Historiographical production within twelfth-century Puglia seems to have been markedly limited, and falls broadly into three categories. Firstly, there are those works which are extant: the ‘family’ of Bari annals, two branches of which (the Annals of Lupus Protospatharius and the Anonymous Barensis Chronicon) stretch their narratives into the first decades of the twelfth century.Footnote 2 Secondly, there are, possibly, other now lost, annalistic sources.Footnote 3 Thirdly, there are important hagiographical texts which were produced in Puglia and which contain some valuable historiographical framing too.Footnote 4 It is a meagre corpus which frustrates attempts to access internal perspectives on a region which played a pivotal socio-political and economic role within southern Italy as it fell under Norman rule, and was subsequently absorbed into the new Kingdom of Sicily in 1130. But then again we must also remember the limited survival of all types of historiographical writing in eleventh- and twelfth-century southern Italy.Footnote 5
It might, however, be possible to bolster the region's twelfth-century historiographical outputs if we were to include a largely overlooked and problematic source, the so-called Fragmentary Troia Chronicle (henceforth the TC). The TC actually seems to combine all three aforesaid Pugliese historiographical categories: (1) it is extant; (2) but only partially, and can therefore be placed in dialogue with evidence for other lost sources; (3) its focus on episcopal activity and information on relics provides a quasi-hagiographical framing. In its current form the TC is a short, hybridized and fragmented Latin text usually considered to be a product of the late twelfth century as a result of its chronological coverage (De Santis, Reference De Santis1986: 259; Loud, Reference Loud2007: 266). It consists of a first section containing an annalistic-style account of political and religious events mostly of relevance to the northern Pugliese city of Troia and its bishopric, and ostensibly covers 1014 to 1124.Footnote 6 It is then merged in a confused arrangement with a second section which initially contains some traces of annalistic material relevant to the later twelfth century and which transitions into an appendix of twelve privileges related to the Church of Troia covering the eleventh and twelfth centuries. It is certainly a problematic, laconic and, often, disorganized account. Its authenticity is questionable. Not surprisingly, the work has often been sidelined in modern scholarship. There are some notable exceptions, particularly Vera von Falkenhausen's valuable examination (1982: 58–65), although her study's focus on the TC was primarily in the context of Norman–Byzantine relations.Footnote 7 As a result, it is important to acknowledge that von Falkenhausen's objectives were not to undertake a holistic assessment of the entire text, particularly the appended documents, nor to situate fully the role of the TC within Pugliese historical writing.
This article therefore offers the first extended analysis of the TC's place within Pugliese historiographical production. It revisits questions around its authenticity, examines potential contexts surrounding its production and content, and provides the first English translation of the narrative section of the chronicle. In so doing, it will argue that we must certainly tread very carefully when using this source, but that the TC's (fragmented, though I prefer the term broken) existence and its main chronological focus could at the very least hold significance as a marker of an enduring remembrance of a vibrant era of episcopal, literary and urban development in the Pugliese city in the eleventh and twelfth centuries.
LITERARY CULTURE AND EPISCOPAL IDENTITY
Before we examine the TC in detail, it is important to situate the content of the text within the localized context in which it purports to be embedded. The TC's content, both its annalistic section and the documents, spans the early eleventh through to the late twelfth century. As we shall see, there are problems with accepting this date range, but we can say at this point that in its surviving form the annalistic section of the TC covering the period c. 1080–1124/7 represents the core of the text. This in itself could be an important inference, because while there is currently no way to conclude definitively that the work was initially produced during that period, either in stages, or perhaps in one phase after 1124/7, we do know that the 50 years from c. 1080 to c. 1127 were a ‘boom’ phase for Troia, both politically and culturally. Indeed, the logical search for the place of production and collation of the information in the TC is of course the city of Troia itself, and (given the TC's content) more specifically the city's bishop and network of officials. If we assess the political and cultural development of Troia across the eleventh and twelfth centuries, compelling evidence emerges to explain why this city was peculiarly well positioned to produce and preserve a work like the TC, whereas so many other Pugliese communities seemingly did not. As such, this also strengthens the case for considering the text as a useful witness to historiographical production and literary culture in medieval Puglia.
Troia was founded in the northern region of Puglia, known as the Capitanata, in c. 1018/19, at the command of the Byzantine emperor. It formed part of a wider wave of new settlements founded in an initiative to consolidate the northern frontier of Byzantine Italy (Mor, Reference Mor1956: 29–36). Early on, the city endured a siege from the German emperor, Henry II, in 1022 which received wide attention in contemporary sources. In the succeeding decades, Troia developed into the pre-eminent urban centre in this region, assisted by its endowment with a sizeable and well-defined rural hinterland and establishment of a bishopric. Unfortunately, we do not know a great deal about the bishopric in its formative stage. Although a Byzantine foundation, it does not seem that the Troian Church was originally served by Greek-rite clergy, and early charter evidence indicates the population was predominantly South Italian Lombard.Footnote 8 It is also not clear why exactly Troia's bishopric was made directly subject to the papal see in 1030 (Chartes de Troia: no. 2, p. 82). At some point in the 1060s Troia passed under the rule of the ascendant Norman Hauteville family and thereafter became a regular place of residence for Robert Guiscard, the new Duke of Apulia (Martin, Reference Martin1990: 188; Oldfield, Reference Oldfield2009: 22, 27). The last quarter of the eleventh century then saw Troia and its Church in ever closer alliance with new forces: the Normans and the reform papacy. The city's bishops accrued more privileges and public revenues, particularly the tithe on ducal income, from the Norman dukes, constructed a new cathedral between c. 1093 and c. 1120, and the city hosted high-profile papal councils in 1093, 1115, 1120 and 1127.Footnote 9 Significantly, some of these bishops also appear to have been external figures, from France and northern Italy, an unusual occurrence in eleventh-century southern Italy and one which indicates the political importance of the city's episcopacy (Martin, Reference Martin1990: 195–8). In the political vacuum that followed the death of Duke William of Apulia in 1127 it should be no surprise that the city received a wide-ranging charter of privileges from Pope Honorius II which articulated the city's aspirations and status (Chartes de Troia: no. 50, pp. 182–5). Some of these privileges were suppressed and modified once Troia passed under the rule of the new Norman king of Sicily, Roger II, in the 1130s, and in the same decade the city was destroyed by the new monarch and temporarily abandoned (Oldfield, Reference Oldfield2009: 57–9, 62–3, 73–4). But Troia revived under the monarchy, and its bishops regained a prominent position in the kingdom's Church in the second half of the twelfth century (Loud, Reference Loud2007: 299, 316–17, 323, 389).
During this same period, there is ample evidence for the development of a burgeoning literary and artistic culture centred on the city. This literary activity equally looked to the city's ‘deep’ past by promoting Troia as heir to the ancient city of Aecae, the ruins of which lay close by. The development of a literary community seems to have been additionally boosted by interconnections with the prestigious Benedictine abbey of Montecassino. One of its most influential abbots, Desiderius (1058–87), developed close affinities with the Norman elites, and consequently the abbey received several donations in the vicinity of Troia where Robert Guiscard was often based.Footnote 10 Bishop Stephen of Troia (c. 1059–80) commissioned the production of an Inventio of St Secundinus, a putative early bishop of Aecae, whose remains were believed to have been rediscovered between 1022 and 1034. Stephen turned to Desiderius for its production, and the latter enlisted one of his monks, Guaiferius, who accomplished the task at some point between 1058 and 1080.Footnote 11 Later, probably in the first half of the twelfth century, an account was produced of the translation of the relics of Sts Eleutherius (another putative bishop of Aecae), Pontianus and Anastasius which arrived in Troia in c. 1105. It was composed by a precentor of Troia named Roffred and dedicated to a Troian bishop named William, who was almost certainly Bishop William II (1106–41) (Poncelet, Reference Poncelet1910). Montecassino again played a part, as the narrative of the translation account assigned roles to one of its monks and to its Abbot Oderisius I (1087–1105), and involved several places connected to the abbey. On this occasion the duke of Apulia, now Guiscard's son Roger Borsa, was also depicted as the supporter of churches and of the relic translation, while the associated miracle accounts situated Troia within a wider pilgrimage network.Footnote 12
Across the eleventh and twelfth centuries Troia was also one of the main production centres for the region's celebrated Exultet Rolls: at least four are known to have been produced in the city (Kelly, Reference Kelly1996: 8). This all fits with evidence for the growing aspirations of the city's episcopacy. As the TC notes too, the last decade of the eleventh century saw the commencement of the construction of the city's cathedral dedicated to St Mary, which was completed by c. 1120. The cathedral boasted two bronze doors, dated to 1119 and 1127, and it seems that the original panels depicted each of the saints recorded in the aforementioned hagiographical texts (Belli D'Elia, Reference Belli D'Elia1999; Bottiglieri, Reference Bottiglieri, Burkhardt and Foerster2013: 119). Indeed, Amalia Galdi (Reference Galdi2010: 78–9) considers Bishop William II (1106–41) to have generated a ‘refined ideological project’ firmly conscious of the mobilizing power of civic identity and ‘marked by an intense cultural activity’. The 1105 translation account, the iconographic programme of the bronze doors, and the surviving list of donations made by the bishop to his cathedral including several books and also some precious objects, are all components of this programme, and evidence of reform and leadership (Braga, Reference Braga, Coulson and Grotans2008; Antonetti, Reference Antonetti, Coss, Dennis, Julian-Jones and Silvestri2020). Indeed, it would certainly be plausible to associate Bishop William II with the commissioning of a work like the TC, and there may be some link between the work and his episcopate.
Finally, Donald Matthew (Reference Matthew, Davis and Wallace-Hadrill1981: 251–3, 256) has made the convincing case that the important twelfth-century chronicle attributed to Archbishop Romuald of Salerno contained a section up to c. 1126 which potentially included traces of a lost set of Troian annals. It is tempting here to make a link with the TC and the abrupt halting of its coverage at the death of Duke William, which it dates to 1124 rather than 1127. It is also tempting to connect these breaks with the ensuing decade or so of conflict on mainland southern Italy as Roger II attempted to establish his new kingdom. This was a conflict in which Troia was heavily involved: in 1133 it was destroyed by Roger II and its inhabitants dispersed to local settlements, while in 1139 the repopulated city was still a key centre of resistance to the new king (Falcone di Benevento: 154, 216, 224, 226; Alexandri Telesini abbatis Ystoria: bk II.49, p. 47). But, as we shall see, while these links are plausible, the problematic manuscript transmission and extant form of the TC render it impossible to make a secure and direct link with either the putative lost Troian annals or the disruption of the 1130s. But we can conclude this section by at least acknowledging that in the last decades of the eleventh and the first decades of the twelfth century, the city of Troia certainly offered the sort of environment in which a local annalistic source, one which focused largely on the city's bishop, could have been produced. Indeed, Graham Loud (Reference Loud2007: 266) suggested that the TC might have the makings of a rudimentary Gesta Episcoporum, a genre of episcopal biography completely absent from twelfth-century southern Italy (Sot, Reference Sot1981).Footnote 13 But, if all of this suggests that the initial roots of the material of the TC could be traced back to this era, unfortunately we cannot conclude definitely that this was so. The problematic manuscript tradition and the condition of the extant edition simply do not allow this.
THE MANUSCRIPT TRADITION, CONTENT AND STRUCTURE
There is no extant medieval manuscript of the TC. It survives now only in a late eighteenth-century edition by the Neapolitan Alessio Aurelio Pelliccia (d. 1823), who, among his many varied works, compiled collections of other historical texts from within the Kingdom of Naples.Footnote 14 Further evidence on provenance is evasive and uncertain. The preface to Pelliccia's edition in De Christianae Ecclesiae primae, mediae et novissimae aetatis Politia only claims that it was drawn from a copy of a parchment codex made by a Giovanni Francesco de Rubeis (olim ex membranaceo codice a Johanne Francisco de Rubeis descriptum), who seems to have been active in the sixteenth century (Chronici Trojani fragmentum: 343).Footnote 15 This copy had apparently been given to Pelliccia by a friend, a Celestine monk named Lodovico Vuolo, ‘into whose hands with other ancient charters by chance fortune it had arrived’ (in cujus manus cum aliis antiquis chartis forte fortuna devenerat) (Chronici Trojani fragmentum: 343). Significantly, however, there is a notable overlapping of content between the TC and the Ristretto dell'Istoria della città di Troja e sua Diocesi dall'origine delle medesime al 1584 (henceforth: RI), a late sixteenth-century vernacular work of the notary Pietrantonio Rosso di Manfredonia, who seems to have died by 1592.Footnote 16 It is clear from a closer examination of the RI that it undoubtedly interacted with the TC in some way. Indeed, at one point the text even mentions a ‘Giovan Francesco de Rubeis. Gentiluomo di questa patria [that is, Troia], dottor letteratissimo e curioso scrittore di cose ed istorie antiche’ (RI: 105).Footnote 17 There are numerous distinctive parallels between the content of parts of the TC and parts of the RI. Among the many one could mention are: the encampments of Hannibal (Castra Hannibalis, Annibalis castra) discussed in the origin accounts of Troia; the Byzantine official who founded Troia is called Bubagiano/Bubagianus, and his alliance with Saracens noted; the story of the burning of Castelluccio by the bishop of Troia; and the mysterious plan/journey of Duke William to see Alexius Comnenus in Constantinople (RI: 105, 106, 108–9, 155, 160, 162) (the TC instead locates the meeting in ‘Pipopolim’). There are more fundamental similarities though. The bishops of Troia are central reference points for several sections of the RI. Biographical data are supplied for the bishops, noting that they were consecrated (often by the pope), providing the dates of their episcopates, and sometimes their geographical origins; and papal synods are also recorded. Often the TC and the RI provide identical data for much of these, or when not it would seem to be a copying error.Footnote 18 Also, as will be discussed shortly, of the twelve appended documents found in the TC, summaries of at least seven are included in the RI while there is overlap with the version of the translation of the relics of Sts Eleutherius, Pontianus and Anastasius in c. 1105.Footnote 19 However, in the RI these document summaries are usually embedded at the appropriate chronological place within the narrative rather than grouped separately at the end of the text as in the TC.
There are of course many differences. The RI is much more extensive, its coverage extends well beyond the 1124 narrative cut-off of the TC and it seems to draw on several other sources. It often includes more information on the same events found in the TC, or more information on events and figures and additional document summaries all of which are omitted from the TC as it currently survives. Vice versa, some information in the TC does not appear in the RI: the references to falling stars, Coybanus (Kerbogha), the rebellion of the Troians in 1081 and William Hauteville's cruelty to Jerusalem pilgrims. There is not scope here to attempt to reconstruct in greater detail a fuller version of the TC based on its interrelationship with the RI, nor would it be entirely possible. Falkenhausen (Reference Falkenhausen1982: 159) may well be correct that the RI utilized the work of Giovanni Francesco de Rubeis from whose codex Pelliccia said he had copied the TC. Certainly, one can speculate that certain additional information found in the RI may well have originally been embedded within the TC. But tracing the manuscript transmission of the RI and assessing the editorial work of Nicola Beccia is also complicated and raises another set of questions (Tateo, Reference Tateo1985: 111).Footnote 20 Addressing these would require a detailed study in itself. Add to this the highly fragmented state of the TC and it proves very difficult to be completely certain about the exact mechanisms of the intertextual links between it and the RI. The precise mutual relationship of dependency is not clear, and one cannot rule out a lost common source too. In short, we cannot say much more than that both the RI and TC at some point interacted either directly or indirectly, and that this indicates that a large amount of the material that formed the basis of the TC was in circulation by the late sixteenth century at the latest. While a deeper investigation is beyond the scope of this article, it is also important to acknowledge that material found in the TC may have been circulating earlier than the RI. Marino Zabbia (Reference Zabbia, Sanfilippo and Rigon2010: 32 n. 35) is correct that, although the TC does focus on some similar Troian events which appear in the late twelfth-century chronicle attributed to Romuald of Salerno, there is no conclusive evidence of a direct link. Thus we cannot turn to that chronicle as certain proof of an early circulation of parts of the TC. Francesco Tateo (Reference Tateo1985: 112) briefly noted that the Italian humanist historian Biondo Flavio (d. 1463) and Marino Freccia (d. 1566) utilized the TC, and the entry for the Troian bishops in Ferdinando Ughelli's Italia Sacra series (originally published in 1644 with a second edition in 1717) does reference both Biondo Flavio and Marino Freccia as sources of information on the ancient origins and foundation of Troia (Italia Sacra: c. 1334). But a much broader study would be required to trace the potential interlocking web of transmission circulating in the late medieval and earlier periods.
Whatever future studies unearth, it is more germane to the present study to emphasize, however, that the manuscript tradition and provenance of the TC is highly problematic. Moreover, as Pelliccia claimed already in the eighteenth century, in both the preface and the very title of his edition (Chronici Trojani fragmentum), the copy he transmitted was a fragmentary version. As mentioned, the RI offers clues in places as to how the TC might have looked in a fuller form. Aside from the obvious point that the annalistic section of the TC regularly misses several years, there are numerous indications of its ‘broken’ form. It is full of omissions, editorial errors, and curious interpolations and synthesizing of text: the reference to the succession of a Bishop John is the first mention of the Church of Troia and yet implies that this followed some earlier information on the bishopric (information incidentally which the RI contains); the abrupt ending of the entry in 1105; short omitted sections in the entries for 1119 and 1122; the year 1119 also seems to combine two separate entries, as consecutive paragraphs open with ‘Anno 1119’.
The entry for 1093, in particular, highlights problems with transmission. It starts by noting a papal synod held at Troia and records the start of the construction of the church of St Mary by Gerard, bishop of the city. Then after noting some falling stars in the sky (often associated with the start of the First Crusade in 1095) a new entry seems to begin which repeats the information about the church of St Mary. It then says ‘after some time’ (post aliquod tempus) Bishop Gerard died and was succeeded by Cenomanicus, while the next sentence starts ‘in the above said year’ (Supra dicto Anno) and seems to record, in an ambiguous and incomplete sentence, how Coybanus (undoubtedly Kerbogha, the atabeg of Mosul) besieged Antioch (an event which occurred in 1098) (TC: 344–5). The year to which this entry now refers is far from clear. Moreover, the subsequent entry, for 1097, starts by noting that Bishop Gerard died in that year, that he was succeeded by another bishop, called Ubertus Cenomanicus (clearly the Cenomanicus from the 1093 entry), and that the city of Troia was struck by a fire. To further confuse matters, this 1097 entry also seems to conclude with a sentence that should actually be part of the following entry for 1101: it says that the bishop of Troia (implying Ubertus Cenomanicus) died and was succeeded by Bishop William Vigoctus. Ubertus did not die in 1097. He was last recorded in a charter of November 1100, the original of which is extant, and the following entry in the TC repeats the same information on Ubertus's death and William's succession under the year 1101, adding that Ubertus had been in office for nearly four and a half years (Chartes de Troia: no. 35, pp. 147–8). So, the first mention of Bishop Gerard's death initially appears to be embedded in the 1093 entry, as does the news on Kerbogha, though neither clearly should be there. Gerard's last mention in a charter is in December 1095, while the sieges of Antioch stretched from October 1097 to June 1098 (Chartes de Troia: no. 31, pp. 141–2). The same information about Ubertus and William Vigoctus is mentioned at the end of the 1097 entry and the start of the 1101 entry. It is clear that information is missing and jumbled up in the 1093, 1097 and 1101 entries, suggesting that another entry from a different source has been partially inserted or only partially omitted.Footnote 21
Further confirming the hybrid, miscellaneous nature of the TC, after the 1101 entry the text is suddenly provided with a subheading: De corporibus sanctis (TC: 345). There follows a short passage on the translations of the relics of Sts Eleutherius, Pontianus and Anastasius in 1105. This is an interesting interlude because the information closely follows the aforementioned account of the translation by Roffred which survives in one twelfth-century manuscript. More than this, the TC includes a couple of passages from the latter almost verbatim, and indicates how copying errors seem to have crept into the TC (Poncelet, Reference Poncelet1910: 414–16). The TC (345) describes the procession that welcomed the relics to the city: cum Troyanis decem millibus utriusque sexus obviam exierunt. The twelfth-century hagiography says: obviam procedentes cum fere decem milibus utriusque sexus exeunt civitatem (Poncelet, Reference Poncelet1910: 422). Similarly, the TC's closing passage (345–6) dates the translation thus: Gloriosa Translatio Santorum [sic] Martryum Eulecterii Pontificis, & Anastasii fuit quartodecimo Kalendas Augusti, anno Dominicae incarnationis 1105. 13. inditione, procurante totius Orbis Monarchia Paschali pp (papa) secundo Imperante et Gloriosissimo, ac sanctar. Preceptoris Ecclesiarum Electore Roggerio nobilissimo. The twelfth-century hagiography says: Quorum gloriosa translatio quarto decimo kalendas augusti anno dominice incarnationis millesimo centesimo quarto, indictione tertia decima, procurante totius orbis monarchiam Pascali papa secundo, imperante etiam gloriosissimo ac sanctarum pre ceteris ecclesiarum cultore Rogerio duce nobilissimo (Poncelet, Reference Poncelet1910: 424). This shows that the compiler did on this occasion draw on an authentic twelfth-century text, either directly or indirectly. The TC also inserts in the midst of this section on the translation a brief reference to an inscription relating to the discovery of the relics by individuals named Perisentia, Gaio and Aurelio. This is not found in the twelfth-century hagiography, while it is worth noting that the RI does on occasion utilize inscriptions and other material evidence to support its narrative. Poncelet (Reference Poncelet1910: 412–13), the editor of the hagiographical account, noted this reference in the TC and the discussion of it, with a fuller transcription which claimed it could be found on the saints’ reliquary, by Stefanelli in his Memorie storiche della città di Troia (Napoli, 1879). The TC then jumps abruptly back to the annalistic entries which recommence with the year 1114.
The TC is also problematic after the last unambiguous annalistic entry for the year 1124. The next section of the TC (pp. 347–8) is then organized under two headings: Dedicatio Ecclesiae S. Vincentii de Troya and Corrigia Troyana. Immediately after the first heading, the text suddenly appears to offer a copy of a short document relating how the bishop of Troia dedicated an altar and church in honour of St Vincent and furnished it with the relics of Sts Eleutherius, Secundinus, Nicholas and Katherine. The document and dedication are dated Sunday 2 May 1169, in the fifteenth year of the bishop's episcopate, but no witnesses or notaries are recorded. Immediately after this document two confusing sentences are inserted: ‘In the year of the Lord 1170 of the incarnation of the Lord in the month of March, the fifteenth day, fourteenth indiction, died the aforesaid William in the year 1190 in the month of November, the eighteenth day, died King William II.’ It is not at all clear what is going on here apart from the seeming reference to the death of King William II of Sicily.
Similarly, after the Corrigia Troyana heading, there appears what seems to be a summary of a document. Dated to 11 May 1182, Bishop William IV of Troia records how he had regained a tenimentum (called corrigia Troyana), which the Church of Troia had previously leased out to a certain Raynaldo, and that two identical documents were made, one for Raynaldo and one for the Church treasury, but it is not entirely clear if these documents relate to the bishop's regaining of the corrigia Troyana or the earlier agreement to lease it out. The implication would seem to be the former. But, consequently, the reference that this occurred in the first year of William's episcopate is ambiguous, because a papal charter to William IV of March 1180 shows the bishop would already have been in office for over two years by May 1182 (Chartes de Troia: no. 97, pp. 289–91).Footnote 22 Then immediately after this notice about William's first year in office we encounter the following two sentences: ‘He died, in the year of the Lord, the lord Bishop William IV in the year of the Lord 1187, in the month of February in the fifth indiction. To the same Bishop William IV succeeded Bishop Roger from the clergy of the same city.’
Thus, in both headings we have what seems to be a summary of a document, followed by a somewhat jumbled remnant of biographical data related to the bishops of Troia and the king of Sicily. This offers a hint that either the narrative section of the TC may have continued into the later twelfth century, particularly as an episcopal biography, but has been substantially lost, or that another narrative source was embedded after the first section of the TC. It is worth noting here that the RI (226) mentions the death of Bishop William of Troia in 1188 and the election of Roger from the Troian Chapter, a similar qualifier for the new bishop found also in the TC.
Finally, we must consider the remaining collection of documents which follow the two aforementioned documents and entries. What do they tell us about the TC's compilation? Including the two aforementioned documents, there are in total twelve documents or apparent summaries of documents. Each is entered under a separate heading in the TC and arranged in the TC in the order given in Table 1. The table also notes the provenance of the document, and if it is present in the RI.
Table 1 Documents included within the Troia Chronicle
a The document numeration has been added for the purposes of the current article: there is no such numbering in the TC; the page numbers refer to their place in Pelliccia's edition of the TC as listed below in the references section.
b The page numbers refer to Beccia's edition of the RI as listed below in the references section.
c Chartes de Troia: no. 85, p. 260.
d Chartes de Troia: no. 36, pp. 149–51.
e Chartes de Troia: no. 33, pp. 144–5.
f Chartes de Troia: no. 16, pp. 108–9.
g Chartes de Troia: no. 27 pp. 133–4. But Martin (Chartes de Troia: p. 133) suggests that ‘Les données chronologiques sont aberrantes’. The TC version omits the lengthy section on boundaries.
h Chartes de Troia: no. 34, p. 146.
i Chartes de Troia: no. 62, pp. 210–11.
j Chartes de Troia: no. 15, pp. 107–8.
k Chartes de Troia: no. 99, pp. 293–6.
l Holtzmann (the editor of Italia Pontificia: no. 5 p. 204), concluded it is spurious, and neither was it included by Martin in his edition of the Chartes de Troia.
The important point to note is that after the first two documents there is no indication of any narrative or biographical data inserted alongside the remaining ten documents. The only organizing principle behind the inclusion of all these documents is that they concern the bishop and Church of Troia; otherwise, they are not arranged chronologically, nor by patron nor donation type. It seems unlikely that they were part of an attempt to create some sort of episcopal cartulary, given that the documents, at least as they are presented in the Pelliccia edition, do not pretend to be like-for-like copies of an original, and would only imperfectly function as tools for memoralization (Bouchard, Reference Bouchard2015: 9–37). They are mostly abbreviated summaries conveying only incomplete signs of authentication (that is, without a full protocol, eschatocol, papal rotas, etc.).
More importantly, and problematically, only one document (no. 4) is still extant in its original form in the Archivio Capitolare. Of the others, two (nos. 2 and 3) are only known via the TC, while the majority, nine, are only now extant in the work of the Troian canon Vincenzo Aceto de San Severo who compiled a two-volume manuscript entitled Troia Sagra in 1728. Unfortunately, for the present study it was not possible to consult this work which remains unpublished within Troia's Archivio Capitolare. Aceto's manuscript collection contains transcriptions of documents still surviving in the Archivio Capitolare and others that are no longer extant.Footnote 23 However, Martin (Chartes de Troia) reasonably considered that a number of those charters preserved by Aceto which no longer survive in their originals were of sufficient authenticity to edit and include them in his modern edition of the Troia charters. Indeed, nine of the ten documents which are extant in Aceto and also appended (often in a moderated form) in the TC appear in Martin's modern edition. While some of the Aceto charters included by Martin carry more complete signs of documentary validation, the ten that overlap with the TC, with the possible exception of nos. 4 and 11, mostly do not. Moreover, in Pelliccia's edition, even nos. 4 and 11 are both notably abbreviated with reduced signs of documentary authentication compared to the Aceto version edited by Martin. Both are missing the arengae and closing admonitions and penalties for breaking the agreement, and in no. 4 only four of the eight witnesses are listed.Footnote 24 Document no. 11 is a particularly abbreviated version of the Aceto charter, with some sentences seemingly elided and their sense altered, and some extra information added, such as a record of a previous ‘privilege of Duke Roger son of Robert Guiscard’ (TC: 355). Of all the TC documents which overlap with Aceto, Martin only omitted document no. 12, no doubt because, in line with Holtzmann's earlier judgement that it was spurious, the document contains erroneous information about the foundation of Troia and an anachronistic reference to the settlement of Foggia (Corsi, Reference Corsi and Mariani1997: 13–17; Martin, Reference Martin1998: 29–38).
It is beyond the immediate scope of this current article to explore any connections between the TC and Aceto further, but future studies of Aceto's manuscript may well shed additional light on the TC. As things stand, the TC might, if we accept that its material was in circulation at least by the late sixteenth century, preserve the earliest, albeit problematic, record of some documents which now only survive in their fullest form in Aceto's much more extensive collection. But we cannot safely accept that they were originally appended in the eleventh or twelfth century, or whenever the narrative section was compiled. That in fact seems rather unlikely, and document no. 3 would seem to contain information that postdates 1231. That documents 1 and 2 also contain very short traces of annalistic material, the apparent ghost of a text, remains intriguing and leaves open the possibility that the narrative was at one point more extended, at least pushing into the late twelfth century.
All of this also raises the question of Pelliccia's own intervention in the text. The edition contains several instances of abbreviations which are accompanied by an expansion in brackets in italics, which seems to indicate Pelliccia's editorial intervention, as do the footnotes.Footnote 25 Beyond this, did Pelliccia simply transmit the text as he received it? When and how did the jumbled annalistic sentences covering the late twelfth century become inserted after the first two documents? Definitive answers are not currently possible. Pelliccia calls the whole text a fragmentum, which acknowledges its incomplete, broken form, and surely full awareness of the disorganized and problematic nature of the work he published.
Despite the significant reservations that therefore must surround the compilation and form of the TC, there are some indications that we should not discount entirely its potential to represent an eleventh/twelfth-century voice on Troia and Puglia. The TC's problematic manuscript transmission is not actually unique for the region. Indeed, the manuscript histories of the other main extant Pugliese historiographical works from the same period are far from straightforward. These are the three aforementioned Bari annals, which are rightly accepted as important sources. The earliest extant manuscripts of the Annales Barenses and the Annales of the so-called Lupus Protospatharius, which cover the years 605–1043 and 855–1102 respectively, only date from the fifteenth century, but their transmission does seem authentic and satisfactory (D'Angelo, Reference D'Angelo2003: 17–25, 198–215; Spiezia, Reference Spiezia and Trombetti Budriesi2009: 261; Antiche cronache di Terra di Bari: 17–25). The transmission of the third, the Anonymous Barensis Chronicon (covering the years 855–1118), however, is somewhat less satisfactory: its codex is lost and it survives only in an edition produced in 1643.Footnote 26 Moreover, the first, annalistic section of the TC up to 1124 does retain some value. Indeed, Falkenhausen concluded that the ‘Chronicle of Troia is more reliable than its reputation’ (1982: 65). In summary, Falkenhausen noted that, while there are inaccuracies with some dates and minor information, possibly owing to editing errors in the transmission of the text, the work itself does convey some authenticity. According to Falkenhausen's assessment, this is due to the combination of two factors: on the one hand to some of the content's corroboration in other South Italian works of the eleventh and twelfth century, and on the other to the TC containing information that does not appear elsewhere, meaning that it might represent autonomous textual traditions. Falkenhausen's study explored some of these comparisons, differences and dating inaccuracies.Footnote 27 There is no need to repeat these in detail — a few illustrative examples among the several which von Falkenhausen explored will suffice:
a) information unattested elsewhere: the entry which is dated 1081 accurately records Robert Guiscard's invasion of the Byzantine Empire on account of the sidelining of his daughter, here called Olimpias, a name not attested in any other source, and the primary subject of Falkenhausen's study; a localized conflict in 1115 between Bishop William II of Troia and a William de Hauteville, a local aristocrat, said to be committing cruelties on Jerusalem pilgrims is not mentioned in other sources (though a charter of 1120 from Troia's Archivio Capitolare supports the potential for dispute between these same two parties) (Chartes de Troia: no. 43, pp. 168–71);Footnote 28 news of the Duke of Apulia's journey in 1116 to the Byzantine emperor — such a journey seems strange if not entirely implausible and Falkenhausen concedes that there is no clear solution.Footnote 29 Finally, certain biographical information relating to the bishops of Troia is not attested in other eleventh- and twelfth-century sources, but charter material supports some of this information as we will see in more detail below.
b) parallels with information provided in other eleventh/twelfth-century sources (though in no case does this indicate direct links with the TC): with William of Apulia's Deeds of Robert Guiscard (the record of a revolt in Troia in 1081); with the Montecassino Chronicle, Annales Beneventani and the Annales of Lupus Protospatharius (reference to shooting stars around the time of the First Crusade); with the Troian material (perhaps some lost Troian annals) found in Romuald of Salerno's Chronicle (the synod at Troia in 1093, fire in the city in 1097, and the translation of relics to Troia, though Romuald's chronicle dates this to 1106 and the TC to 1105).Footnote 30
c) dates: some of those which are obviously askew could be explained by copying errors (as could inaccuracies with some names);Footnote 31 the date of Troia's foundation is given as 1008, but this seems to be a copying error for 1018, the more likely date of the city's establishment; Duke William of Apulia's death is given as 1124 instead of 1127; the date for the siege of Antioch appears in the entry for 1093 rather than 1098 (though this seems, as mentioned above, to be linked to the jumbled insertion or omission of material). Some dates appear to be around a year out, not an entirely uncommon occurrence in local medieval annals which potentially followed different dating methods. The investiture of Duke William dated to 1115 took place in late 1114; the entry for 1116 records a papal synod held in Troia, but this seems to have been held in 1115; Pope Gelasius II is said to have died in 1118, when his death occurred in 1119. Pope Calixtus II was said to have visited Troia in 1119, while charter evidence attests his holding of a Truce of God council in the city in 1120; Jordan II became Prince of Capua in 1120 not 1119 (Chartes de Troia: no. 43, pp. 168–71).
Moreover, Graham Loud (Reference Loud2007: 209) is right to question the TC's claims that Troia's bishops were regularly consecrated. He also notes that Bishop William II was unlikely to have been consecrated, according to the TC, in December 1106 by the pope, because the latter was in France at that point. Perhaps, however, this is another example of inaccurate dating. On the other hand, information on several of Troia's bishops often correlates, at least broadly, with the chronologies known from charter documentation, or provides otherwise unattested evidence that appears plausible. Indeed, Jean-Marie Martin compiled a short biographical appendix of Troia's bishops across the eleventh to the thirteenth century, which synthesized information provided by charter material and information drawn from other sources including the TC. For example:Footnote 32
• Bishop Stephen: last documented attestation = July 1080; death in the TC = 11 October 1080.
• Gerard: last documented attestation = December 1095; death in the TC = 10 January 1097.
• Hubert/Ubertus: first documented attestation = December 1097 and last attested 10 November 1100; consecrated in the TC = 20 June 1097 and death in the TC = 13 December 1101.
• William I: first documented attestation = January 1105; consecrated in the TC = 13 January 1102.
• William II: episcopate documented from 1106 to 1140; the TC mentions him in entries for 1114, 1118, 1119 and 1122.
• William IV: last documented attestation = May 1187; the TC gives his death as February 1187.
So, while we cannot fully understand the interrelationship between the annalistic section and the appended documents, nor ascertain when certain parts of the TC were first compiled, the current evidence suggests that we can tentatively proceed on the basis that the TC's information does contain some reliability in the annalistic section up to 1124/7. But this is offset by inaccuracies particularly linked to the transmission of the text, to skewed dating and names, and to the presence of certain puzzling entries.
CONCLUSION
The aim of this article has been to offer, as far as is currently possible, a holistic understanding of the TC, and to address questions around its authenticity and production. These tasks were undertaken to alert readers in two ways: first, to the potential existence of a Pugliese historiographical text which may do a rather rare thing: shed some light on the region, and textual production there, in the eleventh and twelfth century from a Pugliese perspective; second, to the importance of using this historiographical text with care and due caution. For, as demonstrated, the TC seems to embody the tormented history of a broken text: its authenticity remains open to question, its transmission has been fractured, its core integrity broken with omissions, insertions and problematic compilation. It seems to contain a miscellany of materials. If several questions have been raised in this article for which definitive answers are not currently possible, this in itself is an important finding, and it is hoped it will serve as a platform for further research. Indeed, future studies may clarify more fully some of the questions around dating, around the manuscript tradition, about the interrelationship and provenance of the sources which may form the TC, and about editorial intervention. For now, and perhaps for some time, we must be comfortable with the uncertainty. It may preserve echoes of contemporary eleventh- and twelfth-century voices, or it may not. The current balance of evidence indicates that the TC does have some value both for the study of eleventh- and twelfth-century Troia and also for an understanding of historical writing in medieval Puglia. Moreover, much of the TC's material appears to have been in circulation by the late sixteenth century at the latest. It is clear, above all, that there was an unusually rich corpus of material and resources associated with late-eleventh- and twelfth-century Troia. These could be drawn on (as the TC clearly did on at least one occasion with Roffred's twelfth-century account of the relics translation) by any author, compiler or editor wishing to produce a historical record of the city. Indeed, the late eleventh and early twelfth century represented a particularly vibrant phase of cultural, ecclesiastical and political accomplishment for Troia. An annalistic work centred on its bishops would not be an anomalous product from this city in this era, and at the very least the TC reflects the memory of this phase of increased activity. What the TC does clearly demonstrate, even in its broken and problematic form, is that later generations continued to remember, recycle and rework Troia's medieval past, in an ongoing, but often highly precarious, process.
APPENDIX
THE FRAGMENTARY TROIA CHRONICLE (NARRATIVE/ANNALISTIC SECTION)Footnote 33
In the year 1014, Henry Duke of Bavaria, with the support of the nobles, as well as from hereditary succession, first took on the imperial power at Rome after twelve years and received the crown from Benedict VIII; he fought against the Saracens, and drove them from Capua, and Bubaganus the duke of the Greeks, who supported them to such an extent that he was pursued in battle, was expelled from Troia, which he had founded in Apulia in 1008, having introduced Greek settlers [coloniis] where the encampments of Hannibal had been.
Afterwards the Church of Troia was vacant for one month and ten days and he succeeded from the clergy Bishop Jo[hn?]Footnote 34 whom Pope Benedict IX consecrated, and he was bishop for thirty years, one month and twenty-five days, and died on the sixth day of August.Footnote 35 The Church of Troia was vacant for eight months and twenty-five days, and Bishop Stephen Normannus followed.
Bishop Stephen Normannus was in office for eighteen years, seven months and ten days. He died on the eleventh day of October. The Church was vacant and Walter Frangente succeeded him on the second day of the month of November, and he was consecrated by Gregory VII.
In the year 1081, Alexius was made emperor, and Duke Robert crossed the sea, and started to attack him on account of Olimpias his daughter, whom he had banished from his palace.Footnote 36 In this year the Troians rebelled on the feast of St Mark, because [of?] Count Roger the son of Duke Robert, whom the father had left behind to guard his land, and, having been defeated by the Normans, thereafter cruel death was inflicted on them.
Bishop Walter was in office for eleven years, three months and eight days; he died on the fourth day of the month of August. The Church was vacant for one month and eleven days, and Gerard Placentinus succeeded him on the eighth day of the month of October, and was consecrated by Pope Urban II.Footnote 37
In the year 1093 Pope Urban II held a Holy Synod at Troia with fifty-five bishops and twelve abbots for one week in the month of March.
In this year Bishop Gerard of Troia commenced building the church of St Mary in the month of April. In this year a sign was made in the stars, which seemed to fall from heaven like stones.
In the same year Bishop Gerard of Troia started to construct the church in the month of April and after some time the aforesaid Bishop Gerard died and he was succeeded by Cenomanicus.Footnote 38 In the aforesaid year Coybanus Turcus was led with an innumerable multitude of Christians, who with the Christian people at Antioch in which they had besieged the Christians . . . . .Footnote 39
In the year 1097 Bishop Gerard died on the tenth day of the month of January and he had been in the episcopate for nine years, seven months and eleven days, and he was succeeded by Bishop Ubertus Cenomanicus on the twentieth day of June, who was consecrated by Pope Urban II. In the same year the new city of Troia suffered a fire on the first night of August. After this the Troian bishop died and Bishop William Vigoctus succeeded.
In the year of the Lord 1101, Ubertus Bishop of Troia died on the thirteenth day of the month of December. He sat in the episcopate for four years, five months and twenty-four days and the Church was vacant for thirty days and William Vigoctus succeeded in the episcopate on the thirteenth day of the month of January, [and] who was consecrated by Pope Paschal II.Footnote 40
Concerning Holy Bodies Footnote 41
In the year of the Lord 1105, thirteenth indiction. The bodies of the saints Eleutherius and Pontianus, and Anastasius were translated to Troia from the city of TiveraFootnote 42 on the nineteenth day of the month of July, and standing by with the arriving bodies [were] Bishop William and the bishop of Bovino along with that of Tertiveri, the bishops making a procession, went to meet [the saintly bodies] with 10,000 Troians of both sexes. This was written on the side of the tomb:Footnote 43 I Perisentia, pilgrim, with Gaio and Aurelio my husband found the bodies of the saints Eleutherius the bishop, and Pontianus the pope, near the via framineam in the place V . . . . . . .Footnote 44
The glorious translation of the martyr saints Eleutherius the bishop, PontianusFootnote 45 and Anastasius was on the fourteenth [day before the] kalends of August [that is, 19 July], in the year of the Lord's incarnation 1105, thirteenth indiction, with Pope Paschal II administering the monarchy of the whole world and with the most glorious, and noble, and tutor of the sacred churches, the Elector Roger governing.
In the year 1114, William II Bishop of Troia offered many gifts to the Troian episcopacy, and many books and other ornaments, and he recovered an ecclesiastical province,Footnote 46 which had been lost for twenty-eight years.
In the year 1115, Duke William accepted investiture from Pope Paschal for the Duchy of Apulia at Ceprano,Footnote 47 in this year the Troian bishop besieged Castelluccio, and he burnt it, and he captured within it William of Hauteville with everything [followers/goods?] of his on account of the cruelty which he had done to the Jerusalemite pilgrims.
In the year 1116, Duke William crossed the sea and went for a conference with Emperor Alexius at Pipopolim.Footnote 48 In the above-mentioned year Pope Paschal celebrated a synod at Troia and confirmed truces of up to three continuous years.
In the year 1118, Pope Paschal died; John of Gaeta succeeded him in the papacy, in the same year Bishop William offered the bronze doors which are on the front of the Church of Troia.
In the year 1119, twelfth indiction. In the first year of the pontificate of Lord Pope Calixtus II [and] the ninth of the son of the glorious Duke Roger. William II Bishop of this Troian see in the episcopacy….. who in the twelfth year these doors to be made from his own Church treasury, and also he raised the fabricFootnote 49 [of the Church] almost from its foundations.
In the year 1119, Robert Prince of the Capuans died, [and] JordanFootnote 50 his brother succeeded. In this year Duke William was made the man of Pope Calixtus at Benevento from the month of October. In the same month, the aforesaid pope came to Troia and was honourably received.
In the year 1122, Lord William Bishop of Troia started to build the Titulus Footnote 51 of St Mary and offered many gifts to its Church of Troia. He died…he succeeded.
In the year 1124, Honorius Bishop of TroiaFootnote 52 ordained the abbess called KatherineFootnote 53 in the monastery of St Blasius. In this aforementioned year the said bishop of Troia offered small bronze doors to [the church of?] St Mary. In the same year Duke William died at Salerno.
Dedication of the Church of St Vincent of Troia Footnote 54
In the name of the Father, the Son and the Holy Spirit, I, William III by divine Mercy humble Troian bishop dedicated this altar and church in honour of God and of Blessed Vincent, DeaconFootnote 55 and Martyr, furnishing it with the most blessed relics of the blessed Eleutherius, Bishop and Martyr, and of the blessed Bishop Secundinus, and of Nicholas, Bishop and Confessor, as well as of the Virgin Katherine.Footnote 56 In the year of the Incarnation of our Lord Jesus Christ, from whose side hanging on the cross the Church was built, 1169, in the fifteenth year of our episcopate, Sunday, the month of May, the second day, the second indiction.
In the year of the Lord 1170 of the incarnation of the Lord in the month of March, the fifteenth day, fourteenth indiction, died the aforesaid William in the year 1190 in the month of November, the eighteenth day, died King William II.
Corrigia Troyana
In the year of the Lord 1182, I, William IV by divine mercy humble Troian bishop on the eleventh day of May regained the tenement which is called corrigia Troyana, of which tenement the Troian Church had no use for a very long time, and leased it to lord Raynaldo and his heirs for four ounces annually, and concerning this two identical documents were made, of which one is in the possession of lord Raynaldo, and the other is in the Treasury of the Church, and this was done in the first year of his episcopate. He died in the year of the Lord, the Lord Bishop William IV in the year of the Lord 1187 in the month of February in the fifth indiction.
To the same Bishop William IV succeeded Bishop Roger from the clergy of the same city.
