Hostname: page-component-745bb68f8f-b95js Total loading time: 0 Render date: 2025-02-05T00:20:59.545Z Has data issue: false hasContentIssue false

GLOSSING THE GLOSS: READING PETER LOMBARD’S COLLECTANEA ON THE PAULINE EPISTLES AS A HISTORICAL ACT

Published online by Cambridge University Press:  30 October 2018

PETER O'HAGAN*
Affiliation:
University of Toronto
Rights & Permissions [Opens in a new window]

Abstract

Peter Lombard's influential commentary on the Pauline Epistles, the Collectanea in omnes divi Pauli epistolas, has received little extended analysis in scholarly literature, despite its recognized importance both in its own right and as key for the development of his Sentences. This article presents a new approach to studying the Collectanea by analyzing how Lombard's commentary builds on the Glossa “Ordinaria” on the Pauline Epistles. The article argues for treating the Collectanea as a “historical act,” focusing on how Lombard engages with the biblical text and with authoritative sources within which he encounters the same biblical text embedded. The article further argues for the necessity of turning to the manuscripts of both the Collectanea and the Glossa, rather than continuing to rely on inadequate early modern printed editions or the Patrologia Latina. The article then uses Lombard's discussion of faith at Romans 1:17 as a case study, demonstrating the way in which Lombard begins from the Glossa, clarifies its ambiguities, and moves his analysis forward through his use of other auctoritates and theological quaestiones. A comparison with Lombard's treatment of faith in the Sentences highlights the close links between Lombard's biblical lectures and this later work. The article concludes by arguing that scholastic biblical exegesis and theology should be treated as primarily a classroom activity, with the glossed Bible as the central focus. Discussion of Lombard's work should draw on much recent scholarship that has begun to uncover the layers of orality within the textual history of scholastic works.

Type
Research Article
Copyright
Copyright © Fordham University 2018 

Around the year 1175, the anonymous abbot of a monastery in northern France opined that the commentaries on scripture by Peter Lombard were much to be preferred to those of his own day. In words glowing with admiration, the abbot explained that Lombard's expositions are so learned and clear that anyone who applies himself to them scarcely requires instruction from a teacher.Footnote 1 The abbot's words do not express an anomalous sentiment, as is clear from the phenomenal success of Lombard's Collectanea on the Psalms and on the Pauline Epistles — over two hundred manuscripts of each of these works are extant,Footnote 2 and on Lombard's epitaph we find these two commentaries listed alongside the Sentences as the achievements for which posterity was to remember him.Footnote 3 One of the most illustrious of Lombard's successors in the Paris schools, the future archbishop of Canterbury Stephen Langton, found Lombard's works so helpful that he lectured not on the Glossa “ordinaria” but on the Collectanea when teaching the Pauline Epistles.Footnote 4

Despite the medieval enthusiasm for Lombard's commentaries, modern scholarship on the Collectanea remains underdeveloped and in need of a fresh start. As with so much related to Peter Lombard, it is to Ignatius Brady in the 1960s and 1970s and Marcia Colish in the 1990s that we owe most of our insights into the significance of these commentaries.Footnote 5 Brady's manuscript work on the Collectanea remains fundamental,Footnote 6 and it is thanks to his careful industry that we can appreciate the extent to which Lombard drew on the Collectanea in composing the Sentences.Footnote 7 The only extensive analyses of Lombard's exegetical approach to date have been undertaken by Marcia Colish in a series of articles and in her magisterial Peter Lombard, in which she argues that Lombard's Collectanea met the theological and exegetical needs of his day more successfully than the work of his contemporaries and immediate predecessors.Footnote 8 Lombard's Collectanea has been discussed more briefly in other publications, but it remains a work that requires greater and more sustained attention than it has hitherto received.Footnote 9 In a 2005 survey of scholarship on Lombard, Philipp Rosemann called for “detailed studies” of the Collectanea, a task that has yet to be undertaken.Footnote 10

In this article, I propose an approach to Lombard's Collectanea on the Pauline Epistles that turns to the manuscripts and focuses on Lombard's exegetical method, situating him within a scholastic milieu that is frequently misunderstood. Fundamentally, I propose to take seriously two medieval perspectives, one specifically about Lombard and one more generally about the role of the magister in twelfth-century Paris. First, according to medieval sources, Lombard's Collectanea on the Pauline Epistles was an expansion of the Glossa “Ordinaria” (hereafter Glossa)Footnote 11 on the Pauline Epistles. The connection between the Glossa and Lombard's Collectanea has been noted by scholars, but almost no effort to clarify this relationship has been undertaken. This is a major gap in scholarship that leads to an incomplete picture of Lombard's exegetical methods. Hence, my central concern in this article is to reconstruct Lombard's use of the Glossa and to show not only that he uses it, which is generally accepted, but how he uses it. I will argue that the Glossa is not one source among many on which Lombard draws, but is in fact the text on which he is commenting — that is, that the medieval characterization of his practice is correct, even if it does not capture the entirety of his exegetical methodology.

This shift in perspective will locate Lombard firmly within a scholastic milieu in which lecturing on the Bible frequently meant lecturing on the glossed Bible.Footnote 12 Closely related to this insight is the second medieval perspective that needs to be taken into consideration when approaching the Collectanea — masters of the twelfth-century schools did not usually refer to themselves as “theologians” but expressed their self-understanding through the closely related terms magister in sacra pagina, magister in sacra doctrina, and the like. Their task, as they understood it, was the teaching of sacred scripture, not the construction of a theological system or the writing of theological summae. As a result, the textual witnesses of their activities, namely, their biblical commentaries, collections of sententiae, and theological treatises, are frequently more fruitfully approached as the written records of the oral culture of the classroom than as commentaries and treatises written for a reading audience. Much recent scholarship has emphasized precisely this fundamentally oral culture of the twelfth-century schools, delving into the manuscripts to discern the complicated process of teaching, writing, and editing that lies behind the texts we encounter (see the discussion and bibliography in my conclusion below). At the heart of much of this scholarship lies a reappraisal of how the Glossa was used — namely, as a “taught text” rather than as an encyclopedia or reference work — and of how the magistri actually went about their work — namely, by lecturing on earlier magisterial texts, frequently reworking them in the process. In addition, scholars continue to emphasize the close connection that existed between the different “genres” of scholastic texts — instead of traditional narratives that argue for an increasing separation between “biblical exegesis” and “systematic theology” over the course of the twelfth century, with some magistri more focused on one or the other of this dichotomy, several scholars have argued that we need to reintegrate in our narratives what were never separate in reality. In other words, the scholastic magister was not a theologian producing systematic theology, but a teacher expounding scripture — a “master of the sacred page.”

In the analysis that follows, I hope to show that this shift in perspective is valid and illuminating for Lombard as well.Footnote 13 This becomes apparent if we pay closer attention to his methodology in commenting on Paul, specifically with his use of the Glossa. My suggestion is that this is a new and important avenue for thinking about Lombard that will be fruitful not only for our understanding of his Collectanea but also of his Sentences, of the history and use of the Glossa, and of the scholastic project in general. What follows is only a beginning, a call to look more closely at the manuscripts, and a proposal for a method that has borne fruit in the study of other scholastic figures.

Lombard and the Glossa

According to medieval sources, Lombard's work was an expansion and elaboration of an earlier work, the Glossa on the Psalms and on the Pauline Epistles, attributed to Anselm of Laon. Herbert of Bosham, Lombard's student, in his prologue to the Collectanea on the Psalter, notes that Lombard's intent in composing the work was to “make clear the obscure brevity of the older glosator, namely, master Anselm of Laon.”Footnote 14 Gerhoch of Reichersberg, writing in 1168, links Lombard's “most recent” glosses on the Psalms and Epistles with those of Anselm of Laon and Gilbert of Poitiers, Anselm's one-time student who had based his own commentary on the Glossa.Footnote 15 Robert of Auxerre, writing before 1203, notes that Lombard's work “explained (explicuit) more extensively and clearly” the interlinear and marginal gloss of Anselm on the Psalms and the Pauline Epistles, which had been set out “continuously” by Gilbert of Poitiers.Footnote 16 Vincent of Beauvais, writing in the mid-thirteenth century, makes a similar statement, that Lombard “explained (explicuit) more extensively and openly the glossatura on the Epistles and the Psalter by Anselm, which was divided into interlinear and marginal glosses and was afterwards rendered in a continuous manner by Gilbert; and [Lombard] added many things from the sayings of the saints.”Footnote 17 These claims set Lombard firmly within a milieu of scholastic teaching in which lecturing on the Bible meant lecturing on the Glossa.

The connection between the Glossa and Lombard's Collectanea has been noted by scholars, though little has been done with this information. Beryl Smalley noted, “If the Gloss and the Lombard's Great Gloss are collated, every word of the first, either marginal or interlinear, will be found in the second, which is written out, line by line, as a continuous commentary. The Lombard is simply glossing and completing the Gloss.”Footnote 18 The fact that almost the entirety of the Glossa is contained within the Collectanea impressed Hans Hermann Glunz so much that he argued that the relationship must be the other way around — namely, that the Glossa is not the work of Anselm at all, but is the abbreviation of Lombard's Collectanea.Footnote 19 Smalley argued against Glunz's hypothesis, however, insisting that if we compare the Glossa on the Psalms to the Collectanea on the Psalms and to Herbert of Bosham's comments, it “becomes impossible to maintain that the Gloss could have originated as an extract from the Magna Glosatura. There could be no clearer case of a small work's being expanded into a larger work.”Footnote 20

This remains the scholarly consensus, with accounts of the Collectanea generally noting the link between Lombard and the Glossa. However, this is usually based not on new manuscript work but on reiterations of earlier claims or brief comparisons between the inadequate Rusch edition of the Glossa and Patrologia Latina version of the Collectanea.Footnote 21 Scholars have noted that Lombard expands and adds to the Glossa, but they have not done much to assess how he does this, generally being content briefly to compare a sample of each and present Lombard's work as the Glossa plus some comments and additional authorities.Footnote 22 In contrast, a few scholars have assessed Lombard's exegetical approach, comparing it to that of his contemporaries or to his own Sentences, but their discussions include little or no assessment of what role the Glossa plays in his exegesis.Footnote 23 For scholars of the Glossa and of Lombard alike, the fact that Lombard's Collectanea is based on the Glossa has been something to be noted but not investigated.

If we shift the set of questions we ask, however, along with shifting our focus to the manuscripts, it becomes clear that the Glossa remains central to understanding Lombard's Collectanea. I propose that we ask: how does Lombard approach the task of commenting on Paul? Where does he begin? What are the tools he uses? How does he introduce authorities and questions? Fundamentally, I am interested in treating Lombard's Collectanea as a historical act. This terminology comes from an important article by James Ginther, in which he argues that scholars need to take a different approach than is usual in understanding medieval theology. For Ginther, medieval theology, from lectio to summa, is about the interpretation of a text, and that text is the Bible.Footnote 24 Further, he points out that the medieval master encountered the Bible as a dispersed text — namely, he did not simply read it as a text by itself but also encountered it embedded within a variety of other contexts, from canon law collections to patristic texts to the liturgy. Hence, we need to “treat medieval scholastic exegesis as a historical act. As historians of theology, we need to identify how a scholastic exegete fulfilled his task, and what textual and cultural tools he exploited in the process since it is in these very tools that he also encountered sacred Scripture.”Footnote 25 In contrast to a frequent scholarly conception of the medieval approach to biblical exegesis that implies the existence of three independent elements (reader, Bible, authorities), Ginther offers a different model:

The alternative, then, is to envision a reader, the text under study, and that same text embedded in the sources which he is exploiting. Sometimes that embedding was a physical reality for the whole text, particularly if the reader was using a glossed Bible. As equally important, however, is the fact that the reader would also embrace the portions of the sacred text within other texts, sometimes during his reading, or often as a recollection of past encounters with that text. In other words, the medieval exegete experienced the Bible as a dispersed text.Footnote 26

Perhaps the most significant encounter with the Bible embedded within another text occurs with the use of the Glossa. The first element of understanding Lombard's Collectanea as a historical act, then, is to recognize that Lombard is commenting not simply on the Pauline Epistles, but on the Pauline Epistles embedded within the Glossa. In other words, the Glossa is not a “source” at Lombard's elbow, from which he draws when it has something relevant to say, or which he works to “incorporate” into his commentary on Paul.Footnote 27 Instead, the Glossa is the very text on which Lombard is commenting.

Hence, the first, and perhaps most significant, aspect of my analysis has been to take seriously the medieval claim about Lombard's fundamental approach in the Collectanea. And, as I will show below, it is possible to reconstruct Lombard's use of the Glossa through a careful comparison of the Collectanea with the twelfth-century manuscripts of the Glossa. This comparison will show that Robert of Auxerre and Vincent of Beauvais used exactly the right word when explaining Lombard's relationship with the Glossa: “explicuit.” Lombard “unfolds,” “opens up,” “expands” the Glossa; he “disentangles” it, “sets [it] in order,” and “arranges” it. His first exercise in approaching each verse of the biblical text is sorting through the various glosses and determining how they help in explicating Paul's words. Whether he is using the terse interlinear glosses to explicate the biblical text or clarifying the sense of a particular gloss, this level of his analysis is directed towards the immediate sense of Paul's words. Such an exercise will sometimes mean that Lombard sorts through glosses that appear contradictory or only fit together uneasily. At other times, he expands the Glossa’s quotation of auctoritates, clarifies the Glossa’s terminology, or uses the Glossa as a starting point for further analysis. Futher, Lombard will frequently bolster the Glossa’s interpretations by the deft addition of other auctoritates — the Glossa does not appear to stand as a sufficient auctoritas, as useful as it is for teaching scripture. Fundamentally, however, Lombard's teaching of the Pauline Epistles begins from and is shaped by the Glossa. In the Collectanea, the biblical text is interpreted through the Glossa within which it is embedded.

Lombard begins from the Glossa but he does not remain with it — he moves beyond it in important ways.Footnote 28 This primarily involves (a) drawing on a more extensive array of authorities and (b) raising and discussing many more theological questions.Footnote 29 Analysis of such moments is another key component of reading the Collectanea as a historical act, since they provide further contexts within which Lombard's encounter with the biblical text can be characterized as “dispersed.” Understanding these other contexts can help in understanding why Lombard includes the authorities he does, or why he raises a particular question at a given point.Footnote 30 Lombard is not elaborating the Glossa in a vacuum, but his questions, his answers, and his selection of authorities should also be understood as arising from and responding to the earlier commentary tradition and to discussions occurring in the schools at the time. I would argue, however, that if we begin from the Lombard's direct encounter with the biblical text embedded within the Glossa, we stand a better chance of understanding how these two factors, namely, his raising of theological quaestiones and his abundant quotation of authorities, fit into his work as a whole. We will see how they arise more or less organically through the encounter with the Bible as a dispersed text, whether it is embedded within the Glossa, within the auctoritates that Lombard has at his fingertips, within the debates over theological doctrine within the schools, or within other contexts as yet undetermined. I will show one way in which this can be approached in the case study portion of this article.

Turning to the Manuscripts

At the heart of the difficulty in assessing Lombard's use of the Glossa lies the fact that we possess reliable editions of neither the Glossa nor the Collectanea — most scholars rely on the Rusch editon of the former and the Patrologia Latina edition of the latter for their analyses. It is high time to turn to the manuscripts. Before presenting my case study of Lombard's use of the Glossa, I want to highlight several challenges inherent in this effort, which go a long way to explaining why such an exercise has not been attempted before. This will afford me the opportunity to note my own methodology in the case study that follows, as well as to highlight the areas of research that will require much more extensive analysis.

I would identify three related but separable challenges inherent in this exercise, each of which requires close attention: 1) establishing the text of the Glossa used by Lombard; 2) establishing the text of the Collectanea; 3) identifying the Glossa within the Collectanea. Each of these questions takes us to the heart of questions about textuality, orality, and the manuscript witnesses to scholastic thought, questions that have been brought to the fore in exciting new ways in recent scholarship.Footnote 31 All of these issues have relevance to the questions surrounding the Glossa and the Collectanea — I have only scratched the surface of them in this paper, but my goal is to point towards the work that needs to be done and the proper questions to be pursued.

First, it is necessary to identify the text of the Glossa itself, a task that has not been sufficiently investigated.Footnote 32 The Glossa for each book of the Bible has its own history, ranging from the relatively straightforward — as with the Glossa on Lamentations, for which we know the author and have a stable text; to the extremely complicated — as with the Glossa on Genesis, the Apocalypse, and Matthew, for which several different versions exist.Footnote 33 The Glossa on the Pauline Epistles falls somewhere between these two extremes. Although authorship of this book of the Glossa has been ascribed to Anselm of Laon (d. 1117), this is not at all certain and much more research will be required to sort out questions of authorship, possible relationships to full-length commentaries, and so on.Footnote 34 For my purposes in this article, however, none of this is crucial: what matters is the establishment of a workable text that can be compared to Lombard's Collectanea. In order to do this, I have consulted ten twelfth-century manuscripts of the Glossa on the Pauline Epistles. Two of these manuscripts are among the four oldest we possess, identified by Patricia Stirnemann (“not without hesitation”) as possibly of Laon origin and as dating from before 1140.Footnote 35 The basis of my transcription is one of these, Oxford, Christ Church 95; my readings generally follow this manuscript. Four of the other manuscripts, two of which are mentioned in the same Stirnemann article,Footnote 36 have been dated to the years 1140–50, while the other four have not been dated more specifically than the “twelfth century.”Footnote 37

Manuscripts of the Glossa on the Pauline Epistles Used for This Study

CC = Oxford, Christ Church 95 (bef. 1140?)

Re1= Reims, Bibliothèque municipale, MS 196 (bef. 1140?)

L1= Paris, BNF MS Lat. 14409 (1140–50)

L2= Paris, BNF MS Lat. 312 (twelfth century)

L3= Paris, BNF MS Lat. 313 (twelfth century)

L4= Paris, BNF MS Lat. 654 (twelfth century)

Tr1= Troyes, Bibliothèque municipale, MS 512 (1140–50)

Tr2= Troyes, Bibliothèque municipale, MS 880 (1140–50)

Tr3= Troyes, Bibliothèque municipale, MS 1026 (1140–50)

Vg1= Biblioteca Apostolica Vaticana, vat. lat. 140 (twelfth century)

There are a few key points relevant to my analysis below that have arisen from my comparison of these ten manuscripts. First, the Glossa of the twelfth century is not identical to, even if it forms the core of, the Glossa “Ordinaria” on the Pauline Epistles included in the fifteenth-century editio princeps by Adolph Rusch. The Rusch edition, or later sixteenth- or seventeenth-century printed editions, is still frequently used by scholars who want to reference the Glossa, but none of these editions accurately represents the twelfth-century manuscripts with which Lombard would have been familiar.Footnote 38 Indeed, the Rusch text itself may represent a version of the Glossa to which excerpts from Lombard's Collectanea have been added, as Mark Zier has recently suggested.Footnote 39 An analysis of Lombard's use of the Glossa therefore cannot begin from the Rusch text.

Second, the text of the Glossa is not an ordinaria, if by this term we mean a stable set of glosses arranged in the same way from manuscript to manuscript. Several of the manuscripts contain additional glossing in another hand; several contain unique glosses; there are also differences in the arrangement of glosses, with some glosses appearing at one verse in one manuscript and at a later verse in another; what is a single gloss in some manuscripts may be split into two or three glosses in another; and so on. My goal therefore was not to achieve a critical text that could represent a supposed “original” or “standard” version of the Glossa, but was instead to approximate the text that Lombard was using. Fortunately, there is an identifiable core of glosses that are the same, or nearly so, across all ten manuscripts and that correlate with Lombard's text. It is this core that I have transcribed and against which I have compared Lombard's Collectanea.

The second key stage in my analysis was to establish a better text of Lombard's Collectanea than has usually been used. If the Rusch edition of the Glossa is insufficient, then the version of the Collectanea in Patrologia Latina, volume 191, is equally so. In investigating several manuscripts of the Collectanea, Ignatius Brady came to the conclusion that there were at least two versions of the work, an earlier one (versio primitiva) composed prior to the Sentences, and a later one (textus receptus), which Lombard himself edited, perhaps for the purpose of teaching, at the same time as he was preparing the Sentences.Footnote 40 These do not differ in most aspects, but if we are to get as close to Lombard's actual use of the Glossa as possible, we will want to take this factor into account. I have used twelve manuscripts of the Collectanea, three of which were identified by Brady as being of the earlier version; it is one of these (Paris, BNF MS Lat. 17246) that formed the basis of my transcription and the readings of which I generally follow. Following Brady's criteria for identifying the versio primitiva, I have included five manuscripts that also seem to be the versio primitiva;Footnote 41 the other four are of the textus receptus, including the version prepared by Lombard's student Herbert of Bosham. The variations between manuscripts and between versio primitiva and textus receptus are small enough not to affect fundamentally the nature of my argument.

Manuscripts of Lombard's Collectanea Used for This Study

A1= Avranches, Bibliothèque municipale, MS 31 (vers. prim.)

A2= Avranches, Bibliothèque municipale, MS 32 (vers. prim.)

P1= Paris, BNF MS Lat. 14266 (text. recep.)

P2= Paris, BNF MS Lat. 17246 (vers. prim.)

R1= Reims, Bibliothèque municipale, MS 154 (text. recep.)

TC1= Cambridge, Trinity College B. 4.1 (text. recep.)

TC2= Cambridge, Trinity College B. 4.23 (vers. prim.)

TC3= Cambridge, Trinity College B. 5.6 (Herbert) (text. recep.)

TC4= Cambridge, Trinity College B. 5.20 (vers. prim.)

TC5= Cambridge, Trinity College B. 16.12 (vers. prim.)

V1= Biblioteca Apostolica Vaticana, vat. lat. 144 (vers. prim.)

V2= Biblioteca Apostolica Vaticana, vat. lat. 695 (vers. prim.)

Having established a useable text of the Collectanea, however, we still have to ask: do we have the Collectanea in its original form? Was it originally composed for Lombard's own use, as Herbert says was the case with the Collectanea on the Psalms, or was it composed for use in the classroom? The latter appears likely, given Brady's research, methodological considerations, and certain signs of orality within the text. Nevertheless, the text as we have it is several editorial stages removed from the classroom — it is a much more polished text than we would expect from a reportatio. Hence, disentangling the Glossa from the surrounding text is challenging, especially since almost the entirety of the Glossa is subsumed into the text of the Collectanea, undifferentiated from Lombard's own words or from the words of auctoritates. Any attempt to determine Lombard's use of the Glossa must tease it out through careful analysis. My point is that such can be done — but it requires attentive work. Alongside this, there has to be sourcing of the other auctoritates Lombard uses and analysis of how he treats the authorities contained in the Glossa. A proper sourcing of Lombard will not only identify when he is quoting Augustine, for example, but will also identify when he is commenting on the Glossa.

Case Study: Romans 1:17

I would like to show Lombard's methodology, and the difference that taking account of the Glossa can make, through a close reading of his comments at Romans 1:17, where some of the central themes of Paul's epistle are first stated. In verse 16, Paul defines the gospel as “the power of God unto salvation for all who believe, for the Jew first and for the Greek.” Now he explains, “For the iustitia of God is revealed in eo from faith unto faith, as it was written: ‘The just man lives by faith.’”Footnote 42 This is a pregnant passage that is made even more complicated (and hence fruitful for interpretation) by some ambiguity in its phrasing. First of all, there is the phrase iustitia Dei, which could have either a subjective or an objective genitival sense. Second, there is the claim that this iustitia is revealed in eo — but the antecedent of this pronoun is unclear. Does “in eo” refer to the gospel or the one who believes? Depending on how one answers these two questions, there remains the issue of what is meant by saying that this iustitia is revealed in eo. Finally, what is meant by saying that this revelation occurs ex fide in fidem?

The Glossa contains four marginal glosses on this verse, of which I will discuss three. In the chart below, the left-hand column contains the biblical text with the interlinear glosses written beneath, attached by a letter to the biblical phrase on which they are commenting — this system is not present in the manuscript, but is for ease of reference. The right-hand column contains the three marginal glosses that pertain to this verse. I have numbered them for ease of reference, and the variants are included in the apparatus criticus. In transcribing the marginal glosses, I have generally followed the spelling and readings of MS CC, providing modern punctuation.Footnote 43

The Glossa on Romans 1:17

The first marginal gloss (MGlos.1), taken from Ambrosiaster, defines iustitia Dei as “that which freely justifies the impious man through faith without the works of the law”; the gloss further explains that the gospel reveals this iustitia “when it gives to man the faith through which he who believes that God is just and true in his promises is justified.” Paul is evidently saying this “against the Jews, who deny that Christ is this man whom God promised.” Hence, the movement ex fide in fidem is a movement “from the faith of God who promises unto the faith of the man who believes.”

MGlos.1 contains traces of an ambiguity that it inherits from its source. This becomes clear through a comparison with the entire passage in Ambrosiaster from which this gloss is excerpted. In the chart below, the left-hand column contains the complete passage from Ambrosiaster with the sections excerpted by MGlos.1 in italics; the right-hand column contains MGlos.1.

Ambrosiaster and the Glossa on Romans 1:17

Ambrosiaster appears to provide two different (though related) definitions of iustitia, each of which interprets eo differently. At first, in a statement omitted by MGlos.1, Ambrosiaster explains that God's iustitia is revealed in the believer. Then, Ambrosiaster defines God's iustitia as what makes man just — this iustitia is revealed in the Gospel, since it is the Gospel that gives man the faith that makes him just. This is the interpretation carried over into MGlos.1. Then, however, Ambrosiaster provides another definition of iustitia, as God's veritas in promising — God gives what he had promised through the prophets. This iustitia is revealed in the believer, because God is proven to be truthful by the believer's faith in Christ, the one whom God promised. Ambrosiaster has shifted back to understanding eo as referring to the believer, not the Gospel. Thus, Ambrosiaster's interpretation of ex fide in fidem reads fides in two different senses: God's fidelity to his promises and man's belief in those promises. MGlos.1 includes excerpts from both senses, thus preserving the ambiguity of Ambrosiaster's account. As can be seen from the chart above, the bulk of MGlos.1 is excerpted from sections 1a and 3, in which Ambrosiaster sets out his understanding of iustitia as something imparted to man by God; it is in section 2, which MGlos.1 sums up in the phrase “ex fide dei promittentis in fide hominis credentis,” that Ambrosiaster understands iustitia as God's truthfulness in promising.

Yet another understanding of faith occurs in the second marginal gloss (MGlos.2). Here fides now seems to refer to the content of belief. MGlos. 2 expresses the movement ex fide in fidem by a series of parallel statements: “From the faith of the Old Testament, where one God [is worshipped], into the faith of the New, where Father and Son and Holy Spirit [is worshipped], from the faith of the first coming to the faith of the second,” and so on. Finally, this gloss also explains that iustitia belongs to the person who crosses from the faith of the Old Testament to that of the New and from the faith in words and hope into the faith of things and of sight. Thus, verse 17 is given a twofold sense here — Paul is saying that the iustitia that makes man just comes not only to those who cross from the old dispensation to the new but also to those who cross from the present life to the beatific vision. The third marginal gloss (MGlos.3), from Augustine's Quaestiones Evangeliorum, establishes that this eschatological interpretation of fides is acceptable: “Faith is that by which those things are believed that are not seen. And it is faith when belief is given not to words but to present things, which will be when God will have given himself as the one to be seen. Therefore he says ‘from the faith’ of words by which we believe what we do not see, ‘unto the faith’ by which we obtain what we believed.”

Without calling attention to the discrepancies, these three marginal glosses preserve several different senses of the term fides as Paul uses it in verse 17. These different understandings stem from the Glossa’s use of different auctoritates in expounding the verse, and the tension between them is not addressed. The Glossa’s immediate concern is not the issue of faith per se, but instead the explication of the biblical text through the application of authoritative statements. It seems clear why someone using the Glossa to interpret Romans would have to be careful and attentive if the Glossa was to clarify rather than confuse the issue.

Lombard recognizes these discrepancies in the Glossa and attempts to resolve them. If we compare Lombard's comments to the Glossa, we can see that he is using the Glossa to comment on Paul. In the quotation from the Collectanea below, as in all those that follow, the biblical lemmata are in majuscule letters, the portions of Lombard's text excerpted from the Glossa are in italics, biblical quotations are enclosed in single quotation marks, and quotations from auctoritates are enclosed in double quotation marks. The spelling and syntax generally follow P2, with modern punctuation included; variations between manuscripts are included in the apparatus criticus in the footnotes.Footnote 67

IUSTITIA ENIM DEI IN EOFootnote 68 REUELATUR EX FIDE IN FIDEM SICUT SCRIPTUM EST: IUSTUS AUTEM EX FIDE UIUITFootnote 69 Footnote 70 Footnote 71. Quasi dicatFootnote 72: uere euangelium credenti Footnote 73 est in salutem quia est ei Footnote 74 in iusticiam, que est causa salutis, quod clare ostenditur in ipso euangelio. Et hoc est quod ait. IUSTICIA ENIM DEI REUELATUR IN EO, euangelioFootnote 75. Iusticia Footnote 76 dei est qua Footnote 77 gratis iustificat impium per fidem sine operibus legis, ut alibi dicitFootnote 78 apostolus: ‘Inueniar in illo non habens meamFootnote 79 iusticiam que ex lege est sed illam que ex fide est’Footnote 80 “Hec est iusticia dei que in testamento ueteri uelataFootnote 81 in nouoFootnote 82 reuelatur que ideo iusticia deiFootnote 83 dicitur quia inparciendo eam iustos faciatFootnote 84.”Footnote 85 Hanc autem iusticiam euangelium reuelat, ipso effectu scilicet, dum dat fidem homini per quam iustificatur qui credit deum iustum et ueracem in promissis. “Reuelat etiam hanc iusticiam uerboFootnote 86 dum in eo dicitur: ‘Qui crediderit et baptizatus fuerit saluus erit,’Footnote 87 id est iustus.”Footnote 88

Lombard uses the first interlinear gloss (“credenti est … Euangelio”)Footnote 89 to provide the transition between verse 16 and verse 17; then he begins to gloss verse 17 by indicating explicitly that Evangelium is the antecedent for eo. He then moves immediately to the definition of iustitia Dei from MGlos.1, thus indicating that he sees these two glosses (interlinear gloss “a” and MGlos.1) as providing the same interpretation. Lombard follows MGlos.1's rephrasing of Ambrosiaster, but he adds the quotation from Philippians that the gloss had omitted. Lombard then adds a quotation from Augustine's De spiritu et littera that further explains MGlos.1, namely, the fact that God's justice is so called because it makes man just. Now Lombard turns to the next portion of MGlos.1, in which it explains why the Gospel is said to reveal this justice. Once again, Lombard follows the gloss's paraphrase of Ambrosiaster. He then adds words from Haimo that provide the same point.

For Lombard, then, the first interlinear gloss, plus the first sentence (“Iustitia Dei … in promissis”) of MGlos.1, constitute one interpretation of Romans 1:17a. God's justice makes man just, and this justice is revealed in the Gospel, which gives man the faith by which he can be justified. Now, however, Lombard provides another possibility that he takes from the sections of Ambrosiaster omitted from MGlos.1. In the excerpt below, each of the quotations from Ambrosiaster is in quotation marks; a comparison between this passage and Ambrosiaster's (see chart above) shows how Lombard includes almost everything that had been omitted by MGlos.1, rearranged to provide an alternative reading.

Vel, “iusticia dei est quiaFootnote 90 quod promisit dedit,” que iusticia reuelatur IN EOFootnote 91 “qui creditFootnote 92 se consecutum quod promiserat deus per prophetas suos.” Qui enim hoc credit et confitetur, “iustum deum probatFootnote 93 et testis est iusticie eius.” Et loquitur hicFootnote 94 apostolus contra iudeos qui negant hunc esse christum quem promisit deus.Footnote 95 Secundum hoc ita lege: Bene dixi euangeliumFootnote 96 in salutem esse omni credenti, “quia” IN EO scilicetFootnote 97 qui credit, siue iudeus sit siue grecus, REUELATUR IUSTICIA DEIFootnote 98, id estFootnote 99 iustus apparet deus et uerax. “In eo uero qui non credit iniustus uidetur, negat” enim ueracem deumFootnote 100 Footnote 101 qui non credit eumFootnote 102 dedisse quod promisit.” Vnde subditFootnote 103: EX FIDE IN FIDEM. Quasi dicat: iusticia reuelatur, tendens ex fide dei promittentis in fidem hominis qui credit ei, quia “in credente iustus deus apparet.”Footnote 104

As we have seen, this is the interpretation provided by Ambrosiaster in his second paragraph, which MGlos.1 summarized as “qui credit Deum justum et veracem in promissis.” For Lombard, however, this constitutes a new definition of God's iustitia. This definition fits better with the second half of MGlos.1, which Lombard now quotes (“contra Iudeos … promisit Deus”), explaining that Paul is speaking here to the Jews who deny that Christ is the one promised by God. In following this interpretation, Lombard uses Ambrosiaster again to explain the difference this makes when interpreting 1:17: “According to this, read it thus: well did I say that [the Gospel] is unto salvation for every believer, since the justice of God is revealed in the one who believes, whether he is a Jew or a Greek.” Lombard then makes explicit the link between this interpretation, culled directly from Ambrosiaster, and Ambrosiaster's interpretation of ex fide in fidem as expressed in MGlos.1: “Whence he adds: ‘from faith unto faith,’ as if he is saying: the justice of God is revealed, stretching from the faith of God who promises unto the faith of the man who believes him, since in the one who believes God appears just.”

What can be extracted from the above analysis? In the first place, it is clear that Lombard is proceeding through the biblical text by using the Glossa. Further, he attaches another auctoritas (Augustine, Haimo, and Ambrosiaster) to each segment of the Glossa that he quotes. This may be an indication that, while he uses the Glossa as his guide through Paul's text, it does not stand on its own as an auctoritas but requires bolstering from other recognized auctoritates. Second, Lombard is attentive to ambiguities and discrepancies within the Glossa and the auctoritates, as he makes more explicit what is left unaddressed by both the Glossa and by Ambrosiaster. Lombard recognizes that the sentiments contained in MGlos.1 represent two different interpretations of 1:17. He indicates that they are different and explains the difference more clearly through a deft rearrangement of the words from Ambrosiaster that that gloss had not included. Hence, not only is Lombard using the Glossa as an aid to studying Paul's text but also he is interpreting the Glossa so that it becomes useful in understanding the multiple interpretations of this verse.

This will become even clearer as we continue through Lombard's analysis. He now introduces a third possible understanding of Romans 1:17, contained in the other two marginal glosses, namely, the list of movements ex fide in fidem (MGlos.2) and Augustine's definition of faith (MGlos.3). Under this reading, God's iustitia is described as the “justice of faith,” and is said to belong to the person “who crosses from faith unto faith” (see below). Here, Lombard defines fides as adherence to all the articles of the creed, and he interprets the series of movements ex fide in fidem as expressions of these articles. He clarifies each of these brief statements, which begin by expressing a person's movement from partial to full belief but end by expressing the transition from the present to eternal life. We can see how his procedure below can be characterized as “glossing the gloss”: he briefly explicates each of the terse phrases from MGlos.2, sometimes adding a verb that is lacking in the gloss (“colebatur,” “colitur”), sometimes clarifying (“‘from the faith of the first coming unto the faith of the second,’ so that he might believe each”; “‘from the faith of the first resurrection,’ which is in the soul, ‘unto the faith of the second,’ which will be in [our] bodies”; “‘from the faith of the sowers,’ that is, the prophets, ‘unto the faith of the reapers,’ that is the apostles”). It is here that the medieval characterization of Lombard's method appears, his “explication” (explicauit) of the Glossa:

Vel secundum alteramFootnote 105 premissam sentenciam sic procede: IUSTICIA DEI est EX FIDE, hec autem iusticia fidei est ei homini qui transit EX FIDE IN FIDEM. Hoc dicitFootnote 106 propter omnes partes fidei, ut omnis qui uult per fidemFootnote 107 iustificari uniuersos articulos simboli habeat, ut transeat iudeus et quicumque alius ex fide ueteris testamenti ubi unus deus colebatur, in fidem noui Footnote 108 ubi pater et filius et spiritus sanctus colitur; ex fide primi aduentus in fidem secundi, ut utrumque credat; ex fide prime resurrectionis, que est in anima, in fidem secunde que eritFootnote 109 in corporibus; ex fide promissionis in fidem redditionis, ut credatFootnote 110 deum promisisseFootnote 111 ac reddidisseFootnote 112 uel redditurum fore; ex fide predicatorum in fidem populorum Footnote 113, ut credatFootnote 114 quod creduntFootnote 115 maiores et minores; ex fide seminantium, id est prophetarum, in fidem metencium, id est apostolorum.

The final movement “ex fide in fidem” provided by MGlos.2 is as follows: “Ei est iusticia qui transit de fide in fidem, ut transeat Iudeus et quicumque alius de fide ueteris in fidem noui testamenti, et de fide uerborum et spei in fidem rerum et speciei.” Lombard slightly rephrases this gloss (see the next excerpt from the Collectanea below), leaving out the repetitious reference to the movement from Old to New Testaments, and uses it to transition to MGlos.3, Augustine's definition of faith. Lombard does this because this movement, “from the faith of words and hope unto the faith of things and appearance,” is of a different kind than the earlier movements and expresses the way in which we will move from the faith we possess in this life to the enjoyment, in the next life, of the one in whom we have faith.

Transeat etiam iustificandus uel transeundum fore intelligat de fide uerborum et spei in fidem rerum et speciei. “Est enim fides qua creduntur ea que nonuidentur,” que proprie dicitur fides; “sed tamen est etiam fides rerum, quando” non uerbis sed rebus ipsis presentibus creditur, quod erit cumFootnote 116 per speciem manifestam se contemplandam sanctis prebebit ipsaFootnote 117 dei sapientia.”Footnote 118 “Non ergo” esset iusticia fidei, nisi esset absconditum quod predicatum crederemusFootnote 119 et credendo ad uidendum perueniremus.”Footnote 120 Dicit ergo iusticiam illi esse qui transit “ex fide uerborum, quibus credimus que non uidemus, in fidem rerum quacredita obtinebimus Footnote 121,”Footnote 122 scilicetFootnote 123 “ex fide in qua ministraturFootnote 124 deo in illam ubi” fruatur deo,”Footnote 125 que tamen inproprie dicitur fides.

Lombard's comments begin from the use of Augustine in MGlos.3, but Lombard not only quotes this gloss, but also adds to it from the same locus in Augustine's Quaestiones evangeliorum. This amplification of the Glossa’s auctoritates is a frequent practice for Lombard — he will often follow the Glossa’s wording of an auctoritas, but supplement it from the same source. In this instance, Lombard feels that MGlos.3's choice of Augustinian quotation requires clarification. Augustine distinguishes two kinds of faith, only one of which is properly called faith, according to Lombard. Lombard is interested in clarifying theological language where the Glossa renders it ambiguous or unclear. He does this, first, by expanding MGlos.3's reference, noting where Augustine speaks about faith “properly” or “improperly,” and, second, by introducing another quotation from Augustine that points to the same distinction. This is the quotation from the Enarrationes in Psalmos, in which Augustine explains that iustitia fidei refers to our belief in what is now hidden to us; by believing in this, we will eventually see it (that is, in heaven). Lombard thus uses another auctoritas to clarify the possible ambiguities present in the auctoritas selected by the Glossa. In doing this, Lombard may be taking his cue from the Summa sententiarum, an important theological text, perhaps of Victorine provenance, that was in circulation by 1137/38 and which constitutes an important source for Lombard in both his Collectanea and his Sentences.Footnote 126 In the Summa, the same Augustinian quotation used by MGlos.3 is connected to Romans 1:17 within a discussion of faith as belief in what is not seen.Footnote 127

Lombard now uses the presentation of faith in MGlos.2 and MGlos.3 to speak more generally about faith: “And since here mention is made of faith, it ought to be seen what faith is, and in how many ways faith may be taken, and about what it is concerned.” Lombard's discussion here also depends to a great degree on the discussion of faith in the Summa sententiarum. Here we see that Lombard's concerns extend beyond what the Glossa provides — he wants to investigate faith more deeply at this place. Hence the scholastic culture of the quaestio here becomes apparent, another context within which Lombard operates. Further, he takes his cue for the discussion of faith here from the Summa sententiarum, which also links its discussion of faith to Romans 1:17. The Summa is thus another text within which Lombard encounters the biblical text.

Et quia de fide mentio fit, hic uidendum estFootnote 128 quid sit fides, et quot modis accipiaturFootnote 129 fides, et de quibus sit. Fides est uirtusFootnote 130 qua creduntur que non uidentur. Apparentia enim non habent fidem sed agnitionem. Fides enim est quod non uides credere.Footnote 131 Accipitur autem fides tribus modis, scilicet pro eo quo creditur et est uirtus, et pro eo quo creditur et non est uirtus, et pro eo quod crediturFootnote 132. Fides enim qua creditur cum caritate uirtus est, et hec est fundamentum omnium bonorum in qua nemo perit. Hec fideles facit etFootnote 133 uere christianos. Alia uero demonumFootnote 134 estFootnote 135 et nominetenus christianorum. Nam etFootnote 136 demones credunt et contremiscunt. Hec est informis qualitas mentis, que dicitur informis quia sociam non habet caritatem, que est forma omnium uirtutumFootnote 137. Pro eo autem quod creditur accipitur fides, sicut ibiFootnote 138: “Hec est fides catholica, quam nisi,” etc. Ita et hicFootnote 139 accipitur cum ait EX FIDE IN FIDEM. Fidei enim queFootnote 140 creditur, id est simboli fidei, multe sunt partes, quarum alique sunt hic posite. Eodem modo quoque accipitur fides cum dicitur fides catholica, quasiFootnote 141 quodFootnote 142 uniuersaliter ab omnibus credendum est. Solet autem a quibusdam inquiriFootnote 143 utrum illa informis qualitas mentis que in malo christiano dictur fides, qua credit uniuersa que uere christianus credit, accedente caritate remaneat et uirtus fiat, an ipsa eliminetur et alia succedatFootnote 144 qualitasFootnote 145 que uirtus sit. Ad quod pocius diuinum oraculum implorandum uideturFootnote 146 quam aliquid diffiniendum a nobis; utrumlibet tamen sine periculo dici potest. Est autem fidesFootnote 147 de bonis etFootnote 148 de malis, et de preteritis et de presentibus et de futuris. Spes autem de bonis tantum et de futurisFootnote 149.Footnote 150

Lombard's discussion of faith does not take us very far from Paul's text — he is not interested simply in discussing the nature of faith but in making Paul's words clear. So he begins from the fact that Paul speaks about faith. The definition he provides is the Augustinian definition already mentioned, namely: “Faith is the virtue by which those things are believed that are not seen.” Lombard adds that belief in what is seen is knowledge rather than faith.Footnote 151 Later on, in his Sentences, he will spend a lot of time reiterating precisely this point — the proper object of faith is what is not seen (see below). His commitment to this position is already apparent in the Collectanea, first in his clarification of the Augustinian quotation provided by the Glossa, and now here when he turns to a formal definition.

Then, following Augustine in De Trinitate, Lombard provides three senses in which the term fides can be understood. He has two goals here — first, the clarificiation of theological language about faith, and second, determining to which sense of faith Paul refers at 1:17. Lombard explains that fides can refer to that by which one believes — this can either be a virtue, if it is informed by charity,Footnote 152 or it can not be a virtue, as in the demons who “believe and tremble.”Footnote 153 This latter fides is an “unformed quality of mind” and is not the fides that is the “foundation of all goods, in which no one perishes.” The third understanding of the term fides refers to what is believed, as expressed in the Creed. As Lombard explains, this is how Paul's use of the term should be understood here, in the phrase ex fide in fidem. This fides, which is one, consists of many parts, “some of which are placed here,” namely, in MGlos.2. Lombard's discussion of the three senses of the term fides thus arises from and is directed towards his reading of the biblical text along with Glossa. Now Lombard raises a pertinent question here, addressing a topic of debate within the schools, namely, whether unformed faith remains once it becomes faith formed by charity, or whether it is removed and replaced by a different faith. Lombard is cautious about expressing an opinion, allowing that both may be held.Footnote 154 Finally, Lombard explains that faith can also be differentiated from hope — faith is about things past, present, and future, both good and bad, while hope is only about future good things.Footnote 155 Having thus briefly discussed fides, he then returns to his exposition of the biblical text.Footnote 156

What Lombard provides here as a summary of faith, he will extend into a more thorough discussion in the Sentences, under appropriate headings and separated from the immediate goal of explicating Romans and the Glossa. So, in the Sentences he does not need to engage with the problematic Augustinian auctoritas used by the Glossa — he can simply define fides in its proper sense as “the virtue by which unseen things are believed.”Footnote 157 Then he provides the same three senses of the term fides he had discussed in the Collectanea, though with more extensive analysis of what it means to say that charity is linked to faith.Footnote 158 His discussion of the third sense, referring to the content of belief, is no longer attached to the Glossa’s list of movements ex fide in fidem — instead, Lombard draws on his discussion of Ephesians 4:5 in the Collectanea as well as Augustine's De Trinitate.Footnote 159 Then Lombard introduces several more auctoritates in order to prove the definition of faith with which he began, namely, that it concerns things not seen, in which analysis he explains the difference between faith and hope.Footnote 160 He then also explains how faith becomes the “foundation of all virtues and good works,” namely, through its joining with charity.Footnote 161 What was a simple summary, with a single quaestio, in the Collectanea has now become a more fully explicated depiction of the virtue of faith, bolstered by many auctoritates. Nevertheless, the close link between the two discussions remains.

Through this analysis of Lombard's comments on Romans 1:17, we can see how Lombard uses the Glossa in his assessment of Paul's words, in a manner that we can term “glossing the gloss.” His first approach to teaching the biblical text occurs as an engagement with the glosses on that text. He applies each gloss to the correct word or phrase from Paul's text, adding words when clarification is necessary. He carefully separates disparate elements within the various glosses and applies them to Paul's words in a way that indicates a clear structure, moving smoothly from point to point. He adds authorities to bolster the Glossa’s statements. Further, we can see how he brings to the text his own concerns — specifically, his interest in clarifying theological language, which he achieves both through the introduction of auctoritates and through a brief excursion away from the biblical text in order to investigate the nature of faith more thoroughly. Lombard's starting point is the biblical text with the Glossa, such that he uses the Glossa to identify the multivalent interpretations present in the divine words. He then deepens his engagement with the biblical text by raising further questions, clarifying theological language, and deftly applying new auctoritates to each interpretation.

Conclusion: Glossing the Glossa

The case study just conducted is meant to give a taste of what can be discovered through approaching Lombard's Collectanea as a historical act — specifically through reconstructing his use of the Glossa. Research on this topic is still in its nascent stages, with so little as yet known about the Pauline Glossa itself and with research into the manuscripts of the Collectanea remaining essentially where Brady left it in the 1970s. As case studies go, my example was a fairly straightforward one, since there is little variation in the glosses on Romans 1:17, and thus, Lombard's interaction with the Glossa can be established with a high degree of probability. Other examples are more complicated, wherein the manuscripts of the Glossa display greater variations and thus what Lombard is adding becomes less clear. Nevertheless, the same general pattern emerges, in which Lombard's first moment of engagement with the biblical text is through the Glossa. This engagement is something more than “expansion” and even “explication” — Lombard adds to the Glossa’s authorities, brings in many additional authorities, discusses theological questions, and develops his own insights. Nevertheless, the text he is explicating is not solus Paulus but the glossed Paul.

This analysis raises a question, however, that we have skirted without addressing: for whom is Lombard explicating the glossed Paul? Who is the audience for Lombard's exegetical project? As James Ginther has pointed out, the scholastic master's encounter with the biblical text was only the first activity in the construction of scholastic commentary — the second was the presentation of this work to the students.Footnote 162 The commentary is not simply the result of the master's working alone in his room; it is the result of and directed towards classroom activity. This raises the question, then, of the relationship between the scholastic text and the classroom setting. It is not sufficient for us simply to note that scholastic theology occurred in a classroom setting; this should also impact the way in which we approach scholastic texts. Recognizing that Lombard is in fact addressing himself to the glossed Paul should bring this issue into higher relief for the Collectanea since recent scholarship has shown how the Glossa functioned as a “taught text,” namely, as the basis for lectures in the scholastic classroom.Footnote 163 Is the case similar for the Collectanea?

Herbert of Bosham, in his introduction to his edition of the Collectanea on the Psalms, stated that Lombard composed the work for his own edification, not for teaching; it was only near the end of his career and at the insistence of his students that he began revising the work for lecturing.Footnote 164 Whether the Collectanea on the Pauline Epistles was initially composed for Lombard's own edification rather than for lecturing is not clear from Herbert's comments. As mentioned already, the Pauline Collectanea did have at least two versions: the earlier one (composed probably after 1148 according to Brady, though some scholars argue for an earlier date)Footnote 165 contained several extended theological reflections that were removed in the second version and utilized in the Sentences.Footnote 166 Was this a revision for the purpose of teaching, as some have suggested?Footnote 167 Or had it originally been composed for teaching? And what exactly did this composition look like? Was it Lombard alone working with the Glossa and authorities, or is there a more complicated background of orality and textuality, of classroom teaching, student interaction, and magisterial composition? Lesley Smith suggests that he may have composed the text for the purpose of teaching but did not release it for copying or allow an official reportatio to be circulated — “it was only after his death that Herbert made the texts available.”Footnote 168 The text of the Collectanea as we have it is certainly not a reportatio and appears to be several editorial stages removed from the classroom, although enough signs of orality appear in the text to suggest that the Collectanea is the result of classroom activity, or at least designed for use in the classroom.Footnote 169 All of this only goes to emphasize, however, how essential it is that work on the Collectanea turn to the manuscripts, especially when investigating the relationship between the earlier and later recensions.

The fact that the Collectanea appears as a more edited, polished text than the reportationes or lecture notes of other magistri should not deter us from approaching it as a textual witness to oral activity. Constant Mews, for example, called attention to the complex interaction between textuality and orality present in the various manuscripts containing Peter Abelard's sententiae.Footnote 170 These works are not collections written by Abelard himself as systematic treatises but are the result of a complex interaction between student notes of his lectures, his own editing of this material, and his use of it in later more polished works — a process similar to that which Hugh of St. Victor appears to have followed in composing his De Sacramentis.Footnote 171 The same process may lie behind the Sentences and the Collectanea of Peter Lombard: he may have used student notes of his lectures as the basis for his composition of these works.Footnote 172 These works, then, even when they come down to us in more polished form, are likely the end product of a process of teaching, student report, and editing. The student role in the production of magisterial texts in the schools, in fact, appears to have increased rather than decreased over the course of the twelfth century:Footnote 173 witness, for example, the recent argument of Quinto and Bieniak that Stephen Langton likely never produced a final version of his quaestiones, and that the manuscripts contain various versions of student reportationes of his disputationes.Footnote 174 That a similarly complicated process of teaching and student report lies behind biblical works of the time is also being established with more and more certainty. Mark Clark, for example, has demonstrated that Peter Comestor's Historia scholastica was a school text that was developed through teaching, reaching its final form through its use in the classroom by Peter Comestor's student Stephen Langton.Footnote 175 A text that has traditionally been characterized as “written” and as a “handbook” turns out to have a more complicated interaction of orality and textuality as its origin.

My own contribution to this discussion here should suggest similar possibilities in discussing Lombard's Collectanea. The Collectanea as we have it is not a reportatio — it does not have the flavor of the classroom that we sense from reading Peter Comestor's Gospel lectures or the biblical lectures of Stephen Langton. These latter texts are more clearly witnesses to an oral exercise and as such are not “texts” in the same way we conceive of texts today. Lombard's Collectanea, if it indeed began from a reportatio, was subsequently edited to such a degree that its relationship to the oral classroom setting is much less immediate.Footnote 176 Nevertheless, it seems most likely that it was composed with the classroom in mind — a classroom in which it was the glossed Pauline Epistles that was the text to be discussed. Just as the teaching context is crucial for understanding the nature of the Sentences, so is it crucial for understanding the Collectanea.Footnote 177 If we shift our focus to thinking of Lombard as first and foremost a teacher, of the text he is teaching as the Bible, and of the biblical text itself being embedded within the Glossa, we can gain a better foundation from which to assess his interpretive method, his goals, and the relationship between his biblical lectures and his Sentences. Restoring the centrality of the Glossa to interpretations of Lombard's Collectanea is thus of fundamental importance if we are to understand how the Magister Sententiarum approached his duties as a Magister in sacra pagina.

References

1 “Paucis annis (c. 1175) post mortem Petri Langobardi [sic], quidam anonymus abbatiae probabiliter Aquicinensis (Anchin) diocesis Atrebatensis eum aestimavit Scripturarum expositoribus illius temporis maxime praeferendum ‘eo quod ingenio sagaci et usu assiduo tanta in exponendis Scripturis luce claruerit ut pene magisterio doctoris non egeat qui glosarum ipsius lectioni animum intendere voluerit’” (Brady, Ignatius, Sententiae in IV Libris Distinctae; Tom. I, Pars I: Prolegomena [Grottaferrata, 1971]Google Scholar, 62*).

2 See Stegmüller, Friedrich, Repertorium Biblicum Medii Aevi, 11 vols. (Madrid, 1950–80)Google Scholar, 4:319–38. The title of Lombard's Pauline commentary is variously listed by Stegmüller as Glossa continua, Maiores glossae, Glossatura maior, or Magna glossatura Rom.-Hebr. He also includes the title given in PL 191, the Collectanea in omnes divi Pauli epistolas (Stegmüller, Repertorium, 4:336).

3 “HIC IACET MAGISTER PETRUS LOMBARDUS PARISIENSIS EPISCOPUS, QUI COMPOSUIT LIBRUM SENTENTIARUM, GLOSSAS PSALMORUM ET EPISTOLARUM CUIUS OBITUS DIES EST XIII KAL. AUGUSTI” (Here lies Master Peter Lombard, bishop of Paris, who composed the book of Sentences [and] the glosses on the Psalms and the Epistles, the day of whose death is on the thirteenth day before the kalends of August [i.e., July 20]) (quoted in Brady, Prolegomena, 43*, translation mine).

4 For Stephen Langton's Postillae on the Pauline Epistles, see esp. Dahan, Gilbert, “Les commentaires bibliques d’Étienne Langton: Exégèse et herméneutique,” in Étienne Langton: Prédicateur, bibliste, théologien, ed. Bataillon, Louis-Jacques, Bériou, Nicole, Dahan, Gilbert, and Quinto, Riccardo (Turnhout, 2010), 201–39CrossRefGoogle Scholar.

5 The most thorough recent account of Lombard's life and works is Doyle, Matthew, Peter Lombard and His Students (Toronto, 2016), 1112Google Scholar, which builds extensively on earlier chronological and biographical work by Joseph de Ghellinck, Damien Van den Eynde, Ermenegildo Bertola, Philippe Delhaye, and especially Ignatius Brady. See Doyle's notes and bibliography for these references.

6 See Brady, Prolegomena, 46*–89*.

7 In the footnotes and indices of his edition of the Sentences, Brady meticulously documents the many places in which Lombard draws on the authorities, and frequently the actual discussions, present in his Collectanea. An example of this will occur below in my case study. This close relationship between Lombard's biblical work and his Sentences is strong evidence that the practice of theology in the twelfth century was inseparably linked to the study and teaching of the Bible.

8 For the Collectanea on the Pauline Epistles, see esp. Colish, Marcia L., Peter Lombard, 2 vols. (Leiden, 1994), 1:155–58Google Scholar, 189–225, which builds on her earlier articles: From sacra pagina to theologia: Peter Lombard as an Exegete of Romans,” Medieval Perspectives 6 (1992): 119Google Scholar, and Peter Lombard as an Exegete of St. Paul,” in Ad Litteram: Authoritative Texts and Their Medieval Readers, ed. Emery, Kent Jr. and Jordan, Mark D. (Notre Dame, IN, 1992), 7192Google Scholar. See also her more recent discussion of faith in the Collectanea: Faith in Peter Lombard's Collectanea,” in “Fides virtus”: The Virtue of Faith from the Twelfth to the Early Sixteenth Century, ed. Forlivesi, Marco, Quinto, Riccardo, and Vecchio, Silvana (Münster, 2014), 3951Google Scholar. For her discussion of the Collectanea on the Psalms, see Colish, Peter Lombard, 1:155–88, which builds on her earlier article, Psalterium Scholasticorum: Peter Lombard and the Emergence of Scholastic Psalms Exegesis,” Speculum 67 (1991): 531–48Google Scholar.

9 For some brief discussions of the Collectanea on the Pauline Epistles, see, for example, Doyle, Peter Lombard, 90–94; Levy, Ian Christopher, The Letter to the Galatians, Bible in Medieval Tradition (Grand Rapids and Cambridge, 2011), 5255Google Scholar, with a translation of the PL version of the Collectanea on Galatians 2 at 185–206; Levy, , The Letter to the Romans, Bible in Medieval Tradition (Grand Rapids and Cambridge, 2013), 3334Google Scholar, with a translation of the PL version of Lombard's prologue to Romans at 59–64; Rosemann, Philipp W., Peter Lombard (Oxford, 2004), 4448CrossRefGoogle Scholar. Two extremely useful articles to which I will refer at greater length below are Dahan, Gilbert, “Le Livre des Sentences et l'exégèse biblique,” in Pietro Lombardo: Atti del XLIII Convegno storico internazionale, Todi, 8–10 ottobre 2006 (Spoleto, 2007), 333–60Google Scholar, and Mark A. Zier, “Peter Lombard and the Glossa Ordinaria: A Missing Link?,” in Pietro Lombardo: Atti del XLIII Convegno, 361–409. For the Collectanea on the Psalms, see Patrizia Stoppacci, “Le Glossa continuae in Psalmos di Pietro Lombardo: Status quaestionis; studi pregressi e prospettive di ricerca,” in Pietro Lombardo: Atti del XLIII Convegno, 289–331.

10 Rosemann, Philipp W., “New Interest in Peter Lombard: The Current State of Research and Some Desiderata for the Future,” Recherches de théologie et philosophie médiévales 72 (2005): 133–52CrossRefGoogle Scholar, at 145.

11 The term “ordinaria” is a later term and does not in any case accurately depict the twelfth-century reality of the Glossa. See, for example, the cautionary comments made in Smith, Lesley, The “Glossa Ordinaria”: The Making of a Medieval Bible Commentary (Leiden, 2009), 1216Google Scholar.

12 Cf., for example, Andrée, Alexander, “Peter Comestor's Lectures on the Glossa ‘Ordinaria’ on the Gospel of John: The Bible and Theology in the Twelfth-Century Classroom,” Traditio 71 (2016): 203–34CrossRefGoogle Scholar, at 205; Froehlich, Karlfried, “Christian Interpretation of the Old Testament in the High Middle Ages,” in Biblical Interpretation from the Church Fathers to the Reformation, Variorum Collected Studies (Farnham, Surrey, and Burlington, 2010), 504Google Scholar; Van Engen, John, “Studying Scripture in the Early University,” in Neue Richtungen in der hoch- und spätmittelalterlichen Bibelexegese, ed. Lerner, Robert E., Schriften des historischen Kollegs Kolloquien 32 (Munich, 1996), 1738CrossRefGoogle Scholar, at 24–25.

13 This perspective is certainly not entirely absent from other scholarly discussions of the Collectanea — Colish's discussion of Lombard's methodology is precisely geared towards unpacking how he engaged with Paul in a more robust and attentive manner than other twelfth-century commentaries, while the statement that Lombard lectured on the Pauline Epistles is common (e.g., Brady, , Sententiae in IV Libris Distinctae; Tom. II [Grottaferrata, 1981]Google Scholar, 19*; Colish, Peter Lombard, 1:24; Rosemann, Peter Lombard, 45). Nevertheless, much more detailed work is required if we are to determine exactly how the text of the Collectanea relates to Lombard's classroom lectures and what it might mean to treat the Collectanea as reflecting the oral culture of the classroom.

14 “Nam cum hec opera scriberet, nequaquam, sicut ipsomet referente didici, ipsi venit in mentem, quod in scolis publicis legerentur; solum ob id facta, ut antiquioris glosatoris, magistri videlicet anselmi laudunensis, brevitatem elucidarent obscuram” (Glunz, H. H., History of the Vulgate in England from Alcuin to Roger Bacon: Being an Inquiry into the Text of Some English Manuscripts of the Vulgate Gospels [Cambridge, 1933], 343Google Scholar). It is not entirely clear whether this comment is meant to refer only to the Collectanea on the Psalms or also to that on the Pauline Epistles, although most scholars assume it refers to both. Either way, as will be shown below, it is clear that Lombard's starting point in the Collectanea on the Pauline Epistles is the Glossa on the Pauline Epistles.

15 See Brady, Prolegomena (n. 1 above), 82*–83*, including his argument for placing these works in 1168 rather than in 1142.

16 “Petrus … etiam glosaturam super psalterium et epistolas Pauli ab Anselmo per glosulas interlineales marginalesque distinctam, post a Gisleberto continuative productam, latius apertiusque explicuit” (quoted in Brady, Prolegomena, 74*).

17 “Nam cum [Lombardus] esset inter Franciae magistros opinatissimus, glossaturam epistolarum et psalterii ab Anselmo per glossulas interlineares marginalesque distinctam et post a Gilberto continuative productam latius et apertius explicuit multaque de dictis sanctorum addidit” (quoted in Brady, Prolegomena, 53* n. 1).

18 Smalley, Beryl, “A Collection of Paris Lectures of the Later Twelfth Century in the MS. Pembroke College, Cambridge 7,” Cambridge Historical Journal 6 (1938): 103–13CrossRefGoogle Scholar, at 109.

19 Glunz, Vulgate, 214–15.

20 Smalley, , “Gilbertus Universalis, Bishop of London (1128–34), and the Problem of the ‘Glossa Ordinaria,’Recherches de théologie ancienne et médiévale 7 (1935): 235–62Google Scholar [Part 1]; 8 (1936): 24–60 [Part 2], at 36. According to Smalley, Glunz retracted his thesis in his correspondence with her (Smalley, “Collection of Paris Lectures,” 109). Smalley's arguments against Glunz's hypothesis in the two articles just cited remain convincing. Glunz's argument essentially rests on a belief that it is untenable “that the Lombard had adhered to his original so faithfully as to keep the whole text word for word.” However, as Smalley points out, “the Lombard's attitude is fully explained by the veneration which the twelfth-century scholars, a free lance like Abailard [sic] excepted, felt for Master Anselm. Recent studies too are showing how widespread a practice it was to build on the text of one's predecessors” (Smalley, “Collection of Paris Lectures,” 109). One could also note that Herbert of Bosham and the other medieval witnesses quoted above explicitly state that Lombard was “explicating” the Glossa — if the text he is discussing is the Glossa, and the point is to explain it, then there is nothing odd about that text being fully incorporated into his own. Further, recent work, emphasizing the originally oral nature of twelfth-century biblical commentaries, has demonstrated even more clearly how masters built on the works of their predecessors and how masters lectured on the glossed texts. See, for example, Alexander Andrée, “Peter Comestor's Lectures”; Clark, Mark J., “Peter Comestor and Stephen Langton: Master and Student, and Co-Makers of the Historia Scholastica,” Medioevo 35 (2010):123–49Google Scholar; and Clark, , The Making of the Historia scholastica, 1150–1200 (Toronto, 2015)Google Scholar.

21 Biblica Latina cum Glossa Ordinaria: Facsimile Reprint of the Editio Princepts, Adolph Rusch of Strassburg 1480/1, introd. Froehlich, Karfried and Gibson, Margaret T., 4 vols. (Turnhout, 1992)Google Scholar; PL 191 and 192.

22 For instance, Beryl Smalley expresses the opinion that “Lombard is simply glossing and completing the Gloss” (Smalley, “Collection of Paris Lectures,” 109). Lesley Smith, after comparing the Glossa and the Collectanea on Rom. 3:1–3 in order to give “a sense of the proportion of Gloss to non-Gloss text” in the latter work, similarly concludes that the Collectanea is “on the whole a clarification and enlargement of the Gloss, rather than a departure from it” (Smith, Glossa Ordinaria [n. 11 above], 202, 203). Both Smalley and Smith, of course, are primarily concerned with the Glossa and its use, not with Lombard's methodology. Mark Zier very usefully compares Lombard to the twelfth-century Glossa and to Rusch in order to show how the Rusch text includes extensive material added to the Glossa from the Collectanea; his insights are invaluable pointers towards the necessity of returning to the manuscripts, but they do not contain any analysis of Lombard's use of the Glossa (Zier, “A Missing Link” [n. 10 above]).

23 Marcia Colish is the most thorough in assessing Lombard's approach to biblical exegesis, but her emphasis lies elsewhere than with Lombard's use of the Glossa; she is interested in determining how Lombard's approach improved upon other Pauline exegesis of the time and how it points forward to his approach in the Sentences. As such, while she notes a few places where Lombard's analysis begins from the Glossa, she focuses on the ways his approach moves beyond it (for example, Colish, Peter Lombard [n. 8 above], 1:197, 201; Colish, “Lombard as Exegete of St. Paul” [n. 8 above], 74). Philipp Rosemann characterizes the Collectanea as Lombard's “revision of the Gloss on the Pauline Epistles” and focuses on how many authorities Lombard adds before turning to Lombard's theological discussions and their relationship to the Sentences — again, the Glossa is something to be mentioned and left behind in assessing Lombard's approach (Rosemann, Peter Lombard [n. 9 above], 44–48). Gilbert Dahan provides an illuminating study of Lombard's exegetical approach in the Sentences, which includes a comparison of several excerpts from the Collectanea on Galatians to passages from the Sentences, arguing for a close link between “exegesis” and “theology” in Lombard, but his focus is on the connection between the Collectanea and the Sentences, not with Lombard's approach to explicating Paul (Dahan, “Livre des Sentences” [n. 10 above]).

24 For similar insights pointing to the centrality of the Bible in scholastic theology, see, for example, Andrée, “Peter Comestor's Lectures” (n. 12 above); Clark, Mark J., “Peter Comestor and Peter Lombard: Brothers in Deed,” Traditio 60 (2005): 85142CrossRefGoogle Scholar; Dahan, “Livre des Sentences”; Henri de Lubac, Exégèse médièvale: Les quatre sens de l’Écriture, 4 vols. (Paris, 1959–64), 1:59–60, 111; G. Paré, A. Brunet, and P. Tremblay, La renaissance du XIIe siècle: Les écoles et l'enseignement, Publications de l'Institut d’Études Médiévales d'Ottawa 3 (Paris and Ottawa, 1933), 213; and Saccenti, Riccardo, “The Materia super libros Sententiarum Attributed to Peter Comestor: Study of the Text and Critical Edition,” Bulletin de philosophie médiévale 54 (2012): 155215CrossRefGoogle Scholar.

25 Ginther, James R., “There is a Text in This Classroom: The Bible and Theology in the Medieval University,” in Essays in Medieval Philosophy and Theology in Memory of Walter H. Principe, CSB: Fortresses and Launching Pads, ed. Ginther, James R. and Still, Carl N. (Aldershot, UK, 2005), 3151Google Scholar, at 35.

26 Ginther, “Bible and Theology,” 34.

27 Scholarly comments on the Collectanea can sometimes imply such a view of Lombard's practice, as, for example: “Early in his teaching career, Peter Lombard began making commentaries on the Psalms and Pauline Epistles, using the Gloss as a source, and incorporating all of the Gloss material into his own work” (Smith, Glossa Ordinaria, 78); “Die Quellenanalyse ergibt: Die ‘magna glossatura’ des Lombarden zu Röm. 13, 1–7 ist eine reine Kompilation. Hauptquelle ist die ‘parva glossatura’ Anselms, die Petrus Lombardus fast ganz übernimmt wobei er überwiegend wörtlich zitiert” (Affeldt, Werner, Die weltliche Gewalt in der Paulus-Exegese: Röm. 13, 1–7 in den Römerbriefkommentaren der lateinischen Kirche bis zum Ende des 13. Jahrhunderts [Göttingen, 1969], 158CrossRefGoogle Scholar); “Sicut Glossa Anselmi in Psalmos inter fontes primarios Glossae Lombardi in eosdem … sic et Glossa Anselmi in Apostolum, eius brevitate non obstante, quasi incorporatur in Glossatura Magistri et inde transit etiam in libros Sententiarum” (Brady, Prolegomena [n. 1 above], 75*).

28 Cf. Colish, Peter Lombard, 1:197, 201; Colish, “Lombard as Exegete of St. Paul,” 74.

29 These two aspects of the Collectanea have been the major focal points of most scholarly interactions with it. Colish (Peter Lombard, 1:193), for example, notes that Lombard's extensive quotation from authorities sets him apart from other Pauline commentators and, further, that Lombard goes beyond simple citation to engage critically with these authorities and to resolve apparent contradictions between them (Colish, Peter Lombard, 1:207–8).

30 “By pointing to these textual influences, we must do much more than just simple source criticism, for it is here that the idea of the dispersed text is a force with which we must reckon. We need to dispense with the image of medieval exegesis as a two step process: read the text, and then read the sources. Instead, these two steps could happen simultaneously. We ought to envision in what context the expositor has experienced the text now before his exegetical eye. Is it, for example, a central text to a specific liturgical feast or function? Does the master encounter the text either in part or in whole in the patristic sources he has consulted? In what context does the Glossa ordinaria place the lemma at hand? Is the exegete quoting from a canon law collection, and in what way does that section of canon law focus his exegesis? And finally, in what ways do the guidelines for spiritual exegesis shape his interpretation? … Even if we can answer these questions in part, we may at least be able to see how and why the teaching master has employed other passages of the sacred text as a means of expounding the lemma under scrutiny. We need to have a careful look at how the interpreting portion of Scripture arrives in the mind of the exegete, for as a dispersed text a biblical citation was normally embraced as part of another text or context. Understanding the way in which a biblical lemma is received is a fundamental factor for describing medieval exegesis as a historical event” (Ginther, “Bible and Theology,” 37–38).

31 See discussion and references in my conclusion below.

32 The Glossa Ordinaria in general remains underinvestigated, with most scholarly treatments of it still basing themselves on Beryl Smalley's ground-breaking work in the first half of the twentieth century. More recently, there have been efforts to return to the manuscripts and advance our understanding of the field on a book-by-book basis and through the lens of the Glossa’s use in the classroom. See, for example, the discussion by Alexander Andrée in his “Peter Comestor's Lectures” and the scholarship he cites on the Glossa on Genesis, Lamentations, Ecclesiastes, the Song of Songs, the Apocalypse, and the Gospels of Matthew and John. For Smalley's research, see esp. Beryl Smalley, “Gilbertus Universalis” (n. 20 above); Smalley, , “La Glossa Ordinaria: Quelques prédécesseurs d'Anselme de Laon,” Recherches de théologie ancienne et médiévale 9 (1937): 365400Google Scholar; Smalley, , “Les commentaires bibliques de l’époque romane: gloses ordinaire et gloses périmées,” Cahiers de civilization medievale 4 (1961): 1522CrossRefGoogle Scholar; and Smalley, Study of the Bible. The most extensive recent treatment of the Glossa, which summarizes and builds on Smalley's scholarship, is Lesley Smith's Glossa Ordinaria (n. 11 above); for a discussion and partial critique of some of Smith's claims, see Andrée, Alexander, “Laon Revisited: Master Anselm and the Creation of a Theological School in the Twelfth Century,” Journal of Medieval Latin 22 (2012): 257–81CrossRefGoogle Scholar.

33 See Andrée, “Peter Comestor's Lectures,” 206–7 and the references found there.

34 Smalley sets out the evidence in favor of Anselm's authorship of the Pauline Glossa in the second part of her article “Gilbertus Univeralis.” Although she concludes that “we can hardly deny to Anselm the authorship of the Glossa on the Psalter and St. Paul” (34), she also admits that she was unable to find any manuscripts of the Glossa on the Pauline Epistles that attribute it to Anselm (32). Further, we cannot yet know whether what we have in the manuscripts of the Pauline Glossa is in fact Anselm's own composition, a distillation of his teaching made by his students, or an abbreviation of a continuous commentary perhaps written by Anselm. A recent monograph by Cédric Giraud has shown how the collections of sentences attributed to Anselm were not his compositions but were compiled from his teaching by his students (Giraud, Cédric, Per verba magistri: Anselme de Laon et son école au XIIe siècle [Turnhout, 2010]CrossRefGoogle Scholar); work by Alexander Andrée on the Glossa on John has shown how that text was abbreviated from an earlier continuous commentary from the school of Laon (Andrée, , “Anselm of Laon Unveiled: The Glosae super Iohannem and the Origins of the Glossa Ordinaria on the Bible,” Mediaeval Studies 73 [2011]: 217–60Google Scholar). It seems more likely that the Pauline Glossa was composed in one of these ways rather than that Anselm was its author; in any event, the exact nature of his relationship to the Pauline Glossa has not been established. Cf. Alexander Andrée, “Laon Revisited,” 274: “Thus there is no doubt that Anselm was associated with the glossing of Scripture; but jumping from that to the claim that he produced the Glossa on any particular book is quite a large leap. We just do not know in what precise way he was involved. The case seems to parallel that of the collections of sentences studied by Giraud: though these certainly transmit the teachings and memory of Anselm, the master had very little to do with their actual compilation. Judging by available evidence, it would seem that the Glossa is a similar testimony to the teachings of the Master and his school, and was thus gathered by his students rather than by the Master himself.”

35 The manuscripts in question are Reims, BM 195; Reims, BM 196; Paris, Bibl. Mazarine 125; and Oxford, Christ Church 95. See Stirnemann, Patricia, “Où ont été fabriqués les livres de la Glose Ordinaire dans la première moitié du XIIe siècle?,” in Le XIIe siècle: Mutations et renouveau en France dans la première moitié du XIIe siècle, ed. Gasparri, Françoise (Paris, 1994), 257301Google Scholar, at 261–62.

36 The two manuscripts are Paris, BNF MS Lat. 14409 and Troyes, BM 512 (Stirnemann, “Où ont été fabriqués les livres,” 266–68). The latter manuscript was a part of the set of glossed books donated by “Henry, the king's son” to Clairvaux when he entered the abbey in 1146 (264–65).

37 The more specific dating is from Stirnemann, “Où ont été fabriqués les livres,” 261–68, and Martin Morard, “Principaux manuscrits de la Bible glosée des XIIe et XIIIe siècle,” Glossae.net, Gloses et Commentaires de la Bible au Moyen Âge: Portail de Ressources Numériques, last updated 12 January 2016, http://glossae.net/fr/content/manuscrits-de-la-bible-glos-e-des-xiie-et-xiiie-si-cle. The dates of the other manuscripts are those provided by the catalogues in their respective libraries.

38 Cf. Smith, Glossa Ordinaria, 15. This has not been sufficiently noted in scholarship, with some scholars stating that the Rusch edition contains the early version of the Glossa. See, for example, Margaret T. Gibson, “The Glossed Bible,” in Biblia Latina, 1:ix n. 32; and Smith, Lesley, “The Glossed Bible,” in New Cambridge History of the Bible, vol. 2, From 600 to 1450, ed. Marsden, Richard and Matter, E. Ann (Cambridge, 2012), 363–79CrossRefGoogle Scholar, at 370. Zier's analysis, cited below, has demonstrated that this is not the case. Nevertheless Rusch is still frequently used as a witness to the twelfth-century text, including in comparisons between Lombard and the Glossa — for example, in Smith, Glossa Ordinaria, 202–3; and Woodward, Michael Scott, trans., The Glossa Ordinaria on Romans (Kalamazoo, 2011), xviixxGoogle Scholar.

39 Mark A. Zier, “A Missing Link” (n. 10 above). In comparing the text of the Glossa, Lombard, and Rusch on Rom. 1:1–7, Zier found not only that the Rusch text “is several times the length of the text as it is found in the 12th and 13th century manuscripts,” but that Lombard's Collectanea contains both the text of the Gloss from the manuscripts and “virtually all of the additional text as it appears in the Rusch edition” (Zier, “Missing Link,” 363–64). Hence, “with respect to the manuscripts of the Gloss, Peter seems to have incorporated virtually all of the material found there, either glossing the Gloss or expanding citations that had been abbreviated in the Gloss. But it seems most likely that Peter's text, in turn, became a principal source for the additional text found in the Rusch edition of the Gloss.” How this occurred is not at all clear, since there is “an apparent lack of manuscript witnesses to the full text of the Rusch edition” (Zier, “Missing Link,” 380–81).

40 Brady, Prolegomena (n. 1 above), 65*. Looking at manuscripts in Paris and in the Vatican, Brady identified four that appeared to represent the versio primitiva and fifteen of the textus receptus. For descriptions of these manuscripts, see Brady, Prolegomena, 66*–71*.

41 The differences between the two versions are generally minor; however, he did identify a few brief treatises on theological topics that are present in the versio primitiva but not in the textus receptus — they had been removed and instead utilized in the Sentences (see Brady, Prolegomena, 86*–88*). In particular, Brady identified three such treatises in Romans — a lengthy treatise on the incarnation of the Word at 1:3, a treatise on the Trinity and the sin against the Holy Spirit at 1:20, and a treatise on the procession of the Son and the Holy Spirit at 11:36. Using several manuscripts not consulted by Brady, I divided these into versio primitiva or textus receptus based on the presence or absence of these three treatises. Brady provides transcriptions of the treatise De processione Filii et Spiritus Sancti in the Prolegomena to his first volume of the Sentences (90*–93*), and of De incarnatione in the Prolegomena to his second volume (54*–77*); the treatises on the Trinity and on the sin against the Holy Spirit at Romans 1:20 are provided in footnotes to the two places in the Sentences where Lombard draws on them (1.34.4 and 2.43.11, respectively).

42 Non enim erubesco evangelium. Virtus enim Dei est in salutem omni credenti, Iudaeo primum et Graeco. Iusticia enim Dei in eo reuelatur ex fide in fidem, sicut scriptum est: “Iustus ex fide uiuit.”

43 Critical text and apparatus here includes CC, 2r; L1, 1v; L2, 6v; L3, 3r; L4, 2v; Re1, 7r; Tr1, 5r–5v; Tr2, 8v; Tr3, 2r; Vg1, 2v–3r.

44 est] om. L3

45 in] om. L1 L2

46 iusticiam] iustitia L3

47 est] om. Tr2

48 non ex lege] om. CC

49 ex fide … ex lege] ex fide est iusticia et ita salus, sicut abacuc dicit: “Iustus est ex fide,” et ita uiuit uita aeterna CC ex fide est iusticia et ita salus, sicut abacuc dicit Tr1 ex fide est iusticia et ita salus, sicut abacuc dicit L4

50 iustus est … vita] “Iustus est ex fide,” et ita uiuit aeterna uita non ex lege Tr1 “Iustus est ex fide,” et ita uiuit aeterna uita L4 non ex lege add. in alia manu L4

51 qua] quae CC Tr1 qua corr. ad quae L3 Tr3 qua uel quae L2

52 quam2] quem L1

53 qui credit ei] credentis CC Tr1

54 ex fidei … credit ei] part of gloss 2 CC separate gloss Tr1

55 fidem] fide L4 Vg1

56 redditionis] redemptionis L1 Tr2 redemptionis corr. ad reditionis Tr1 reditionis corr. ad redemptionis L2

57 predicatorum] predicantium Re1

58 ei] eius Re1

59 in fidem rerum] in fide rerum L1

60 qua] quam CC

61 creduntur quae] creduntur ea quae Vg1

62 credimus] om. Tr1

63 quae non uidemus] quod non uidemus L2 L3 Re1 Tr3 Vg1

64 fides est … obtinebimus] om. L1

65 Ambrosiaster, In Epist. ad Rom. 1:17, recension ƴ, sec. 1–3 (CSEL 81:37–39).

66 Phil. 3:9.

67 Critical text here includes A1, 11ra–11va; A2, 6vb–7ra; P1, 14va–15rb; P2, 7r–7v; R1, 6v–7r; TC1, 10va–11rb; TC2, 7va–8ra; TC3, 15rb–15vb; TC4, 10ra–10rb; TC5, 8r–8v; V1, 11rb–vb; and V2, 6vb–7rb.

68 IN EO1] om. P1

69 EX FIDE UIUIT] EX SUA FIDE UIUIT P2

70 ENIM DEI … FIDE UIUIT] om. TC1

71 EX FIDE IN FIDEM … FIDE UIUIT] om. TC3

72 Quasi dicat] IUSTITIA ENIM DEI, etc. Quasi dicat TC2 TC 4

73 credenti] omni credenti TC3 TC 4 TC 5

74 ei] om. V2

75 euangelio] scilicet euangelio A1 P1 P2

76 Iusticia] Iusticia enim P1 TC4 5

77 qua] quod TC1

78 dicit] ait P1

79 meam] illam R1 TC3

80 Phil. 3:9.

81 uelata] uelata est A1 P1 P 2 TC1

82 nouo] nouo testamento A2 TC4

83 dei] om. V1

84 faciat] facit TC2

85 Augustine, Spir. et litt. 11.18 (CSEL 60:170–71).

86 uerbo] uerbo ipso TC4

87 Matt. 16:16.

88 Cf. Haymo of Halberstat (incorrect attribution; actually Haimo of Auxerre), Expositio in divini Pauli epistolas: In epistolam ad Romanos (PL 117:361–508, at 372D): “Justitia Dei, id est justificatio qua justificat in se credentes, in Evangelio manifestatur dum dicitur. Qui crediderit et baptizatus fuerit, salvus, id est justus, erit.”

89 Interlinear gloss (a) in chart above.

90 quia] qua TC4 V2

91 IN EO] IN EO id est in homine P1

92 credit] crediderunt A1

93 probat] probat et ueracem in promissis TC1 TC2 TC3 TC4 TC5 V1

94 hic] hoc TC1 V1

95 deus] deus pater P2

96 euangelium] om. P1 R1 TC1 TC3 V1

97 scilicet] om. A1 V2

98 IUSTICIA DEI] om. V1

99 id est] om. P2

100 ueracem deum] ueracem esse deum TC4

101 deum] christum A1

102 eum] deum TC2

103 subdit] subditur V1

104 Ambrosiaster, In Epist. ad Rom. 1:17, recension ƴ, sec. 1–3 (CSEL 81:37–39).

105 alteram] aliam V2

106 hoc dicit] id est A2

107 per fidem] fide TC1

108 noui] noui testamenti P1

109 erit] est TC3

110 ut credat] ut scilicet credat V1

111 deum promisisse] deum uerum promisisse P1

112 reddidisse] redisse TC4

113 populorum] apostolorum P2

114 credat] credant A1 R1 TC1 TC 2 TC 3

115 credunt] crediderunt TC1

116 cum] quando TC1

117 prebebit ipsa] prebebit et ipsa TC1

118 Cf. Florus of Lyons, Expositiones epistolarum beati Pauli apostoli ex libris sancti Augustini doctoris eximi a quodam Floro collecte (Paris, BNF MS Lat. 11575, 3va): “Intelligitur quidem fides qua creduntur ea quae non uidentur: sed tamen est etiam fides rerum quando non uerbis sed rebus ipsis presentibus creditur. Quod futurum est: cum iam per speciem manifestam se contemplandam prebebit sanctis ipsa dei sapientia per quam facta sunt omnia.” From Augustine, Quest. ev. 2.39.2–8 (CCL 44B): “Quod dixerunt discipuli: Domine, auge nobis fidem (Luke 17:5), potest quidem intellegi hanc fidem sibi eos augeri postulasse, qua creduntur ea quae non uidentur; sed tamen dicitur etiam fides rerum, quando non uerbis sed rebus ipsis praesentibus creditur, quod futurum est, cum iam per speciem manifestam se contemplandam praebebit sanctis ipsa dei sapientia per quam facta sunt omnia.” See also Summa sententiarum septem tractatibus distincta (PL 176: 41–174, at 44D): “Idem Augustinus in libro Quaestionum Evangeliorum: Est etiam fides rerum, quando non verbis, sed rebus ipsis praesentibus creditur; cum jam per speciem manifestam se contemplandam praebebit sanctis ipsa Dei sapientia. De qua fide rerum lucisque ipsius praesentia forsitan Paulus dicit: Justitia enim Dei revelatur in eo ex fide in fidem.”

119 crederemus] credemus V1

120 Augustine, Enarrat. Ps. 59.8.5–7 (CCL 40).

121 obtinebimus] obtinebamus P1

122 Augustine, Quaest. ev. 2.39.15–17 (CCL 44B). Cf. Florus, Expositiones, 3va.

123 scilicet] om. TC1

124 ministratur] ministraui V1

125 Cf. Augustine, Quaest. ev. 2.39.39–43 (CCL 44B): “Quid haec pertineant ad id quod dictum est: ‘Domine, auge nobis fidem’ (Luke 17:5), difficile apparet, nisi intellegamus ex fide in fidem, id est ex fide ista qua ministratur deo in illam fidem eos significasse transferri ubi fruantur deo.” Cf. Florus, Expositiones, 3va.

126 The Summa’s authorship has been disputed, with some scholars opting for Otto of Lucca and others settling for the “Victorine Anonymous.” See Colish, Marcia L., “Otto of Lucca, Author of the Summa sententiarum?,” in Discovery and Distinction in the Early Middle Ages: Studies in Honor of John J. Contreni, ed. Chandler, Cullen J. and Stofferahn, Steven A. (Kalamazoo, 2013), 5770Google Scholar. The Summa has been characterized as “a succinct compendium of Hugh [of St. Victor]’s teaching supplemented with significant patristic authorities as expounded by Anselm of Laon” that was “often critical of ideas being promoted by Peter Abelard” (Constant Mews, J. and Monagle, Clare, “Peter Lombard, Joachim of Fiore and the Fourth Lateran Council,” Medioevo 35 [2010]: 81122Google Scholar, at 95–96).

127 “Videntur tamen quaedam auctoritates velle quod fides etiam de iis quae videntur sit, ut in Joanne Nunc autem dico vobis priusquam fiat, ut cum factum fuerit credatis (John 13:19)… . Idem Augustinus in libro Quaestionum Evangeliorum: Est etiam fides rerum, quando non verbis, sed rebus ipsis praesentibus creditur; cum jam per speciem manifestam se contemplandam praebebit sanctis ipsa Dei sapientia. De qua fide rerum lucisque ipsius praesentia forsitan Paulus dicit: Justitia enim Dei revelatur in eo ex fide in fidem (Rom. 1:17). Sed potest dici quod Augustinus hoc dicat, non de sacramento fidei, sed de re fidei. Est enim ipsa fides qua nunc Deum cernimus per speculum in aenigmate, sacramentum illius futurae visionis qua Deum videbimus facie ad faciem (1 Cor. 13:12)” (PL 176:44C–45A).

128 est] om. A2

129 accipiatur] dicatur A2

130 uirtus] uirtus mentis TC4

131 Cf. Summa (PL 176:44B–C): “Illud quoque sciendum est quod fides est solummodo de iis quae non videntur. Gregorius Homilia tertia: Apparentia non habent fidem, sed agnitionem. Item in Dialogo: Cum Paulus dicat: Fides est substantia rerum sperandarum, argumentum non apparentium (Heb. 11:1); hoc veraciter dicitur credit, quod non valet videri. Nam credi jam non potest, quod videri potest. Thomas aliud vidit, aliud credidit; hominem vidit, et Deum confessus est dicens: Deus meus et Dominus meus (John 20:28). Credimus ut cognoscamus; non cognoscimus, ut credamus. Quid est enim fides, nisi credere quod non vides? Fides ergo est quod non vides credere; veritas quod credidisti videre.”

132 et pro eo quo creditor … quod creditur] om. A2

133 et] om. TC1 TC3 TC5 V1 V2

134 demonum] demonium TC2

135 est] om. P2 V1

136 et] om. A1 R1

137 uirtutum] om. V2

138 ibi] hic P2

139 Ita et hic] Ita hic TC3 Et ita et hic A1 Et ita P1 R1 TC3

140 que] qua R1 TC1 V2

141 quasi] om. TC1

142 quod] quidem P2

143 inquiri] queri A2

144 succedat] succedit A1 TC2

145 qualitas] caritas A1 TC2 caritas corr. ad qualitas A2

146 uidetur] dicimus A2 TC4

147 autem fides] autem et fides V1

148 et] om. P2 TC1

149 Spes … futuris] add. in marg. P2

150 Cf. Summa (PL 176:44A–B): “Hoc distat inter fidem et spem, quod fides est de praeteritis, ut sunt nativitas et passio Christi; et de praesentibus, ut quod in altari est verum corpus Christi; et de futuris, ut est immortalitas. Spes autem de futuris tantum est; item fides est de bonis et malis; spes de bonis tantum adipiscendis.”

151 Cf. Summa (PL 176:44B–C): “Illud quoque sciendum est quod fides est solummodo de iis quae non videntur. Gregorius Homilia tertia: Apparentia non habent fidem, sed agnitionem. Item in Dialogo: Cum Paulus dicat: Fides est substantia rerum sperandarum, argumentum non apparentium (Heb. 11:1); hoc veraciter dicitur credit, quod non valet videri. Nam credi jam non potest, quod videri potest. Thomas aliud vidit, aliud credidit; hominem vidit, et Deum confessus est dicens: Deus meus et Dominus meus (John 20:28). Credimus ut cognoscamus; non cognoscimus, ut credamus. Quid est enim fides, nisi credere quod non vides? Fides ergo est quod non vides credere; veritas quod credidisti videre.”

152 Cf. Summa (PL 176:45A–B): “Solet quaeri de fide utrum sit virtus. Prosper ex dictis Augustini: Tres sunt, inquit, summae virtutes, fides, spes, charitas. Et Apostolus: Credidit Abraham Deo, et reputatum est ei ad justitiam (Gal. 3:6). Item fides habet meritum; sed nihil habet meritum nisi virtus. Quidam tamen dicunt quod non sit virtus, propter illud Apostoli: Si habuero omnem fidem, charitatem autem non habeam, nihil sum (1 Cor. 13:2), ubi innuit quod fides sine charitate possit haberi; sed nulla virtus sine charitate potest esse. Sane tamen potest dici quod fides per dilectionem operans sit virtus, sine dilectione non est virtus. Si opponatur, tunc sunt duae fides: non sunt duae, sed eadem aucta, unde illud: Adauge in nobis fidem (Luke 17:5), ut enim dicit Hieronymus: Quantum credimus, tantum diligimus; et e diverso.”

153 Cf. Summa (PL 176:44A): “Ut enim dicit Augustinus in lib. De fide et operibus: Fides quae per dilectionem operatur, fundamentum est: non fides daemonum qua ipsi credunt et contremiscunt.”

154 When discussing the same question in the Sentences, although he reiterates his claim that “utrumlibet sine periculo dici potest,” Lombard indicates that he prefers the former solution: “Mihi tamen videtur quod illa qualitas quae prius erat, remaneat, et accessu caritatis virtus fiat” (Lombard, Sent. 3.23.5 [n. 13 above]).

155 Cf. Summa (PL 176:44A–B): “Hoc distat inter fidem et spem, quod fides est de praeteritis, ut sunt nativitas et passio Christi; et de praesentibus, ut quod in altari est verum corpus Christi; et de futuris, ut est immortalitas. Spes autem de futuris tantum est; item fides est de bonis et malis; spes de bonis tantum adipiscendis.”

156 Marcia Colish (Peter Lombard, 1:195) has emphasized that Lombard's theological “excursions” arise organically and do not take him very far from the biblical text. We can see precisely this in this example; in addition, however, we also see how his attentiveness to the biblical text is attentivenesss to biblical text + Glossa and follows the Summa sententiarum, within which he would have encountered this same verse embedded within a discussion of faith.

157 “Fides est virtus qua creduntur quae non videntur” (Lombard, Sent. 3.23.2). Translations from the Sentences are from Lombard, Peter, The Sentences, trans. Silano, Giulio, 4 vols. (Toronto, 2007–10)Google Scholar.

158 Lombard, Sent. 3.23.3–5.

159 Lombard, Sent. 3.23.6.

160 Lombard, Sent. 3.23.7–8.

161 Lombard, Sent. 3.23.9.

162 Ginther, “Bible and Theology” (n. 25 above), 38.

163 Andrée, “Peter Comestor's Lectures” (n. 12 above), 205, 229.

164 Cf., e.g., Doyle, Lombard and Students (n. 5 above), 36, 107; and Rosemann, Peter Lombard (n. 9 above), 43.

165 For Brady's reasoning, see Prolegomena (n. 1 above), 83*–88*. His dating is accepted by, e.g., de Hamel, Christopher, Glossed Books of the Bible and the Origins of the Paris Booktrade (Woodbridge, Suffolk, 1984)Google Scholar, 7 n. 54; Doyle, Lombard and Students, 90; Rosemann, Peter Lombard, 225 n. 36; and Zier, Mark A., “Peter Lombard,” in Dictionary of the Middle Ages, 13 vols. (New York, 1987), 9:516–17Google Scholar. Damien Van den Eynde dates the first recension earlier (“Essai chronologique sur l'oeuvre littéraire de Pierre Lombard,” in Miscellanea Lombardiana [Novara, 1957], 45–63, at 45, 53–55), in which he is followed by, e.g., Colish, Peter Lombard (n. 8 above), 1:23 n. 26.

166 See n. 41 above.

167 See n. 13 above.

168 Smith, Glossa Ordinaria (n. 11 above), 200.

169 For instance, “Peter warns his readers to ‘be careful that you understand in this way’ and to ‘therefore distinguish it in this way’” (Doyle, Lombard and Students, 91).

170 Constant Mews, J., “The Sententie of Peter Abelard,” Recherches de théologie ancienne et médiévale 53 (1986): 130–84CrossRefGoogle Scholar; repr. in Constant Mews, J., Abelard and His Legacy (Aldershot, UK, 2001)Google Scholar.

171 Mews, “Sententie,” 163. One of Hugh's students, Laurence, would take notes of Hugh's lectures, which Hugh would then correct each week; Hugh then used these corrected sententiae in composing the De sacramentis (Mews, “Sententie,” 160).

172 As suggested by Riccardo Saccenti in “Materia” (n. 24 above), 185–86, with examples.

173 “The increasing role of the reportatio as a means of literary production is certainly linked to the necessities and requirements of the students, but it is also a consequence of the progressive definition of the nature of the work of the masters. The system of the reportatio proved to be an efficient means of literary production for the preservation of a master's teaching and its diffusion” (Saccenti, “Materia,” 186).

174 Quinto, Riccardo and Bieniak, Magdalena, “General Introduction,” in Stephen Langton: Quaestiones Theologiae Liber I (Oxford, 2014), 366Google Scholar, at 26–30 and 37–42.

175 See the references to Clark in n. 20 above.

176 The fate of the Collectanea can be compared to that of Lombard's other biblical glosses. There is no doubt that Lombard lectured on books of the Bible other than the Psalms and Pauline Epistles. Clark, Mark J. summarizes the evidence for this assertion in “The Biblical Gloss, the Search for Peter Lombard's Glossed Bible, and the School of Paris,” Medieval Studies 76 (2014): 57113Google Scholar, at 60–76, drawing on the earlier work of Ignatius Brady and Beryl Smalley (cf. Clark, “Search,” 57 n. 2). At his death, Lombard's personal library contained “glossed books of the entire New Testament and many of the books of the Old Testament” (Doyle, Lombard and Students [n. 5 above], 42); subsequently, these books were lost. Scholars have debated whether these refer to copies of the Glossa “Ordinaria” on those books, or whether they might refer to Lombard's own glosses, representative of his teaching. Operating on the latter assumption, both Beryl Smalley and Ignatius Brady searched for extant copies of these works without success (Clark, “Search,” 76–81). Recently, Mark Clark has suggested a new method for finding Lombard's lectures, namely, through the investigation of Lombard's students’ lectures — Clark (“Search,” 81–113) has located the presence of Lombard's introduction to the Gospel of John in Peter Comestor's lectures on that Gospel, and has argued more recently that Stephen Langton's lectures on the Old Testament are based on Lombard's and thus contain parts of these lost lectures (Clark, , “Peter Lombard, Stephen Langton, and the School of Paris: The Making of the Twelfth-Century Scholastic Biblical Tradition,” Traditio 72 [2017]: 171274CrossRefGoogle Scholar). Unlike the Collectanea, which was edited and released for general consumption at some point, these glosses seem to have existed only as lecture notes, used by other magistri in their own lectures but not receiving the same kind of editorial work.

177 Classroom teaching as the context for the Sentences has recently been emphasized by Lombard's English translator, Giulio Silano: “It makes little sense to separate the work of teaching from the effort to identify and point out the coherence of the Christian tradition. An appreciation of the importance of teaching seems preferable to the view that the undertaking in which the masters were engaged was the elaboration of systematic theology. For one thing, the expression ‘systematic theology’ is not a twelfth-century term and would make little sense to the masters; for another, it betrays a degree of abstraction quite foreign to them and may lead us to speak too readily of their doctrine rather than of their teaching. The enterprise in which they were engaged was a deeply personal one; if it also became rational, scientific, or whatever else one may wish to call it, it was because these features of their activity were effective in making the tradition alive and relevant to their students and the larger communities whom those students would serve. It was not out of ideological presuppositions that they prized technique, rationality, or dialectic, but because, without these, they would not be offering their students what was required for the lively understanding and reduction to present normativeness of the massive inheritance they had received from earlier times” (Silano, Sentences [n. 134 above], 1:xxiv).