and literatures. For example, it might be interesting to investigate the concept of authority: how authoritative are authorities such as Pliny for the encyclopedists who cite them? How important is the name of the authority? And is this name-dropping or scholarly argument? Another bigger-picture topic is the nature of argumentation in encyclopedic works. Does an accumulation make an argument, and if so, how does it work? In other words, both burrowing down and gazing up are worth doing in our mutual enterprise of bringing texts to readers.

I mentioned at the outset the capacity of editions to create links between editor and text, between scholarly generations, and between fields. The links I had in mind at that point were conceptual, but what I hope for the future of editing, is that the connections enabled by texts and editors acquire substance and durability: that they become infrastructure, so to speak. Not floating piers like those of the lovely installation on Lago d'Iseo shortly before the 2016 Ars edendi conference, which were in the water for a few short weeks, but the Pont du Gard, which one can still traverse after two millenia.

# **Appendix 1**

Passages of the *Hortus sanititatis* taken from the *Speculum naturale* of Vincent of Beauvais and derived from Pliny, *NH* 9, with alterations by Vincent including:


*HS* 4.1.5 (= *SN* 17.29.5 > *NH* 9.60): Apud antiquos *piscis* nobilissimus habitus accipender, unus omnium squamis ad os versis contra AQUAM NANDO meat. Nullo nunc in honore est, quod quidem miror, cum sit rarus inventu. Quidam eum elopem vocant.

*HS* 4.2.2 (= *SN* 17.31.2 > *NH* 9.74): Anguillae octonis vivunt annis. Durant et sine aquis diebus senis aquilone spirante, austro paucioribus. AT HIEMEM in exigua aqua non tolerant neque in turbida. Ideo circa Vergilias maxime capiuntur, fluminibus tum praecipue turbidis. Pascuntur noctibus. Exanimes piscium solae non fluitant.

*HS* 4.2.2 (= *SN* 17.31.2 > *NH* 9.73): **Porro** vice pedum *pinnae datae sunt binae piscibus omnino longis* ut anguillis et congris.

*HS* 4.2.3 (= *SN* 17.31.2 > *NH* 9.160): Anguillae scopulis se atterunt ea**que** strigmenta vivescunt; nec alia est earum procreatio

*HS* 4.2.3 (= *SN* 17.31.2 > *NH* 9.177): Diutius **autem** CETERIS *vivunt anguillae* aquis exemptae

*HS* 4.2.3 (= *SN* 17.32.3 > *NH* 9.155): Aculeo spinae in dorso *obnoxius est ac venenosus*

*HS* 4.4.4 (= *SN* 17.32.3 > *NH* 9.145): *Aries est piscis qui* grassatur ut latro: nunc grandi navium in salo stantium occultatus umbra, si quem nandi voluptas invitet expectat, nunc elato extra aquam capite piscantium cymbas speculatur, occultatusque adnatans mergit.

*HS* 4.14.2 (= *SN* 17.34.2 > *NH* 9.12): **Balaenas** in Gaditano oceano non ante brumam conspici tradunt, condi autem AESTATIS temporibus in quodam sinu placido et capaci, mire gaudentes ibi parere; hoc scire orcas, infestam his beluam et cujus imago nulla representatione alia possit exprimi quam carnis immensae dentibus truculentae.

*HS* 4.14.2 (= *SN* 17.34.2 > *NH* 9.13): Irrumpunt ergo in secreta vitulos**que** illarum aut fetas vel etiamnum gravidas morsu lacerant incursuque ceu Liburnicarum rostris fodiunt. Illae **vero** ad flexum immobiles, ad repugnandum inertes ac pondere suo oneratae, tunc quidem in utero graves pariendive poenis invalidae, solum auxilium noverunt in altum perfugere se**que** toto oceano defendere. Econtra orcae laborant occurrere seque opponere et CAVATIS angustiis trucidare, in vada urgere, saxis illidere. Spectantur hujusmodi proelia ceu ipso mari sibi irato, nullis in sinu ventis, fluctibus vero ad anhelitus ictusque quantos nulli turbines volvunt.

*HS* 4.14.2 (= *SN* 17.34.2 > *NH* 9.16): Ora balaenae habent in frontibus ideoque summa aqua annatantes in sublime nimbos efflant.

*HS* 4.14.2 (= *SN* 17.34.2 > *NH* 9.19): Balaenis branchiae non sunt, **sicut** nec delphinis. Haec duo genera *fistulis* spirant quae ad pulmonem *pertinent*, balaenis a fronte, delphinis a dorso.

*HS* 4.14.3 (= *SN* 17.34.2 > *NH* 9.41): Quae pilo vestiuntur animalia pariunt, ut pristis balaena vitulus.

*HS* 4.20.4 (= *SN* 17.104.4 > *NH* 9.46): in Gange Indie Statius Sebosus haud modico miraculo affert vermes branchiis binis, lx cubitorum, ceruleos, qui nomen a facie traxerunt; his tantas esse vires ut elephantos ad potum venientes mordicus comprehensa manu abstrahant in profundum**.**

*HS* 4.22.1 (= *SN* 17.44.1 > *NH* 9.102): Concharum genera firmioris sunt testae, in quibus magna varietas *appare*t ludentis naturae, tot colorum differentiae, tot figurae.

*HS* 4.22.1 (= *SN* 17.44.1 > *NH* 9.106): *inter quas* PRINCIPATUM tenent margaritae.

*HS* 4.23.1 (= *SN* 17.45.1 > *NH* 9.101): *Cochl*eae aquatiles terrestresque sunt exerentes se domicilio, binaque ceu cornua protendentes contrahentesque. Oculis carent, **ideoque** corniculis *earum* praetentant iter.

*HS* 4.24.2 (= *SN* 17.46.1 > *NH* 9.73): *Conger est piscis longus ut anguilla vel murena*, flexuoso inpulsu corporis *ita mari utens natando ut serpens terra rependo*.

*HS* 4.24.2 (= *SN* 17.46.1 > *NH* 9.185): *Conger et murena* mutuo odio flagrant, caudam inter se praerodentes.

*HS* 4.24.2 (= *SN* 17.46.1 > *NH* 9.57): *Congri mensibus hiemis jacent in speluncis conditi; nec capiuntur nisi diebus aestatis.*

*HS* 4.24.2 (= *SN* 17.46.1 > *NH* 9.73): *Pinnae*, quae vice pedum datae sunt piscibus, binae **sunt** omnino longis, ut anguillis et congris.

*HS* 4.24.2 (= *SN* 17.46.1 > *NH* 9.87): Polippum sua rodere brachia falsa opinio est

*HS* 4.24.2 (= *SN* 17.46.1 > *NH* 9.185): Id enim a congris evenit ei: **nam** polippum lacerant congri. Polippum **vero** locusta pavet, locustam conger.

*HS* 4.24.2 (= *SN* 17.46.1 > *NH* 9.73): IN SICCUM repunt etiam congri.

*\*HS* 4.26.3 (= *SN* 17.114.3 > *NH* 9.82): Draco marinus captus et immissus in arenam cavernam sibi rostro mira celeritate excavat.

*HS* 4.33.3 (= *SN* 17.52.3 > *NH* 9.62): Escaro – vel scaro – principatus nunc datur, qui solus piscium ruminare dicitur herbisque vesci, non aliis piscibus.

*HS* 4.33.4 (= *SN* 17.52.3 > *NH* 9.62): Carpathio mari maxime frequens, promontorium Troadis Lectum numquam sponte transit. Inde **autem** advectos A TIBERIO CLAUDIO PRINCIPE OPTATUS elipertus, praefectus classis, inter Ostiensem et Campaniae oram sparsos disseminavit. Quinquennio fere cura **est** adhibita ut capti redderentur mari. Postea frequentes inveniuntur *in litoribus Italiae*, non ante ibi capti.

*HS* 4.36.3 (= *SN* 17.49.4 > *NH* 9.79): HUNC ARBITRATUR ARISTOTELES PEDES HABERE ita posita pinnarum similitudine.

*HS* 4.39.3(= *SN* 17.54.3 > *NH* 9.81): *Ficis est piscis qui colorem suum* mutat : reliquo tempore candida, vere **autem** varia. Eadem sola piscium ex alga nidum facit atque in **ipso** nido parit.

*HS* 4.40.4 (= *SN* 17.55.4 > *NH* 9.54): Est parvum animal scorpionis effigie, aranei magnitudine; hoc se et thinno et ei qui vocatur gladius, CEREBRO delphini magnitudinem EXCEDENS, sub PINNO aculeo affigit tantoque dolore infestat ut naves saepenumero exiliant.

*HS* 4.40.6 (= *SN* 17.55.6 > *NH* 9.58): Quidam **enim** *pisces aestate impatientes* sexagenis diebus mediis latent fervoribus, ut *glaucus et azellus.*

*HS* 4.43.2 (= *SN* 17.57.2 > *NH* 9.49): Amiam sive hamiam vocant piscem cujus incrementum diebus singulis intelligitur. Cum thinnis haec et pelamides in Pontum intrant gregatim ad dulciora pabula, *cum* suis quaeque ducibus.

*HS* 4.49.2 (= *SN* 17.61.4 > *NH* 9.155): VENENA DIRA NON CESSANT IN LEPORE. Qui in Indico mari, etiam tactu pestilens, vomitum dissolutionemque stomachi protinus creat.

*HS* 4.51.2 (= *SN* 17.119.2 > *NH* 9.83): *Loligo est piscis mollis sanguine carens sicut et sepia. Caput habet etiam, sicut et sepia, inter ventrem et pedes.*

*HS* 4.51.3 (= *SN* 17.119.2 > *NH* 9.83): Sepiae et loligini pedes duo *ex his* longissimi sunt et asperi, quibus ad ora cibos admovent et in *fluctibus* se velut anchoris stabiliunt.

*HS* 4.51.3 (= *SN* 17.119.2 > *NH* 9.84): Loligo etiam volitat, extra aquam se efferens.

*HS* 4.51.4 (= *SN* 17.119.2 > *NH* 9.164): Sepia **quidem** in terreno parit inter arundines *ac in alto* CONSERVATA OVA *edunt loligines.*

*HS* 4.51.5 (= *SN* 17.119.2 > *NH* 9.158): *Item loligines, sicut et sepiae*, coeunt linguis, brachia inter se componentes et in contrarium *natantes*; ore **quoque** pariunt.

*HS* 4.51.6 (= *SN* 17.119.2 > *NH* 9.93): Loligines etiam in litore mirae magnitudinis sunt, sed in mari nostro quinque cubitorum. **Nec ossa nec spinas habent.**

*HS* 4.52.3 (= *SN* 17.63.2 > *NH* 9.95): Locustae *pisces* crusta fragili muniuntur; in eo genere quod caret sanguine latent mensibus quinis. Similiter *et* cancri, qui eodem tempore occultantur.

*HS* 4.52.4 (= *SN* 17.63.2 > *NH* 9.95): Et ambo veris principio senectutem more anguium exuunt renovatione tergorum. Cetera in undis natant ; locustae reptantium more fluitant

*HS* 4.52.5 (= *SN* 17.63.2 > *NH* 9.95): Si nullus ingruat metus, recto meatu, cornibus, quae sunt r*otunditate* praepilata, ad latera porrectis, **et** eisdem erectis in pavore oblique in latera procedunt. Cornibus inter se dimicant.

*HS* 4.52.6 (= *SN* 17.63.2 > *NH* 9.96, 9.4): Vivunt **autem** *hujusmodi locustae* in petrosis locis. **l**n Indico mari locustae quaterna cubita implent.

*HS* 4.52.7 (= *SN* 17.63.2 > *NH* 9.185): Polippum in tantum locusta pavet ut, si juxta *viderit*, omnino moriatur. Locustam **quoque** conger *inimicus lacerat.*

*HS* 4.54.3 (= *SN* 17.65.3 > *NH* 9.57): Praegelidam hiemem omnes sentiunt pisces, sed maxime qui lapidem in capite existimantur habere, ut lupi et pagri. Cum**que** asperae f*uerint* hiemes, multi capiuntur caec*i.*

*HS* 4.54.3 (= *SN* 17.65.3 > *NH* 9.61): Luporum **vero sunt** laudatissimi qui appellantur lanati, candore mollitiaque carnis.

*HS* 4.54.3 (= *SN* 17.65.3 > *NH* 9.168–69): *Meliores sunt lupi* in amne Tiberi inter duos pontes *quam alibi*.

*HS* 4.54.4 (= *SN* 17.65.3 > *NH* 9.162): Piscium lupus bis in anno parit.

*HS* 4.54.4 (= *SN* 17.65.3 > *NH* 9.185): Nigidius est actor lupum mugili caudam praerodere eosdemque mensibus statis concordes esse.

*HS* 4.55.5 (= *SN* 17.66.5 > *NH* 9.82): Milvus SUBIT IN SUMMA MARIA, PISCIS EX ARGUMENTO APPELLATUS LUCERNA: LINGUAQUE IGNEA PER OS EXERTA TRANQUILLIS NOCTIBUS RELUCET.

*HS* 4.55.6 (= *SN* 17.66.5 > *NH* 9.82): Attollit e mari *sesquipedalia* fere cornua AB HISQUE NOMEN TRAXIT.

*HS* 4.56.2 (= *SN* 17.67.2 > *NH* 9.59): Mugilum natura ridetur, in metu capite abscondito totos se occultare credentium. Iisdem **tamen** tanta facilitas ut in Phoenice et Narbonensi provincia, coitus tempore, e vivariis marem, linea longinqua per os ad branchias religata, emissum in mare **eodemque** linea retractum feminae sequantur ad litus rursusque feminam mares partus tempore.

*HS* 4.56.2 (= *SN* 17.67.2 > *NH* 9.29): **In** Narbonensis provinciae et in Nemausiensi agro, *stagno scilicet laterna appellato*, ubi homines DELPHINI SOCIETATE ET MUGILUM piscantur.

*HS* 4.56.2 (= *SN* 17.67.2 > *NH* 9.185): Mugilus et lupus odio flagrant mutuo.

*HS* 4.56.2 (= *SN* 17.67.2 > *NH* 9.54): *Mugiles aliorum piscium vim timentes, super naves exiliunt* et tam praecipuae velocitatis sunt ut transversa navigia interim superjactent

*HS* 4.57.3 (= *SN* 17.74.3 > *NH* 9.186): *Amicitiae* sunt exempla musculus et balaena: quandoque praegravi superciliorum pondere obrutis ejus oculis, infestantia magnitudinem *in* vada praenatans demonstrat oculorumque vice fungitur.

*HS* 4.57.5 (= *SN* 17.74.5 > *NH* 9.63): *Mustelae pisces mensae deputantur; quas inter Alpes lacus* Rethiae Bigrantinus aemulas murenis generat.

*HS* 4.58.3 (= *SN* 17.73.3 > *NH* 9.160): *Animalia marina* quae sunt *durae* testae, ut murices **aut** purpurae, salivario lentore *proveniunt*, sicut acescente humore culices, atque spuma maris incalescente cum admissus est imber.

*HS* 4.58.3 (= *SN* 17.73.3 > *NH* 9.164): Et hae pariunt in vere.

*HS* 4.59.2 (= *SN* 17.71.2 > *NH* 9.72, 9.40): Murena est piscis longus, molli cute intectus**.**

*HS* 4.59.2 (= *SN* 17.71.2 > *NH* 9.57): Hieme latet nec capitur nisi diebus aestatis, sicut conger et omnes saxatiles.

*HS* 4.59.2 (= *SN* 17.71.2 > *NH* 9.71): Exit in terram, sicut polippus. **Huic aemula et mustela**.

*HS* 4.59.2 (= *SN* 17.71.2 > *NH* 9.72): *Quibusdam* piscibus datae sunt **pinnae** vice pedum, *quibusdam vero* nullae sunt, ut murenis, quibus nec branchiae: haec flexuoso impulsu corporum ita mari utuntur, ut serpentes terra ; in sicco quoque repunt.

*HS* 4.59.2 (= *SN* 17.71.2 > *NH* 9.76): Murena quocumque mense parit, cum ceteri pisces statuto tempore pariant. Ova ejus citissime crescunt. *In sicco litore* elapsas vulgus coitu serpentium impleri putat.

*HS* 4.63.2 (= *SN* 17.75.2 > *NH* 9.88): Nautilos, *qu*i ab aliis vocatur pompilos, inter praecipua miracula *maris computatur*. **Nam** supinus in summa aequorum pervenit, ita se paulatim *sub*rigens ut, emiss*a* per fistulam aqua, velut exoneratus sentina facile naviget. Postea, prima duo brachia retorquens, membranam inter illa mirae tenuitatis extendit: qua velificante in *auras*, ceteris brachiis subremigans, media cauda ut gubernaculo se *dirigit*. Ita vadit alto, Liburnicarum gaudens imagine, si quid pavoris interveniat, hausta se mergens aqua.

*HS* 4.66.3 (= *SN* 17.75.2 > *NH* 9.143): *Silicea testa inclusis fatendum est* nullum inesse sensum, *ut ostreis.*

*HS* 4.66.3 (= *SN* 17.76.5 > *NH* 9.160): Putrescente limo *proveniunt* ac spuma circa navigia diutius stantia defixosque palos et *ligna*. Nuper compertum *in ostrearia* humorem fetificum effluere *in modum* lactis.

*HS* 4.67.3 (= *SN* 17.78.3 > *NH* 9.57): Omnes quidem pisces praegelidam hiemem sentiunt, sed maxime qui lapidem habere in capite aestimantur, ut lupi et pagri. Cum**que** asperae *fuerint* hiemes, multi ex eis caeci capiuntur, *sicut de lupo marino dictum est superius*.

*HS* 4.67.7 (= *SN* 17.78.6 > *NH* 9.160): Pectines sponte naturae arenosis proveniunt.

*HS* 4.70.1 (= *SN* 17.79.1 > *NH* 9.142–43): *Pinna est piscis* concharum generis. Nascitur in limosis, subrecta semper nec umquam sine comite, quem pinnocherem vocant, alii pinophilacem, *id* est squilla parva, alibi cancer, dapis assectator. Pandit **autem** se, luminibus orbum corpus intus minutis piscibus praebens. Assultant illi protinus et, ubi licentius audacia crevit, implent eam. *At illa, ore compresso*, quicquid *concluserit* exanimat partemque socio tribuit. Quo magis miror existimasse quosdam aquatilibus nullum inesse sensum.

*HS* 4.74.1 (= *SN* 17.82.1–2 > *NH* 9.126): **Purpuras autem** vivas capere contendunt, quia cum vita *sua* succum illum evomunt et majoribus quidem *conchis* detracta concha auferunt, minores **vero** cum *testis* frangunt, ita demum *rorem* eum EXCIPIENTES TYRII.

*HS* 4.74.1 (= SN 17.82.1–2 > *NH* 9.128): Lingua purpurae longitudine digitali, qua pascitur perforando reliqua conchilia, tanta aculeo duritia. *Aqua dulci enecantur* et sicubi FLUMINI IMMERGUNTUR alioquin captae etiam diebus quinquagenis vivunt saliva sua.

*HS* 4.77.1 (= *SN* 17.86.1> *NH* 9.72): *Rumbus est piscis planus, sicut passer et solea. Hi a passeribus tantum situ corporum differunt. Dexter hic* resupinatus est illis, passeri levus.

*HS* 4.84.3 (= *SN* 17.84.3 > *NH* 9.78): *Squatina est ex piscium genere qui pro spina cartilaginem habent*. Omniaque talia CARNOSIORA sunt, et supina vescuntur.

*HS* 4.84.3 (= *SN* 17.84.3 > *NH* 9.161): Diversa piscium genera non coeunt *insimul* praeter squatinam et ranam. Ex quibus nascitur *piscis* parte priori ranae similis. Et nomen ex utroque compositum apud Graecos trahit.

*HS* 4.84.3 (= *SN* 17.84.3 > *NH* 9.162): *Piscium quidam ter in anno pariunt, quidam bis, vere et autumno; ex planis sola squatina bis,* autumno occasuque Vergiliarum.

*HS* 4.84.3 (= *SN* 17.84.3 > *NH* 9.165): SQUATINA **vero** intra se parit ova praemollia in alium locum uteri transferens ibique excludens. Simili modo **et** omnia quae appellavimus cartilaginea. Ita fit ut sola piscium et animal pariant et ova concipiant.

*HS* 4.86.1 (= *SN* 17.88.1 > *NH* 9.145): Scolopendrae *marinae*, terrestribus similes, quas centipedes vocant, hamo devorato omnia interanea evomunt donec hamum egerant, deinde resorbent.

*HS* 4.86.4 (= *SN* 17.88.4 > *NH* 9.99): *Cancri mortui* in scorpiones *figurantur* in sicco.

*HS* 4.90.1 (= *SN* 17.95.1> *NH* 9.183): Stellam in mari sapientia claros autores video mirari. **Cujus** figura **haec** est, parva admodum caro intus, extra duriore callo. Huic tam igneum fervorem esse tradunt ut omnia in mari contacta adurat, omnem cibum statim peragat.

*HS* 4.92.3 (= *SN* 17.96.2–3> *NH* 9.78): *Torpedo, sicut et squatina, est ex genere piscium* qui cartilaginem habent pro spina. Omniaque talia CARNOSIORA sunt et supina vescuntur.

*HS* 4.92.3 (= *SN* 17.96.2–3 > *NH* 9.165): Torpedo *ova intra se parit praemollia*, in alium uteri locum transferens, *ibique* excludens. OCTOGENOS HABET FETUS.

*HS* 4.92.3 (= *SN* 17.96.2–3 > *NH* 9.143): Novit torpedo vim suam, ipsa non torpens mersaque in limo se occultat.

*HS* 4.102.3 (= *SN* 17.99.2 > *NH* 9.146): Urtica noctu vagatur noctuque MUTATUR. Vis ejus pruritu mordax eadem quae terrestris urticae.

*HS* 4.102.6 (= *SN* 17.99.4 > *NH* 9.145): Vulpes marinae *in periculo capturae* glutiunt *non tantum hamum sed* amplius usque ad infirma lineae quae facile praerodant.

# **Bibliography**

Editions

Pliny:


Vincent of Beauvais:


*Hortus sanitatis*:

Jacquemard, C., *et al*., eds, *L'Hortus sanitatis:* Livre IV*: Les poissons* (Caen, 2013). https://www.unicaen.fr/puc/sources/depiscibus/ (accessed 17 April 2020).

Secondary literature:


# **Beyond the Edition: On the Linguistic Annotation of Vernacular Texts**

*Odd Einar Haugen*

## **1. Annotation as an editorial enterprise**

Preparing an edition of a classical or medieval text is more often than not a long-term project for an editor. The aim of the exercise is the *constitutio textus*, the bringing together of divergent and often contradictory sources ideally into an unmitigated whole. The edition provides a point of entry to the text, making the text accessible in its variety, although at the same time it incorporates the editor's critical understanding of the sources and his or her recommended routes through them. The publication of the edition is not necessarily the end of the editorial work, but beyond this point, the editor will have the company of other contributors in the form of scholars who draw attention to specific aspects of the text by commenting upon it, by annotating it in various ways, or by contextualising it more generally.

The focus of this chapter will be on annotation, and specifically on the linguistic annotation of medieval texts. Although the examples given here are drawn from the Nordic vernaculars, the principles are in the main the same for other language families, and there are in fact several annotation projects that span a wide variety of texts that are diverse with respect to their language, their provenance, their dating and their contents. After an initial discussion of the representation of manuscript text, looking in particular at vernacular texts, the chapter will move on to two central types of linguistic annotation, those of morphology and of syntax. While many projects concerning medieval texts have been annotating morphology, there are far fewer that have included syntactic structures in their annotation.

#### **How to cite this book chapter:**

This lecture was given at the *Ars edendi* concluding conference 18 August, 2016.

Haugen, O. E. 2020. Beyond the Edition: On the Linguistic Annotation of Vernacular Texts. In: Kihlman, E. and Searby, D. (eds.). *Ars Edendi Lecture Series. vol. V*. Pp. 67–88. Stockholm: Stockholm University Press. DOI: https://doi.org/10.16993/bbd.e. License: CC-BY

In a digital workflow, annotation is so interwoven with the encoding of a text that it may be useful to see annotation as an integral part of the whole editorial process. The annotation establishes a link to a number of external resources such as dictionaries and grammars, and it makes the text accessible for a broader user base. As I will argue below, morphological and syntactic annotation is not a strictly linguistic endeavour. It also means that literary and historical investigations of a much wider scope can be conducted.

## **2. Multi-level representation of a text**

Some works are only preserved in a single manuscript, a situation that is probably more common in medieval than in classical literature, and this is certainly the case in Nordic vernacular literature. The majority of the Eddic poems, for example, have been handed down to us in a single manuscript, and the same goes for many other of the most prominent early Nordic works.1 A work preserved in a single manuscript makes life simple for the editor, but whenever a work has been preserved in more than one manuscript, the editor has to decide whether the various manuscripts should be transcribed *in extenso* or whether some of them, perhaps all of them except the *Leithandschrift*, should be reflected solely through variants in the apparatus of the edition.

**Figure 1.** An admonition in the Old Norwegian Homily Book. Copenhagen, The Arnamagnæan Collection, AM 619 4º, f. 8v, l. 23–26.2 Copyright: The Arnamagnæan Collection, License: CC BY-NC-ND.

<sup>1</sup> The major Eddic manuscript is Reykjavík, The Árni Magnússon Collection, GKS 2365 4º (dated ca 1270), often referred to simply as *Codex Regius*. Another example is the translation from Old French of the *lais* by Marie de France, *Strengleikar*, in Uppsala, University Library, De la Gardie collection, DG 4–7 fol (dated ca. 1270). This is the only manuscript of *Strengleikar*, apart from a fragment, Copenhagen, The Arnamagnæan Collection, AM 666 b 4º, which was once part of the same codex.

<sup>2</sup> The text of the Homily Book was edited by Gustav Indrebø, *Gamal norsk homiliebok* (Oslo: Kjeldeskriftfondet, 1931); the passage quoted here is on p. 17, l. 15–19 in his edition. The text of the Homily Book is also available with full colour facsimiles in *The Medieval Nordic Text Archive* at http://clarino.uib.no/menota/catalogue.

As for the actual representation of the source, the text may be represented with various degrees of fidelity, ranging from a close reproduction of its graphical form to an extensive regularisation of its orthography. In the encoding of vernacular sources, I suggest that three focal levels can be identified along this axis: a facsimile level, a diplomatic level and a normalised level.3 This multi-level rendering can be illustrated with a short extract from the Old Norwegian Homily Book, with my rather literal translation into English at the end:

Sꝩa ſem drıupanda [< drıupande] hunang ero ꝩarrar poꝛt kono. 7 bıartara ꝩıð ſmıoꝛꝩı hálſ hænnaɼ. en hınır æfſto lutıɼ hænnaɼ ero bıtrıɼ ſem æıtr. 7 o lyfıan. *ok* hꝩaſſer ſem tíu æggıat ſꝩærð. fǿtr hennaɼ ſtıga nıðr til dꜵuða. 7 lıggıa gꜵtuɼ hennar tıl hælꝩıtıſ.

Sva sem driupanda hunang ero varrar port kono. *ok* biartare [< biartara] við smiorvi háls hænnar. en hinir æfsto lutir hænnar ero bitrir sem æitr. *ok* o lyfian. *ok* hvasser sem tuí [< tíu] æggiat sværð. fǿtr hennar stiga niðr til dꜵuða. *ok* liggia gꜵtur hennar til hælvitis.

Svá sem drjúpanda hunang eru varrar portkonu, ok bjartari viðsmjórvi hals hennar. En hinir ǿfstu hlutir hennar eru bitrir sem eitr ok úlyfjan, ok hvassir sem tvíeggjat sverð. Fǿtr hennar stíga niðr til dauða, ok liggja gǫtur hennar til helvítis.

Like dripping honey are the lips of a harlot, and her throat is brighter than oil. But her end is bitter as wormwood and poison, and sharp as a twoedged sword. Her feet lead down to death, and her road to hell.

This admonition, which ultimately is a faithful rendering of Proverbs 5.3–5, was quoted in Ch. 18 'De castitate' of Alcuin's *De virtutibus and vitiis*, written between 801 and 804.4 Alcuin's immensely popular text was translated into Old Norwegian around 1200 and became part of the Homily Book.

The three levels exemplified above can be seen as three perspectives on the same passage in the text. They differ in what might be termed granularity, i.e. the degree to which they adhere to the source.

<sup>3</sup> Recently discussed in Odd Einar Haugen, 'Levels of Granularity: Balancing Literary and Linguistic Interests in the Editing of Medieval Vernacular Texts,' in *Philology Matters! Essays on the Art of Reading Slowly*, ed. by Harry Lönnroth (Leiden: Brill, 2017), pp. 118–135.

<sup>4</sup> Cf. Paul E. Szmarach, 'A Preliminary Handlist of Manuscripts Containing Alcuin's *Liber de virtutibus et vitiis*', *Manuscripta*, 25 (1981), 131–40. Alcuin's text is published in *Patrologia Latina* vol. 101, cols 613–638. A new edition of the text remains a desideratum.

On the facsimile level, all characters, including diacritical marks and abbreviation signs, are copied in their position along the base line. On the diplomatic level, a smaller number of characters are used, so that allographs, such as the round and the straight 'r', are represented by a single character, 'r', and abbreviations are expanded. On the normalised level, the orthography is regularised according to the standard grammars and dictionaries of the language. The latter level is in many respects unique for Old Norse (i.e. Old Icelandic and Old Norwegian) texts; there are no comparable standard orthographies for most other European vernaculars. Even for medieval European vernaculars without a standard orthography, however, a certain degree of regularisation is not uncommon, such as the introduction of punctuation (often according to modern rules), the capitalisation of proper names and the first word after a full stop, and perhaps the ironing out of minor spelling variations.

It should be underlined that the three levels illustrated here are not variants of the text, since variants would be quoted on the same level of granularity, and there are no variants unless there are at least two manuscript witnesses of the same passage. The levels exemplified here are representations of a single source. They are alternative ways of seeing and representing a specific manuscript rather than a work as such.

As can be seen from the square brackets in the Old Norwegian transcription above, there are several comments and corrections that the editor might wish to add to the text. The first is visible on the facsimile level and points to the fact that the scribe had obviously corrected an 'e' to an 'a' in the third word of the first line. On the diplomatic level, there is another comment regarding the form of the comparative *biartare* 'lighter, more luminous', which is spelt *biartara* in the manuscript. However, this form is wrong according to Old Norse grammars, which specify the plural form of the comparative as *bjartari*, or, in the case of a manuscript with vowel harmony, as *bjartare*. 5 At some point, the editor might want to correct the text here, although not necessarily on the diplomatic level. In fact, this is an early reflection of the merger of the endings of comparatives that took place in Old Norwegian

<sup>5</sup> Vowel harmony was a characteristic trait of Old Norwegian and meant that the height of an unstressed vowel in the ending of a word was controlled by the height of the stressed root vowel. For example, in a word with a high root vowel, such as *líf* n. 'life', the unstressed vowels would also be high, *lífi* and *lífum*, while in a word with a non-high root vowel, such as *lof* n. 'praise', the unstressed vowels would also be non-high, *lofe* and *lofom*.

during the thirteenth century, and should probably be transcribed as such. Finally, the form *tíu* in the manuscript has been analysed as a metathesis of *tuí* 'double', part of the adjective *tvíeggjaðr* 'two-edged'. This correction seems obvious and also helpful for the users of the edition.

One of the strengths of the multi-level type of editing exemplified here is that it allows uncorrected and corrected text to live side by side, and it can also draw a useful distinction between what may be termed scribal interventions, such as the correction *driupande* > *driupanda*, and editorial interventions, such as *tíu* > *tuí*. At the facsimile level, the text is rendered "as is", in an uncorrected state, apart from corrections made by the scribe himself, while on the diplomatic level, and even more so on the normalised level, editorial interventions are allowable. Furthermore, this division into levels has ramifications for the annotation of the text, especially, as we shall see below, with respect to linguistic annotation.

For classical scholars, the focus on minute variation may seem odd. Why immerse oneself in the accidentals of a text, when there are substantives to behold?6 The answer lies in the fact that vernacular texts are important sources for the history of the early stages of a written language. When the provenance and dating of a source have been established, the orthography has its story to tell. Often, it is a conflated story, since the orthography of a manuscript has to be understood in the context of the orthography of the exemplar, the linguistic norm of the scribe, and, sometimes, even the intended audience.7 In order to use the orthographical representation of any vernacular manuscript, these influences need to be identified and isolated. In some cases, this can be done with a high degree of certainty. In other cases, the language of the source appears inconsistent, which usually is understood as a conflict of norms, between the copy and the exemplar, or between the orthography of the manuscript and the internalised orthographical norms of the scribe. A normalisation of the orthography will remove these traces of norm conflict and lessen the source value of the

<sup>6</sup> These terms were introduced in the much-quoted essay by Walter Wilson Greg, 'The Rationale of Copy-Text', *Studies in Bibliography*, 3 (1950–51): 19–37. Although referring to the editing of printed works, the distinction is equally relevant for classical and medieval works, in which the textual variation may be as plentiful and as complex as in printed works.

<sup>7</sup> It has been argued that a number of manuscripts produced in Iceland in the late fourteenth century were written by Icelandic scribes in Old Norwegian orthography since these manuscripts were intended for export to Norway, cf. Stefán Karlsson, "Islandsk bogeksport til Norge i middelalderen," *Maal og Minne* 1979: 1–17.

manuscript. This might be fine for some scholars, but certainly not for those who are trying to extract the linguistic norm from it. In the model discussed above, the facsimile level and to some extent the diplomatic level offer the necessary level of granularity here, while the normalised level moves the text closer to a representation of the work and away from the source itself.

## **3. Encoding procedures**

Most editorial projects nowadays will situate their work in an open, digital environment, encoding their texts in an interchangeable format. In recent years, this has become more or less equivalent with XML, *Extensible Markup Language*, an open and stable format for a variety of texts.8 This format was chosen for the archive that the author of this chapter heads, the Medieval Nordic Text Archive.9 A great number of other text archives also use XML, and many follow the guidelines set up by the Text Encoding Initiative (TEI).10

It has to be admitted that texts encoded in XML look forbidding as they stand. While the raw XML is not intended for most users of an archive, the editor must nevertheless understand and be comfortable with the basics of it. The upside is that when a text has been encoded in an open and to a great extent self-documenting format like XML, it will be accessible to a wide range of users hopefully over a very long period of time. An XML file is a straightforward text file, as simple as they come, and it will be readable as long as basic, unformatted text files can be read.

The *Guidelines* published by the Text Encoding Initiative, now in its fifth version, specify the encoding of a wide variety of sources, prose as well as poetry.11 However, in our experience handwritten medieval sources require a number of additional specifications. *The Menota* 

<sup>8</sup> There are numerous introductions to XML, e.g. "What is XML and why should humanists care" by David J. Birnbaum, http://dh.obdurodon.org/what-is-xml.xhtml. Another introduction is Ch. 2, "Text encoding using XML" in v. 3.0 of *The Menota handbook*, http://www.menota.org/handbook.xml.

<sup>9</sup> Cf. http://www.menota.org.

<sup>10</sup> Among high-profile archives are the British National Corpus, Deutsches Textarchiv, Digitales Wörterbuch der deutschen Sprache, and the Perseus Digital Library of Latin and Greek texts. There are a host of other archives, of early and modern texts, and the list grows continuously. Many of the archives using the TEI guidelines, among them the Menota project, are listed by TEI at http://www.tei-c.org /Activities/Projects/.

<sup>11</sup> *TEI: P5 Guidelines*, https://tei-c.org/guidelines/p5/.

*Handbook* is intended to supplement the TEI Guidelines explaining and exemplifying how XML encoding as recommended by the Text Encoding Initiative can be used for the specific purpose of encoding vernacular medieval documents.12

For a linguistic annotation of a text, there are two categories that need to be clearly marked up: sentences and words. In XML, the <s> element groups each sentence, and the <w> element each word. To this pair of elements, a <pc> element should be added for punctuation characters, such as full stop, comma, colon and the like. Each element states its opening, e.g. <s>, and its end, e.g. </s>. So, this is how a sentence is contained in the <s> element, and the words in the <w> element and punctuation in the <pc> element:

```
<s>
  <w>This</w>
  <w>is<w>
  <w>a</w>
  <w>sentence</w>
  <pc>.</pc>
</s>
```
A multilevel edition can be encoded as parallel readings for each word, using elements such as <facs>, <dipl> and <norm> for the three levels exemplified above.13 The very first word would receive the following encoding:

```
<w>
  <facs>Sꝩa</facs>
  <dipl>Sva</dipl>
  <norm>Svá</norm>
</w>
```
On the <facs> level, the usage of an Insular form of the 'v' is recorded. On the <dipl> level, this character is merged with the ordinary 'v', and on the <norm> level, an accent is added to the vowel to indicate that

<sup>12</sup> *The Menota Handbook* was published in v. 1.0 on 20 May 2003 and in v. 3.0 on 12 December 2019. See http://www.menota.org/handbook.xml for an overview of versions.

<sup>13</sup> Since the elements <facs>, <dipl> and <norm> are not part of the standard TEI repertoire, they have been prefixed with 'me:' (for Menota-specific elements) in the XML encoding, but this prefix has been left out in the examples here for the sake of conspicuity.

it is phonemically long. In order to make the multi-level structure explicit, the element <choice> states that the contents of this element are alternatives:

```
<w>
  <choice>
 <facs>Sꝩa</facs>
 <dipl>Sva</dipl>
 <norm>Svá</norm>
  </choice>
</w>
```
It should be underlined that the encoding examples given here are not meant to be typed, character by character, by a transcriber. They are the representations of transcriptions that would be done by various input methods.14

The actual encoding of a text on several levels is no guarantee that it can easily be displayed in a manner that is accessible for any non-technical reader. In the Menota archive, the display is based on the Corpuscle application.15 As shown in Fig. 2, this application allows a text to be displayed at up to three parallel levels, including a photographic facsimile. The Corpuscle application is used for several other archives, some of which are rather close in scope and format to Menota, such as the Georgian National Corpus, covering Old, Middle and Modern Georgian.16

While the characters on the normalised level and largely on the diplomatic level can be rendered by almost any font, many of the characters on the facsimile level require specialised fonts. Until a few years ago, this meant that users of the archive had to install a font containing the necessary characters. Such fonts have been offered by the Medieval Unicode Font Initiative since 2004, and several of these fonts can be

<sup>14</sup> *The Menota Handbook* exemplifies this in its Tutorial, which was introduced in v. 3.0 of the handbook, https://www.menota.org/HB3\_T1.xml.

<sup>15</sup> The Corpuscle application is a corpus management system for annotated texts developed by Paul Meurer at the University Library, Bergen, cf. http://clarino.uib.no /korpuskel.

<sup>16</sup> The Georgian National Corpus is available at http://clarino.uib.no/gnc/page. For other corpora, see Paul Meurer, 'Corpuscle – a new corpus management platform for annotated corpora', in *Exploring Newspaper Language: Using the Web to Create and Investigate a large corpus of modern Norwegian*, ed. by Gisle Andersen (Amsterdam: John Benjamins, 2012), pp. 29–50.

**Figure 2.** Display of a multilevel transcription of an Old Norwegian law fragment. Oslo, National Archives, Norse Fragment 7, f. 5r, col. A (ca 1320–1350). The three columns render the text on a facsimile, diplomatic and normalised level, based on a single transcription in XML. A photographic image of the fragment complements the display. Copyright: Odd Einar Haugen, License: CC BY-NC-ND.

downloaded, installed and used free of charge.17 In spite of the availability of free fonts, Menota could not offer a "plug and play" solution, and some users were surely put off by missing characters, boxes or question marks in the web display. It was a great step forward when the Web Open Font Format (WOFF) was officially launched in 2012.18 This means that any recent browser can display all necessary characters on the fly, irrespective of which fonts happen to be installed on the user's computer.

Having established this simple (although admittedly verbose) model of representing sentences and words on more than one level in XML, the next question is how to enhance the text with additional information.

<sup>17</sup> The Medieval Unicode Font Initiative was established in 2001 and has published several recommendations for font usage as well as several free Unicode fonts that have a good selection of medieval characters. See http://www.mufi.info.

<sup>18</sup> The Web Open Font Format was developed in 2009 and made a recommendation by the World Wide Web Consortium (WC3) in December 2012, cf. https://www.w3.org /TR/2012/REC-WOFF-20121213/.

# **4. Annotating a text**

There is a plethora of textual features that can be identified and annotated: motives and themes, allusions, rhetorical devices, names of persons and places, stylistic features, metrical properties, allusions to or readings from other texts, and so on. From a linguistic point of view, phonological, morphological, syntactic and lexical features are all relevant, but, in my experience, morphology and syntax are particularly well suited for annotation.

## 4.1 Morphological annotation

A morphological annotation will as a minimum specify the lemma, i.e. headword, of each running word in the text, and usually also the grammatical form. In the Menota XML, this information is added to each word by way of attributes to the <w> element. In the example of the adverb *svá* (as the entry is spelt in an Old Norse dictionary), the lemma attribute is simply *svá*, as shown in a slightly simplified encoding:

```
<w lemma="svá">
  <facs>Sꝩa</facs>
  <dipl>Sva</dipl>
  <norm>Svá</norm>
</w>
```
Adverbs like *svá* are not inflected, so the only grammatical information in addition to the lemma will be its word class (part of speech). In the Menota project, the *msa* attribute (the full form being *me:msa*) specifies the morpho-syntactic analysis of the word. An adverb will receive the value *xAV*, in which *x* signifies word class and *AV* adverb:

```
<w lemma="svá" msa="xAV">
  <facs>Sꝩa</facs>
  <dipl>Sva</dipl>
  <norm>Svá</norm>
</w>
```
Words with more grammatical categories, such as nouns, adjectives and verbs, have a longer list of values for the *msa* attribute, but the principle is the same, so that, for example, the noun *varrar* 'lips' will, in addition to its word class, be annotated for gender, case, number and species (the latter category has two values, indefinite as in *varrar* 'lips' or definite as in *varrarnar* 'the lips'):


**Figure 3.** A morphological analysis of the last sentence in Fig. 1. The first line has the diplomatic reading of each word, the second line the word class, the third line the grammatical categories, and the fourth line the head words in a standard, Old Norse dictionary. Copyright: Odd Einar Haugen, License: CC BY-NC-ND.

```
<w lemma="vǫrr" msa="xNC gF cN nP sI">
  <facs>ꝩarrar</facs>
  <dipl>varrar</dipl>
  <norm>varrar</norm>
</w>
```
The *msa* attribute contains one or more name tokens, each specifying a grammatical category; in this case *xNC* for "noun common", *gF* for "gender feminine", *cN* for "case nominative", *nP* for "number plural" and *sI* for "species indefinite".19

Assuming that each word of a text has received morphological annotation, the information can be displayed in various ways, including tabular displays such as the one in Fig. 3. The actual encoding and the display is more or less self-explanatory and draws on a long tradition of traditional dictionary archives based on the venerable index card.20

It should be pointed out that the usefulness of a linguistic annotation is dependent on the variability of the orthography of the texts. For a literature in which the language of the texts is highly regularised, such as in most corpora of modern texts, a morphological analysis can to a

<sup>19</sup> In v. 3.0 of *The Menota Handbook*, levels are discussed in Ch. 4 and linguistic annotation in Ch. 11.

<sup>20</sup> One of many archives of this type is the *Gammelnorsk Ordboksverk* [Old Norwegian Dictionary Project], which worked with index cards from the 1950s until the early 1980s. Many of these archives have been converted into digital databases, but still present part of their material in facsimiles of the original index card collections, such as the *Ordbog over det norrøne prosasprog* [Dictionary of Old Norse Prose] and *Gammeldansk Ordbog* [Old Danish Dictionary], both in Copenhagen.

high degree be done by semi-automatic analysis. For medieval vernacular texts in which the orthography is variable not only between sources but even within sources, a linguistic annotation really comes into its own. By the same token, the annotation is more time-consuming and less suited for automatisation. The inflection of many Old Norse words illustrates this point, e.g. the verb *verða* 'become', which has these (amongst other) forms:


The initial *v*- is dropped in several forms, and the root vowel shifts between *e*, *a*, *u*, *y* and *o* due to a combination of Ablaut and Umlaut. This degree of variation within a paradigm is known from many other languages, but what really complicates the analysis here is the fact that each form could be spelt in more than one way, sometimes in a frustratingly high number of ways. For example, the regularised form *urðu*, 3rd person preterite indicative of *verða*, might (at least in theory) be spelt *urðu*, *urðv*, *vrðu*, *vrðv*, *urþu*, *urþv*, *vrþu*, *vrþv*, *urþu*, *urþv*, *vrþu*, *vrþv*, *urþo*, *urðo*, *vrþo*, *vrðo*, and more. On the normalised level, there would only be the form *urðu*, but on the diplomatic, and especially the facsimile level, there will be many more forms.

An added difficulty is that while Old Norwegian and Old Icelandic (i.e. Old Norse) have a normalised orthography, this is not the case for Old Swedish or Old Danish. The Old Norse normalised orthography was established in the late nineteenth century and is used with minor variation in standard grammars, dictionaries and in many editions.21 No similar norm exists for Old Swedish and Old Danish, even if Old Swedish texts in particular might be suited for normalisation.22 The Old Danish language is less conducive to normalisation, partly due to the fact that it evolved so quickly in the Middle Ages and partly due to the sparsity of sources up to ca 1300.23 A similar lack of orthographic

<sup>21</sup> Ivar Berg, 'Om normalisert norrønt', *Arkiv för nordisk filologi*, 129 (2014), 21–54.

<sup>22</sup> Henrik Williams, 'Normalizing Old Swedish texts: Why Not?', in *Beyond the Piraeus Lion: East Norse Studies from Venice,* ed. by Jonathan Adams and Massimiliano Bampi (Odense: Syddansk universitetsforlag, 2017), pp. 51–58.

<sup>23</sup> For Old Swedish and Old Danish, see the discussion in Odd Einar Haugen, 'Normalisering av vest- og østnordiske middelaldertekster', in *A Copenhagen Miscellany: Studies in East Nordic Philology*, ed. by Simon Skovgaard Boeck and Seán D. Vrieland (Copenhagen: Syddansk Universitetsforlag, 2019), pp. 161–181.

norm applies to other European vernaculars, and as a consequence, morphological annotation is a desideratum across the board for the European vernaculars.

In spite of these difficulties, morphological annotation of the type discussed above is a fairly simple undertaking and it is not the object of many linguistic controversies. After all, the aim of the annotation is to link texts to existing resources such as grammars and dictionaries, and, consequentially, the grammatical categories will be traditional. There is some variation in matters such as the lemma orthography and the grammatical categories, especially word classes, but at least within Medieval Nordic philology, these problems are relatively small.24

#### 4.2 Syntactic annotation

Syntactic annotation is more of a challenge than morphological annotation. Competing syntactic models have evolved over the years, and there is a varying degree of compatibility between them. For a language of comparatively free word order, such as the Medieval Nordic languages, it seems that dependency analysis is a suitable and fairly simple syntactic model.25 In dependency analysis, each word is described by its function and hierarchical position within the sentence. This is typically displayed in a tree with labels for each word specifying its function, as shown in Fig. 4. It is a characteristic and perhaps unexpected trait of dependency analysis that words rather than phrases are assigned syntactic functions. There are some non-intuitive consequences of this, for example that conjunctions are analysed as heads (as in Fig. 4) and for this reason have full sentences as their dependents. However, the internal hierarchy of the coordinated sentences, each having a predicate as its head, is not affected by the fact that the conjunction has been elevated, as it were, to the position of a head.

While morphological annotation easily can be incorporated in the XML discussed above, syntactic annotation is better carried out in a separate module. The PROIEL project developed at the University of

<sup>24</sup> *The Menota handbook* v. 3.0, Ch. 10 and 11.

<sup>25</sup> For a general introduction to dependency analysis, see Igor A. Mel'čuk, *Dependency Syntax: Theory and Practice* (Albany: State University of New York Press, 1988). An implementation for Old Norwegian is offered by Odd Einar Haugen and Fartein Th. Øverland, *Guidelines for Morphological and Syntactic Annotation of Old Norwegian Texts* (Bergen Language and Linguistic Studies, vol. 4:2, Bergen, 2014).

**Figure 4.** A dependency analysis of the last, coordinated sentence in Fig. 1. The conjunction *ok* 'and' functions as the head of the two coordinated sentences, while each of these has a predicate as its head, *stiga* 'go' and *liggia* 'lie'. The subject of the first sentence is *f*ǿ*tr* 'feet', with *hennar* 'her' as an attribute, and the subject of the second is *gaotur* 'roads' also with *hennar* 'her' as an attribute. The preposition *til* 'to' in both sentences is heading an oblique specifying the goal of the predicate, dominating *daouða* 'death' and *hælvitis* 'hell' respectively, and finally, the adverb *niðr* 'down' specifies the direction of the predicate *stiga* 'go'. Copyright: Odd Einar Haugen, License: CC BY-NC-ND.

Oslo offers exactly this type of annotation environment.26 PROIEL initially undertook a syntactic analysis of the New Testament in five old Indo-European languages, the original Greek and early translations into Latin, Gothic, Armenian and Old Church Slavonic. Through cooperation with other projects, the annotated corpus in the PROIEL format has later been extended to many more languages, among them

<sup>26</sup> The original project was funded by the Norwegian Research Council in the period 2008–2012 and was led by Dag Trygve Truslew Haug. The core group also included Hanne Martine Eckhoff and Marius Larsen Jøhndal.

Old Icelandic, Old Norwegian and Old Swedish.27 While the PROIEL project was originally designed for the study of information structure, all texts were annotated for morphology and on the basis of this also for syntax. The result is a deep annotation with considerable linguistic information about each text, organised in the form of a treebank.28 The original PROIEL treebank project and the subsequent projects now form what may be called a treebank family of early attested Indo-European languages, ranging from classical to medieval, and in some cases modern, stages in their development.29 In total, approx. 1,6 million words have been annotated manually at a high level of accuracy. For Old Norwegian and Old Swedish, there are so far no other treebanks than those in the PROIEL family.30

The dependency model illustrated here is in some respects close to lexical-functional grammar (LFG), but it contrasts with especially phrase structure models. However, as can be seen in the examples here, the functional categories used in dependency analysis are by and large familiar, such as predicate, subject and object, although some categories, especially the obliques and the external objects, are less familiar. Even so, there seems to be a sufficiently high degree of recognition between central syntactic models. The major criterion in such cases is, I believe, that an insight is only a fruitful insight if it can be transferred from one model to another; if not, it may be an insight solely into the model, not

<sup>27</sup> These projects include the ISWOC project for Old English and several Romance languages, the TOROT project for Russian, the Menotec project for Old Norwegian, and the MAÞIR project for Old Swedish. PROIEL has also added many Greek and Latin texts to the original New Testament texts. The exact number of languages depends on the classification; five major language families are represented, i.e. Armenian, Germanic, Greek, Romance and Slavic, and all but Armenian have several branches. The total of languages (or linguistic stages) covered by PROIEL is 18.

<sup>28</sup> *Treebank* is a term that reflects the fact that a syntactic analysis typically takes the form of a tree, and that a collection of such trees make up a bank. The term gained usage in the 1990s after the Penn Treebank Project was established in 1989. Cf. Mitchell P. Marcus, Beatrice Santorini and Mary Ann Marcinkiewicz, 'Building a Large Annotated Corpus of English: The Penn Treebank', *Computational* 

*Linguistics*, 19/2 (1993): 313–30. 29 Hanne Martine Eckhoff, Kristin Bech, Gerlof Bouma, Kristine Gunn Eide, Dag Trygve Truslew Haug, Odd Einar Haugen and Marius Larsen Jøhndal, 'The PROIEL treebank family: A standard for early attestations of Indo-European languages', *Language Resources and Evaluation* (2017): 1–37, https://doi.org/10.1007 /s10579-017-9388-5.

<sup>30</sup> The Old Norwegian texts are are part of the Menotec collection accessible through the INESS portal at http://clarino.uib.no/iness, and they are in the process of being published on the *Syntacticus* website at http://www.syntacticus.org.

what the model purports to explain. A case in point is the fact that a dependency tree (with some modifications) can successfully be converted to an LFG representation, and the other way round; in other words, dependency and LFG models are able to express similar analytic insights.31

# **5. The appeal of annotation**

A text annotated for morphology and syntax is indeed a boon for the linguist and the language historian. For many other scholars, for example of a literary or historical inclination, the annotation is one of several resources for a better understanding of opaque or ambiguous passages in a text. The annotated text is a close cousin of the commentary; while the latter can go into extensive detail and list a number of interpretations, the annotated text makes a decision and is usually unambiguous, unless the categories themselves have been designed to be polyvalent.

Old Norse poetry offers a host of complex and enigmatic passages. In the Eddic poem *V*ǫ*luspá* 'The Prophecy of the Seeress', stanza 2.4–6 is still unresolved. As Fig. 5 from *Codex Regius* shows, the poem is written in continuous lines, and the script is somewhat difficult to read here.

**Figure 5.** Stanza 2.5–8 of *V*ǫ*luspá* in Reykjavík, The Árni Magnússon Collection, GKS 2365 4º, f. 1r, l. 4–5. The words "nio man ec heima" can be read (with some difficulty) at the end of the first line in the photograph (i.e. l. 4 in the manuscript), while the remaining part follows on the next line, "nio iviþi [?] miot uið mæran f*yr* mold neðan". Copyright: The Arnamagnæan Collection, License: CC BY-NC-ND.

Below, the stanza is quoted in the edition by Gustav Neckel and Hans Kuhn32 and the translation by Henry Adams Bellows.33

<sup>31</sup> Dag [Trygve Truslew] Haug, 'From dependency structures to LFG representations', in *Proceedings of LFG12*, ed. by Miriam Butt and Tracy Holloway King (Stanford: CSLI Publications, 2011), pp. 271–291, and Paul Meurer, 'From LFG structures to dependency relations', in *The very model of a modern linguist – in honor of Helge Dyvik*, ed. by Victoria Rosén and Koenraad De Smedt (Bergen: Bergen Language

and Linguistic Studies, vol. 8, 2017), pp. 183–201. 32 Gustav Neckel and Hans Kuhn, eds, *Edda: Die Lieder des Codex Regius nebst verwandten Denkmälern*, Vol. 1: *Text*. 5th ed. (Heidelberg: Winter, 1983).

<sup>33</sup> Henry Adams Bellows, transl., *The poetic Edda*, Scandinavian classics, vol. 21–22 (Princeton: Princeton University Press, 1936).


1 I remember yet 2 the giants of yore,


A much-debated question concerns the reading and understanding of the phrase "nio iviþi" in stanza 2.6 of the poem. This has been taken as normalised *níu í viði* 'nine in the tree' in many editions, since there commonly was no word division between a preposition and its complement, and *viði* is a *bona fide* accusative of *viðr* m. 'wood, tree'. However, after studying the manuscript closely, some philologists conclude that there is an almost invisible abbreviation character after "viþi", in the form of an *ur* sign. If this is correct, the reading of the word becomes "iviþiur", normalised *ívíðjur*, meaning 'giantesses'. This reading happens to be supported by another manuscript, *Hauksbók* in Copenhagen, The Arnamagnæan Collection, AM 544 4º, f. 20r, l. 3, so even if the reading in GKS 2365 4º is pending, *ívíðjur* 'giantesses' has been adopted in the latest edition of the Eddic poems.34

In the annotated version of this poem, the latter interpretation has been selected, as can be seen from the morphological analysis in Fig. 6. 35


**Figure 6.** The morphological annotation of *V*ǫ*luspá* stanza 2.5–8. Copyright: Odd Einar Haugen, License: CC BY-NC-ND.

As for the syntax of the half-stanza, the analysis in Fig. 7 juxtaposes one full sentence, *níu man ek heima*, with two elliptical sentences, in

<sup>34</sup> *Eddukvæði*, ed. by Jónas Kristjánsson and Vésteinn Ólason, vol. 1, p. 291 (*Konungsbók* [*=* Codex Regius]) and p. 308 (*Hauksbók*) (Reykjavík: Hið íslenzka fornrítafélag, 2014).

<sup>35</sup> See PROIEL at http://foni.uio.no:3000, under "Old Norse". This text may eventually be moved to the *Syntacticus* website, http://syntacticus.org.

**Figure 7.** The syntactic annotation of *V*ǫ*luspá* stanza 2.5–8. The main predicate is *man* (of *muna*) 'remember', which is present in st. 2.5 and must be seen as covert in the two other sentences, 2.6 and 2.7–8, represented by an encircled "V", and slashed lines pointing to *man* in order to indicate identity with the overt predicate. Copyright: Odd Einar Haugen, License: CC BY-NC-ND.

which there is no overt predicate or subject*, níu [man ek] íviðjur* and *mj*ǫ*tvið [man ek] mæran fyr mold neðan*.

In this analysis, *íviðjur* 'giantesses' is explicitly analysed as the object in *níu [man ek] íviðjur* 'nine [I remember] giantesses', and *níu* 'nine' as an attribute. It is an analysis that makes rather heavy assumptions about ellipted words, but, assuming that *íviðjur* is a noun rather than a preposition and its complement, the present analysis seems to be the best one.

The point of this example is not that a linguistic annotation offers the definitive answer to an enigmatic reading. That would be presumptuous. Rather, what it offers is a clear and consistent analysis of each word in each sentence of the poem in question, not skipping over any difficult passages, and as such it has a high degree of explanatory potential. It offers a point of reference for any interpretation of the text.

## **6. Costs and benefits**

Nobody would deny that annotation enhances an edition and makes it accessible for a wider audience. What is a matter of discussion is the cost of annotation compared to the benefits it offers. For a literature where many texts are still awaiting an edition (or a sufficiently good edition), the priority would be to edit the remaining texts, unless these texts by common consent were regarded as being of too little value even for the most avid scholar of the period. This is not the case for the field this author is most familiar with, Old Norse literature. The great majority of Old Norse works have been edited, many several times, and there are few works that are not available in a decent edition. Most works are preserved in more than one manuscript, but the really large manuscript traditions known from e.g. classical scholarship are few and far between. In Old Norwegian literature, the 15 preserved manuscripts of *Barlaams saga ok Jósafats* count as a rather large tradition, and *Konungs skuggsjá* with 60 preserved manuscripts (the majority being younger Icelandic ones) is one of the largest manuscript traditions, only surpassed by the law code of Magnús Hákonarson. A notable trait is the fact that almost all of the earlier manuscripts of these work are fragmentary, so that the text of each work has to be pieced together from several textual witnesses. And, as stated above, these manuscripts typically have different orthographies, representative of their time and locale.

In comparatively small textual traditions it makes sense and is in many cases feasible to transcribe each of the manuscripts. Any critical edition can only offer a glimpse of the textual variation through its apparatus, and while for many scholars it is fine to have an apparatus that only contains the substantive variants, for other users, the variation in accidentals is as interesting. Even in a fairly small vernacular manuscript tradition, there are simply too many variants for a workable apparatus, so the only way to record them is to edit each manuscript as an individual witness to the work. These transcriptions will preferably be digital ones published in text archives and searchable within these archives.

Many editors would be happy with an edition of the text as it appears in a single, typically best, manuscript. Probably the majority of texts published in an archive like the Medieval Nordic Text Archive will remain at this level. However, some editors would like to add annotation in order to open up the text, and other scholars would like to contribute by annotating editions by previous editors. Incrementally, more and more texts will be annotated. On the whole, this process is likely to be self-regulatory, although canonical texts are more prone to receive annotation than other texts. GKS 2365 4º, the major source for the Eddic poems, is a prime example of this type of text. The question of cost will ultimately be decided on the background of the canonicity and thus the general interest in the texts to be annotated.

# **Bibliography**


*Essays on the Art of Reading Slowly*, ed. by Harry Lönnroth (Leiden: Brill, 2017), pp. 118–135.


#### **Manuscripts**

Copenhagen, The Arnamagnæan Collection/Den Arnamagnæanske Samling

AM 619 4º AM 666 b 4º AM 544 4º

Oslo, National Archives Norse Fragment 7

Reykjavík, The Arnamagnæan Collection/Handritasafn Árna Magnússonar GKS 2365 4º

Uppsala, University Library De la Gardie collection, DG 4–7 fol

## **Web resources (last accessed 3 April 2020)**

The *Corpuscle* application: http://clarino.uib.no/korpuskel *Gammeldansk Ordbog*: http://gammeldanskordbog.dk The INESS portal: http://clarino.uib.no/iness The Medieval Nordic Text Archive (website): http://www.menota.org ——— (catalogue): http://clarino.uib.no/menota/catalogue The Medieval Unicode Font Initiative: http://www.mufi.info *Ordbog over det norrøne prosasprog*: http://onp.ku.dk The PROIEL Treebank: https://proiel.github.io Projects using the Text Encoding Initiative: http://www.tei-c.org/Activities /Projects/ The Syntacticus web site: http://www.syntacticus.org The Text Encoding Initiative, P5 Guidelines: https://tei-c.org/guidelines/p5/ The Web Open Font Format: https://www.w3.org/TR/2012 /REC-WOFF-20121213

# **Challenges of Editing Latin Patristic Texts: A report from inside the** *Corpus Scriptorum Ecclesiasticorum Latinorum*

*Dorothea Weber*

Google 'Ars edendi' on a computer in the Salzburg area, and one of the first sites to pop up is 'Ars edendi – the art of eating', by Monika Speier, a nutritionist from near Munich. Though eating and editing in Latin do not share the quantity of the first vowel, they do have some things in common.1 For example, an editor is used to making collations, and so does everyone in Italy when having breakfast, the *prima collazione*. To ruminate is to both chew food and to think over a problem again and again – something that everyone producing an edition knows very well from experience, and finally the delight that results from a nice meal can be similar to the delight a well-produced edition gives to its user. The enjoyment of well-produced editions is not the least outcome of the Ars edendi project at Stockholm University. This large and excellent project is now reaching its end, and has made editors better aware of the possibilities, aims, risks and limits which they encounter in the editing of medieval texts.

## **Editing Latin Texts from Late Antiquity**

With texts from Late Antiquity, some editorial issues are quite the same as with medieval, but others are not. Only within the last 100 years have scholars come to understand Late Antiquity as a period of transformation when in politics, religion, arts, and literature, Europe and especially the Mediterranean areas turned away from the usual paths

#### **How to cite this book chapter:**

This lecture was given at the Ars edendi concluding conference 19 August, 2016.

<sup>1</sup> As Michael W. Herren also commented in volume 2 of *Ars Edendi Lecture Series*.

Weber, D. 2020. Challenges of Editing Latin Patristic Texts: A report from inside the *Corpus Scriptorum Ecclesiasticorum Latinorum*. In: Kihlman, E. and Searby, D. (eds.). *Ars Edendi Lecture Series. vol. V*. Pp. 89–108. Stockholm: Stockholm University Press. DOI: https://doi.org/10.16993/bbd.f. License: CC-BY

and patterns.2 The period in question began around the year 200 and, depending on regional developments in the different regions of the former Roman Empire, ended between the late fifth and the early seventh century.3 Classical philologists in the German speaking countries agree nowadays that the responsibility for editing and commenting on the Latin texts of that period, whether they be Christian or pagan, belongs to the field of classical philology, whereas previously only theologians studied the Christian literature of Late Antiquity. Moreover, that period was judged to be a time of deterioration and degeneration; researching it was regarded as investing time and energy in a culture that produced no art but only objects devoid of artistic value.4

Defining historical periods may often create more problems than it resolves, it is true. We may even suspect that establishing periods means no more than projecting the opinions of contemporary art criticism

<sup>2</sup> See Alexander Demandt, 'Die Spätantike als Epoche', in *Spätantike. Mit einem Panorama der byzantinischen Literatur,* Neues Handbuch der Literaturwissenschaft, Bd. 4, ed. by Lodewijk J. Engels and Heinz Hofmann (Wiesbaden: Aula Verlag, 1997), pp. 1–28. – Hartmut Leppin, *Das Erbe der Antike*, C.H. Beck Geschichte Europas 1 (München: C.H. Beck, 2010). – Alexander Demandt, *Die Spätantike. Römische Geschichte von Diocletian bis Justinian 284 bis 565 n. Chr.,* Handbuch der Altertumswissenschaft III/6 (München: C.H. Beck, 20072 ). – *Idem*, *Geschichte der Spätantike. Das Römische Reich von Diocletian bis Justinian 284–565 n. Chr*. (München: C.H. Beck, 20082 ). – *Idem*, 'Die Zeitenwende von der Antike zum Mittelalter', in *Von der Spätantike zum frühen Mittelalter: Kontinuitäten und Brüche, Konzeptionen und Befunde*, ed. by Theo Kölzer (Ostfildern: Thorbecke, 2009), pp. 17–34. – Averil Cameron, *The Mediterranean World in Late Antiquity: AD 395–600* (London-New York: Routledge, 20122 ; first edition 1993)..

<sup>3</sup> Lodewijk J. Engels, 'Spätantike und lateinisches Mittelalter – ein rezeptionshistorischer Ausblick', in *Spätantike. Mit einem Panorama der byzantinischen Literatur,* Neues Handbuch der Literaturwissenschaft, Bd. 4, ed. by Lodewijk J. Engels and Heinz Hofmann (Wiesbaden: Aula Verlag, 1997), pp. 601–633. On different concepts and models of Late Antiquity see Scott Fitzgerald Johnson, 'On the Uniqueness of Late Antiquity', in *The Oxford Handbook of Late Antiquity*, ed. by Scott F. Johnson (Oxford: Oxford University Press, 2012), xi–xxiv.

<sup>4</sup> See, for example, what Eduard Munk wrote about Sulpicius Severus in 1861, i.e., five years before CSEL vol. 1 (Sulpicii Severi Opera) came out: 'Die Noth der Zeit wirkte selbst erschlaffend auf die kirchliche Literatur. Die großen Kirchenlehrer des vorigen Jahrhunderts hatten keine würdigen Nachfolger. Das Mönchstum fand immer mehr Anhänger und Lobredner und mit mönchischer Beschränktheit erfaßten Sulpicius Severus … und Paulus Orosius … in ihren geschichtlichen Schriften die Vergangenheit und Gegenwart' (Geschichte der römischen Literatur, Dritter Teil: Geschichte der nach-classischen Literatur der Römer, Berlin, 1861, p. 294). – For a recent analysis of that concept of Late Antiquity cp. Clifford Ando, 'Narrating Decline and Fall', in *A companion to Late Antiquity*, ed. by Philipp Rousseau (Chichester: Wiley Blackwell, 2009), pp. 59–76, and Stefan Rebenich, 'Latin Antiquity in Modern Eyes', *ibid*., pp. 77–92.

onto the past. Nevertheless, defining Late Antiquity as part of Antiquity, thus separating it from the Middle Ages, is somehow convenient from the viewpoint of research: tools which are most useful for doing research in classical Latin literature, for example, in the fields of history of literature, grammar, and language,5 are useful as well in dealing with phenomena of Late Antiquity. On the other hand, research on the culture of Late Antiquity does not differ much from research on Medieval culture, since both profit from and in fact are undertaken with the focus on interdisciplinarity. Apart from classicists, there are scholars of linguistics, history, theology, and Jewish Studies who are specialists in the issues of Late Antiquity.

For a classicist specialising in editing, Late Antiquity has some interesting new features: due to historical and cultural circumstances that favoured textual transmission, the extant literature of Late Antiquity represents far more varied artistic levels than in the case of classical literature. We have highly ambitious literature, for example, panegyrics to emperors or the pretentious and sophisticated letters of Ennodius, as well as primitive literary products such as, as early as from the end of the fourth century, the so-called *Peregrinatio Egeriae* or *Aetheriae*, 6 a description by a nun named Egeria or Aetheria of her pilgrimage to the Holy Land, or monastic rules (*Regula Magistri*, *Regula Benedicti*, *Regula Donati*, etc.) from the sixth century onwards. In other and new aspects of everyday life a specific terminology developed, for example, the legal terminology necessary for the administration of the Roman Empire and the subsequent Germanic kingdoms,7 or the homiletic terminology. Each of these new developments generated specific customs and language patterns. Furthermore, borders between literary genres, formerly neatly separated, began to be blurred. Remember, for example, the *Satura Menippea* as used by Boethius in his *Consolatio Philosophiae*

<sup>5</sup> Cf. the *Thesaurus linguae Latinae* (München, 1900ff.) and the Latin grammar by Manu Leumann, Johann Baptist Hofmann and Anton Szantyr, *Lateinische Grammatik,* 3 vols., Handbuch der Altertumswissenschaft 2, 2 (München: Beck, 1972–19792 ; reprint 1997).

<sup>6</sup> This text is closely associated with the famous Swedish Latinist Einar Löfstedt, who as early as 1911 used the *Peregrinatio* as a basis for his fundamental analysis of the lexis, morphology, and syntax of Late Latin: *Philologischer Kommentar zur Peregrinatio Aetheriae: Untersuchungen zur Geschichte der lateinischen Sprache* (Uppsala: Alquist & Wiksell, 1911; several reprints).

<sup>7</sup> E.g. *Digestae*, *Institutiones*, *Novellae*, *Lex Romana Visigothorum*; see Peter E. Pieler, 'Die Rechtsliteratur', in *Spätantike. Mit einem Panorama der byzantinischen Literatur,* Neues Handbuch der Literaturwissenschaft, Bd. 4, ed. by Lodewijk J. Engels and Heinz Hofmann (Wiesbaden: Aula-Verlag, 1997), pp. 565–599.

by which he refines the *De nuptiis Mercurii et Philologiae* by Martianus Capella.8 New genres appeared, as can be seen in Augustine of Hippo's *Confessiones,* which combine biographical elements and Bible exegesis. For some of these shifts, the emergence of Christianity seems to have been decisive. Thus, simple style and language were made respectable by the *sermo piscatorius* of the Bible. The genres of speech before a court or before a political gathering transformed into homilies, and the philosophical dispute eventually became a dispute against heretics.

It is evident, however, that in many respects Christian literature underwent the same changes as the non-Christian literature of the time. In fact, the distinction between them is nowadays judged as inappropriate because both reflect their position in regard to classical antiquity and recur to it in an almost identical way. This similarity results from the uniform institutional schools which were spread all over the Roman Empire. Since the contents of the curricula remained mostly as they had been and had not been Christianised, there was only one canon of literature and educational subjects that could be taken as default from Spain to Illyria and from North-Africa to Gaul.9 The contacts with Greek literature of that time, however, gradually became looser, as the knowledge of Greek declined.

## **The Foundation of CSEL in 1864**

The *Corpus Scriptorum Ecclesiasticorum Latinorum* (CSEL), therefore, has not limited itself to editing Christian Latin texts of Late Antiquity on the assumption that there is any essential difference in editing these texts and editing classical texts. On the contrary, the limitation to Christian text of Late Antiquity is due to the state of classical scholarship around the middle of the nineteenth century, when the long-term project of the CSEL was founded.10 It was precisely at that time that the

<sup>8</sup> See, e.g., Danuta Shanzer, 'Interpreting the Consolation', in *The Cambridge Companion to Boethius*, ed. by John Marenbon (Cambridge: Cambridge University Press, 2009), pp. 228–254 (esp. 233sq.).

<sup>9</sup> Cf. the comprehensive study of Henri Irénée Marrou, *Histoire de l'éducation dans l'antiquité* (Paris: Seuil, 1948), and Konrad Vössing, *Schule und Bildung im Nordafrika der Römischen Kaiserzeit* (Bruxelles: Latomus, 1997), pp. 7–22.

<sup>10</sup> For the history of the CSEL see Rudolf Hanslik, '100 Jahre Corpus Scriptorum Ecclesiasticorum Latinorum', in *Anzeiger der philosophisch-historischen Klasse der Österreichischen Akademie der Wissenschaften,* 101 (1964), 21–35; Michaela Zelzer, 'Ein Jahrhundert (und mehr) CSEL. Evaluation von Ziel und Veröffentlichungen', *Sacris Erudiri,* 38 (1998), 75–99; Dorothea Weber, '150 Jahre Corpus Scriptorum Ecclesiasticorum Latinorum', in *Edition und Erforschung lateinischer patristischer* 

huge project of the *Thesaurus linguae Latinae* was in its planning stage. The scholars in charge did not feel comfortable with the idea that they would have to rely on outdated uncritical editions for Christian literature. This, in fact, was a problem since the 'Thesaurus' was designed to be a tool covering all the Latin literature of Antiquity. This is why the former imperial Academy of Sciences in Vienna, which was a partner in the 'Thesaurus' project, initiated a series of critical editions of the Latin church fathers. As Volume 1 it published Sulpicius Severus,11 with good reason. Apart from the *Vita S. Martini,* the works of Sulpicius are preserved in a single manuscript. Roughly the same is true for Arnobius' *Adversus nationes*, Volume 4.12 Thus, the new series started off at a rapid pace. Prior to the publication of Volume 1, a catalogue of older manuscripts of the Latin Church Fathers was published to enable editors to save time when searching for manuscripts.13 The catalogue, it is true, has been a useful tool for editors, although we are now far more cautious in attributing the highest value to the oldest manuscripts. I will come back to this. The editions of Sulpicius and Arnobius however are not typical for the work done at the CSEL. Especially with the big names, i.e. Ambrose, Jerome, Augustine and Gregory the Great, but even with others such as Juvencus' biblical epos, an editor has to deal with innumerable text witnesses. This is one of the major challenges we have to face. I will also return to this. In sum, it has become evident, I think, that the methods of the so-called New Philology14 are not adequate for our texts, which in most instances are not texts that

*Texte. 150 Jahre CSEL (FS Kurt Smolak)*, ed. by Victoria Zimmerl-Panagl, Lukas Dorfbauer and Clemens Weidmann (Berlin: De Gruyter, 2014), ix–xi.

<sup>11</sup> Sulpicius Severus, *Opera*, ed. Karl Halm, CSEL 1 (Wien: Österreichische Akademie der Wissenschaften, 1866).

<sup>12</sup> Arnobius, *Adversus nationes*, ed. August Reifferscheid, CSEL 4 (Wien: Österreichische Akademie der Wissenschaften, 1875).

<sup>13</sup> Karl Halm, *Verzeichniss der Älteren Handschriften lateinischer Kirchenväter in den Bibliotheken der Schweiz* (Wien: Gerold in Comm., 1865).

<sup>14</sup> The term was coined in consequence of Bernard Cerquiglini's *Éloge de la variante* (Paris: Seuil, 1989; trans. in 1999 into English: *In Praise of the Variant*), and since then has been discussed among medievalists, see, e.g., R. Howard Bloch, 'Introduction: The New Philology Comes of Age', in *Rethinking the New Medievalism*, ed. by R. Howard Bloch [et al.], (Baltimore [MD]: Johns Hopkins University Press, 2014), pp. 1–11.; Gabrielle M. Spiegel, 'Reflections on The New Philology', *ibid.* 39–50. For a critical evaluation with regard to German Studies, see Freimut Löser, 'Postmodernes Mittelalter? New Philology und Überlieferungsgeschichte', in *Kulturen des Manuskriptzeitalters*, ed. by Arthur Groos [et al.], unter Mitarb. von Jochen Conzelmann (Göttingen: V & R Unipress, 2004), pp. 215–236. The concept of 'New Philology', though, has not had much impact on Late Antiquity studies.

underwent systematic rescription, quite the contrary: they were held in esteem as authoritative and were thus not adapted freely. As a consequence, we generally apply stemmatic methods in our editions, even when it is impossible to design a stemma, as happens with text traditions that remain hopelessly contaminated.

## **Glimpses of the Work done at CSEL**

I am now going to address some problems we have encountered in our most recent editing projects, some experiences we had and some solutions we found. I will do this without any systematic order and, of course, without the intention of completeness. Rather, every edition has challenges and problems of its own.

Whereas nowadays editors have at their disposal digital images of almost all manuscripts, our predecessors had to content themselves with microfilms. About 50 years ago, however, Rudolf Hanslik, at that time head of the CSEL, was in a far better situation since, when he ordered a microfilm of a *Regula Benedicti* codex from a library in Southern Italy, after a few weeks he received the manuscript itself in the post. In any case, the editors of Augustine, at that time the focus of the CSEL, in particular had to invest much time and energy in investigating the pertinent manuscripts. Therefore, Hanslik founded a series of publications intended to increase the speed of production and also the reliability of the editions. The series offers catalogues of manuscripts that transmit works attributed to St. Augustine. The series thus goes so far as to include writings falsely ascribed to Augustine. Starting with Italy, the series now comprises eleven volumes, thereby covering almost all European countries except France.15 The volumes aim at being exhaustive and are

<sup>15</sup> Die handschriftliche Überlieferung der Werke des Heiligen Augustinus, vol. I/1 and I/2: 'Italien' by Manfred Oberleitner, Vienna 1969 and 1970 (SB ÖAW 263 and 267); vol. II/1 and II/2: 'Großbritannien und Irland' by Franz Römer, Vienna 1972 (ibid. 281 and 276); vol. III: 'Polen (Anhang: Die Skandinavischen Staaten Dänemark – Finnland – Schweden)' by Franz Römer, Vienna 1973 (ibid. 289); vol. IV: 'Spanien und Portugal' by Johannes Divjak, Vienna 1974 (ibid. 292); vol. V/1 and V/2: 'Bundesrepublik Deutschland und Westberlin' by Rainer Kurz, Vienna 1976 and 1979 (ibid. 306 and 350); vol. VI/1 and VI/2: 'Österreich' by Dorothea Weber, Vienna 1993 (ibid. 601); vol. VII/1 and VII/2: 'Tschechische Republik und Slowakische Republik' by Clemens Weidmann, Vienna 1997 (ibid. 645); vol. VIII/1 and VIII/2: 'Belgien, Luxemburg und Niederlande' by Marie-Theres Wieser, Vienna 2000 (ibid. 685); vol. IX/1 and IX/2: 'Schweiz', by Sarah Janner and Romain Jurot, Vienna 2001 (ibid. 688); vol. X/1 and X/2: 'Ostdeutschland und Berlin', by Isabella

compiled on the basis of the autopsy of the manuscripts themselves; the catalogues are not simply compiled from other catalogues. This was intended in order to avoid reproducing any errors in previous catalogues. The project has not only provided a useful tool for editors in the volumes of this series, but has also led to new discoveries of some importance. Johannes Divjak, for example, discovered 29 new letters by Augustine – nowadays known as the *Epistulae Divjak*, in two French libraries, Paris and Marseille.16 Nine years ago, Isabella Schiller, while working on the volume of the former German Democratic Republic, identified six new sermons by Augustine in a twelfth-century manuscript in Erfurt.17 These sermons have subsequently been edited by Isabella Schiller, Clemens Weidmann and myself.18 To make an *editio princeps* of a sermon by an author as famous for his preaching as Augustine is a fascinating task. We have tried to reconstruct when, where, and under what circumstances the sermons were delivered, and to analyse their characteristics, the lines of thought and the structure of the arguments. We have had to deal with problems of textual criticism, it is true, but at the same time we had to help the readers to understand the spirit and temperament of the text by punctuating and organising it in paragraphs.

Remarkably enough, new texts have been discovered not only in very old manuscripts. To give two examples, Brian Møller Jensen discovered a new sermon by Augustine in a manuscript in Piacenza from the twelfth century,19 and the above-mentioned codex found in Marseille with the *Epistulae Divjak* originated as late as the fifteenth century. Thus, it becomes evident that we might miss new texts or at least good variants if we confine ourselves to the oldest text witnesses alone. Moreover, with pre-Carolingian manuscripts it may be the case that the texts they contain are even more corrupted than those that underwent correction in accordance with the Carolingian reforms. Even the famous *Codex Petripolitanus* from the early fifth century transmitting Augustine's *De* 

Schiller, Vienna 2009 (ibid. 791); vol. XI: 'Russland, Slowenien und Ungarn', by Irina Galynina [et al.], Vienna 2010 (ibid. 809).

<sup>16</sup> Paris, Bibl. nat., lat. 16861 (twelfth century), and Marseille, Bibl. mun. 209 (fifteenth century).

<sup>17</sup> Erfurt, University Library, Dep. Erf. CA. 12° 11.

<sup>18</sup> *Wiener Studien* 121 (2008), 227–284 and 122 (2009), 171–213.

<sup>19</sup> Ms. Piacenza, bibl. cap. 60, from the second half of the twelfth century; the text was edited by C. Weidmann in CSEL 101 (Wien: Österreichische Akademie der Wissenschaften, 2015), 31–44, as Augustini Sermo 59A.

*doctrina Christiana* has some evident blunders, although it was written during Augustine's lifetime.20

Yet with the works transmitted to us in some old as well as in many younger manuscripts, it may prove impossible to collate all of them. In such cases we take sample collations in order to identify those manuscripts that transmit variants not attested by the older ones, since the younger manuscripts may be copied from older ones now lost. To establish text families, it may be useful not only to look at variant readings but also at the corpora of texts contained in the manuscripts, since the arrangement of texts was often retained. But even indirect text transmission may be very valuable. For example, when the Venerable Bede has a variant reading in a quotation from Augustine, we know that this variant is pre-Carolingian that may be lost in the extant Augustine manuscripts. At the same time texts which we are working on may cite others so that they attest variants not preserved by direct text transmission. Handling facts of this sort requires complex methodology and can lead to interesting results. I will present examples taken from three of our projects and which differ in their sets of problems and in the appropriate strategies for resolving them. I have deliberately chosen examples on a macro scale and others that pertain to single words only.

The most spectacular example is the recent discovery of the commentary on the Gospels by Fortunatianus,21 bishop of Aquileia in Northern Italy around the middle of the fourth century. Until then only three quotations of the text were known, two of them stemming from an exegetic compilation handed down in a manuscript from around 1100 in Troyes, the other from an exegetic anthology in a ninth century manuscript in Angers. In each of them the quotation is attributed to Fortunatus or Fortunatianus respectively.22 For all other information on Fortunatianus Jerome's *De viris illustribus* ch. 97 was the only source. In 2012, when Lukas Dorfbauer, scholar at the CSEL, was browsing

<sup>20</sup> Lat. Q. v. I. 3, fifth century, described, e.g., in: Elias A. Lowe, *Codices Latini Antiquiores*, vol. 11 (Oxford: Clarendon Press, 1966), 1613; Michael M. Gorman, *The Manuscript traditions of the works of St. Augustine* (Firenze, 2001), 260; Bernhard Bischoff, *Katalog der festländischen Handschriften des neunten Jahrhunderts (mit Ausnahme der wisigotischen), Teil II: Laon – Paderborn. Aus dem Nachlaß hrsg. von Birgit Ebersperger* (Wiesbaden, 2004), 2317b.

<sup>21</sup> Edited by Lukas J. Dorfbauer, Fortunatianus Aquileiensis, *Commentarii in Evangelia*, CSEL 103 (Wien: Österreichische Akademie der Wissenschaften, 2017).

<sup>22</sup> Ms. Troyes, Bibl. mun. 653 (s. XI/XII): *expositio Fortunati episcopi in eodem evangelio* and *expositio Fortunati episcopi ex eodem evangelio*; Ms. Angers, Bibl. mun. 55 (s. IX): *nunc vero de libro beati Fortunatiani Aquilegense episcopum aliqua testimonia scerpsimus quae hic congruit intimare*.

the digital library of the Dombibliothek at Cologne, he stumbled over a Gospel commentary lacking any indication of its author in codex 17 from the early ninth century. The text however could easily be dated to Late Antiquity, because quotations from the bible had the pre-Vulgate wording. Dorfbauer spotted the three quotations mentioned (cf. note 22), each of them in its appropriate context. Since the language of the text and its wording have striking characteristics, he was able to attribute the text beyond any doubt to Fortunatianus. Interestingly enough, the text seems to have been very much appreciated by later writers: it was cited quite frequently, but because the citations do not give any author, only now could they be identified.23

The *constitutio textus* becomes a really intricate task when the text to be edited consists of excerpts taken from other and earlier texts, as is the case of the monastic rule for nuns written by Donatus in the middle of the seventh century. This rule consists mostly of excerpts from the monastic rules of Caesarius of Arles, Columbanus and Benedict, which, apart from slight adaptations in the grammatical gender or number, are cited *ad verbum*. In the case of Donatus' rule, the main line of textual transmission, which consists in a single manuscript, the so-called *Codex regularum*, develops alongside with an extensive secondary transmission. However, the editor, Victoria Zimmerl-Panagl,24 took not only into account the wording of the preceding rules from the available editions but also all their variant readings, because it was impossible to identify the manuscripts used by Donatus. One textual problem which, though inconspicuous, can serve as an example of intertextuality as well as paratextuality is found in ch. 64 of Donatus's rule which consists of a single sentence by which, it seems, headgear is regulated. It reads: *Capita numquam altiora ligent nisi quomodo in hoc loco mensuram* 

<sup>23</sup> It is interesting to observe that Lukas Dorfbauer dealt with problems of orthography similar to those Michael W. Herren ('Is the Author Really Better than his Scribes? Problems of Editing Pre-Carolingian Latin Texts', in *Ars Edendi Lecture Series, vol. 2*, ed. by Alessandra Bucossi and Erika Kihlman (Stockholm: Stockholm University Press, 2012), pp. 83–105) has pointed to with respect to Latin texts composed after c. 600 and before the Carolingian writing reforms and what Robert Burchard Constantijn Huygens, *Ars Edendi: A Practical Introduction to Editing Medieval Latin Texts* (Turnhout: Brepols, 2000), p. 41, recommended for editing medieval texts: 'There are quite a number of factual, and even grammatical errors the editor should not correct, since it is by no means certain that the author himself cannot have made them.' See Dorfbauer, CSEL 103, pp. 89–98.

<sup>24</sup> The edition is published in CSEL 98: *Monastica 1 (Donati Regula, Pseudo-Columbani Regula monialium [frg.])*, ed. Victoria Zimmerl-Panagl (Wien: Österreichische Akademie der Wissenschaften, 2015).

*de incato fecimus* (the edition by Holstenius from 1661 has *incausto* instead of *incato*). 25This is almost completely identical with Caesarius' rule, ch. 56: *Capita numquam altiora ligent quam in hoc loco mensuram de incausto* (var. lect.: *incato*) *fecimus*. The word *incatum* is not attested elsewhere. Evidently, it is a variant of *encaustum*, the Latin form of the Greek ἔγκαυστον, 'ink'. Whether it was Donatus himself or the scribe of the *Codex regularum* who wrote *encatum* can hardly be determined. Moreover, even in Caesarius both forms, *encaustum* and *encatum*, are transmitted, and interestingly enough *incatum* is the reading of the very same *Codex regularum* which is also the oldest extant manuscript of Caesarius' rule. Thus, the editor had to venture a guess: did the scribe of Donatus' rule write *incatum* because this was the reading of the manuscript he was copying? Or did he substitute *incaustum* with *incatum* according to the word he found in the rule of Caesarius? Or the other way round, did the scribe correct the word in the rule of Caesarius according to the rule of Donatus? Or does *incatum* occur in both rules only because this was the form the scribe was accustomed to, and *incaustum* is correct in both rules? In this case, the editor indicated with an asterisk in the critical apparatus that the decision to print *incatum* is still doubtful.26 Besides, some manuscripts display a line in the margin of that paragraph. We can guess that it was meant to indicate the height of the headdress. Alas, the length of the lines differs from manuscript to manuscript. In the *Codex Regularum*, it is 3.25 cm in the Donatus text and 5 cm in the Caesarius text. In another Caesarius manuscript in Tours which is now lost, it is said to have been 9.5 cm.27

During my own work on Augustine's *Contra Iulianum,* I encountered a somewhat similar problem. Augustine wrote this work in the last decade of his life, as part of the discussion he had with Julian of Aeclanum, a follower of Pelagius, on original sin and on whether men can be completely free from sin or not. The first two books contain a collection of quotations from patristic writings that are meant to support the doctrine of original sin. In some of these quotations, the manuscripts of *Contra Iulianum* have the very same variant readings that can be found in the manuscripts transmitting the cited works. Some of these accordances might have come about by chance, it is true. For example, in *Contra Iulianum* 2,10 a quotation from Ambrose, *In Lucam* 7,142 is inserted that does not match exactly what seems to be the original text:

<sup>25</sup> 'They, i.e., the nuns, are not allowed to bind their heads higher than is indicated

here with ink.' 26 In the preface, the editor discussed this problem at p. 114f.

<sup>27</sup> *Ibid.*, p. 114.

Augustinus, Contra Iulianum 2,10: Rursus in eodem opere, cum de spiritali atque incorruptibili loqueretur (*scil*. Ambrosius) cibo: "Etenim misericors cibus mentis est, inquit, praeclaraque alimonia suavitatis, quae membra non oneret neque in naturae pudenda, sed ornamenta convertat, cum libidinum volutabrum commutatur in dei templum diversoriumque vitiorum sacrarium incipit esse virtutum. …

*cf*. Ambrosius, in Luc. 7,142:<?>

 … etenim ratio cibus mentis est praeclaraque alimonia suavitatis. Quae membra non oneret neque in naturae pudenda, sed ornamenta convertat, cum libidinum volutabrum conmutatur in dei templum deversoriumque vitiorum sacrarium incipit esse virtutum. …

misericors] ratio *Cod. Orléans BM 162 (s. IX)* oneret] haeret *Cod. Orléans BM 162 and eight other old mss*.

\_\_\_\_\_\_\_\_\_\_

ratio] *var. lect*.: rationibus, oratio oneret] *var. lect.:* haeret

\_\_\_\_\_\_\_\_\_\_

Instead of *oneret,* one Ambrose manuscript as well as several of Augustine's *Contra Iulianum* have *haeret* which does not make sense. The error might have its origin in the preceding *non* or rather in its abbreviation, *n¯*, and it may have come about in both text traditions independently (*n¯ oneret* misread as *non eret*). This type of explanation, however, is not appropriate for the other textual problem in the same passage. Since all the manuscripts of Augustine's work have *etenim misericors cibus mentis est* as the first words cited, only a single codex, though one of the oldest extant manuscripts, Orleans, Bibl. mun. 162 from the ninth century, cancelled out *misericors* and wrote *ratio*  instead, which surely is the original reading in Ambrose. Most of the Ambrose manuscripts have *ratio*, few others have *oratio* or *rationibus*, but none have *misericors* or any other word which might even faintly resemble *misericors*. From this evidence we may conclude that *misericors* in Augustine is a *varia lectio* already present in Late Antiquity and that the scribe who corrected *misericors* to *ratio* in the Orleans manuscript did so because he had a manuscript of Ambrose at hand. In *Contra Iulianum* the correct text is therefore *misericors*, 28 though

<sup>28</sup> Giovanni Paolo Maggioni discussed substantially similar problems in Iacobus' de Voragine Legenda aurea, see: Barbara Crostini – Gunilla Iversen – Brian Jensen (ed.), *Ars Edendi Lecture Series, vol. 4* (Stockholm: Stockholm University Press, 2016), pp. 26–49.

from the viewpoint of the Ambrose text this word is not what Ambrose wrote.

As already said, a good edition also must make the structure of the text visible. This can, but need not, be part of the *constitutio textus* in its traditional meaning. In this context, I am going to give examples from a large-scale project of the CSEL, the edition of Augustine's corpus of the *Enarrationes in psalmos*. 29 I will not, however, deal with problems connected to the orality of the *Enarrationes* which consist largely of homilies. A preacher may, of course, make anacolutha and mental leaps, or it is possible that during the homily he spontaneously departs from his draft due to an unforeseen event – something that is normally difficult to reconstruct.30 All of this is a major challenge for an editor, it is true. However, I just want to present one passage from *Enarratio in Psalmum* 65, 2–3 (commenting on ps. 65, 1–2: [1] *Iubilate deo omnis terra* [2] *psallite autem nomini eius*) in order to show the extent to which punctuation and the insertion of paragraphs determines the character of the text. Many *Enarrationes* follow one and the same pattern: first a verse from the psalm – let it be called verse A – is cited, commented upon, and cited for a second time. Then the same pattern is applied to verse B: citation, commentary, citation, and so on. This is precisely the way the text of the *Enarrationes* is structured in vol. 36 of the monumental, though uncritical edition the Maurists made in Paris towards the end of the seventeenth century which was, with minor changes, reprinted in CC.SL 39:

(2) ... Inde coepit: *Iubilate Deo. Qui? Omnis terra.* Non ergo sola Judaea. Videte, fratres, quemadmodum commendetur universitas Ecclesiae toto orbe diffusae; et non solum dolete Judaeos qui gratiam istam Gentibus invidebant, sed plus haereticos plangite. Si enim dolendi sunt qui collec-

<sup>29</sup> So far nine out of twelve volumes are published in CSEL: Augustinus, *Enarrationes in psalmos 1–32 (expos.)*, by Clemens Weidmann, CSEL 93/1A (2003); 18–32 (*Sermones*), by Clemens Weidmann, CSEL 93/1B (2011); 51–60, by Hildegund Müller, CSEL 94/1 (2004); 61–70, by Hildegund Müller, CSEL 94/2 (2020); 101–109, by Franco Gori, CSEL 95/1 (2011); 110–118, by Franco Gori, CSEL 95/2 (2015); 119–133, by Franco Gori, CSEL 95/3 (2001); 134–140, by Franco Gori, CSEL 95/4 (2002); 141–150, by Franco Gori, CSEL 95/5 (2005).

<sup>30</sup> This happened, for example, when Augustine delivered sermo 313G (=Erfurt 6; see Dorothea Weber, 'Sechs neue Augustinuspredigten. Teil 1 mit Edition dreier Sermones', *Wiener Studien*, 121 [2008], 280–282). – On problems with editing medieval sermons see Nicole Bériou, 'Written Sermons and Actual Preaching: A Challenge for Editors', in *Ars Edendi Lecture Series, vol. 2*, ed. by Alessandra Bucossi and Erika Kihlman (Stockholm: Stockholm University Press, 2012), pp. 9–34.

ti non sunt, quanto amplius qui collecti divisi sunt? *Iubilate Deo, omnis terra. Quid est, Iubilate? In vocem erumpite gaudiorum, si non potestis verborum.* Non enim verbis iubilatur; sed solus gaudentium sonitus redditur, quasi parturientis et parientis cordis laetitiam in vocem rei conceptae, quae verbis explicari non possit. *Iubilate Deo, omnis terra: nemo iubilet in parte. Nemo, inquam, iubilet in parte: omnis terra iubilet, Catholica iubilet.*  Catholica totum tenet: quicumque partem tenet, et a toto praecisus est, ululare vult, non iubilare. *iubilate Deo, omnis terra.*

(3) *Psallite autem nomini eius.* Quid dixit? Psallentibus vobis benedicatur nomen eius. Quid sit autem psallere, heri dixi, et credo meminisse Caritatem Vestram. Psallere est organum etiam assumere quod psalterium dicitur, et pulsu atque opere manuum vocibus concordare. Si ergo iubilatis quod Deus audiat, psallite etiam quod homines et videant et audiant; …

The text, however, does no longer resemble a didactic and somewhat tedious commentary and gains coherence when one inserts the paragraphs in slightly different positions, i.e. immediately after the commentary of a verse and before the repetition of the citation. Thus, the psalm is not cut into pieces totally isolated from each other, but has a continuous thread in which Augustine, as Hildegund Müller has shown convincingly, was far more interested than in the meaning of single verses. Then, verse A and B together introduce the commentary of verse B:31

(2) … Inde coepit: **iubilate deo**. qui? **omnis terra**: non ergo sola Iudaea. Videte, fratres, quemadmodum commendetur … Catholica totum tenet: quicumque partem tenet et a toto praecisus est, ululare vult, non iubilare. iubilate deo omnis terra, (3) **psallite autem nomini eius.** Quid dixit? Psallentibus vobis benedicatur nomen eius. Quid sit autem psallere, heri dixi et credo meminisse Caritatem vestram. Psallere, est organum etiam assumere, quod psalterium dicitur, et pulsu atque opere manuum vocibus concordare. Si ergo iubilatis quod deus audiat, psallite etiam quod homines et videant et audiant. …

Another startling discovery regarding the arrangement of the text was the outcome of Clemens Weidmann's work on the *Enarrationes*. 32 This Corpus, as already said, consists mostly of preached homilies; for psalms 1 to 32, however, it contains commentaries dictated by Augustine.

<sup>31</sup> Text according to Hildegund Müller, *Eine Psalmenpredigt über die Auferstehung: Augustinus, Enarratio in psalmum 65. Einleitung, Text, Übersetzung und Kommentar* (Wien: Verlag der Österreichischen Akademie der Wissenschaften, 1997).

<sup>32</sup> The following example is taken from: Augustinus, *Enarrationes in Psalmos 1–32 (expos.)*, by Clemens Weidmann. CSEL 93/1A (Wien: Österreichische Akademie der Wissenschaften, 2003).

Within this group, psalms 1 to 14 are commented upon verse by verse, whereas 15 to 32 are rather paraphrased briefly in one or two sentences; only now and then an allegorical explanation is added. Clemens Weidmann was struck by the syntactical chaos and disorder of many of the paraphrastical explanations, e.g. in 29/1,7sq.:

(7) Ego autem dixi in abundantia mea, Non movebor in aeternum: ego autem ille populus, qui ab initio loquebar, dixi in mea abundantia, iam nullam patiens egestatem: Non movebor in aeternum. (8) Domine, in voluntate tua praestitisti decori meo virtutem: sed hanc abundantiam, Domine, non ex me mihi esse, sed in voluntate tua praestitisse te decori meo virtutem ex eo didici, Avertisti autem faciem tuam a me, et factus sum conturbatus; quod avertisti aliquando a peccante faciem tuam, et factus sum conturbatus, recedente a me illuminatione notitiae tuae.

*Sed hanc abundantiam* does not fit well with the preceding psalm verse (*Domine, in voluntate tua praestitisti decori meo virtutem*), and the logical structure of the rest, i.e. *ex eo didici: Avertisti autem faciem tuam a me, et factus sum conturbatus, quod avertisti aliquando a peccante faciem tuam*, is not clear at all. Passages like this make perfect sense if one removes those quotations that are meant as lemmata, so that the text under discussion reads: *ego autem – ille populus, qui ab initio loquebar – dixi in mea abundantia iam nullam patiens egestatem: 'non movebor in aeternum'. Sed hanc abundantiam, domine, non ex me mihi esse, sed in voluntate tua praestitisse te decori meo virtutem ex eo didici, quod avertisti aliquando a peccante faciem tuam.* It is only thanks to this that *quod avertisti* can be identified so as depending on *ex eo*. Furthermore, in a manuscript from Echternach from the eleventh century,33 Clemens Weidmann found the remains of a numeric reference system. On this basis, he was able to reconstruct the text as follows: in *Enarrationes* 15–32 each psalm is first quoted in its full length, its verses are numbered all the way through. Then, in the explanation Augustine refers to the pertinent verse not by citing it but only by its number. We know that Augustine used this reference system in his *Contra sermonem Arrianorum* and the *Breviculus collationis cum Donatistis*. Hence, in CSEL 93/1A the text is constituted as follows:

<sup>33</sup> Paris, BnF, lat.1983 (P11 in Weidmann's edition).


Again, it is worth noting that only one manuscript from the eleventh century has preserved part of the original reference system. If the editor had considered only the manuscripts predating the year 1000 – about 20 from before the eleventh century – he would not have been able to reconstruct the original arrangement of the text.

## **The organisation of the CSEL**

Up to 2012 the CSEL was funded and the editions were published by the Austrian Academy of Sciences. Since 2012 the CSEL has been part of the Department of Classics at Salzburg University, and the publisher is De Gruyter. The office, however, is still in Vienna. Our focus is on Augustine and Ambrose, but only as a guideline. Thus, within recent years, we have also published editions of an anonymous commentary on Job, two volumes on Prosper of Aquitaine, and the volume on Donatus mentioned above. The CSEL has volumes edited by the staff as well as by external editors. External editors receive advice and supervision whenever needed. Each volume is reviewed first by the CSEL staff, then by the advisory board, before it goes into print. Last but not least, the staff regularly gives courses on palaeography and editorial work in order to hand down the relevant skills and ultimately the pleasure of producing editions to the next generation of researchers.

## **Bibliography**

Ando, Clifford, 'Narrating Decline and Fall', in *A companion to Late Antiquity*, ed. by Philipp Rousseau (Chichester: Wiley-Blackwell, 2009), pp. 59–76.


# **Editing in Three Movements: Eriugena, Coleridge, Eliot**

*David Greetham†*

A strange triumvirate. What can Eriugena, Coleridge, and Eliot and the editions in which their work is enshrined have in common for me to bring them here to the closing conference of *Ars edendi* and then to the commemorative volume? Or, can the *differences* in the editing of the three inform our sense of the opportunities and challenges yet remaining in scholarly textuality? And, given the musical allusion in my title, is there some way in which each of these authors and their texts can be thought of as separate 'movements' each contributing to the total text (if you like, the harmony) that is editing and textual criticism?

Bearing in mind A. E. Housman's very neat formulation − 'textual criticism is a science, and, since it comprises recension and emendation, it is also an art. It is the science of discovering error in texts and the art of removing it,'1 − can all three authors and the editions of their works fit comfortably in the art/science dichotomy that we have inherited from Housman? Of course, the conference and the organisation sponsoring it, *Ars edendi*, recognize just one element of Housman's formulation, the 'arts' of editing, but perhaps for the moment it will suffice to recall that Housman's balanced formula was offered as a corrective to the Germanic dominance of *Altertumswissenschaft* and the imposition of a strict philology that left little room for the 'thought' in Housman's title. We have to remember that, based on a strict application of *recen-*

#### **How to cite this book chapter:**

On account of illness this lecture was not delivered by the author but read by one of the *Ars edendi* members at the concluding conference 17 August, 2016. In the final stages of editing this volume we received the sad news that David Greetham had passed away.

<sup>1</sup> A.E. Housman, 'The Application of Thought to Textual Criticism', *Proceedings of the Classical Association*, 18 (1921), 67−84, p. 68.

Greetham, D. 2020. Editing in Three Movements: Eriugena, Coleridge, Eliot. In: Kihlman, E. and Searby, D. (eds.). *Ars Edendi Lecture Series. vol. V*. Pp. 109–122. Stockholm: Stockholm University Press. DOI: https://doi.org/10.16993/bbd.g. License: CC-BY

*sio*, Lachmann could claim that he could reconstitute not just the text but also the folio breaks in the manuscript archetype of Lucretius *De Natura Rerum*. Lachmann's 1850 edition of Lucretius demonstrated to his satisfaction that this archetype contained 302 pages of 26 lines per page and that this in turn was a copy (no longer extant) of a manuscript written in a minuscule hand, derived itself from a copy of a fourth-fifth-century manuscript written in rustic capitals.2

In the face of such sureties, Housman was very concerned that editors had been 'readily duped by […] scientific criticism or critical method',3 and that the Germans had mistaken textual criticism for mathematics.4 His promotion of 'thought' could be seen, if you like, as an early example of English compromise in the face of Teutonic *System*.

But why, indeed, these three authors and their texts? All three are representative of recent editing of documents from three periods − late classical, romantic, and modern. The edition of Eriugena by Édouard Jeauneau was published in 1996,5 the Coleridge by J.C.C. Mays in 2001,6 and the Eliot by Christopher Ricks and Jim McCue in 2015.7 As a group, they thus embody current theories and practices of the late twentieth and early twenty-first century. And while the methods and principles vary a good deal, all three stand at the end of a long tradition of editing in their respective fields and all three aim to provide a comprehensive, indeed exhaustive, access to the extant documents of their authors. This can be immediately seen in the range and sheer heft on display in the texts of Eriugena, Coleridge, and Eliot.

The Jeauneau edition of Eriugena's *Periphyseon* runs to five sturdy volumes, for a total of 3 293 pages, often with multiple parallel texts, where blank spaces indicate a gap in the text of that witness. The Mays edition of the *Poetical Works* of Coleridge occupies four volumes of the poetry in section sixteen of the Bollingen *Complete Works of Coleridge*,

<sup>2</sup> K. Lachmann, *In T. Lucretii Cari De rerum natura libros commentarius* (Berlin: Georg Reimer, 1850).

<sup>3</sup> A.E. Housman, *Selected Prose*, ed. by J. Carter (Cambridge: Cambridge University Press, 1961), p. 37.

<sup>4</sup> Cf. Housman, *Selected Prose*, p. 132.

<sup>5</sup> *Iohannis Scotti seu Eriugenae Periphyseon: editionem novam a suppositiciis quidem additamentis purgatam, ditatam vero appendice in qua vicissitudines operis synoptice exhibentur*, ed. by É. Jeauneau, 5 vols, CCCM 161−165 (Turnhout: Brepols, 1996−2003).

<sup>6</sup> *The collected works of Samuel Taylor Coleridge* 16: *Poetical Works* I: *Poems*  (*Reading Texts*); *Poetical works* II: *Poems* (*Variorum Texts*), ed. by J.C.C. Mays (Princeton, NJ: Princeton University Press, 2001).

<sup>7</sup> *The Poems of T.S. Eliot: [the annotated text]*, ed. by C. Ricks and J. McCue, 2 vols (London: Faber, 2015).

with two volumes of a 'reading text' and two volumes of a 'variorum text', for a total of 2 800 pages, in which, for example, *The Rime of the Ancient Mariner* takes up thirty-five pages of text historical collation, and notes; and while the Ricks and McCue Eliot seems almost abstemious with a mere 667 pages just on *Practical Cats*, and 1 311 pages on *Collected and Uncollected Poems*, the three publications cannot avoid being labelled monumental. These are weighty tomes, and it is unlikely that they will be superseded any time soon by rival editorial projects.

Given these determinations *and* uncertainties, can we find any cultural logic in regarding my first author, Johannes Scottus Eriugena, the ninth-century Irish philosopher, as emblematic of the early medieval period; Coleridgean plenitude as appropriate for the Romantic period, and the recent Eliot edition (as distinct from the Eliot corpus) for early twenty-first century textuality?

There is one immediate distinction that can be seen and that is important in the recent history of editing: that the Eliot edition exists at all is a reflection of the often fraught relations between the estates of modernist authors (notably Joyce, Laurence, and Eliot) and scholarly editors. Jonathan Bate records that '[f]or forty years the Eliot estate, in the form of the poet's widow, rigorously restricted quotation and limited cooperation with scholars,'8 leading Peter Ackroyd to explain in his biography, that 'I am forbidden by the Eliot estate to quote from Eliot's published work,'9 a restriction that Bate believes largely explains 'why Eliot's reputation took such a severe battering over the decades around the turn of the century.'10 Clearly, there were no such restrictions placed on the editing of the texts of Eriugena or Coleridge, so there was a celebratory mood about the Ricks and McCue edition of Eliot, a mood that was appropriate for the concluding conference of *Ars edendi*.

First, let us examine the format of these three editions. In an earlier discussion of Jeauneau's edition of Eriguena for a special commemorative issue of the *American Catholic Philosophical Quarterly*, 11 I made much of the fact that Jeauneau reflected a current concern with a 'rolling' postmodernist text in which the lure of the 'definitive' had been replaced by an 'open' edition *in qua uicissitudines operis synoptice exhibentur,* an exemplification of Bernard Cerquiglini's *L'éloge* 

<sup>8</sup> J. Bate, 'Fair enough?', *The Times Literary Supplement*, 6 August, 2010, 14–15, p. 15.

<sup>9</sup> P. Ackroyd, *T.S. Eliot* (London: Hamilton, 1984), p.10.

<sup>10</sup> Bate, p. 15.

<sup>11</sup> D. Greetham, 'Édouard Jeauneau's Edition of the Periphyseon in Light of Contemporary Editorial Theory', *American Catholic Philosophical Quarterly* 79 (Fall 2005), 527−548.

*de la variante: Histoire critique de la philologie* (Paris: Seuil, 1989), and is to be compared with the immediately precedent edition by I.F. Sheldon-Williams, a volume in the series *Scriptores Latini Hiberniae,* in which the assertion *(desideratum?* hope? illusion?) was that 'the present edition attempts to present the text with which the author finally came to be satisfied, and at the same time to exhibit the stages of its development, through the creation of a "positive apparatus criticus".'12 Such desires for 'satisfaction' (the 'making complete' of a work as it progresses toward that most thorny of contemporary textual states, 'authorial final intention'), move Sheldon-Williams's edition in an epistemological direction in the reverse of Jeauneau. Similarly, the desire to construct a 'positive' apparatus (again, for the sake of 'fullness') and the expressed confidence that the 'scribal blunder' can be effectively distinguished from authorial idiosyncrasy are both evidence for the movement in the opposite direction of Jeauneau. Where Sheldon-Williams aims for teleology and completion, authorial and transmissional, the proliferation of textuality in the Jeauneau edition (what we might see as counterpoint or polyphony) sets out the 'critical' edition as only one state in the presentation of textual variation, and allows the 'synoptic apparatus' to become the 'fullest' part of the editorial enterprise (that is, four of the five volumes). Furthermore, what Sheldon-Williams hopefully enlists as a 'positive apparatus criticus' is nonetheless in his edition put in a conventional 'inferior' textual space, in reduced type at the *bottom* of the page, so that its positivism is in fact a mark of its degenerative status, again a conventional assumption. On the contrary, by freeing the synopticism from this mark of the 'inferior' and placing 'Versiones I–II, Versio II, Versio III, Versio IV, and Versio V' in a visual and spatial equality with one another (and by implication, with the 'critical' edition that sets the whole procedure in motion) Jeauneau forcibly moves the readerly eye (or ear) away from the plainchant of a single utterance into a polyphony with multiple voices and variance as a normative condition.

Thus, while he may be working from basically the same raw materials as Sheldon-Williams (no new witnesses of any significance have appeared of late, and Jeauneau's description of the redactions of *Periphyseon* is not substantially different from that of his immediate

<sup>12</sup> I.P. Sheldon-Williams, ed., *Iohannis Scotti Eriugenae Periphyseon*, 4 vols, Scriptores Latini Hiberniae 7, 9, 11, 13 (Dublin: Dublin Institute for Advanced Studies, 1968−1995).

predecessor), Jeauneau holds the earlier editorial aims of 'satisfaction' and 'fullness' in abeyance, if they are accorded any value at all, in the face of textual fragmentation and proliferation. It is in the section of Jeauneau's editorial introduction designated as 'Les remaniements du Periphyseon' that the methodological and even ideological distance from Sheldon-Williams's 'satisfaction' is clearly marked:

Le *Periphyseon* nous a été transmis sous différentes formes, recensions ou versions. L'étude attentive de ces différentes versions nous permet de suivre, au moins en partie, l'évolution de l'œuvre. Nous sommes en présence, non d'un produit fini, mais d'une matière en fusion; non point d'un texte établi et fixé de façon canonique, mais d'un texte en perpétuel devenir.13

It is in this sense that *all* the variants, negative as well as positive, contribute to the *perpétuel devenir* of a text, that, as we have seen, a blank space (indicating a section that is not present in one of the versions) in the Jeauneau is just as much an authoritative variant as is one with a *different* text, even though it may be disturbing to the reader's navigation. A blank space in the Jeauneau is just as much a *presence* as is a positive textual variant. In the Eriugena edition for example, there are numerous occasions when the blank space is recorded as a 'negative reading' in one of the texts in parallel display, facing a passage from another version for which there is no equivalent in other witnesses.

And because Jeauneau believes that Eriugena 'se corrigeait constamment, ajoutant, retranchant, modifiant son texte'14 (in a manner that, as we shall see, is similar to Coleridge), Jeauneau's edition is the most expansive of the three authors and editions under review, since it aims to record the *complete* several different versions that cumulatively make up the documentary history of *Periphyseon*. 15

Jeauneau's decision to present the texts of Eriugena in facing-page parallel columns does of course place burdens on both the editor and the reader, and such parallel texts are not likely to become popular. The basic problem of creating and negotiating a parallel text is similar to that involved in the 'full score' of music, in which, as Roland Barthes notes, the reader has to negotiate both the horizontal (paradigmatic)

<sup>13</sup> Jeauneau, vol. 1, p. xix.

<sup>14</sup> Jeauneau, vol. 1, p. xxii.

<sup>15</sup> Chaucerians will already be familiar with the device of the parallel text, as can be seen in the F and G versions of the *Prologue* to the *Legend of Good Women*, and parallel texts have a long and distinguished history, from Origen's *Hexapla* and on to the *Complutensian Bible*.


**Figure 1:** Pages 254–255 in *Iohannis Scotti seu Eriugenae Periphyseon: editionem novam a suppositiciis quidem additamentis purgatam, ditatam vero appendice in qua vicissitudines operis synoptice exhibentur,* ed. by É. Jeauneau, vol. 5, CCCM 165 (Turnhout: Brepols, 2003), License: CC BY-NC-ND.


for melody and the vertical (syntagmatic) for harmony.16 The reader confronted by such a 'full score' has to work in two modes, scanning *down* one text until a 'gross constituent'(in the terminology of structuralism) unit has been recognized and then moving *across* the page break to compare the gross constituent unit with a similar unit across the page example from Jeauneau. The movement downwards can be regarded as a search for *harmonics,* in a full score, and the horizontal movement a serial or narrative manoeuvre. A parallel text display works best when there is a basic accord in substantial units of a text, but not so much to obscure variation *within* a text or specific document.

As an experiment, a while back I constructed a parallel text based on the 1805 and 1850 versions of Wordsworth's *Prelude*, in which I attempted to show the complex (and not necessarily linear) relations between the two states.17 If the result is indeed a 'full score' it is a score in which elements duplicate each other, interrupt across the 'bar of difference,' and continually prevent a neat resolution of the harmonics. Such an interrupted score is to be seen in the five substantive versions of Eriugena.

Coleridge shares with Eriugena an almost obsessive involvement with change and revision. Indeed, Jack Stillinger adopts Coleridge as the *type* of the revising author, with an antitype in Keats, who very rarely returned to a poem after publication. And in Mays's edition of Coleridge, there is a similar acceptance of plenitude, reflecting the editor's view of Coleridge's composition techniques; 'Coleridge's mind operated at several levels, in several ways, and moved easily between them. An edition should display − not obscure − the variety and vitality of *his mind working*.'18 Mays's concentration on 'Coleridge's mind' is partly a reflection of the earlier Anglo-American intentionalism and partly an attempt to show that mind as it negotiates various social pressures, some quite intimate. Mays argues that '[a]n edition that displays Coleridge's working brain cannot use a "strictly synoptic procedure" because that won't do in charting that mazy mind, where "deliberation alternates with chance, and different intentions exist side by side" (cxx) or they shift and mutate haphazardly. "There is no clear tendency which

<sup>16</sup> See for instance R. Barthes, *Elements of Semiology*, transl. by A. Lavers and C. Smith (London, Cape, 1967).

<sup>17</sup> D. Greetham, *Theories of the Text* (Oxford: Oxford University Press, 1999), p. 316–319. [Editor's note: The Oxford University Press does not allow any content to be used under any form of open access license, which is why we have not been able to include an illustrative image of the result of Greetham's experiment.]

<sup>18</sup> Mays, *Poetical Works* I: *Poems (Reading Text)*, Part 1, p. lxxxviii (emphasis added).

could provide the basis of a rule" (cxxi).'19 Furthermore, Mays argues that 'Coleridge's materials are unruly. The editor must therefore be, like the poet, "fluid and opportunistic" (xv) and like the reader, "fixed in a permanent state of multiple vision"'.20 Even the 'reading text' does not imply finality, for it is often merely that text reflecting 'Coleridge's concern, up to the time he lost interest (as he so often did) (cxlvi).'21 The 'reading text' is simply a device to aid in the charting of variance in the historical collation and the 'variorum text'.

Thus, the insistent revision, according to Stillinger, shows Coleridge as a 'compulsive, wilful, out of control' tinkerer with his texts.22 Given what Mays characterizes as Coleridge's 'capaciousness,' with, for example, sixteen versions of the *Ancient Mariner*, ironically the advocate of 'unity' in poetry may be one of the most 'scattered and *dis*unified poets in English literature.'23 Stillinger speculates that 'perhaps he conspicuously featured his poetry's textual instability in order to imply that his poems were always in progress toward a never-to-be attained but increasingly approached perfection. Perhaps he wished to suggest that the perfect poem is a chimera and that authority itself is therefore a fiction. Perhaps he kept changing his poems to show that he was not dead.'24 In these circumstances, a multiform edition, like that of Mays, becomes the only honest way of representing Coleridge and his methods of composition.

The poem 'Dejection: An Ode', for example, first appeared in a nascent form in the so-called 'Letter' of 1802, which Mays insists should be seen as the muddled, unstructured first inspiration, 'never intended for publication.'25 The 'reading text' in the Mays edition, because it is clear text, does not allude to this muddle, but simply prints it as poem 289, the 'Letter' containing the stanza 'O Sara! we receive but what we give/And in *our* Life alone does Nature live,' which becomes 'O Lady' in poem 293 'Dejection: An Ode.'26 In the 'variorum text,' the very

<sup>19</sup> J. McGann, *A New Republic of Letters: Memory and Scholarship in the Age of Digital Reproduction* (Cambridge, MA: Harvard University Press, 2014), p. 116.

<sup>20</sup> McGann, p. 117.

<sup>21</sup> McGann, p. 118.

<sup>22</sup> J. Stillinger, *Coleridge and Textual Instability: The Multiple Versions of the Major* 

*Poems* (Oxford: Oxford University Press, 1994), p. 117. 23 Stillinger, p. 117, original emphasis. 24 Stillinger, p. 117. 25 Mays, *Poetical Works* I: *Poems (Reading Text)*, Part 2, p. 677. The poem in this form, called 'A Letter to -', is on pp. 679–691. 26 Mays, *Poetical Works* I: *Poems (Reading Text)*, Part 2, p. 689 and pp.695–702.

much shortened poem lists an array of dedicatees: 'O Wordsworth,' 'Edmund,' 'William,' 'Edmund,' and 'Lady.' Coleridge was fortunate that all of these choices are disyllables, none of which would disturb the metrics of the line.27

In my correspondence with Mays, he tells me that the 'Letter' is part of the 'circular progression (Sara Hutchinson to Wordsworth in various guises and Edmund for the generic 'poet' − i.e. Spenser) and can be understood as the muddled, unstructured first inspiration reconceived and re-projected as 'Lady,' a version of the same imaginative ideal that is at the heart of the poem.'28 But while the poem, from the 1802 'Letter' to the 1817 first publication and beyond, may indeed have an 'imaginative ideal that is at the heart of the poem' each of the variant addressees represents a significant part of Coleridge's poetic conception and reconception.

Now, all of this cumulative variance does not completely answer the question already raised of why some authors can never let go of a text and continue to re-enter the text on numerous compositional stages. Just as Beethoven would frequently add another measure to an already engraved score from his publisher, so Coleridge would not regard a print proof as anything but a way-station and not the 'final word'. But it does provide evidence of the 'perpétual devenir' that Coleridge shares with Eriugena.

It would seem that the editors of both the Eriugena and the Coleridge have similar aims: to make available to the reader an enormous corpus of variants; but the methods are very different, Jeauneau opting for a series of parallel texts containing the various authoritative witnesses in full, but without attempting a reading text or a base manuscript for collation. But Mays creates (or selects) a 'reading text' and then provides a full documentary record of variants. It is important to note that the selection of the base manuscript for Coleridge does not confer any specific authority on this witness: it is simply a device to aid in the charting of variance in the historical collation and the 'variorum text'.

As I take titles seriously, I have to believe that Ricks and McCue's decision to call their edition 'the annotated text' rather than a 'critical' text or some such makes a statement about editorial policy and procedures.

<sup>27</sup> Poem 293 in Mays, *Poetical works* II: *Poems (Variorum Texts)*, p. 890. [Editor's note: Princeton University Press does not allow any content to be used under any form of open access license, which is why we have not been able to include illustrative images of the different variants of the poem.]

<sup>28</sup> J.C.C. Mays in private correspondence.

And because it is an 'annotated' edition not a 'critical' edition, the Ricks and McCue volumes do not present variance in the texts presented.

The Eliot corpus is a relatively invariant body of poetry, partly reflective of Eliot's diffidence about revision. As Ricks and McCue remark: 'Although Eliot was reluctant to revise after publication, examination has shown that even repeated impressions of the same edition diverge to an unexpected extent.'29 Variance exists in despite of authorial intention. Eliot is equally aggrieved about the state of his first editions: 'I have never succeeded in getting a first edition of one of my own books printed without some errors in it, and I sometimes find that when those are corrected new errors appear.'30

Why does this distinction of purposes and practice matter in this discussion of Eriugena, Coleridge, and Eliot? As has already been shown, the five volumes of Eriugena, displaying a vast array of texts and variants, fall in line with the usual procedures of the 'literary' editions, though we should be very clear that the accumulated evidence of these five volumes is not put in service of the establishment of a singular, authoritative text. Each of the versions recorded is given unique and specific authority and is not subordinated to the support of a 'critical' or 'eclectic' text. At best, they might be likened to *Leitmotifs* recognizable in various parts (and versions) of a musical text, but not establishing a specific authority. Or, they might be the variations (like Beethoven's Diabelli, Elgar's Enigma, or Bach's Goldberg) without the theme to which they would usually relate. The Coleridge is different, depending on where you look. The volumes of the 'variorum text' in the Coleridge consist entirely of the variants (there is no 'standard' text to which they can be compered), whereas the 'reading text' provides precisely that, a clear-text presentation of an editorially preferred text, unencumbered by the sort of record of variance in the 'variorum text.'

The Eliot volumes have a clear text up front (the equivalent of the 'reading text' in the Coleridge), followed by a series of interpretative texts recording the documentary history and a commentary which is the basic rationale for the edition, announcing itself as the 'annotated text.'

The Eliot edition is not immune from the sort of incorrigible error that often plagues standard editions, as when the last line of Part II ('with eyes I dare not meet in dreams') of 'The Hollow Men' is repeatedly omitted in several subsequent editions.

<sup>29</sup> Ricks and McCue, vol. 1, p. xii

<sup>30</sup> Letter to Djuna Barnes 15 Oct 1936, quoted in Ricks and McCue, vol. 1, p. xii.

Similarly, the error in the dedication to Jean Verdenal in 'Prufrock' (1889 instead of 1890) is retained, although the mistake was acknowledged by Eliot, on the grounds that 'the dedication has stood for almost a century and it has been thought best not to alter it'31 an argument that might seem to support the 'socialization' of the text or could be put down to editorial weariness.

Eliot was in general receptive to Pound's advice: 'He cut out a lot of dead matter. I think that the poem as originally written was about twice the length. It contained some stanzas in imitation of Pope, and Ezra said to me "Pope's done that so well that you had better not try to compete with him" which was sound advice'.32 Actually, Eliot's memory has exercised a little self-censorship, for Pound's intercession was less delicate than 'not try to compete'. What he actually warned Eliot was that 'you cannot parody Pope unless you can write better verse than Pope – and you can't.'33

Perhaps the most striking example of divergence between published text and drafts is the opening page for the 1922 *Waste Land* versus the 'same' text in the 'editorial composite'.34 Thus, we would not know from the clear text that the original title for the poem was 'He do the police in different voices' (a quotation from Dickens, *Our Mutual Friend,* bk 1, ch. xvi),35 though this reading is confirmed in the facsimile edition containing Eliot's original typescript or manuscript, Pound's annotations and the very occasional marginal comment by Vivienne Eliot ('Wonderful').36 The decision to omit any reference to pre-publication manuscript readings in Ricks and McCue is, if you like, a very powerful example of a 'final intentions' ideology, privileging later print over manuscript draft. It does not take much imagination to speculate on the likely social and canonical status of the poem if it were still called 'He do the police', and is another example of the importance of titles.

But there is more to it than just the title. We should not imagine that there is a linear progression from the so-called 'manuscript' to the print

<sup>31</sup> Ricks and McCue, vol. 1, p. xiii. 32 Ricks and McCue, vol. 1, p. 581.

<sup>33</sup> T.S. Eliot, *The Waste Land: A Facsimile and Transcript of the Original Drafts, Including the Annotations of Ezra Pound*, ed. by Valerie Eliot (London: Faber and Faber, 1971), p. 127.

<sup>34</sup> Ricks and McCue, vol. 1, pp. 55–71 (published text), and pp. 324–346 (editorial composite). [Editor's note: Images of these pages have not been possible to include here since we could not obtain an Open Access license from Faber & Faber.]

<sup>35</sup> Ricks and McCue, vol. 1, p. 324.

<sup>36</sup> Eliot, *The Waste Land: A Facsimile*, pp. 10–11. [Editor's note: Images of these pages have not been possible to include here since we could not obtain an Open Access license from Faber & Faber.]

edition of 1922. For example, the passage beginning 'the typist home at teatime' occurs twice in the manuscript in very different contexts and in one such is the occasion for some of Pound's more caustic comments ('perhaps be damned', 'Make up your mind' and 'inversions not warranted by any real exegience [*sic*] of metre.').37 While Eliot was generally very receptive to the sort of revisions suggested by Pound; 'the typist' section, despite its dual prominence in the manuscript, is not carried forth into 1922, which remains the *de facto* terminus for the evolution of the poem and is thus very different from the sort of variance encountered in Eriguena and Coleridge.

This potted history now means that I have to retract my earlier assumptions that the sheer weight of these three editions would mean that editing now stops. In fact, it might very well be the exact opposite: that the availability of the texts in these forms will encourage further textual examination based on the new information (specifically for Eliot). But there is another factor: before embarking on this investigation, I had worked on several previous editorial projects, separate from my writings on textual and editorial theory and history. These had included work on a) the collaborative edition of John Trevisa's *On the Properties of Things*, the fourteenth-century encyclopedia of, well, everything, in which the aim was to reconstruct the archetype lying behind and above the extant manuscripts, and b) the editing of Thomas Hoccleve's *Regement of Princes*, through the construction of an idiolect drawn from the practices in several autograph manuscripts of *other* Hoccleve's works. Neither of these projects is in any way similar to the conditions of the documents encountered in Eriugena, Coleridge, and Eliot, so I have learned a great deal in doing the research for this paper. The one underlying principle has been, and will continue to be, fidelity to the author, who, despite the dire claims of Roland Barthes, is very much alive, but alive through the loyalties and devotions (and sheer hard work) of editors working in the long tradition extending from the Alexandrian librarians to the present day. It is perhaps in this sense of a long tradition that I see these loyalties and devotions so much evident in the range and enthusiasms of this conference, and I am very grateful to have been able to add to these enthusiasms in working on this paper. I am now in a different place from before I started work on this report and thank you all for giving me the opportunity to demonstrate what I have learned from this experience.

<sup>37</sup> Eliot, *The Waste Land: A Facsimile*, pp. 44–45.

### **Bibliography**

Ackroyd, P., *T.S. Eliot* (London: Hamilton, 1984).


#### STUDIA LATINA STOCKHOLMIENSIA *Published by Stockholm University* Nos. 1–22

#### Editor: Dag Norberg


#### STUDIA LATINA STOCKHOLMIENSIA *Published by Stockholm University*

#### Nos. 23–46

#### Editor: Jan Öberg


#### STUDIA LATINA STOCKHOLMIENSIA *Published by Stockholm University* Nos. 47–48 Editor: Monika Asztalos


#### STUDIA LATINA STOCKHOLMIENSIA *Published by Stockholm University*

#### Nos. 49–53

#### Editor: Hans Aili and Gunilla Iversen


### STUDIA LATINA STOCKHOLMIENSIA *Published by Stockholm University* Nos. 54–58

#### Editor: Hans Aili

54. *Alexander Andrée, Erika Kihlman*, ed. Hortus troporum. Florilegium in honorem Gunillae Iversen. A Festschrift in Honour of Professor Gunilla Iversen on the Occasion of her Retirement as Chair of Latin at Stockholm University. Stockholm 2008. Pp. XIX+384; 28 pl.


#### STUDIA LATINA STOCKHOLMIENSIA *Published by Stockholm University*

#### Nos. 59–

#### Editor: Maria Plaza


#### CORPUS TROPORUM

#### *Studia Latina Stockholmiensia* (SLS)*/Corpus Troporum* (CT)

CORPUS TROPORUM I, Tropes du propre de la messe. 1. Cycle de Noël, éd. Ritva Jonsson. SLS 21. Stockholm 1975.

CORPUS TROPORUM II, Prosules de la messe. 1. Tropes de l'alleluia, éd. Olof Marcusson. SLS 22. Stockholm 1976.

CORPUS TROPORUM III, Tropes du propres de la messe. 2. Cycle de Pâques, éd. Gunilla Björkvall, Gunilla Iversen, Ritva Jonsson. SLS 25. Stockholm 1982.

CORPUS TROPORUM IV, Tropes de l'Agnus Dei. Edition critique suivie d'une étude analytique par Gunilla Iversen. SLS 26. Stockholm 1980.

CORPUS TROPORUM V, Les deux tropaires d'Apt, mss. 17 et 18. Inventaire analytique des mss et édition des textes uniques par Gunilla Björkvall. SLS 32. Stockholm 1986.

CORPUS TROPORUM VI, Prosules de la messe. 2. Les prosules limousines de Wolfenbüttel, Herzog August Bibliothek Cod. Guelf. 79 Gud. lat., par Eva Odelman. SLS 31. Stockholm 1986.

CORPUS TROPORUM VII, Tropes de l'ordinaire de la messe. Tropes du Sanctus. Introduction et édition critique par Gunilla Iversen. SLS 34. Stockholm 1990.

CORPUS TROPORUM IX, Tropes for the Proper of the Mass. 4. The Feasts of the Blessed Virgin Mary. Edited with an Introduction and Commentary by Ann-Katrin Andrews Johansson. CT. Stockholm 1998.

CORPUS TROPORUM X, Tropes du propre de la messe. 5. Fêtes des Saints et de la Croix et de la Transfiguration. A Introduction et commentaires (Pp. 647 + carte + 13 photos). B Édition des textes (Pp. 560). Par Ritva Maria Jacobsson. CT. Stockholm 2011.

CORPUS TROPORUM XI, Prosules de la messe. 3. Prosules de l'offertoire. Édition des textes par Gunilla Björkvall. CT. Stockholm 2009. Pp. X + 254.

CORPUS TROPORUM XII, Tropes du Gloria, vol 1, Introduction et édition des textes, (428pp.), Vol. 2. Aperçu des manuscrits (2438pp.+32 planches), Par Gunilla Iversen, SLS 61, Stockholm 2014.

*Pax et Sapientia*. Studies in Text and Music of Liturgical Tropes and Sequences, in Memory of Gordon Anderson, ed. Ritva Jacobsson. SLS 29. Stockholm 1986.

*Recherches nouvelles sur les tropes liturgiques.* Recueil d'études réunies par Wulf Arlt et Gunilla Björkvall. SLS 36. Stockholm 1993.

This is the fi fth and fi nal volume of lectures on textual criticism and classical philology – broadly understood – given within the framework of the Ars edendi research programme (2008 – 2015).

Two of the six papers in this volume stem from a 2015 workshop on editorial theory and method, the theme of which dealt with fragments and the writing of commentaries. As regards the former, S. Douglas Olson problematizes the creation and continuation of scholarly knowledge concerning texts that have only come down to us in a fragmentary state, emphazising the challenges and pitfalls that lay in wait for the editor. Benjamin Millis offers a nuanced homage and apology for the traditional text edition with a scholarly commentary, especially underscoring its importance as a connective pathway between text and reader as well as the impetus it can give to scholarly research.

The other four lectures were given at the concluding conference of the Ars edendi programme, held in August 2016. In a case study Cynthia Damon shares her refl ections on how to digitally edit Pliny's *Natural History* in a form that will provide this work's rich reception history and at the same time its extensive use of sources, many of which are now lost. The digital component is also prominent in Odd Einar Haugen's contribution in which he shows that digital mark-up is also an editorial enterprise and how it can be useful for the textual scholar. Dorothea Weber gives an insider's view of the *Corpus Scriptorum Ecclesiasticorum Latinorum*, an editorial project on-going since 1864, and especially how improved cataloguing has led to numerous discoveries of texts by St. Augustine. As a conclusion to the volume, David Greetham, one of the founders of the Society for Textual Scholarship, refl ects on three different methods for editing texts that have undergone various degrees of rescription, namely the *oeuvres* of Eriugena, Coleridge, and Eliot.

Erika Kihlman and Denis Searby eds Ars Edendi Lecture Series Vol. V

STUDIA LATINA STOCKHOLMIENSIA

LXIII

Volume V

Edited by Erika Kihlman and Denis Searby