Creeping up on the Roman Provincial

Abstract

Around the year 1200, the court of the Roman pope produced a list of all the bishops – and hence of all the cities – in the Christian world. For the next three centuries this text was copied and updated by kings, priests, lawyers and academics across Europe. The numerous surviving manuscripts of this text – the so–called ‘Roman provincial’ – have received some attention from scholars, but the sheer number of manuscripts has meant that any attempt to catalogue and study them en masse is all but destined to fail. This article suggests a different approach: that the most interesting feature of the provincial manuscripts is their differences; the ways in which copyists changed the ecclesiastical and political geography of Europe to meet their own preferences and expectations. Political geographers and modern historians have long been aware of ‘contested cartographies’ and battles over borders on maps; by studying the Roman provincial we can apply such lenses to the medieval world too. Thirteenth-century kings were quite as aware as we are that maps and lists constitute, rather than just describe, political realities.

Lists are interesting. Although some colleagues who grapple with the excitement of legal cases, battles, internecine village disputes or high politics might yet dispute this statement, many would now accept it.¹ Lists structure our world: they both describe and constitute reality (at least in the minds of the readers). For Jack Goody, they were characteristic of a literate rather than oral culture: how could a list exist in a purely oral society? Lists encourage categorisation and division, hence reflecting the concerns of the list-maker, irrespective of how peculiar such divisions might seem to readers from other cultures.²

Many societies have lists; the Middle Ages were no different. Around 1200, the papal court – the courtiers, administrators and clerics who formed the jurisdictional apex of medieval Christendom – put together a text. This was the so-called Roman provincial, the Provinciale Romanum, a list of all the bishops in the Christian world. As a list of all the bishops, the Roman provincial was also a list of all the cities in Christendom. Of course, this was not the first such list: earlier lists of all the bishops and all the cities in the world had been written and copied since antiquity.³ But around 1200 a new list was made. This list, the Roman provincial, would continue to be copied, altered and updated until 1500.⁴

The Roman provincial listed all the bishops in the world. The bishops were divided up both geopolitically and by province. So, for example, most manuscripts of the provincial read:⁵

There is a geopolitical heading (In Anglia – ‘In England’) and then divisions by provinces – Canterbury and York. The cities of England and Wales are enumerated in order. The internal logic of the provincial is thus fairly simple: the list is divided by polity and by ecclesiastical province; by both secular and ecclesiastical unit; by both king and archbishop. This logical division, as well as being observable from the manuscripts, is also confirmed by one of our earliest literary references to the provincial. Gerald of Wales, in one of his frequent attempts to get the diocese of St David's recognised as an archdiocese (and as the metropolitan of the other Welsh dioceses, none of which were – according to Gerald – subject to Canterbury), appealed to Pope Innocent III. When Gerald was in Rome, around 1199, Innocent called on his copy of the Roman provincial to check the status of St David's.

Innocent

ordered his register to be brought, where the Cathedral Churches, both metropolitans and their suffragans, from all the world of the faithful and every kingdom are enumerated in order. And when he turned to the kingdom of the English, it was written there in this manner and read: ‘The metropolitan of Canterbury has these suffragan churches: Rochester, London’, and so on in order. But with the suffragan churches of England all enumerated, [and] with a rubric ‘concerning Wales’ [De Wallia] interposed, it continued in this manner: ‘In Wales, the Church of St David's, Llandaff, Bangor and St Asaph’. When this was heard, the pope suggested, as if taunting and mocking, ‘See! The Church of St David's is numbered with the others’.

Gerald replied: ‘But that [Church] and the others of Wales are not numbered in the same manner as the others, that is to say, in the accusative, as the suffragans of England. If they had been, then they could indeed have been considered subject’.

To which the pope said ‘You saw that well. But there is something else which similarly works for you and your church, that is to say, the inserted rubric, which is never added to the register, except where there should be a transition either from kingdom to kingdom or from metropolitan to metropolitan’. ‘That's true!’, said Gerald, ‘And Wales is a part of the Kingdom of England, not a kingdom itself’.⁶

We can see fairly clearly that this register was a provincial. The bishops under a metropolitan archbishop are subject to that archbishop, and a rubric denotes either a change in political unit, or a change from archiepiscopal province to archiepiscopal province.

Although, in the example above, the provincial manuscript merely includes an abbreviation mark at the end of each diocese (e.g. Londonien’), those manuscripts that do not abbreviate overwhelmingly seem to give place names rather than groups of people. So, for example, Bamberg, Staatsbibliothek [BS], Msc. Can. 91 gives Londoniensem – ‘London’, a city, rather than ‘the Londoners’. The provincial – from the manuscripts I have looked at – envisages bishops as bishops of dioceses rather than as bishops of groups of people. In this article, I have only expanded the place names where the original manuscript has done so.

The provincial was divided geopolitically. The overall layout of the provincial is very consistent across the manuscripts. Most (although not all) manuscripts of the provincial begin with a list of all the churches in the city of Rome which had cardinals, beginning of course with the cardinal-bishops, and then list the archbishops and bishops of the provinces of Italy, Sicily and Istria, then the rest of Croatia and the Dalmatian coast; Hungary; Poland; Germany, Austria and Livonia; Burgundy and Lotharingia; modern-day France; Spain and Portugal; England; Scandinavia; Scotland and Ireland; Sardinia; and then the dioceses ultra mare – beyond the sea, in Jerusalem and the Holy Land.⁷

When one considers that any list betrays the worldview of its compiler, it is easy to see how this layout gave rise to the view that the provincial created,

in the form of lists and tables, an administrative order establishing continuity with the center via concentric circles, like a “new empire, in which papal authority radiated outward, urbi et orbi, over all the provinces and bishoprics of Christendom.⁸

The recent scholarship on the provincial – from, inter alia, Fabrice Delivré and Florian Mazel – has tended to take this view: that the layout of the provincial makes it a conceptual tool of papal monarchy, whereby Rome is at the centre of the world and Christendom – under the authority of the pope – radiates out to the periphery. Delivré, the scholar who has focused most on the provincial itself and its manuscripts in recent years, has significantly advanced our understanding of the genesis and use of the text.⁹ Delivré's work on the genesis of and precursors to the provincial, and his overview of the fortunes of the provincial are excellent, in terms of identifying manuscripts and proving that the provincial was first compiled around the year 1200. Florian Mazel, on the other hand, was less interested in the provincial per se, and more focused on whether and how it contributed to his arguments as to how the medieval Church pioneered the creation of ‘space’ (identifying social identities or political authority with particular territory). For Mazel, the provincial represented the theocratic papacy's new territorial understanding of Christendom; the imperial papacy saw itself at the head of the Church and the World: ‘not only … a mental image of the orbis christianorum ruled by Rome, but also … a true instrument of government at the service of the pope and the Curia, as evidenced by its [the provincial's] copying into a series of administrative tools with diplomatic or fiscal purposes’.¹⁰ The provincial is, for Mazel, a conceptual and practical tool of papal monarchy, part of the muscular, activist, post-Gregorian papacy's desire to govern the world from Rome.

There is value to that view – the view that the purpose of the provincial was to make Christendom ‘legible’ to the pope; a sort of James Scott avant le lettre.¹¹ Certainly, I am willing to believe that such could have been the intention of the original compiler of the provincial at the papal court, in the years before 1200.

Let us say, however, that the intent of the provincial was indeed to lay out all the bishops and cities of Christendom and to emphasise Roman centrality and papal supremacy, to make the world legible to the City. Even if that is the case, papal control of that image vanished as soon as the first copy of the provincial left the Curia. Copyists outside the Roman court – and most manuscripts of the provincial must have been copied outside the immediate papal sphere – could change and alter the picture presented by the provincial to fit their own views of what the world should look like.

Manuscripts of the provincial are found everywhere: royal courts, episcopal and archiepiscopal administrations, lawyers’ handbooks, works of theology, in monasteries, and in chronicles. Most of these copyists were not papal scribes. As soon as the first visitor to the papal court copied the first provincial, and carried it outside Rome, and had it copied again, the papal court lost control of the spread of the provincial. And, as well as losing control of the spread of the provincial, the Curia lost control of what the provincial said. When a scribe in England or France made a copy of the Roman provincial, they had the ability to alter the political and ecclesiastical geography of Europe to bring it into line with what they thought it ought to be.

The provincial is arguably interesting as a tool of papal power, but the many different manuscripts of the provincial are also interesting because they tell us about ‘contested cartographies’: battles over borders, and differing ideas about what precisely is inside a particular political or ecclesiastical unit. A provincial copied in one place can have a very different idea of what Europe looks like compared to a contemporary provincial copied somewhere else.

This methodology – of ‘Cartographic Struggle’ rather than ‘seeing like a state’ (to put it crudely) – is fairly easy to understand: lists, like maps are not objective statements of the world but political claims, ways of acting upon and changing the world, at least partially constitutive of political realities.¹² An article in the Washington Post in 2020 drew attention – apparently with surprise – to Google Maps’ policy of, for example, altering the borders of Kashmir depending on whether one accesses Google Maps from within India or not.¹³ The scholarship on cartographic struggle over the last thirty or so years has offered fascinating insights into how modern societies use maps as weapons; we would be foolish to assume that pre-modern societies did not do the same with both lists and maps.¹⁴ Mapping and ‘counter-mapping’ – the ‘appropriation of the cartographic discourse of a dominant power to oppose and challenge its views on territories’ – are present within the provincial manuscripts.¹⁵ This dynamic – conflicts over what the world is supposed to look like, played out through different manuscripts of the Roman provincial – is to me a much more profitable way of approaching the provincial than purely as a tool of papal authority.

This approach also frees us from aiming to construct stemmatic diagrams, or to reconstruct the ‘original’ ur-Text of the provincial.¹⁶ While it is useful still to identify provincial manuscripts which seem to share a common exemplar, a full stemmatic diagram is probably impossible; I intend to move forward by studying different manuscripts and attempting to explain their differing worldviews.

It should be admitted that the two viewpoints – cartographical struggle and seeing like a state – are not necessarily opposed. If one accepts that different copyists could change the picture of the world presented in the provincial to bring it into line with the world as they thought it ought to be, then logically the copyists at the papal Curia could do exactly that, and use the text to curate an image of the centrality of papal monarchy.¹⁷ Some provincials, Roman provincials in the geographical rather than nominative sense, must indeed present the papacy as the centre of world and ‘papal authority radiat[ing] outward, urbi et orbi, over all the provinces and bishoprics of Christendom’.

What, however, does taking a more localist view – believing that local copyists could simply change what the provincial said – do to our view of the supposedly monolithic ‘papal monarchy’? Really all it does is bring it into line with how most scholars of the papacy see the period formerly known as the ‘papal monarchy’. The pope was not, and never could be, an absolute ruler. Papal letters and documents were issued at the initiative of petitioners. As soon as a papal missive left Rome, its realisation was left up to those in the provinces. Papal instructions could be ignored, or enforced in selective ways, as local agents wished.¹⁸ The pope could do nothing to prevent it. As with the papal monarchy in general, so with the provincial: local agency matters most.

The sources

I do not know how many manuscripts of the provincial are still extant: certainly more than a hundred, probably hundreds. The earliest manuscripts survive from the early thirteenth century – soon after the list was first compiled – and the latest from the late fifteenth century. Without a full catalogue of all the surviving manuscripts – a vast task, manageable only by a well-funded team working for years – I can only speak about those I have seen, around fifty or so manuscripts. Fifty seems to me to be a significantly large sample from which to draw some conclusions about the provincial, as I have done in this article.

Large numbers of provincials survive in the Bibliothèque nationale de France [BnF], the British Library [BL] and the Bibliotheca Apostolica Vaticana [BAV]. Smaller numbers exist in various other archives, such as the Staatsbibliothek in Munich. Then there are numerous archives with one copy of the provincial: St John's College, Cambridge, for example. The provincial was certainly copied as far northwest as Scotland, and probably as far east as Poland.¹⁹ The nature of medieval manuscript survival means that, unsurprisingly, more fifteenth-century provincials seem to survive than fourteenth-century provincials, and likewise for fourteenth- and thirteenth-century provincials, although this is only my impression.

In general, provincials appear in manuscripts bound up with other texts. Provincials are not infrequently found in manuscripts of Martin of Troppau's Chronicle of the Popes and Emperors, presumably because both aspired to some sort of universal scope.²⁰ Worcester, Cathedral Library, F 38 contains a provincial as well as texts on pastoral care (e.g. the Speculum iuniorum) and Buonaguida of Arezzo's Summa on dispensations and privileges. Hereford, Cathedral Library, P.VII.3 copies a provincial into a manuscript which also contains glosses and commentary on Gratian. The register of Andrea Sapiti, a fourteenth-century proctor at the papal Curia, contains a provincial.²¹ Formularies of the various departments of the papal Curia often contained a provincial, especially the chancery, the chamber and the penitentiary.²² Occasionally, however, one finds a provincial in a stand-alone libellus, such as Cambridge, St John's College, G.9.

As well as appearing as a text in itself, medieval writers used and mentioned the provincial in other works, as we have already seen with Gerald of Wales. Gervase of Tilbury mined it for geographical information for his Otia Imperialia (s.xiiiⁱⁿ).²³ Alexander Minorita found it a helpful resource when he wrote his commentary on the Apocalypse (c. 1235–40s).²⁴ Jean de Saint-Victor used his house's provincial manuscripts when compiling the Memoriale historiarum (s.xiv¹).²⁵ In 1417, at the council of Constance, the English and French delegations argued over the voting rights of the English, whether England was a true ‘nation’. Both sides drew on the provincial to buttress their case.²⁶ At the very end of the Middle Ages, Mercator used it for his Atlas.²⁷ These uses of the provincial – in many different texts, for many different purposes – all potentially demonstrate both the ubiquity and authoritativeness of the provincial. These writers all had access to a text (or multiple texts) of the provincial, and they all thought it was sufficiently ‘right’ to be mined for information to support their arguments.

Interestingly, Latin was not the only language of the provincial: Munich, Bayerische Staatsbibliothek Cgm 1112 (dating from c. 1475) is in German; and BL, Egerton 1500 (c. 1321–4) contains a Provençal or Occitan version of the provincial. Based on an analysis of the other texts in the Egerton codex, Catherine Léglu and Alexander Ibarz have plausibly suggested that Egerton 1500 was produced for ‘an Occitan-speaking court with anti-English, pro-French interests, that was linked by marriage to the family of Robert of Naples’.²⁸ This provincial then was copied for a lay audience, not for churchmen. On Léglu's point of having anti-English interests, it is worth noting that the Occitan translation of the list of kings who have the privilege of being crowned and anointed (a list very often found in the provincial) misses out the king of England completely; normally the English king is one of the four kings who can be crowned and anointed, along with Sicily, France and Jerusalem.²⁹ Perhaps this was an intentional change to downgrade English royalty.

The diversity of contexts for surviving provincial manuscripts makes it hard to generalise about who copied them and for whom. Secular rulers definitely kept copies, most obviously the thirteenth-century Capetian kings of France, as I will show later in this article, and the court affiliated to Robert of Naples which produced Egerton 1500. The different organs of the papal Curia kept many copies, that is clear, and those copyists must have been papal scribes and notaries (or, in some cases, freelancers working at the Curia). Many – perhaps most, at least most surviving – copies probably come from ecclesiastical institutions: abbeys, monasteries and episcopal chanceries, copied with chronicles, formularies or other texts. Whether that is because medieval ecclesiastical archives in general tend to have a better survival rate than secular archives is uncertain.

Does the creation of the provincial around the year 1200 attest to a new desire to list the world at that time? Certainly, it does seem that listing was a particular interest: the proliferation of estate surveys from the twelfth and thirteenth centuries attests to that. Tom Bisson's argument in The Crisis of the Twelfth Century that there was a growth in accountancy and accountability from the later twelfth century has a corollary that running accounts must have been kept to enable ‘true’ accounting.³⁰ Bisson's argument, however, does not really apply to the provincial. The provincial is not a new tool, but an improvement on existing lists; the provincial text composed around the year 1200 is not the first attempt to list all the bishops in Christendom. Fabrice Delivré has explored the immediate precursor to the Roman provincial, the ‘Florence List’. The Florence List, in turn, made use of the Notitia Galliarum (composed around 400) as well as other texts.³¹ Thus the provincial is not a new paradigm but essentially an improved version of similar lists which had already been being compiled for centuries. Rather than being part of sudden desire to list the world, the provincial exists as the latest in a long series of attempts to codify the bishops and cities of Christendom. Florian Mazel would, I think, point out that the creation of a new list at this time – even if not fundamentally different in conception from existing lists of bishops and cities – is itself worthy of comment, attesting to a desire or need at this time for the papacy to ‘create’ space; the provincial is, for Mazel, part of the territorialisation and imperialisation of the medieval Church and papacy which occurred following the Investiture Contest.³²

Battles in the manuscripts

If, as I argued above, cartographical struggle rather than ‘seeing like a state’ is the more profitable approach, then the next question is: can we actually see and explain any cartographic struggles in our provincial manuscripts? Without wanting to multiply examples to the point of tedium, I will spend most of the rest of this article going through some case studies where I think one can see precisely such struggles and differences.³³ First we will look at how France is portrayed in thirteenth-century provincials; then we will turn to Scotland across the entire later Middle Ages; then the development of how Iberia and Brittany were described; and finally the view of Christendom offered by an Eastern provincial, from Wrocław. These sections, though on the face of it unconnected, illustrate the different ways in which the world in the provincial could change: the rubrics and layout could be intentionally changed, with the aim of advancing political claims. The provincial could be accidentally changed, to bring a real but surprising aberration into line with perceived normalcy. Over time, the descriptions of parts of the world might become more developed and detailed. And finally (albeit perhaps obviously), provincials written in one part of the world might display considerable interest and knowledge about their locality, but utter ignorance about a place far away about which the copyist knew nothing. That trend probably even afflicted the papal Curia itself – we err in assuming that the pope really knew chapter and verse about his most distant dioceses.

Francia in the thirteenth century: royal politics

Several of our earliest provincials survive in the milieu of the Capetian royal court. In the registers of Philip II Augustus (r.1190–1223) there is a provincial, probably copied around 1220. The editors of Philip's registers believed that the provincial in BnF Latin 6191 (also copied in 1220) was copied from a similar exemplar to the version in Philip's registers.³⁴ The provincial in BnF Latin 6191 bears a number of similarities to a provincial copied into BL Additional MS. 34254. The most obvious similarity is that, halfway through the provincial proper (the list of all the dioceses in Christendom), BnF Latin 6191 includes a list of which English dioceses are contiguous with each other. BL Additional 34254 also includes this list after the provincial. So, for example: Notandum est quod Cantuariensis diocesis conterminatur Roffen’, Cicestren’, Lindonien’ (sic) – ‘Note that the diocese of Canterbury borders Rochester, Chichester and London’. The order of dioceses is the same in both BnF Latin 6191 and BL Additional 34254: Canterbury; Rochester; London; Chichester; Winchester; Bath; Exeter; Salisbury; Worcester; Hereford; Chester; York; Durham; Carlisle; Norwich; Ely – BnF Latin 6191 misses out Worcester and Durham, however. This list is pretty uncommon – the only other provincial manuscript I have seen which includes it is BnF Latin 4910, a fourteenth-century manuscript which I suspect is copied from the same exemplar as BnF Latin 6191.³⁵ Consequently, I suspect that BnF Latin 6191, BL Additional 34254 and the provincial in Philip's registers all share a fairly close common exemplar (as probably does the later BnF Latin 4910).

They are also close to each other in time. BnF Latin 6191 and the provincial in Philip's register date from 1220. The provincial in BL Additional 34254 was probably copied in c. 1221–2. The same hand that copied the provincial has also made a list of all the previous bishops of every English see by name. Although later hands have updated the list, the original hand appears to name the bishops down to 1221–2 (Eustace de Fauconberg, r.1221–8, is listed as the current bishop of London, and Ranulf of Warham, r.1217–22, as current bishop of Chichester), hence a date of around 1221–2 seems likely.³⁶ The interest, in this manuscript, in English ecclesiastical affairs (listing the past bishops of English dioceses) suggests that this manuscript was copied in England by an English scribe, although I would not care to be any more specific than that.³⁷

So, we have here three provincials, probably sharing a fairly close common exemplar, copied fairly close in time, but two originating from the Capetian court and one from England. The closeness of timing and transmission makes the difference between them – or between their conceptions of European geopolitics – all the more jarring.

In BnF Latin 6191, copied for the Capetian chancery, there is a heading In Francia – ‘In France’ – and then are listed continuously the archiepiscopal cities of Lyons, Sens, Reims, Rouen, Tours, Bourges and Bordeaux, and all the cities of the suffragan bishops of those archdioceses. Auch and Narbonne are listed separately – under the heading ‘In Gascony’ – but otherwise Francia in this manuscript looks fairly similar to l'hexagon, modern-day France. The provincial in Philip Augustus's registers takes an even more expansive view of Francia.³⁸ Under the heading In Francia comes every archdiocese from Lyons to Narbonne (and hence all of their suffragan cities and bishops). BL Additional 34254 does not quite share this interpretation. To the copyist of BL Additional 34254 – probably based in England – only Lyons, Sens and Reims (and their suffragan cities) were In Francia. Rouen and Tours (and their suffragans) were ‘In Normandy’; Bordeaux, Bourges and Auch were ‘In Aquitaine’, and Narbonne was In Gothia – the lands in the south also known as Septimania or Languedoc.

There may be any number of reasons why such variation in rubrics occurred. A fairly obvious one springs to mind, however. Prior to 1202–5, the king of England had also been duke of Normandy, duke of Aquitaine and count of Touraine and Poitou. Philip Augustus of France, however, seized Normandy, Touraine, and Anjou and Poitou north of the Loire, and both Philip and his successor Louis VIII (r.1223–6) had designs on Aquitaine. The king of England would be reduced to a rump duchy of Gascony.³⁹

Around 1220, therefore, the Capetian court preferred to obliterate the existence of Normandy as a separate political unit, and Aquitaine was likewise either to be erased or reduced down merely to ‘Gascony’. Those territories which had been under the authority of the English kings and their sons – broadly corresponding to the ecclesiastical provinces of Rouen (Normandy), Tours (Touraine and Brittany) and Bordeaux (Aquitaine/Gascony), most obviously – were now merely part of Francia. The copyist of our English provincial, however, preferred an older view, where Normandy and greater Aquitaine were still distinct from Francia; perhaps one day to be reclaimed.

And it is not simply a case of the Capetian provincials reflecting reality and the English provincial looking back to a lost past. The Capetian provincials put the Poitevan bishoprics – Poitiers, Saintes, Angoulême, Périgueux (all suffragans of Bordeaux) – unequivocally in Francia, whereas in BL Additional 34254 they were ‘In Aquitaine’ – the king of England's duchy. A few years after these provincials were written, in 1224, Louis VIII made war on Poitou, capturing La Rochelle and receiving the allegiance of most of the county.⁴⁰ Louis realised the claim made in the Capetian provincials in 1220: that Poitou was part of Francia, not Aquitaine; part of his domain, not the king of England's.

The political conflicts of the early thirteenth century were played out through diplomacy and on the battlefield, but also through the Roman provincial. Did Normandy and Aquitaine still ‘exist’ or was Francia the prime political unit of Western Europe? Of what precisely ‘France’ – Francia – consisted varied, depending on which provincial one looked at.

At this stage, we might ask whether the headings in the provincial manuscripts actually mattered to the kings of France and England. Fortunately, Florian Mazel has found evidence that they did, at least later in the thirteenth century.⁴¹ In 1297, the King of France, Philip IV (r.1285–1314), was seeking to incorporate the city of Lyons within his sphere of influence. To do this, he had to find justifications and proofs that the city had always been under the authority of the French king. One of his scribes hit upon the evidence of the Roman provincial. Around August 1297 the French administration sent a list of proofs that Lyons was part of the kingdom of France to Pope Boniface VIII. One item stated:

It is found in the register of the Roman Court – a transcript of which is said to be among the registers of the French court – that the archdiocese of Lyons is in the kingdom of France, and is enumerated among the archdioceses of the kingdom of France.⁴²

Since the Roman provincial said that the archbishop of Lyons was in France, then the whole city of Lyons must be too. Philip IV appealed to the authority of the Roman provincial when seeking to expand the royal reach. The rubrics of the provincial mattered, and could be interpreted as reflecting geopolitical and regnal divisions, as well as ecclesiastical ones.

The Scottish Church in the provincial: Church politics

BL Additional 34254, BnF Latin 6191 and the provincial in Philip Augustus's registers are remarkable for another reason: their presentation of the Scottish Church, which all of them get ‘wrong’, at least as far as historians of medieval Scotland would probably judge. Those familiar with the Scottish Church will know that, between c. 1192 and 1472, most of the Scottish dioceses were anomalous within the universal Church: St Andrews, Glasgow, Caithness, Dunkeld, Dunblane, Brechin, Aberdeen, Moray, Ross and possibly Argyll and Whithorn too (de facto) were exempt dioceses. They did not owe obedience to any superior other than the pope. This unusual ecclesiastical exemption was granted to them by the pope at the end of the twelfth century in a privilege, Cum universi. By this they were freed from the archbishop of York, who claimed jurisdiction over the Scottish dioceses.⁴³ In 1472 St Andrews was raised to the rank of archdiocese, followed closely by Glasgow, and the structure of the Scottish Church fell into line with that elsewhere in Christendom: bishop – metropolitan archbishop – pope.⁴⁴

Some early provincials get this right. BS, Msc. Can. 91, for example, lists the eleven Scottish dioceses, and next to each one writes qui est domini pape – ‘this belongs to the lord pope’, that is to say, these dioceses were exempt. The Capetian provincials take a different tack, however. Both claim that Argyll is exempt, but none of the other Scottish dioceses. Instead, St Andrews is raised to the rank of an archdiocese and the other Scottish bishops are listed as suffragans of St Andrews. This might be the moment to reiterate that these provincials were copied in c. 1220, some 252 years before St Andrews was actually made an archbishopric. It is possible that this was a knowing and intentional change by a copyist who was supportive of St Andrews’ claims to be an archdiocese, but it is equally likely that our copyist (unaware of the complexities of the Scottish situation) was simply tidying up, bringing Scotland into line with what they knew the ecclesiastical hierarchy ought to look like.

BL Additional 34254 is more subtle. If we turn back to the geographical order of the provincial (which I gave in the introduction), we see that almost every provincial manuscript lists England, then Scandinavia (Denmark – Norway – Sweden), then Scotland. BL Additional 34254 has – uniquely – moved Scotland to immediately after England. Thus BL Additional 34254 reads:

As with the Capetian provincials, BL Additional 34254 claims that only Argyll is exempt. Unlike the Capetian provincials, however, BL Additional 34254 does not make St Andrews an archdiocese. The logical conclusion to draw, if one follows the internal organisation of the provincial, is that Argyll is exempt but all the other Scottish dioceses are subject to the most recent preceding archbishop, since the heading for that archdiocese would read: ‘The archbishop of X has these suffragans’ and the Scottish bishops would follow underneath. Normally, that would mean the Scottish bishops were subject to the archbishop of Uppsala in Sweden, since in most provincial manuscripts Scotland immediately follows Sweden. But, because the copyist of BL Additional 34254 has moved the Scottish bishops to come immediately after England, this has the effect of making all the Scottish bishops (bar Argyll) suffragan to York. The list, as given above, seems unequivocal: ‘the archbishop of York has these’, then Durham and Carlisle; then the heading ‘In Scotland’ to show we are moving into a different political unit, a different kingdom; then the ten Scottish bishops, all implicitly dependent on York. The copyist of BL Additional 34254 has undone the terms of Cum universi and made the Scottish Church once more subject to York. They have not, however, claimed that Scotland is subject to England: the heading In Scotia makes it clear that it is a separate political unit, but not a separate ecclesiastical unit.

Admittedly, in the light of Gerald of Wales's testimony above, we should note that the English suffragan dioceses of York are given in the accusative and the Scottish dioceses in the nominative – might this have given a clever advocate of Scottish ecclesiastical independence a line of argument for their position? Potentially yes: grammatically the Scottish dioceses do not agree with the hos … [suffrageneos] subject to York. They might not be covered. Perhaps the copyist neglected to change the Scottish dioceses from the nominative to the accusative when they moved the Scottish bishops (which was rather careless). Nonetheless, I do think that here, in BL Additional 34254, there has been an attempt to rewrite the ecclesiastical history of Scotland and make York (doubtless rightfully, in our copyist's eyes) into the metropolitan archbishop of Scotland.

By the fifteenth century at the latest, Scottish copyists of the provincial were happy to play the English at this game. Cambridge, Corpus Christi College [CCCC], 171A is the first volume of the manuscript of Walter Bower's Scotichronicon, written in Inchcolm in the 1440s.⁴⁵ As well as the Scotichronicon – the chronicle of the Scots – CCCC 171A contains a provincial. Originally, this provincial, like most provincials, listed the Scottish dioceses as exempt. However, a later hand has noted the 1472 elevation of St Andrews to an archdiocese and included the dioceses of the Orkneys and the Isles as ‘In Scotland’ and under St Andrews. Even more interesting, however, is a comment in the original hand regarding the English dioceses.

The copyist here decided to assert that either the territory of the bishop of Carlisle, or the bishop of Carlisle himself, ought to be subject to Glasgow. This was a claim in keeping with the rest of the Scotichronicon. Book XI of the Scotichronicon, in its account of the pleading of Scottish proctors before Pope Boniface VIII in 1301 that Scotland was not subject to England, noted that:

when King David of Scotland [David I, r.1124–53] died at Carlisle in peaceful possession of the counties of Cumbria, Northumbria and Westmorland … King Henry of England [Henry II, r.1154–89] violently invaded and occupied the said counties … and on his own authority, as it seems, erected a cathedral church at Carlisle, a place which previously had been of the diocese of Glasgow …⁴⁶

The claims of Glasgow over Carlisle had a history older than the Scotichronicon. The chronicle of Lanercost in its entry for 1258, recounted the attempts of John de Cheam, Bishop of Glasgow, to expand Glasgow's authority over Carlisle in the 1260s. The parts of the Lanercost Chronicle up to 1297 are believed to be based on a now lost Franciscan chronicle (subsequently added to by a canon of Lanercost).⁴⁷

John de Cheam succeeded Bishop William of Bondington in the Church of Glasgow, having been collated by the pope and consecrated in the curia; [he was] born in southern England, but greatly hostile to England, for in his last days, with increasing cupidity, he was claiming an ancient right in Westmorland, in prejudice to the church of Carlisle, saying that up to the Rere Cross in Stainmore pertained to his diocese …⁴⁸

The Rere Cross, or Rey Cross, is a stone cross between Barnard Castle and Penrith, close to the borders between Cumbria, North Yorkshire and County Durham. If Glasgow's episcopal authority stretched to the Rey Cross, then a huge swathe of northwest England would fall under the ecclesiastical jurisdiction of the bishop of Glasgow. Probably not coincidentally, this area, Cumbria and Westmorland, was an occasional claim of the kings of Scotland. David I had ruled Cumbria, and even exercised authority over a ‘a greater Scoto-Northumbrian realm extending to the Ribble and the Tyne’, although Henry II then regained control of northern England in the second half of the twelfth century.⁴⁹ Alexander II (r.1214–49) seems to have sought rule of the same counties when he joined the baronial revolt against King John in 1215–17. The twenty-five barons of the Magna Carta security clause recognised Alexander's claims to the northern counties, and Alexander was apparently invested with Northumberland. Practically, Alexander exercised considerable power in the north, especially in Cumbria, until driven out by King John.⁵⁰ Scottish royal hegemony over northern England was far from unimaginable.

The legacy of these claims to northwest England is written into the Scotichronicon's provincial. The claim of the provincial is not too extreme – Carlisle ‘ought to be subject to the bishop of Glasgow’ – but it reflects a long-standing belief that the diocese of Carlisle rightly belonged to Glasgow, and perhaps also reflected the Scottish kings’ claims to Northumberland, Cumbria and Westmorland. Again, it was apparently thought that the provincial was the place to record those claims; the authoritative text for the political and ecclesiastical geographies and jurisdictions of Europe. What was woven through the text of the provincial was woven through the lands of Christendom.

Spain (and Brittany) over the centuries

When we turn our attention to the rubrics used for the Iberian peninsula, we see not so much conflict between competing interpretations, but more development over time. The early provincials, from the first half or so of the thirteenth century, tended not to separate out the Iberian kingdoms. Until the fifteenth century, there were a number of Christian kingdoms on the peninsula: Portugal; Castile-León (divided 1157; reunited 1230); Aragon-Catalonia and Navarre. Of course, there were also the Islamic states in the south, but these were not included in the Roman provincial. During the first half of the thirteenth century, there were also four ecclesiastical provinces in Iberia: the archbishop of Braga in Portugal; Tarragona in Aragon-Catalonia; and Toledo and Compostela in Castile-León. Seville was reconquered in 1248, and in the early fourteenth century Zaragoza was also raised to the level of metropolitan archbishop by John XXII.⁵¹ The early thirteenth-century provincials do not really bother to note which dioceses and archdioceses were in which kingdom; the regnal divisions of Iberia did not seem to concern our copyists.

BL Additional 34254 gave all the Iberian dioceses under the headings ‘In Catalonia’ (for the province of Tarragona) and ‘In Spain (Hispania)’ for the provinces of Braga, Compostela and Toledo. BnF Latin 6191 simply listed Tarragona, Toledo, Braga and Compostela as In Hispania; and BnF Latin 5011 also put all four ecclesiastical provinces under Hispania. Ditto for BnF Latin 8874. Paris, Bibliothèque Sainte-Geneviève, 792 – which probably dates to the first half of the thirteenth century – noted that Tarragona was In Hyspania, but Toledo, Braga and Compostela were in Tolleranus, presumably meaning Tolletanus – Toledo.⁵² This peculiar division might have been intended to emphasise the claims of Toledo to be primate of Spain, and hence holding jurisdiction even over the archbishops of Braga and Compostela.⁵³ Hispania was hardly an unknown term in the thirteenth century: the Visigoths had called their kingdom Hispania.⁵⁴ By the eleventh century Gregory VII could address Alfonso VI of Castile-León as ‘King of the Spains’ (rex Hispaniarum), presumably reflecting both the break-up of the peninsula into multiple kingdoms – multiple ‘Spains’ – and the fact that Alfonso dominated all of them.⁵⁵ Innocent III's chancery in 1210 sent a privilege to Rodrigo Jiménez de Rada, Archbishop of Toledo and ‘primate of the Spains’ (Hyspaniarium primas), confirming his primacy ‘through the kingdoms of the Spains’ (per Hyspaniarium regna).⁵⁶ Whether of course this ‘Spain’, or these ‘Spains’, included the Islamic states in the south is open to question. Nonetheless, Hispania, or Hispaniae, seems to have been a term covering the Christian Iberian realms in general. Initially therefore, in the thirteenth-century provincials, it appears that there was very little attempt to map the specific Iberian regnal divisions onto the Spanish and Portuguese Church, and into the provincial.

A century later that had changed. The provincial in Venice, Bibliotheca Nazionale Marciana, Lat. Z. 399 (=1610) – part of Paolino Veneto's Compendium gestarum rerum regnorumque originem, illustrated in Avignon and probably dating from between 1321 and 1326 – offers a rather more complicated picture of Iberia.⁵⁷ First of all, there are multiple rubrics: the archdioceses of Tarragona, Zaragoza and, oddly, Toledo are In regno Aragonie et Cathalonie – ‘In the kingdom of Aragon and Catalonia’. Compostela is In regno Castelle et Legionis – ‘In the kingdom of Castile and León’ – and Braga and Seville are In regno Portugallie – ‘in the kingdom of Portugal’. Even below those general rubrics, there is more specificity. Several of Zaragoza's suffragan dioceses are noted as being isti erant Tarragon’ – ‘these belonged to Tarragona’; that is, they were subject to the archbishop of Tarragona until the elevation of Zaragoza to archdiocesan status. Among those bishops subject to Braga, five – Coimbra, Viseu, Ourense, Lamego and Tui – all have noted next to them In regno Legionis – ‘in the kingdom of León’. Although the accuracy of all this detail is questionable (was Coimbra really in León? was Seville really in Portugal?), there has clearly been an attempt here to indicate the regnal divisions of fourteenth-century Iberia.

BAV, Ott. Lat. 333, dating to around 1346, makes a slightly better fist of the Iberian peninsula.⁵⁸ Tarragona and Zaragoza are ‘In Aragon and Catalonia’; Toledo is ‘In Spain’ (In Yspania); Compostela is ‘In the kingdom of Castile and León’ and Braga and Seville are ‘In the kingdom of Portugal’. Within the provinces, the dioceses of Huesca, Tarazona, Calahorra, Segovia and Segorbe-Albarracín were seemingly noted as ‘In the kingdom of Navarre’; Evora is ‘In the kingdom of Portugal’, as are Lisbon and Guarda; while Lamego, Tui, Lugo, Astorga and Mondoñedo are all ‘in the kingdom of León’. There are still quibbles (surely Huesca and Tarazona were actually in Aragon-Catalonia?) but there is a pretty clear recognition of the peculiar reality of peninsula politics; ecclesiastical provinces did not match up fully with regnal polities; the suffragan bishops of an archbishop in one kingdom were not necessarily all in the same kingdom.

A similar process is visible when provincials came to the archdiocese of Tours and the duchy of Brittany. Although Tours was – obviously – in Touraine, most of the suffragan bishops of Tours were in Brittany. Fourteenth-century provincials often noted this, but it is clearest in the later fifteenth-century provincial in BAV, Ott Lat 65: Le Mans and Angers are listed as subject to Tours; Saint-Brieuc, Saint-Malo, Dol and Rennes followed, but were noted as In Brittania Galllicana – ‘in Gallic Brittany’. The remaining dioceses – Nantes, Quimper, Vannes, Léon and Tréguier – were ‘in Britannic Brittany’ (in Britania Britonice or in Brittania Brittonizante). The distinction here is a linguistic rather than a political one: Britannic Brittany corresponds to Lower Brittany, where Breton was spoken, while Gallic Brittany is the French-speaking Upper Brittany. Whether Nantes could be said to be in Breton-speaking Brittany is open to question, but otherwise Quimper, Vannes, Léon and Tréguier correspond to the (apparently slightly arbitrary) four modern dialects of Breton (Kernev, Vannetais, Léon and Treger).⁵⁹ Indeed, the modern decision to use dioceses to divide Brittany linguistically seems to have a medieval precedent in the provincial. Fourteenth-century and later provincials sought to map secular political (in Iberia) and linguistic (in Brittany) divisions onto ecclesiastical provinces.

The view from the east

The provincial manuscript from the furthest east (at least, the furthest east known to me) is Wrocław, Wrocław University Library [WUL], R 262. It dates from the first half of the fifteenth century and comes from the Bibliotheca Rehdigeriana, the library collected by the Wrocław native and bibliophile Thomas Rehdiger in the sixteenth century.⁶⁰ The picture which WUL R 262 presents is the reverse of that often found in Western provincial manuscripts: WUL R 262 has a pretty good idea of what the ecclesiastical province of Gniezno should look like, but a slightly more questionable view of what, for example, the ecclesiastical province of Canterbury looks like.

WUL R 262 is very nearly the only provincial I have seen which even has the correct number of dioceses under the archbishop of Gniezno, six: Cracow; Poznań; Wrocław; Płock; Włocławek; Lebus. As a comparison, BL Additional 34254, the English provincial from c. 1221–2, lists nine bishops under Gniezno, as does the mid-fourteenth-century provincial in the registers of Pope Clement VI (Vatican City, Archivum Apostolicum Vaticanum [AAV], Reg. Avin. 57, ff. 446r–450v). The extra dioceses are invariably accidental duplicates. BL Additional 34254 has, for example: Włocławek; Wrocław; Lebus; Włocławek (again, as Kuyavia); Płock; Cracow; Poznań; Płock (again, as Mazovia); and either Kamień Pomorski or Włocławek yet again (as Pomerania). Kamień Pomorski is not listed in WUL R 262, probably because (as the copyist of the Polish provincial presumably knew) it was an exempt diocese, not subject to the archbishop of Gniezno at all, and politically Kamień was not part of the kingdom of Poland but more-or-less the diocese of the duchy of Pomerania.⁶¹

The western provincials almost invariably were unaware of the status of Kamień Pomorski as an exempt diocese, with the exception of the provincial of Clement VI. Even Clement's provincial, however, duplicated some of the Polish dioceses: Płock is listed twice (once as Mazovia) and there is a bishop of Pomerania, even though both Włocławek and Kamień Pomorski are listed separately. WUL R 262 is one of the very few provincials I have seen which get Gniezno province right at all.

However, WUL R 262 makes an interesting error in its description of England. The cities of Coventry and Lichfield are listed consecutively but with a note next to them: isti sunt coniuncti et exempti – ‘these are joined and exempt’. Coventry and Lichfield were indeed unified – it is not uncommon for provincials to note dioceses which have been unified – but they certainly were not exempt. I do not know of any source which suggests that the bishop of Coventry and Lichfield was exempt from the jurisdiction of the archbishop of Canterbury. What we have here is simply the inverse of the English provincials which had only a vague idea how many bishops there were in Poland; Polish provincials were unclear on whether all the English bishops were actually subject to Canterbury and York. Indeed, since almost every provincial I have seen gets the number of Polish bishops and cities wrong – including the provincial in the registers of Clement VI and the list of tribute-paying ecclesiastical institutions in the 1192 papal Liber censuum (which might have had the same base list as the c. 1200 provincial) – it seems almost certain that even the Curia did not know how many dioceses there were supposed to be in Poland.⁶² That is quite a radical – though very plausible – inference.⁶³

The conclusion here might seem obvious: provincials were more likely to get their local geography ‘right’ and more distant geography ‘wrong’. But that is an important conclusion; it demonstrates that different manuscripts did vary depending on where they were written. If the provincial from Wrocław corrected the ecclesiastical geography of Poland, what else did the copyist change? If the English provincial totally misunderstood how many bishops there were in Poland, what else did he get wrong? Only further research will tell.

Conclusion

This essay might appear to be a collection of slightly disconnected observations; but these illustrations are intended to show the variation and richness of the provincial manuscripts; no two are the same; the reasons for changes are multiplex; and it is the changes which matter. The analysis I am advocating here is, of course, complicated. It is very difficult to locate the time and place of production for most provincials accurately. This means that trying to work out who might have altered the text – and their motivation to do so – is often speculative. Nonetheless there are cases where we can be fairly confident of the who and the when, which allows us to put forward an argument for the why. And that is more or less how most historians always work.

The provincial was the authoritative text for the geopolitical and ecclesiastical organisation of Christendom between the thirteenth and fifteenth centuries; that was why copyists sought to amend it, to bring it into line with what they thought, or ‘knew’, the world ought to look like. This is one avenue of future research on the provincial, and I believe it to be potentially the most interesting.

However, pursuing this approach also necessitates a holistic approach to the provincial manuscripts: what can the other material in the manuscripts tell us about (1) the copyist and their milieu; (2) what their interests and aims might have been in keeping (and altering) a provincial? The obvious example is Bower and the Scotichronicon, discussed above. We can reasonably link his account in the chronicle proper about Glasgow's claims to Carlisle with the statement in his provincial asserting that Carlisle should be subject to Glasgow.

The number, richness and variation of manuscripts which include a provincial will provide plenty of avenues for comparison of differences between provincials, and evidence for why copyists made changes. The task is a huge one – it seems likely that the surviving number of provincial manuscripts is in triple figures. The potential rewards, however, are significant: an insight into the way that many people in the high Middle Ages saw their world – or how they wanted to see it.