1Servatus Lupus ranks as one of the leading classical scholars of mid-ninth-century Francia. He understood himself as a pupil of Einhard and Hrabanus, who abetted and oversaw, respectively, his studies during an extended visit at Fulda in the mid-830s1. From late 840 until at least mid-862, when he abruptly quits the historical record and doubtless died, he interacted with other leading figures of the day as the abbot of his monastic home at Ferrières. His renown rests on a small but significant group of writings, on some two dozen extant manuscripts (especially of classical works) that show his distinctive text-critical technique, and on the remembrance of his gifted student, Heiric of Auxerre in his Collectanea.

2What knits the evidence into a roughly settled Carolingian Gestalt are the 128 surviving letters found in ms. Paris, BNF, lat. 2858[-I], fols. 1v^o-63r^o (here-after P), which three scribes recopied as a memorial corpus shortly after Lupus’s death2. They drew on the deceased abbot’s personal archives (still intact), and thus followed priorities fixed, at least in part, by Lupus himself. P is Lupus’s letter-book. Lightly annotated and perhaps tapped by Lupus’s pupil, Heiric, for his hagiographical works on St. Germain d’Auxerre, the collection eventually found its way to Fleury (with other manuscripts connected to Lupus and Heiric), where it lost, at some point, one or more quires ahead of the surviving final one (less its last leaf). The large loss took the tailpiece and start of two letters (115 and 115bis) and the whole of those in between3. During the eleventh century, the collection received a now erased Fleury ex-libris, and the book remained there, seemingly unused, until the monastic library was sacked in A.D. 1562. The new owner, Pierre Daniel, foliated and annotated what became the first of two slices of Raubgut; he mated it with his separately annotated «Ripoll letters» (saec. xi, copied at Fleury)4. These still follow Lupus in the subsequent, late-sixteenth-century reliure à la fanfare (elaborate gold stamping on crimson Moroccan). By 1575, the Protestant Daniel, based at Orléans, had conveyed P (purported at the time to have come from Ferrières) to the Catholic publicist, Papire Masson, who deployed the Letters by name in his anti-Reformation scholarship, first against the Huguenot François Hotman, then ten years later to debunk the myth of Pope Joan. Masson finally published the Letters in his error-ridden editio princeps (hereafter m) of 15885. It can be shown that Masson used a very faulty copy of P, which I call β, to produce: 1) his early excerpts from Lupus; 2) the edition m; and also 3) a partial manuscript Ersatz sent to Daniel, evidently as a «piètre récompense» for P. The Ersatz survives in Bern, and is hence called ms. B6. These matters are important because Lupus’s Letters were regarded as treasure and sumptuously rebound, but then at some point after 1636, a further partly damaged leaf (missing its lower outer corner with some text) —fol. 36 in Daniel’s numbering— was definitively removed7. To the foregoing testimony, we can and must add the careful collation of P made by André Duchesne to produce his emended, or more exactly, re-wrought 1636 text of the Letters (hereafter c), which greatly improves on m, always strictly on the basis of P8. Duchesne transmits excellent evidence for fol. *36, and certainly used P before it was compromised9. Summing all the relevant witnesses allows, then, for clear rules to heal, as has not yet occurred, the injured text from P’s lost fol. *36 based on all the derivatives. In the absence of P, where c B agree against m, or c m against B, agreement points to β and suggests the lost face of P. More-over, the mended text of the lost folio shows Lupus in telling relations with his local bishop and metropolitan, Wenilo of Sens. Anyone who would reedit, for now the eighth time, a body of Letters that scholars still rank as treasure and have used for important historical purposes since their discovery, must have sound reasons to do so. In terms of the manuscripts, the tradition, the text and the light it sheds on the author, his sources, and his history, there are good reasons to reedit the Letters of Lupus10.

3It is impossible to cover here more than a tiny slice of the rich and varied world conjured by the 126 whole items and two fragments that still survive thanks to P. With the sole exception of ep. 100, an Admonitio (or circular exhortation) that Lupus must have composed for use by his metropolitan, Wenilo of Sens, none of the manuscript’s extant content has left an outside parallel trace. To be sure, Heiric echoes diction from Lupus’s opening letter to Einhard (ep. 1) to begin and elsewhere in his poetic and prose masterpieces on St. Germain11. That opening letter was at once an homage and program rooted in personal association which Heiric doubtless made his own, and he may have been a moving force behind producing the collection in P, but he did not copy or annotate it12. The hitherto unnoticed parallel documents, it seems, an instance of tandem archiving by Bishop Wenilo and Lupus as his ghostwriter. Their rapport was always tense, and even poisoned by the time of the very late, penultimate letter of the collection (ep. 126, A.D. 862). There Lupus must have sensed, one is tempted to think, that time or circumstances left him little to lose by speaking bitter truths about Wenilo’s disorders and betrayals. From the epitaph he wrote for himself (pre-served elsewhere by Heiric), Lupus always looked steadily to his earthy demise13. Yet he did not do so without preserving the words he lent to Wenilo, whether for his own credit or protection. They later served Odorannus of Sens (ca. 985-1046) as the script for a future ghostwritten exhortation14.

4New parallels and insights about Lupus’s Letters often turn on episcopal connections, because bishops preponderate as interlocutors in the letter-book. Scholars have repeatedly parsed the collection to arrive at tallies of things sent and received, including some written for —and a very few by— others. The mix lends coherence to responses, and documents letter-types and thematic fixations (e.g., books and learning, royal conduct, the priory at Saint-Josse) and also personae that Lupus assumed and archived. Here is not the place to repeat the effort at parsing wholesale, but it bears saying that closely reassessing the individual texts and their social milieu and logic corrects some impressions, and has even produced seven hitherto unsuspected persons. Efforts to focus Lupus at Reims in the ambit of Archbishop Hincmar must, for instance, be balanced by the abbot’s arguments to his own bishop, Wenilo of Sens, to mount a credible rationale for securing permission for monastic retirement for two engaged secular priests (ep. 29). A perpetually absentee abbot had no ethical standing to argue such a case, and would have lacked even the practical contact to press it. Likewise, the intimate friendship Lupus highlights with his «kindred spirit» (unanimis suus), Abbot Wulfad of Soissons/Rebais (ep. 121), must affect how one reads Lupus’s rapport with Hincmar, who relentlessly persecuted Wulfad, to no final effect (he became archbishop of Bourges in 866), owing to the irregular circumstances of his ordination by Hincmar’s own deposed and disgraced predecessor, Ebo. Lupus relied on and worked with Hincmar, but mostly at a distance. The fact and texture of the four letters to him in P show that (epp. 42, 44, 49, 76), and the further extrauagans to Hincmar (ep. A3) on the predestinarian question underscores the physical and mental distance, and the abbot’s need for his advocacy. Lupus was not wholly in his pocket.

5Any prior tally of episcopal recipients also inevitably comes up short, since a correct listing of the 30 items sent in his own right by Lupus to a bishop has lacked, till now, the name of a probably significant person in the abbot’s life: Vulfinus of Die. As Lupus prepared ca. 848 to leave for Rome on a mission whose purpose remains unknown, he wrote to «Vu.» as P records it (ep. 67)15. The letter states that his diocese (urbs) lies on the road leading to Italy (so Die, as in the Itinerarium Burdigalense), and Lupus also identifies him as being born of the same homeland (patria) as himself. Both Theodulf of Orléans (d. 821) and Florus of Lyon (d. after early 855) interacted with Vulfinus over his lifetime: Theodulf knew him as a skilled metrical poet, and Florus, as the «Orléans grammarian» (grammaticus Aurelianensis —an unnecessary reminder of place if he were still there)16. As bishop, Vulfinus redacted a prose and created a metrical Life of St. Marcellus of Die17. Yet it seems also plausible that Vulfinus may have been the early teacher of grammar whom Lupus evoked in his first letter to Einhard. He was promoted to Die by A.D. 835, though the Monumenta Germaniae Historica conciliar editors omit to note that18. It seems to me very probable that he lent Lupus, his compatriot and perhaps ex-pupil, a now well-known grammatical manuscript, and therewith attempted, without success, to ensure its return to «St. Marcellus» by an elaborate full-page ex-libris. Integrating the evidence Lupus’s Letters allows us to see Vulfinus as he was for Lupus, and to sort out the itinerary of that manuscript much more clearly, but adds no luster to Lupus’s standing as a trustworthy borrower19.

6In writing to Vulfinus, Lupus resorted characteristically to what I call auto-effective rhetoric to press desires. He meant to leave no ambit for refusal and deployed words cunningly to impose his wishes on his episcopal acquaintance. In this case, they took the decisive form:

I reckon it idle, moreover, to forewarn you about actively helping me with what custom demands, since I should in no way belittle your good judgment, but trust mightily in your generosity. Strive thus to show us the hospitality of the region where you live and the fellow feeling of the homeland where we were born20.

7Reading the text and absorbing its meaning meant compliance. That was a broader strategy tailored here to work with Vulfinus.

8Elsewhere, Lupus elaborated and used a set-piece rhetoric to press wishes with other bishops. In the letters that survive in P, he deploys the adverb episcopaliter («with bishoply generosity») four times, and always to a bishop whose help he means to secure (epp. 16.2.2; 21.2.1; 29.8.2; 125.3.5). The word is uncommon, but must build on the New Testament usage of 1Tim. 3:2 (oportet ergo episcopum inreprehensibilem esse, […] hospitalem, doctorem). The adverbial form also appears once in Augustine, concerning Ambrose, at Conf. 5.13.23: peregrinationem meam satis episcopaliter dilexit. Whether or not the Augustinian usage was ready to mind, any interlocutor so addressed would have had to pause and recognize the adverb’s quasi-definitional Pauline resonances and obligations.

9For select bishops with well-stocked libraries and larders, Lupus reserved certain Vergilian echoes and tags. In a famous letter to Ursmar of Tours that culminates in a circuitous request to borrow a papyrus codex of Boethius, On Cicero’s Topica, Lupus begins by mating his adverb with a resonance from Vergil: «… I give thanks and feel great gratitude to you that your hallowed self did not shrink to vie freely with favours in receiving the aforesaid brother with bishoply generosity and kindly supporting him, and also inviting my lowly self most warmly to visit your loving person21.»

10The distant allusion to Vergil, Aen. 1.548 (officio nec te certasse priorem) takes clearer form in a like-minded request for books to Bishop Wigmund of York (ep. 61.2.2, 848-849: curauimus […] priores certare officio). Somewhat earlier (in early 847), Lupus also inverted the Vergilian tag in a remarkable fanfare of begging for supplies to Hincmar of Reims:

… To be sure, we vied first with dutiful attention, but we did not request so much as we offered, namely an opportunity to fulfill the precept of charity. Thus, since we now are long suffering from a lack of everything «that normal existence requires,» rush to help us, so that the desired outcome of your bounty fully matches the benevolence of your promises.

We, for our part, are being lodged at an estate that is ironically called Viniacus and lies more or less one mile distant to the southwest from Attigny. I say this, so your signal prudence will also give thought to cartage, since we likewise lack the means to provide for it22.

11Marshall noted the one-off literal quotation of Vergil, Ecl. 2.71 (quorum indiget usus), but Lupus’s thrice repeated, obviously programmatic resort to Aen. 1.548 has been missed. Whatever he expected Wigmund to understand, the dialogic inversion to the first person with Hincmar may be the first irony of ep. 49. Lupus may have voiced beforehand some statement about «vying first with dutiful attention», perhaps in an unretained letter. In a remarkable envoi (ep. 44, early 847), Lupus had already evoked to Hincmar his own dismal failure to secure from the King the return of the sequestered Priory of Saint-Josse: even Vergil, so Lupus, would secure no hearing if he returned with the panoply of his three-fold art (he means pastoral, rural, and epic poetry). If that would fail with the court (itself a remarkable critique), Lupus still appealed to Bishop Hincmar via Vergil. A further irony of ep. 49 is that Lupus, who seems to have possessed no real Greek of his own, there uses the correctly formed Greek adverb antifrasticos, its first documented occurrence, albeit in Latin transliteration. Whether he knew it from Graeci (Carolingian-speak for «Irishmen trained in Greek») at Laon or Reims, we cannot say. It has at least proved possible, for the new edition, to pin down where exactly Lupus resided —the lieu-dit Bigny, about a Gallic league (ca. 2.5 km) south of Attigny— and made ready to receive Hincmar’s desperately needed gifts.

12In writing to his own bishop and metropolitan, Wenilo of Sens, the work of pressing desires often amounted for Lupus to a white-knuckled struggle conducted with words. There are no appeals, in what survives at least, to Vergil or to poignant or exotic adverbs in Greek or Latin, and Lupus writes to Wenilo seven times in P, as against four times to Hincmar. The new edition of the Letters has produced fresh sectioning (which follows the rhythmic cursus), corrected mistaken or incomplete readings from P, and thus improved our textual foundation and extended its historico-contextual reach23. The effects for Lupus and Wenilo are considerable at the now excised fol. 36 in P and for the two letters that mostly covered it.

13I have mentioned above the applicable witnesses and rules for healing the compromised text. After reading the testimony accurately, but before settling on solutions, I plotted the material from fol. *36 with the probable abbreviations. The arrangement produced matching diagonal slices across the lower outside corner for both the recto and verso of the lost folio. What once lay beyond the diagonal remains irrecoverable, but one can map what survives and work through its difficulties —woeful in m and B, much remedied by c— to possible, grounded solutions (not the guesswork of Traube and Dümmler for the MGH). Thanks to modern facsimiles and ready travel, this is more than early editors readily could do, and what Marshall should have done for visiting the witnesses, but did not. It is worth underscoring that epp. 73 and 74, are not «corrupt» except insofar as editors have failed24. Without encumbering the result here with apparatus, the rules named above and a dose of reflection based on stylistic familiarity and some historical context produce what follows (with remaining gaps shown by bullets):

[73.] Ad Guenilonem

[1] Reuerentissimo antistiti Gueniloni, ‹Lupus in Domino salutem›.

[2.1] ••• cuius lectione uestro uictus amore paene ••• ••• per communem propinquum nostrum Ab. Ra••• ••• dirigite. [2.2] Quem quantopere diliga‹m› ••• ••• quod eum praeter uos nulli unquam passus sim ‹commodari›.

[3.1] ••• autem ecclesiae nostrae quod fanum appr••• ••• aequissimum iam decreuistis, et ••• ••• |P 36^v| quaerere ipsi optime intellegitis. [3.2] Quamquam in proximo conuentu uobis gratias retulerim (putans penitus litem esse sopitam), noueritis nihil esse definitum, sed ad uotum eorum qui etiam iniuste non erubescunt, immo appetunt esse uictores, occasione dilationis presbitero uestro praeparatum quod quaesiuit emolumentum ac nostro quam fastidit iam spem inanem relictam. [3.3] Proinde uestra insignis prudentia —quae didicit ex euangelio non personaliter iudicare (see Luke 20:21; John 7:24), instructa etiam illo ueteri praecepto: Iuste iudica proximo tuo (Lev. 19:15)— iudicii ueritatem, ut coepit, constanter et cito perficiat, ne (quod absit) abutantur uestra nobilitate qui cupiditatis patrocinium suscepere. [3.4] Nam et ipsa illis ad subuersionem ecclesiae nostrae oratorium nouum commenta est, et ne ueritati adquiescant unanimitatemque caritatis dissoluant adhuc indefessa hortatur.

[4] Peto etiam ut ad supplicationem Gerohaldi presbiteri Lan. diaconum, eius propinquum, in ipsius titulo dignemini ordinare, quoniam difficultate uisus ••• ••• frequenter non sufficit sacerdotale munus implere.

[5] Cete‹rum de› ••• ••• perfectione, quaeso, me fieri e uestigio certiorem ‹dignamini, ut in›digno, tamen qualicumque paruitatis meae discedatis •••.

[6] ‹Cupio uos› ualere feliciter.

[74.] Ad eundem

[1] ‹Reuerentissimo prae›suli Gueniloni, Lupus in Domino salutem.

[2] Titum Liuium per hunc ‹nuntium ••• ••• diri›gite, quia illo non mediocriter indigemus.

[3.1] Bene‹ficium› ••• •••, ‹quod› tamen postulauimus, dignamini implere •••. [3.2] ‹Per sacro›sanctos ordines, largiemini participem c‹om- ••• et diaco›ni senum infirmitas illius promotione ••• •••.

[4] ‹Collectan›eum quod elaboraui de libero ar‹bitrio et praedestinatione atque› |P 37^r| ueritate pretii quo redempti sumus, ipse uobis elegi ostendere quam per quemlibet dirigere, ut otio nobis diuinitus collato tantarum rerum subtilitatem facilius mecum possitis aduertere.

14Readers are free to turn to the usual editions to compare. My English translation will clarify in due course some inevitable structural and lexical ambiguities.

15For the moment, I want to insist that, lacunae notwithstanding, the methodically restored text is eminently comprehensible in all its critical details25. Ep. 73 concerns, to start, a manuscript that Wenilo has not returned, which is surely not the Livy codex mentioned in ep. 74. Both letters likewise jointly press the need for Wenilo to ordain replacements (or at least substitutes) for parish clergy who can no longer adequately function owing to poor sight (73.4 and 74.3). Ep. 73 seems to point, in closing, to some sort of task or travel by Wenilo and to Lupus’s desire to pay him due respect upon completion or departure. What is truly remarkable about ep. 73 is the long central section’s graphic detail about what is plainly a contest between the abbot and bishop over the control of a church that, with its people and revenues, belongs to Ferrières. To squeeze the abbot’s priest, the favorites of the bishop have «unjustly» set up a new church, and doubtless ordered the local faithful to attend. The episode is a rare counterpart to the now less solitary example that shows how Hincmar of Reims likewise worked, in essence as a church reformer, to deal a subordinate out of a proprietary church26. Lupus gives the view from below.

16Most significant of all is the final section of ep. 74, since Lupus did not write elaboraui delibero, as the witnesses all superficially «agree». Rather, with P’s habitually nested prepositions, the text evokes one of Lupus’s important, but obviously awkward compositions concerning the contest over Predestination that flared around Gottschalk of Orbais in and after 849. Except for his vaguely related ep. 30 (to Gottschalk, on the nature of human sight in heaven; the second longest letter in P), Lupus’s dealings with the subject of Predestination seem to have been skipped as matter for inclusion in P. Lupus’s predestinarian dossier of four works, including the Collectaneum, was certainly available otherwise at the time of P’s creation. Its contents had been compiled, composed, and published by mid-850. Given that Lupus virtually quotes a description of his threefold concerns as given in his Liber de tribus quaestionibus, it is frankly hard to conceive how seven editors have missed the allusion and near self-citation27. The parallel helps to restore the lacuna with confidence. Lupus published his related Collectaneum only later in response to the command for clarifications from Charles the Bald (ep. A4), but his words to Wenilo suggest that the Collectaneum existed in some form as he drafted his Liber de tribus quaestionibus sometime in first half of 849. In ep. 74, we see Lupus cagily baiting and refusing to convey a risky compilation of theological positions to his bishop, who had, perhaps fortunately, failed to return something else. Lupus had a convenient pretext apart from plain theological reservations. Not least important, the juxtapositions allow us to date both epp. 73 and 74 closely to 849-850, against the uselessly vague range of 842-862 proposed by Dümmler, and the twice revised, twice mistaken suggestion of 858 given by Levillain28.

17The date and reference to Livy (74.2) also proves that Lupus’s interest with that and other Latin Classics ran the whole length and breadth of his carrier as abbot at Ferrières and gently obsessed him as much as property disputes, theological differences, or petitions for help or remedy that led him to address letters to bishops and others. Without exhausting the topic, a brief look at Lupus’s writings to bishops reveals unsuspected people, places, connections, and language. The progress is textual and contextual, and above all, amply rewarding for an author we knew well, but can still understand better and more fully.