The medieval English and Latin writings of Wulfstan, archbishop of York, represent a captivating literary case study with the potential to reveal much not only about Christian religious perspective during an era of Viking raid and Scandinavian settlement, but also in terms of multilingual writing practices and textual reception during an emergent era of literacy. In this study, I present first a brief overview of Wulfstan’s cultural context as well as an appraisal of his general creative approach as a writer/orator. Second, I provide a specific examination of Wulfstan’s series of four homilies on the baptismal rite, with the aim of producing a characterization of his engagement with multilingualism, both as a literary device and as a component of his overall creative process. Ultimately, through comparative reference to Wulfstan’s source material and analogy with modern multilingual instance, I develop notions from previous research into medieval bilingual sermon writing and argue that Wulfstan’s pattern of bilingual behavior corresponds with and even emblematically underscores significant noted aspects of his rhetorical style.

Wulfstan’s Cultural Context and Creative Approach

After becoming bishop of London in 996, Wulfstan would later serve simultaneously as bishop of Worcester and archbishop of York from 1002. Coupled with his religious authority and homiletic writing, Wulfstan held political influence during a particularly pivotal period, with his characteristic writing style identified in the legislatures of Kings Æthelred and Cnut (see first: Whitelock, 1942, 1955). The overlapping nature of these roles has created difficulties for modern commentators in terms of editing and categorizing Wulfstan’s various writings as religious or legal, as indeed Patrick Wormald (2004, p. 17) characterizes: “The simple truth is that his earlier laws are heavily homiletic, and his later homilies are very like laws.” Elsewhere, Wormald (2000, p. 205) makes the related observation that, “The significance of Wulfstan’s later homilies is not that they fit neither homiletic nor legal categories but that they fit both.”

This perspective on Wulfstan’s overlapping roles makes for an expedient point of entry into the critical analysis of Wulfstan’s body of work, as from the outset it reminds the modern reader of the artificiality of present-day critical divisions between genres. If such distinctions were not actively recognized by Wulfstan, his audiences, or his compilers, then one equally ought to show a bit of hesitance toward anachronistic differentiation. Moreover, an unwavering insistence upon distinction and finished products would entirely belie Wulfstan’s overall agenda and modus operandi. The no-frills, efficient balancing and mixing of responsibilities actually serves as a neat metaphor for his writing practice, general style, and engagement with multilingualism.

At a fundamental linguistic level, Wulfstan’s own vernacular writing has been characterized as most probably “conventional late West Saxon,” though it contains a comparatively large number of Scandinavian borrowings, e.g., his consistent preference for “lagu” [law] over “æ” [law] (Bethurum, 1957, pp. 49–54). It would be easy to assume that these borrowings are a result of influence from interactions with speakers in his Danelaw archdiocese of York, which “had a sizeable Anglo-Scandinavian population made up of the descendants of late-ninth- and early-tenth-century settlers” (Lionarons, 2010, p. 75). The explanation is not entirely straightforward, however.

Sara Pons-Sanz (2007, pp. 193–230) notes that while Wulfstan’s Scandinavian usages do seem to increase later in his career, it is virtually impossible to know precisely how much time Wulfstan physically spent in York and how much contact with Norse speakers he would have had there. Pons-Sanz further identifies that in counterpoint to more mundane Norse borrowings, which increasingly crop up in the written record from the Middle English period onwards, Wulfstan’s own usages are altogether less commonplace and might more reasonably be interpreted as resulting from contacts at the Danish court of King Cnut. This supports Richard Dance’s (2004, p. 53) interpretation that Wulfstan’s Scandinavian usages are less “the result of any particular dialectal association or attempt at ethnic bonding” and more the sort of technical vocabulary one would expect from a writer familiar with contemporary legal and administrative procedure. I would suggest that this evidence need not be understood in a negative light. As an individual language user, it seems not unreasonable that Wulfstan reproduced language he encountered in daily interactions.

It is necessary to understand Wulfstan as a participant in a more fluid linguistic continuum. Matthew Townend (2000, 2002) describes the everyday language situation as something of a functionally bilingual society made up of Old English speakers and Old Norse speakers, though not necessarily with many fully bilingual individuals. Specifically, Townend (2002, p. 183) argues that the two languages would likely not have been totally mutually intelligible (i.e., “complex sentences [being] spoken and understood in all their lexical variety and syntactic fullness”), but that many speakers would, to a certain degree, have shared a sort of “adequate or pragmatic intelligibility” (i.e., “the ability to understand individual words” and complete “face-to-face and day-to-day transactions” without recourse to becoming totally bilingual or making use of interpreters).

A first-language user of English, Wulfstan possessed Latin proficiency commensurate with his roles as a high-ranking church official and statesmen. He also demonstrated a conscious awareness of linguistic differentiation and emblematic value, as evidenced in nuce by his Latin cognomen “Lupus” [wolf], deriving from the first element of his name and famously attached to the Sermo Lupi ad Anglos. Andy Orchard (2004, pp. 68–69) has argued that Wulfstan’s purposeful distinction between languages can be perceived even at the physical level of manuscript writing and textual mise-en-page, pointing to a manuscript containing Wulfstan’s “Commonplace Book” (Copenhagen, Kongelige Bibliotek 1595, fol. 66v, ll. 16‒31), where Wulfstan’s hand seems to demarcate the two languages of the text (cf. Cross & Tunberg, 1993). This demarcation is in apparent counterpoint to the now well-established contradictory evaluation put forward by Neil Ker (1971, p. 316), the scholar responsible for identifying Wulfstan’s hand: “The most obvious point about Wulfstan as a scribe is his disregard for the conventional distinctions of letter-form according to whether the language used is Latin or Old English.”

Rather than aligning with either side of this debate, I would instead recommend that reconciliation of both realities provides a more revealing insight into Wulfstan’s creative mindset. Wulfstan did take language differentiation into account while producing a text but did not view distinction merely for distinction’s sake as a particularly high priority. My view is reinforced by the example of another manuscript that contains Wulfstan’s handwriting, Oxford, Bodleian Library, Hatton 20. Fols. 1r‒1v feature a linguistically revised version of the prefatory letter to the Alfredian translation of Gregory’s Pastoral Care, a text which by Wulfstan’s period was already more than a century old and accordingly made use of earlier West Saxon language and spellings. The emendations are extensive and encompass changes to sense, morphology, and orthography. However, as Dance (2004, p. 39) has demonstrated, these changes lack consistency; some elements and spellings are updated by Wulfstan’s hand, while others are “left untouched.” These unaltered instances appear to be ones which would not pose a problem to comprehension, despite being atypical and differing from Wulfstan’s own late West Saxon. In summary then, what one sees dominating the writings that are associated with the archbishop at the foundational levels of language use and physical production of texts is a marked emphasis on practical utility and immediate accessibility.

With respect to his language choice and style, one branch of scholarship views Wulfstan as a lingering exemplar of a writer working from within Germanic oral traditions. Orchard (1992, p. 259) groups Wulfstan stylistically with the likes of Aldhelm and the Beowulf poet as examples of “literate Anglo-Saxons who chose to compose in the traditional oral style of vernacular verse,” as opposed to “the more modern literate and literary methods of authors like Bede and Alcuin and Ælfric” whose methods would eventually come to eclipse those of the former group. Major features of Wulfstan’s style include alliteration, occasional rhyme, and his characteristic two-stress rhythm which Milton Gatch (1977, p. 20) highlights as producing greater urgency than “the more leisurely four-stress pattern of Ælfric.” This rhythm tied in with Wulfstan’s frequent use of comparative and contrastive pairings, (e.g., “to hæbbenne 7 to healdenne” [to have and to hold]; “bryne 7 blodgyte” [burning and bloodshed]), as well as the persistent inclusion or addition of intensifying words (e.g., “georne” [eagerly]; “swyðe” [very, exceedingly]) and compounds with first elements intended to serve as intensifiers (e.g., “þeod-” as in “þeodwita” [man of great wisdom] and “þeodlicetere” [arch-hypocrite]) (see further: Bethurum, 1957, pp. 87–98). There is also a great deal of repetition throughout the Wulfstan canon, not only in terms of word choice and use of stock phrases, but also of theme.

Wulfstan’s rhythmic prose can effectively straddle the modern dividing line between poetry and prose, as well as the line between written text and oral sermon. Orchard (1997, p. 103) writes that,

The difficulty of distinguishing between ‘oral’ and ‘written’ elements in Old English literature is therefore a fundamental one, and underlies a range of other tensions between opposing concepts which help to define the field, notably those between verse and prose, secular and religious, and vernacular and Latin; such tensions are inevitably reflected in a literature which depicts the transition from an illiterate and secular society […] to one in which literacy was introduced by a Christian religion which depended on the Book for its authority, and which encouraged (and to a large extent controlled) the spread of reading and the written record.

As important as this orality is, Orchard’s (1997, p. 120) general principle that “the Latin text requires to be read, the English to be heard,” demands a degree of skeptical qualification.

Orchard’s phrasing is certainly not meant as the hard and fast injunction it seems to be when stripped of context, but it is nevertheless akin to the “hasty correspondences” which Jan Ziolkowski (1991, p. 193) warns against when confronting medieval Latin Literature in any cultural context. Even when appropriately grasped in a theoretical sense, it can be difficult to avoid inadvertently foisting modern perceptions of Latin as a dead language backwards in time as part of a modern reading. It is meaningful, therefore, to pause and actively acknowledge on-the-ground linguistic realities. Ziolkowski (1991, p. 195) further cautions the modern reader that,

Latin was a learned language, but none of the ancient or medieval cultures in which it was used were fully literate. […] To be taught effectively to children who came from these transitional cultures, Latin had to be both an oral and a literate/textual phenomenon.

Simultaneously, Herbert Schendl (2013, p. 153) offers the general reminder that, “we should approach medieval written texts as the products of a literate multilingual group of writers and scribes mainly targeted at a bilingual readership.” Thinking then about possible audiences for Wulfstan’s homilies, a central question becomes: How much of a prolonged diatribe in Latin could people, even clergy, legitimately understand in a purely aural, listening context?

Comments relevant to Latin competence made by Wulfstan’s contemporary, Ælfric of Eynsham, paint a somewhat conflicting picture of the situation (Lionarons, 2010, pp. 79–81). On the one hand, Ælfric bemoans the rampant and at times nigh absurd impropriety of both laity and priests. On the other, he elsewhere seems to assume not only the priests’ ability to perform the liturgy correctly, but also their possessing a well-stocked library and sufficient Latin competence to follow the difficult Latin tags and antiphons he uses in his instructions. Perhaps the most telling evidence is that both Ælfric and Wulfstan saw fit to translate exegetical texts and produce sermons in the vernacular addressed to the clergy, since “not all priests understood Latin, and pastoral instructions, biblical exegesis, and other religious educational materials must therefore be translated for them” (Lionarons, 2010, p. 81). A middle-ground of varying levels of fluency seems the most probable reality.

To a degree, it is even possible to call into question Wulfstan’s own Latin expertise. Regarding the full designation of Wulfstan’s representative work, Sermo Lupi ad Anglos Quando Dani Maxime Persecuti Sunt Eos Quod Fuit Anno Millesimo XIIII ab Incarnatione Domini Nostri Iesu Cristi [sermon of the wolf to the English when the Danes greatly persecuted them, which was in the 1014th year from the incarnation of our Lord Jesus Christ], it has been noted that the syntax favors an Anglicized word order where Classical Latin or a deliberately Classicizing Latinist might prefer “eos persecuti sunt” (Orchard, 2007, p. 313). In addition, there is a separate instance of marginally suspect Latin verse that was presumably not authored by Wulfstan, but which does name and praise the archbishop and was apparently copied in his hand in London, British Library, Cotton Vespasian A. xiv. Orchard (2007, p. 330) highlights here the unrevised use of the active infinitive “comere” [to adorn] akin to English, where a passive infinitive would more reasonably conform to Latin grammatical norms if not to the line’s scansion (Orchard suggests “comeri*” but perhaps “comi” [to be adorned]?). Native language syntax, pronunciations, and spellings commonly would influence regional Latin usages (see further Ziolkowski, 1991, p. 205). This phenomenon can be seen, for instance, in the earlier period polygraphic renderings of Latin words on the rear panel of the Franks Casket,Footnote 1 or indeed, in such historically distant contexts as the modern foreign language or TESOL classroom, so-called ‘language transfer’ or ‘L1 interference’ (see e.g., Jarvis & Pavlenko, 2008) Aesthetically, it may be hard to argue with the opinion that “Wulfstan tended to express his rhetorical genius more fully in the vernacular, notwithstanding his own sense of what is appropriate in Latin” (Orchard, 2004, p. 83). Yet, ignoring or discounting the archbishop’s uses of Latin unhelpfully glosses over a critical component of his creative process.

Reading Wulfstan’s writing reveals him to have interacted with source texts in both languages in a very similar way, all throughout the different stages of his creative process. Lionarons (2010, p. 111) succinctly characterizes the archbishop’s procedure as: “first compiling a set of Latin quotations on a subject; then translating selections from these into English, augmented by further quotations or by his own writing; and finally composing a longer, more developed homily making use of both the Latin and English texts.” Whether Latin texts of the Church Fathers, works of Ælfric, or his own previous writings, Wulfstan was constantly in the process of revisiting, revising, and revamping source materials. No matter the language of his source text, Wulfstan embellished through the introduction of “echo-words” (often alliterative) and through linked repetition (Bethurum, 1957, p. 32; Orchard, 2004, pp. 72–73). Again, these tendencies seemingly bring Wulfstan more closely into alignment with vernacular, oral traditions. However, Don Chapman (2002) has persuasively presented that Wulfstan shows an equally keen awareness of the entirely different structures and styles already present in his Latin source material and works to preserve such in his adaptations where doing so suited his purposes. Chapman (2002, pp. 18, 2) argues that Wulfstan’s writings embody a “convergence of two well-established verbal traditions,” and that Latin learning provided Wulfstan with a “degree of linguistic self-awareness” that did not overpower or co-opt his native idiom, but instead strengthened and enhanced it.

It is central to understand Wulfstan and his writing in terms of this both/and rather than either/or. Yet I would again urge a certain measure of caution, given that Wulfstan’s tendency was never purely ars gratia artis. Whatever source material Wulfstan used, his homilies contain very little in the way of metaphor, simile, or allegory. He was inclined to omit proper names, symbolism, and literary digression, preferring moral generalization to specific instance (see further: Bethurum, 1957, pp. 62, 91; Orchard, 2007, p. 335). It is these two features—(1) the reconciliation across a perceived dichotomy, and (2) utilitarian distillation—that can be seen also to govern the archbishop’s practical and literary engagement with multilingualism. Turning now from general principles to specific example, Wulfstan’s series of homilies addressing the baptismal rite provides an informative illustration when considered together in sequence.

Sermo Sancti Augustini de Baptismo non Iterando

The first text survives uniquely in the Copenhagen Royal Library manuscript already referenced, appearing on fols. 60v‒62r, where it is rubricated as Sermo Sancti Augustini de Baptismo non Iterando. This “Sermon of St. Augustine on not repeating baptism” was not authored by Wulfstan in the modern sense. It is neither Wulfstan-original material nor a direct translation, but instead a distillation of excerpts from the writings of Augustine on baptism, with “a little adaptation of word in certain phrases, and addition of linking passages and a final benediction” (Cross & Tunberg, 1993, p. 21). The result is a smooth abridgement or, as it were, a prosodic ‘bullet point list’ of pertinent and related segments from the Augustinian texts on which it draws, namely De baptismo contra Donatistas libri septem VI.i.l‒ii.4 and In Iohannis Evangelium Tractatus CXXIV XI.6‒11 (see further: Hall, 2004, p. 100). The text has also largely been neglected due to its exclusion from the two major editions of Wulfstan’s homilies (Bethurum, 1957; Napier, 1883).Footnote 2 Despite not being in accord with the critical methods or authorializing aims of those two editions, this text provides a foundation that helps to shape an understanding of Wulfstan’s multilingual tendencies and creative process.

The composite text first compares baptism with physical birth to explain why baptism (i.e., spiritual birth) cannot rightly be repeated. It then details the ways in which Christians are ‘born,’ contending that both good and bad followers of Christ have resulted from baptisms administered by worthy and unworthy baptizers. Examples of New Testament figures are cited (Ananias and Paul, Philip and Simon Magus), and an extended analogy is constructed to Old Testament patriarchs and the offspring they produced with legitimate wives and with maidservants. While the words and concepts stem from Augustine, the resultant text foregrounds central features of Wulfstan’s multilingual production process.

First, the text logically could not exist unless at some point while organizing and articulating views on baptism, Wulfstan placed value on there being a collection of meaningful excerpts. Medieval writing in general had a more flexible definition of what constituted scripture, but Gatch (1977, p. 121) has pointedly noted that Wulfstan was one writer who displayed “a sense that one could and ought to discriminate among theological sources” (see also Wright, 2007). Thus, even when a source text was that of a renowned auctor, such as Augustine, it did not possess a sanctity that rendered it immutable (on medieval notions of authorship, see Minnis, 1988). Though not a feature unique to Wulfstan, it is significant to note that he felt at liberty to extract, reorder, and rework his source materials to serve a new function distinct from that of the original text(s). Whoever the initial audience(s) for this series of untranslated and virtually unaltered Latin quotations may have been, even if—or especially if!—only Wulfstan himself, the text demonstrates Wulfstan’s utilitarian linguistic perspective via negative space.

The fact that the excerpts are kept in Latin is a result of Wulfstan performing the multilingual operation of functional allocation. It is not out of a desire to preserve Augustine’s words verbatim, since the text does not do so. Questions of Latin competence aside, Wulfstan was clearly comfortable enough to perform this highly efficient medieval equivalent of ‘copy-and-paste,’ rather than go through the trouble of rewriting in the vernacular, as the text was possibly not intended for very broad usage. Instead of one-way translation or linguistic passivity, Wulfstan interacted with both languages concurrently and adjusted his usages accordingly to suit the particular purposes of the text(s) he happened to be reading, writing, or compiling. The simultaneity of his bilingual mode is further proven by the fact that in addition to his own work, Wulfstan returned again to Latin source materials later while writing and rewriting homilies in the vernacular.

Incipit de Baptismo

The next homily in the series, Incipit de Baptismo (Bethurum VIIIa), has conventionally been viewed as a sort of Latin outline text in preparation for the later Old English homilies treating with baptism.Footnote 3 Interestingly, however, the supposed outline text does not contain a reference to the incorruptibility of the holy sacraments of Baptism and Communion in correspondence with a lengthy passage of the Old English Sermo de Baptismate (VIIIc):

Twa ðing syndon þurh Godes mihte swa myccle 7 swa mære þæt æfre ænig man ne mæg ðæron ænig ðing awyrdan ne gewanian, fulluht 7 huslhalgung. Nis se mæssepreost on worulde swa synful ne swa fracod on his dædan, gyf he ðæra þenunga aþere deð swa swa ðærto gebyreð, þeah he sylf ælc unriht dreoge on his life, ne byð seo þenung þæs na þe wyrse. Ne eft nis ænig swa mære ne swa haliges lifes þæt aðor ðæra þenunga gegodian oððon gemycclian mæge. Do swa hwylc swa hit do, Godes sylfes miht byð on þære dæde þurh halig geryne.

(Bethurum, 1957, pp. 177, ll. 36–44)

Two things are through God’s might so great and so excellent that no one can ever corrupt nor diminish anything thereof, Baptism and Holy Communion. There is no mass-priest in the world so sinful nor so vile in his deeds—if he does either of the services as befits thereto—though in his life he himself performs every wickedness, the service is none the worse for it. Nor again is anyone so great nor of such a holy life that either of the services can be enhanced or increased. Whosoever do it, God’s own might is upon the deed through holy mystery.Footnote 4

While Bethurum (1957) notes the similarity to Augustine’s De baptismo contra Donatistas, Thomas Hall (2004, p. 100) has brought the threads more fully together by pinpointing the theme in a relevant passage from Wulfstan’s Sermo Sancti Augustini:

dico sacramentum baptismi et bonos et malos posse habere, posse dare, posse accipere, et nihil interest ad baptismi sanctitatem, quantequisque peior habeat, quanto peior tradat, potest tradere separatur, si illud baptismum fit in nomine Trinitatis sub trina mersione. Amen.

(Hall, 2004, p. 137)

I say that the sacrament of baptism can involve both the good and the bad—both can administer it, both can receive it— and nothing interferes with the sanctity of baptism, no matter how much worse the one who receives it may be than the one who administers it, and no matter how much worse the one who confers it may be than the one who is baptized, he can still confer it if he is made ready, and if the baptism is performed in the name of the Trinity with triple immersion. Amen.

(Hall, 2004, p. 139)

I highlight Hall’s connections here for the two significant bearings they have upon Wulfstan’s multilingualism. First, quite simply the connections demonstrate how Wulfstan worked across languages and returned to source materials even after producing an outline text. Second, and more intriguingly, it begs the question of why the concept should have been omitted from the Latin homily which allegedly functioned as an outline for the later Old English sermons. The answer which I propose is in line with Wulfstan’s observed tendency to disregard theoretical generalities and speculative theology, instead habitually preferring straightforward explanations of spiritual practice that would prove immediately useful and applicable to a particular sermon’s intended audience.

Incipit de Baptismo exemplifies a different phase of Wulfstan’s production process than Sermo Sancti Augustini. It begins with an explanation of how, in order to receive the sacrament, believers must be instructed so that they properly understand the fundamental tenets of the Christian faith and the rite itself, so that the new believer’s body be rid of the Devil and prepared as a resting place for Christ. After this, the text lists the physical actions of a traditional baptism, explaining their significance and symbolic meaning. The sermon then lays out the first prayer for God to be present and for the Holy Spirit to rest in the water after expelling the Evil One. The ritual is compared to the Israelites’ flight from Egypt, before ultimately explaining how the three baptismal immersions done in the name of the Trinity restore one to grace after threefold lapse into sin. The homily closes with discussion of the clothing the new believer is dressed in and the chrism placed on their head following the ritual, vestments and symbolic crown described as fitting attire for a child of the Heavenly King.

Interpretation from a multilingual perspective helps to clarify the possible rationale for omitting talk of the sacrament’s incorruptibility in the Latin text. Following the suggestions of Hall (2004, pp. 95–96) and Lionarons (2010, p. 42), it is possible to envisage the Latin Incipit de Baptismo not as merely an outline, but as actually having been preached to a gathering of clergy members. If listeners had sufficient Latin competence for the words to have had any meaning, then they were surely more learned priests or higher-ranking church officials who would also have been involved with the sacrament’s administration. In Wulfstan’s view, they may have stood to gain little in terms of practical utility from the sentiment. For a more mixed assembly, however, the reappearance of the notion that worthy Christians can come from all walks of life regardless of the nature of their baptizer would certainly provide appealing reassurance.

By way of contrast, theologically denser elements of the Latin Incipit de Baptismo disappear entirely from the Old English texts. For instance, Wulfstan’s Incipit de Baptismo contains the following passage explaining the relationship between the three immersions of baptism and three avenues of sin:

Baptizatur autem sub trina mersione in nomine sancte trinitatis, id est, patris et filii et spiritus sancti. Et recte ut homo, qui ad imaginem Dei conditus est, per invocationem sancte trinitatis ad eandem renouetur imaginem et, qui trino lapsu peccati, id est, consensu, suasione, delectatione, cecidit in mortem, trino gradu eleuatus de fonte per gratiam resurgat ad vitam.

(Bethurum, 1957, p. 171, ll. 47–53)

One is baptized, moreover, with triple immersion in the name of the Holy Trinity, that is, of the Father, and of the Son, and of the Holy Spirit. And rightly that a person, who is fashioned in the image of God, may be restored to that same image through invocation of the Holy Trinity, and one who has fallen into death by threefold fall of sin—that is through consent, persuasion, pleasure—may rise again to life through grace having been raised from the font by threefold step.

These conceptually denser lines represent a reversal of the preceding example, as they do not reappear in Wulfstan’s Old English sermons. The inclusion in Latin of details about the physical process of baptism and its signification as part of the sacrament conforms again to the pattern of functional allocation already elicited. Connections to modes of lapse into sin may have seemed particularly useful to include for clergy that were responsible for administering the sacrament and providing moral direction to the laity but been deemed more superfluous for a more mixed audience.

Wulfstan also preserves an intriguing rhetorical feature encountered in one of his probable sources, Jesse of Amiens’s Epistola de Baptismo (see further Bethurum, 1957, pp. 303–304; Cross, 1989). Lines 5‒6 read: “Catecuminus Grece, instructus dicitur Latine”[Catecuminus (catechumen) in Greek, instructus (one who has been prepared) it is called in Latin]; similarly, lines 30‒31 offer: “Baptismum Grece, unctio Latine interpretatur” [Baptismum (baptism) in Greek, unctio (anointing) it translates in Latin] (Bethurum, 1957, pp. 169–170). The technique of comparative and explanatory etymology is by no means unique, and yet I perceive it as carrying a peculiar weight in this context. Wulfstan’s linguistic setting was undeniably a blended one, but it is unlikely that he and the majority of his intended audience had any advanced ability in or extensive exposure to Greek. The isolated single-word usages are followed by explanatory Latin definition; they do not depart from the grammar of the Latin sentences, nor do they even present particularly rare or technical terms. “Catecuminus” appears thrice in quick succession at the beginning of the sermon, preceding (l. 1) and following (l. 6) its definition on line 5. “Baptismum” appears regularly throughout. These words then are more akin to assimilated loan borrowings than genuine code-switches between languages, and their being singled out as Greek in origin serves a specific rhetorical purpose. Etymological explanation allows Wulfstan essentially to say the same thing twice and drive home the main points of the sermon. Moreover, this form of multilingual duplication mirrors his general practice of offering two roughly synonymous terms where his source material contained only one.

Wulfstan’s Vernacular Baptismal Homilies

Turning to the Old English sermons Dominica IIIIa vel Quando Volueris (VIIIb) and Sermo de Baptismate (VIIIc), one encounters the most prominent example of Wulfstan’s attentiveness to lexical difference across languages.Footnote 5Incipit de Baptismo begins “Primo necesse est ut paganus caticuminus sit” [First it is necessary that a paganus (pagan, heathen) become a catechumen] (Bethurum, 1957, p. 169, l. 1). Correspondingly, the longer Sermo de Baptismate is comfortable explaining how one can come to Christianity “of hæþendome” [from heathendom] (Bethurum, 1957, p. 175, l. 9). Dominica IIIIa, however, seems altogether reluctant to offer the seemingly obvious translation of “hæþen” for the Latin “paganus” (on Wulfstan’s use of “hæþen” and related terms, see Meaney, 2004). Yet this reluctance makes sense in the context of Dominica IIIIa, as Lionarons (2010, pp. 126–127) has argued, because Dominica IIIIa is a sermon focused on infant baptism. Wulfstan may have been unwilling to label unbaptized infants with the same word he consistently used for those practicing a non-Christian religion and those Christians whose conduct he deemed to be out of line with correct belief. This distinction paints a picture of Wulfstan as an individual alert to contextual, semantic, and pragmatic non-equivalence across languages even at the microscopic level of individual word selection.

In addition to the previous omission, Dominica IIIIa and the longer Sermo de Baptismate both leave out a segment relating the Israelites’ Exodus from Egypt found in the Latin sermon. Superficially, this exclusion accords with Gatch’s (1977, p. 20) suggestion that “Wulfstan made it a point to avoid theological subtlety, to drop exempla and most traces of allegorical interpretation and, usually, to delete specific historical allusions.” On another level, inclusion in the Latin and omission from the Old English versions reinforce the idea that Wulfstan did not merely simplify and generalize, but rather actively adapted his content and language to suit his current purpose and audience. In the baptismal sermons, his purpose seems not to have been increased comprehension of theology and symbolism for their own sake, but rather enhanced accessibility aimed at encouraging moral behavior and facilitating practical performance of the rite.

Thus, one witnesses a similar manner of treatment with respect to the Pater Noster and the Creed. Dominica IIIIa and Sermo de Baptismate contain the following corresponding passages:

He is ealra fæder, 7 þæt we geswuteliað þonne we singað ure pater noster. […] Be þam we magon ongitan 7 oncnawan þæt we synd ealle gebroðra 7 eac geswustra þonne we ealle to anum heofenlicum fæder swa oft clypiað swa we ure pater noster singað.

(Bethurum, 1957, p. 173, ll. 56–66)

He is the Father of all, and we show that when we sing our Pater Noster. […] By that we may recognize and know that we are all brothers and also sisters when we all so often call to one heavenly Father as we sing our Pater Noster.

He is ure ealra fæder, 7 þæt we swuteliað þonne we singað, Pater noster qui es in celis sanctificetur nomen tuum, et reliqua. Be ðysum we magon gecnawan þæt we syn þurh cristendom ealle gebroðra þonne we ealle to anum heofonlicum fæder swa oft clypiað swa we pater noster singað.

(Bethurum, 1957, p. 180, ll. 91–96)

He is the Father of us all, and we show that when we sing, Pater noster qui es in celis sanctificetur nomen tuum, et reliqua. By this we may know that we are through Christendom all brothers when we all so often call to one heavenly Father as we sing Pater Noster.

Lines are interlarded in Dominica IIIIa’s passage, but the above sentences are close to identical in both sense and structure. The ability to recite the Pater Noster and the Creed is important in its own right, but in the texts’ streamlining of the baptismal process, memorization is also partly reduced to a mandatory prerequisite. Wulfstan notes in Sermo de Baptismate that it is after mastering the Pater Noster and Creed and understanding proper belief, “þonne bið he wyrðe þæt he fulluht underfo” [then is he worthy to receive baptism] (Bethurum, 1957, p. 176, ll. 19–20). Wulfstan encourages older listeners that do not already have the Lord’s Prayer and the Creed memorized to study actively, because one unwilling to learn cannot justly undergo baptism (Bethurum, 1957, p. 184, ll. 145–149).

By contrast, the emphasis on understanding takes a necessary backseat for infant baptism, with the child to be taught as soon as they acquire speech (“æfre swa þæt cild raðost ænig ðing specan mæge”; Bethurum, 1957, p. 182, l. 142). This distinction is significant to note alongside my earlier remarks about Wulfstan’s hesitance to label children “hæþen.” It overwhelmingly foregrounds one of the archbishop’s overarching goals across his homiletic and his legal writings, namely to “construct the faith of his congregation and in doing so [bind] them together into a single political-religious community” (Lionarons, 2010, p. 12). A child’s comprehension is less crucial than that of an adult convert, since a child born to Christians is ipso facto already a member of Wulfstan’s political-religious community.

Despite producing vernacular translations of the Lord’s Prayer and the Creed, Wulfstan is still insistent on learning both in Latin if one possessed the capacity to do so, both in the baptismal homilies and elsewhere. To be sure, Latin learning helped to maintain parity with continental practice, but I would additionally present the use of Latin as having an insulating effect amidst Wulfstan’s blended linguistic setting. Referencing the Old English Carta Dominica homilies, Townend (2002, p. 179) has presented the case that “whether or not they believed the Old Norse language to be genuinely unintelligible to English speakers, the Old English adaptors […] certainly believed that it should be unintelligible.” Linguistic differentiation made for a convenient way to draw an ideological line in the sand between Christian and heathen. Yet rather than simply reinforcing a divide, the weight that Wulfstan puts on learning these standard Latin expressions embodies a form of identity management. It offers shared ‘linguistic membership’ into his envisioned Christian community, regardless of language or dialect. This notion of society building and community establishment is key to Wulfstan’s multilingual uses and emblematic differentiation, and now by way of conclusion I will demonstrate its direct conceptual governance over Wulfstan’s intratextual code-switches between languages.

‘Code-switching’ refers to “the change from one language (or variety) to another within one act of communication” (Schendl, 2000, p. 77). In response to outmoded perceptions of linguistic laziness, John Edwards (1994, p. 3) presents that in much code-switching practice, “language changes made are non-random, that a switch signifies something.” Edwards (1994, p. 78) delineates such rationale as “fitting the word to the topic, finding a word with a nuance unavailable in the other variety, helping out a listener, strengthening intimacy,” and he highlights also the repetition that is frequently involved in the process. Applying similar principles to an exploration of late medieval bilingual sermons, Schendl (2013, p. 160) has determined there to be two levels at which Latin-vernacular code switches often function—one being the concrete, stylistic level, and the other a more abstract level where they fulfill a “social function of establishing group membership […].”Footnote 6 Similarly, Roberts and Tinti (2020, p. 190) have examined Latin/Germanic language code-switching and identified “instances of explicit linguistic consciousness in charters,” arguing that there the vernacular could be purposefully invoked “to engender social inclusion or exclusion […].” Though inversely pertaining more to Latin, it is precisely this notion of group reinforcement and inclusivity by means of code-switching and transplantation that unlocks the emblematic multilingualism of Wulfstan’s Old English baptismal sermons.

The first instance, near the beginning of Dominica IIIIa, displays how shifts in language play directly into the establishment of community.

And þonne se sacerd him ætforan singð, Credo in Deum patrem, þonne getrymmað he his geleafan, 7 mid þam leafan he gescyrt 7 he gewædað his hus; þæt is, þæt he gegearwað his heortan Gode on to wunianne.

(Bethurum, 1957, p. 172, ll. 25–29)

And when the priest sings before him, Credo in Deum patrem, then he confirms his faith, and with that belief he clothes and he equips his house; that is, that he prepares his heart for God to dwell within.

There are only slight changes in Sermo de Baptismate:

And ðonne se sacerd him ætforan singð, Credo in Deum, þonne trymeð he his geleafan 7 mid ðam geleafan gefrætewað 7 gewædað his hus; þæt is, þæt he gegearwað his heortan Gode on to wunianne.

(Bethurum, 1957, p. 178, ll. 56–59)

And when the priest sings before him, Credo in Deum, then he confirms his faith, and with that faith adorns and equips his house; that is, that he prepares his heart for God to dwell within.

In actuality, these examples and most of Wulfstan’s switches are more ‘transplantations’ than ‘code-switches.’ The instance in particular underscores his utilitarian approach to multilingualism, since it reproduces what a priest would literally be saying while conducting the rite. Usage matches situation.

It is also beneficial to reemphasize that any audience would have possessed varying degrees of Latin proficiency. To this point, Alan Fletcher (2013) has stressed the possible difficulties in understanding later medieval bilingual sermons when preached aloud. Wulfstan’s transplantations and switches into Latin, however, provide a counterpoint. They are not only comparatively brief,Footnote 7 but as I have already detailed, each and every member of Wulfstan’s audience was expected and encouraged to learn to recite the Pater Noster and the Apostles’ Creed in Latin. Consequently, these short interpolations—“Credo in Deum (patrem)”—would be more than understandable, they would be intimately familiar. Coupled with simple utility then, they also served the emblematic function of strengthening community bonds and associatively reaffirming cohesive group identity.

Another instance of cross-languaging, found only in Sermo de Baptismate, operates similarly but requires a bit more interpretation for a modern reader:

Þa ðreo dyfinga on fontbæðe getacniað þæt we beoð geedcennede þonne to ecan life ðurh þa halgan þrynnesse, patris et filii et spiritus sancti, þæt is eal an soð Godd.

(Bethurum, 1957, p. 179, ll. 78–81)

The three immersions in the font-bath signify that we are regenerated then to eternal life through the Holy Trinity, patris et filii et spiritus sancti, that is all one true God.

The shift in language here is not presented as transposed discourse but is instead embedded into the structure of the sentence itself. At first glance, it appears classifiable as an intrasentential switch where “the writer’s thought moves forward without glossing, quoting, translating, or announcing a coming development” (Wenzel, 1994, p. 22). Grammatically, however, something both more and less complicated appears to be at work.

The instance is less complicated, because it again seems to be a transplantation more than an authentic code-switch. Rather than a transition with Latin genitives dependent on Old English noun, an easier explanation is that the Latin phrase is transplanted from one of the common “religious formulae” (Schendl, 2018, p. 44; i.e., the Trinitarian formula as in the Great Commission [Matt. 28:19]). The instance is, however, perhaps more complicated than preceding instances, because it does not appear to exhibit the same intrinsic practicality. Schendl (2018, p. 42) observes generally that “code-switched [Latin biblical] quotations add scriptural authority to the [Old English] text, but sometimes also have an organizational function […].” Likewise, in Middle English bilingual sermons, a “function of such Latin switches is to give additional authority to the statement (and the speaker)” (Schendl, 2013, p. 155). In addition to associatively affirming community, I would present that this use of Latin by Wulfstan attains greater rhetorical impact value through language switch and association with the familiar benediction.

In connection with this benediction, one situation in which the Trinitarian formula would presumably have seen use is in accompaniment with the Sign of the Cross. Wulfstan approaches this subject in perhaps the only true code-switch in Sermo de Baptismate:

Þonne se sacerd cristnað, þonne orðaþ he on þone man, þonne hit swa gebyrað, in modum crucis, 7 þonne wyrð þurh Godes mihte sona deofol swyðe geyrged

(Bethurum, 1957, pp. 176–177, ll. 30–33)

When the priest christens, then he breathes on the person, then it thus takes place, in modum crucis, and then immediately through God’s might the devil becomes very terrified

The corresponding passage in Dominica IIIIa provides intriguing counterpoint:

Þonne se mæssepreost cristnað ærest þæt cild, þonne orðað he þry on an on hit, þonne hit swa gebyrað, on Cristes rode tacne 7 on his ansyne; 7 þonne wyrð sona þurh Godes mihte se deofol geyrged 7 utdrifen

(Bethurum, 1957, p. 172, ll. 15–18, emphasis added)

When the mass-priest first christens the child, then he breathes thrice upon it, then it thus takes place in the sign of Christ’s cross and upon its face; and then immediately through God’s might the devil becomes terrified and driven out

The most noticeable modification is the substitution of the Latin “in modum crucis” for the Old English “on Cristes rode tacne,” a divergence that also finds form in CCCC 201’s Sermo de Baptismate where a corresponding interlinear insertion (“on rode tacen”) appears above the phrase. While it is difficult to determine the degree, it seems plausible that this Latin formula may have been slightly harder for some audience members to pick up on than words from the steadfastly memorized Pater Noster and Creed, also with its non-English syntax. That said, this code-switch does have an accompanying physical action that might yield clarification and shorten the mental leap for Wulfstan’s audience.

Compared with previous examples, this instance of code-switching also does not strike me as having similar practical explanatory value. Nor perhaps, is it a usage familiar enough to achieve communal intimacy in the same effective way through instant mutual comprehension. If this less intuitive code-switch was non-random, then it obliges a slightly different line of reasoning. I would argue that while still inferable enough to allow for understanding, the cross-languaging draws heavily on associations an audience—and particularly an audience of the laity or less cosmopolitan clergy—might have with the Latin language itself. Latin’s palpable distinction from the vernaculars, combined with its intimate connection to church practice and political/legal spheres, must have associatively colored its usage in the minds of many listeners (again, cf. Cain, 2016; Schendl, 2018). Drawing on those connotations, this usage could have gained a bit of linguistic gravitas for an action and symbol already laden with religious significance, not wholly unlike Modern English’s associative use of fossilized Latin (which I have deliberately made a point of employing ad nauseam throughout this article).

In closing, I will present one final Latin inclusion in Sermo de Baptismate:

Ac utan understandan hwæt ða twa word mænan, abrenuntio 7 credo, þe man æt fulluht-þenunge on gewunan hæfð. Abrenuntio, þæt is on Englisc, ic wiðsace heononforð æfre deofles gemanan. Credo, þæt is on Englisc, ic gelyfe on God ælmihtigne þe ealle ðing gescop 7 geworhte.

(Bethurum, 1957, p. 181, ll. 120–124)

But let us understand what the two words mean, abrenuntio and credo, that one has in custom at the baptismal service. Abrenuntio, that is in English, I reject forever henceforth the devil’s fellowship. Credo, that is in English, I believe in God Almighty who formed and created all things.

Renouncing the devil and confessing faith in God are foundational religious concepts that for Wulfstan represent the first steps on one’s way to conversion and baptism. Here, beyond the borrowing of associative authority and linguistic connotations, Wulfstan can be seen to come full circle and combine all of the aspects at work in the examples of this multilingual case study.

Recalling his customary practice with source materials, Wulfstan has once again taken what he deems most necessary for an audience to understand and condensed it to two simple words: “abrenuntio” and “credo.” Recalling his patented repetition, Wulfstan is able to ‘hammer home’ the two confessions of faith by repeating them a total of three times each in two languages over three sentences, not including their earlier mention in the homily from line 10. Recalling his handling of the Pater Noster and Creed, these Latin pactiones are presented to the audience with an accompanying explanation as to their meaning and significance. “Abrenuntio” and “credo” are not simplistically translated as “ic wiðsace” and “ic gelyfe,” but are rhetorically embellished in the Old English in a way that underscores Wulfstan’s adaption of his language usage to suit his context, audience, and purpose. “Credo” would be no less familiar here than above thanks to the familiarity of the Creed, while “abrenuntio” might be somewhere in between that and a heightened emblematic functionality.

Were this sermon to be given in a present-day setting, these words might very well be projected onto an overhead monitor for the congregation to see. They function as totemic summary words, as ‘emblems’ that an audience can take away and remember. In doing so, these emblems also representatively encapsulate Wulfstan’s multilingualism. They simultaneously constitute a borrowing of authority, a distillation, an adaptation, and a repetition, all with the aim of presenting a subject that had immediate practical utility for the audience.