Source Translation: Histories from Tenth-Century Catalonia

Thomas Bisson called Catalonia a land of ‘unheroed pasts’. The deep south of the West Frankish kingdom was a land where, for almost three hundred years, historians like Flodoard, Adhemar, or Thietmar were conspicuous by their absence. Nonetheless, I (and many others; these aren’t new thoughts) think we can see a tenth-century tradition of responding to the immediate past of the whole kingdom, and today I’d like to present three texts on that theme.

The first comes from the Abbasid writer al-Masudi, writing in the mid-tenth century. Whilst in Egypt, al-Masudi got hold of a chronicle of the Frankish kings given to the ruler of al-Andalus al-Hakam a few years earlier. The Latin prototype of this text does not survive, but we do have al-Masudi’s Arabic text:

Gotmar of Girona, Chronicle, as given by al-Masudi, The Meadows of Gold, cap. XXXV.

In a book that fell into my hands in Fustat, in Egypt, in the year 336 [=947], given as a gift by ‘Armaz [i.e. Gotmar], bishop of the city of Girona, one of the cities of the Franks, to al-Hakam in 328 [=940]… heir apparent to his father Abd al-Rahman, currently ruler of al-Andalus…

The first king of the Franks was Clovis, and he was a pagan. His wife Clothildis converted him to Christianity. Then his son Clovis ruled after him, then, after Clovis, his son Dagobert ruled. Then after him his son Clovis ruled, then after him his brother Carloman ruled. Then after him his son Charles ruled. Then his son Pippin ruled.

Then Charles son of Pippin ruled, and his reign lasted twenty-six years. This was in the days when al-Hakan was ruler of al-Andalus. After him, his sons fought and disagreements reached the point that the Franks were destroying themselves on account of them.

Then Louis, son of Charles, ruled. He became master of their realm and ruled for twenty-eight years and four months. He was the one who came to Tortosa and besieged it.

After him, Charles, son of Louis, ruled. He is the one who sent gifts to Muhammad [I of al-Andalus]. Muhammad used to be addressed as ‘imam’. His reign lasted for thirty-nine years and nine months.

Then his son Louis ruled for six years.

Then a leader [qa’id] of the Franks named Qumis [= comes, Lat. ‘count’?] rose up against him. He became king of the Franks,  and he remained in power for eight years. He made peace with the pagans in regard to his country for seven years, with six hundred rotl of gold and six hundred rotl of silver, to be given to them by the ruler of the Franks.

After him, Charles, son of Taqwira [= ‘of Bavaria’], ruled for four years.

Then after him, another Charles ruled, and he remained for thirty-one years and three months.

Then after him, Louis, son of Charles, ruled, and he is king of the Franks to this day, in the year 336 [=947]. He has ruled the realm for ten years according to the information that has reached us.

[For those of you wondering, hey Fraser, you don’t speak Arabic, how’d you translate this, what I did was, I took an English, French, and Catalan translation, ran the original Arabic through Google Translate, and compared all four, feeding any Arabic words I wanted to dig into into a dictionary. Reader, it took ages, and I could have happily put up König’s translation to much the same effect; but I wanted to be sure…)

This short passage has received a lot of attention, and trying to disentangle it is quite complicated. The Merovingians and early Carolingians are extremely garbled. This is unusual in the Frankish world, where knowledge of this history tends to be pretty good because of the authority of early Carolingian histories and the prime role played by Merovingians in the history of many major abbeys; but makes sense in a Spanish March only really brought under the Frankish aegis in the later eighth century or after.

Then, we have the reign of Charlemagne. The idea that his sons became embroiled in civil war might also seem like a confusion, but it’s not unique to the March: no less educated a tenth-century abbot then Adso of Der placed the Bruderkrieg between Charlemagne’s reign and Louis the Pious’. We then follow smoothly from Louis the Pious to Charles the Bald to Louis the Stammerer. Then we have this ‘Qumis’ figure. This would appear to be King Odo, and indeed the Arabic qa’id nicely captures the ambiguity of the Latin dux, an appropriate descriptor for Odo by the mid-tenth century. Odo is succeeded by Charles, son of Taqwira, who appears to be Charles the Fat; succeeded in turn by Charles the Simple and thence Louis IV. Ralph of Burgundy is, as so often, elided.

Some historians have wished that we had more of this text than what al-Masudi gifts. I think this is all there was, and that it’s fairly faithful; and the best way to show this is to move on. (We will consider this more when we deal with the texts as an ensemble later on.)

This comes from the Roda Codez (Madrid, Real Academia de la Historia MS 78), a late tenth/early eleventh century book from Navarre. This collection of historical works includes a genealogy of the Frankish kings. It’s not the longest or most complex genealogical text in the work – unsurprisingly, that honour goes to genealogies of the kings of Pamplona and counts of the neighbouring counties. Nonetheless, you may find some of it quite familiar:

De reges Francorum from The Roda Codez, ed. by José Maria Lagarra, ‘Textos navarros del Códice de Roda’, pp. 61-2.  

Concerning the Kings of the Franks

I: Emperor Charle[magne] reigned for 48 years and 3 months.

II: Louis, son of the same, reigned for 13 years.

III: King Lothar reigned for 2 years.

IV: Charles, brother of the same, reigned for 38 years and 3 months.

V: Louis his son reigned for 6 years.

VI: Carloman reigned for 6 years.

VII: Charles of Bavaria reigned for 4 years.

VIII: King Odo reigned for 10 years.

IX: After his death, Charles reigned 32 years and 3 months. Then we were without a king for 7 years, after which Louis reigned for 17 years. And afterwards Lothar his son reigned.

The first folio of the Roda Codex (source)

If al-Masudi’s text was short, this is even more brief. We’ve lost the Bruderkrieg, anyone before Charlemagne, and all incidental detail other than reign lengths. We’ve also gained Lothar I and a hiatus in place of the reign of Ralph of Burgundy. Let’s leave that fairly bland summary for now, and move onto our final text.

This is a Chronicle of the Kings of the Franks from the Liber iudicum popularis of Bonhom, a judge of Barcelona, around the year 1000. The Liber iudicum popularis was a legal compilation based on the Visigothic law which Bonhom – a figure well-known from charter evidence – composed as a reference guide and update to the older law. The Chronicle here is part of the updated material, framing the law as part of Barcelona’s history, which is a regnal one. The text goes as follows:

Chronicon regum Francorum in Bonhom, Liber iudicum popularis, p. 307

The known years when the lord king Louis took Barcelona. In Era 839 [i.e. A.D. 801], in the reign of the lord emperor Charles, in the 34^th year of his ordination as king, his son King Louis entered into the city of Barcelona, having expelled the whole Saracen population (who were keeping it) therefrom.

The aforesaid Charles reigned 47 years and 3 months.

His child Louis reigned for 24 years.

King Lothar reigned for 2 years.

His brother Charles reigned for 39 years and 3 months.

Louis, son of the same, reigned for 6 years.

King Charles the Great reigned for 7 years.

Charles of Bavaria reigned for 4 years.

King Odo reigned for 10 years.

King Charles reigned for 32 years and 3 months.

After his death, they did not have a king for 8 years.

Afterwards, Charles’ child Louis reigned for 19 years.

After his death, his son Lothar reigned for 32 years and 5 months.

After his death, his son Louis reigned for 1 year and 6 months.

Afterwards there reigned Hugh, who had previously been duke, and who snuck his way into a position of rule, and he reigned for 10 years in the realm of the Franks.

After his death, there reigned his son Robert; and he jailed Charles and his sons, who were from the royal family.

[Later additions:

He [Robert] sat in the realm for 36 years.

Henry, his son, reigned for 29 years.

Philip reigned for 49 years.

Louis his son reigned for 29 years.

Louis the Younger reigned for 44.

Philip his son reigned 20.]

*This text can also be found in a Carcassonne manuscript, BNF MS Lat. 5256, with some abbreviations and a different set of later additions.

Again, we don’t have a huge amount of historical information here, but the list and the regnal dates are enough for us to compose a table which will help get a grasp on these texts as a whole:

Gotmar (length of reign)	Roda Codex (length of reign)	Bonhom (length of reign)
Charlemagne (26 years)	Charlemagne (48 years, 3 months)	Charlemagne (47 years, 3 months)
<Bruderkrieg>
Louis the Pious (28 years)	Louis the Pious (13 years)	Louis the Pious (24 years)
	Lothar I (2 years)	Lothar I (2 years)
Charles the Bald (39 years, 9 months)	Charles the Bald (38 years, 3 months)	Charles the Bald (39 years, 3 months)
Louis the Stammerer (6 years)	Louis the Stammerer (6 years)	Louis the Stammerer (6 years)
Qumis (8 years)		Charles the Great (7 years)
Charles son of Taqwira (4 years)	Charles of Bavaria (4 years)	Charles of Bavaria (4 years)
	Odo (10 years)	Odo (10 years)
Charles the Simple (31 years, 3 months)	Charles the Simple (32 years, 3 months)	Charles the Simple (32 years, 3 months)
	Interregnum (7 years)	Interregnum (8 years)
Louis IV (10 years, up to the present)	Louis IV (17 years)	Louis IV (19 years)
	Lothar (unspecified, up to the present)	Lothar (32 years, 5 months) Louis V (1 year, 6 months) Hugh Capet (10 years) Robert the Pious (up to the present; later addition: 36 years)

Putting this data in tandem like this makes me think that all of them represent the same historiographic tradition. The reign lengths given are either identical, or easily accountable by scribal error or slightly different calculations of when the year starts. They also give mostly the same order, including when that order is eccentric. (This, for the record, is why I think Gotmar’s Chronicle wasn’t much longer than what we have, because I think it is an elaborated version of this genealogical tradition rather than a longer text which al-Masudi has abbreviated.) There are a couple of exceptions. The first is the substantial difference in regnal length for Charlemagne and Louis the Pious. Here I think there could be a couple of explanations. On one hand, Gotmar could have dated Charlemagne’s reign to some point based on when he conquered Spain (where Bonhom’s Chronicle begins even if he doesn’t do that with regnal dates); on the other hand, Louis the Pious’ reign as sub-king of Aquitaine during the reign of Charlemagne could be screwing up the calculations somehow. The second point where the texts deviate from each other somewhat are the placing of Qumis in Gotmar and the addition of a Charles the Great, Karolus Magnus, in Bonhom. Qumis’ identity as Odo I think is fairly clear; Karolus Magnus is, I think, Carloman II (r. 879-884). What seems to be happening is that our authors are, in their various ways, getting confused about the genealogical churn of Carolingian rulers between 877 and 888 (I am reasonably sure that the six-year regnal length for Louis the Stammerer comes at least in part from eliding him and his son Louis III). If so, these guys wouldn’t be the only ones.

If these texts do represent different outflows from a more-or-less coherent river of tradition, what can we say about the ideology here? Well, first of all it’s distinctly southern. This is explicit in Bonhom, who starts his account with the conquest of Barcelona (something shared by the so-called Alaó Memorial, a text on the history of the counts of Pallars I chose not to translate); but if I’m right about the odd regnal dating given to Charlemagne in Gotmar’s text, it’s also true there. It is also indicated by the presence of Lothar I. Years ago, I commented that supporters of King Pippin II of Aquitaine against Charles the Baldseem really to have been attached to Lothar, by whom they dated their charters for years after the 843 Treaty of Verdun which divided up the Carolingian Empire; and that he was remembered as ruling in the March suggests a regional affection which persisted into the tenth century.

Lothar I’s presence also hints at the text’s views on dynasticism. Genealogies are inherently dynastic by virtue of being structured by father-to-son descent, but these authors nonetheless find room for Lothar I and Charles the Fat, as well as outright non-Carolingians like Odo, so an uncomplicated Carolingian descent towards the reigning West Frankish monarch such as the West Frankish king Lothar was propagating at this time is not obvious in play. Nonetheless, even if not dynastic per se, there is a clear sense in which the tenth-century Carolingian kings have become the repository of proper regnal order against the high-political disruptions of the time. Charles the Simple is given regnal dates which encompass his entire reign (his coronation as anti-king in 893 to his death in 929, rather than his uncontested reign 898-923), and 929-936 is given as an interregnum rather than allowing Ralph of Burgundy to be king. This broadly tracks with charter evidence, which shows plenty of Catalans at the time acknowledging Charles as king until his death, and then having a very tenuous relationship with Ralph. Then, Hugh Capet and Robert the Pious (especially the former) are treated very snobbishly. Anti-Capetian sentiment in Catalonia wasn’t as widespread or intense as sometimes portrayed – like a lot of West Frankish opposition to the Capetians, it took the form of foot-dragging or snide comments – but the historical texts here track with charter dating clauses such as that which mentioned Hugh ‘who was duke but assumed the beginning of a reign’.

Unheroed as the Spanish March may have been, then, it nonetheless did have a recognisable historical tradition in the tenth century. Would it be great if we had more of it? Yes! Absolutely! Nonetheless, what we have is enough to identify a distinct, regional, approach to the history of the kingdom and a way of caring about the regnal unit of which the March was a part.

Patrimonialisation

The transformation of office into property is one of *the* big socio-political changes of the Late Carolingian and Post-Carolingian periods. The model in question is roughly this: under Charlemagne, if you want to be a count, the king has to appoint you; but by the time of Philip I, if you want to be a count, you need to be the (legal or biological) heir to a county. This isn’t to say that packing up your kit-bag as a bright-eyed peasant to seek your fortune amidst the bright lights and (erm) big city at Aachen is going to lead anywhere – all of the major High Carolingian aristocrats are aristocrats, people of noble descent. What it does mean is that a count’s son c. 850 can expect to get a county, but one c. 1000 can expect to receive the specific county or set of counties ruled by his father.

We have to distinguish between several different processes here: 1) the simple fact of sons who rule their father’s honores; 2) changes in practice; 3) changes in norms; and 4) how all these things changed with regard to different scales and types of honor. I distinguish between 1) and 2) because the simple existence of sons (or other relatives) ruling the exact same counties as their relatives doesn’t have to imply anything about norms or practices. The Neustrian March is probably the best example here: Robert the Strong ruled it, and then so in turn did his sons Odo and Robert of Neustria. However, all of these people were appointed by the ruling king in their role as close political allies, and their close family relationship tells us nothing about encroaching hereditisation. Equally, I want to chuck in 4) because my interest here lies at the very top of society. Individual estates had always been inheritable, it’s offices, and especially very senior offices, where things really change.

This process remains obscure, though, for a couple of reasons. First, there’s an at-least-in-principle importance of royal involvement into at least the early eleventh century. Even Count Odo II of Blois-Chartres-Tours in the 1020s, as hereditary a magnate as you could ask for, was willing to allow Robert the Pious at least a nominal role in control the transmission of his own hereditary honours. Second, and inversely, under the eighth- and ninth-century Carolingians, a degree of inheritability was common: the county of Paris, inherited in one family from the reign of Charlemagne down to the mid-ninth century, is the example historians usually go to.

We can see the shifting balance between the principle that specific honores could be inherited and the principle that their distribution should be controlled by the king play out at length in the East Frankish/German kingdom over the course of the tenth and eleventh centuries. David Bachrach has recently made an, erm, uncompromisingly written case that comital office never became hereditary, arguing that previous historians have never identified any texts where an Ottonian ruler expresses de iure approval of the principle. This strikes me as having missed the point: presumably what Bachrach is thinking of is something like the 877 Capitulary of Quierzy (we’ll get to it) or, like, the Golden Bull; but what historians aren’t saying that the Ottonians legislated in favour of the direct transmission of honores, but describing a de facto process. Certainly, no-one denies that Ottonian rulers could both appoint people to high office and deprive them of it. One thinks of Lotharingia: Otto the Great appointed Conrad the Red as the region’s duke in the 940s and then stripped him of office in the 950s. What such a bald sentence fails to cover, though, is that stripping him of office was difficult. Conrad’s fall came at the end of a lengthy and multi-year civil war that was the final major crisis of Otto’s reign – it doesn’t stand as evidence that the king could deprive people of honores at will. Equally, there is an observable trend in the Ottonian realm for at least some comital honores to be directly transmitted father-to-son – the example I know best is the county of Holland, transmitted to the descendants of the early tenth century Count Dirk of Holland until the line went extent around 1300. Overall, though, the general picture suggests both elements were firmly in play during the Ottonian period and beyond.

The West Frankish kingdom is, as usual, weird. Heredity is everywhere, including at sub-comital level, from an early point. The presence of clearly defined, inheritable, vicecomital lines such as the Thibaudines from the early tenth century onwards does not, as far as I’m aware, have East Frankish parallels. On the flip side, stripping counts of their office looks like it completely ceases to be thinkable during the tenth century. This makes comital honores much closer to family property, which is rarely stripped even from rebels.

That final point, on the end of removal from office, requires a bit more going into, because I’m sure some of you are thinking ‘well, sure, these kings are just too weak to do that’, and I want to head that sort of argument off at the pass. First, in the strict sense of the term, it begs the question – ‘strong kings don’t let aristocrats inherit honores, therefore kings who let aristocrats inherit honores are not strong kings’ – when counterexamples are trivially easy to produce (basically every Early Modern European state). Second, not being in a position to carry out a threat doesn’t necessarily mean that a king won’t make the threat. During his war against Odo, for instance, Charles the Simple ordered Robert of Neustria stripped of the abbey of Saint-Martin in Tours in favour of Archbishop Fulk of Rheims, a decision which was pure posturing. Equally, the East Frankish king Conrad I, at the 916 Council of Hohenaltheim, ordered that his Alemannian opponent Duke Erchengar go into perpetual monastic exile, implicitly stripping him of all his honores. Actually achieving this was probably optimistic – Conrad had tried and failed to exile him a year earlier. Third, instances where the West Frankish kings enjoy a dominant position over rivals don’t involve the kind of stripping of honores we see in the Ottonian case. The closest is probably the succession to William Longsword at Rouen in 943. Louis spends a fair amount of time in the subsequent years fighting over who gets the usufruct, and, per Dudo, the notion that Louis might have been planning to disinherit him was a politically important idea; but Flodoard makes it clear that Louis’ first action in the aftermath was to give (the verb is dedere, an unambiguous construction) the ‘land of the Northmen’ to William’s young son Richard the Fearless. Then there’s the Synod of Ingelheim in 948: with the backing of an unmatchable Ottonian army, Louis IV and Otto the Great threaten Hugh the Great to come to terms, but they don’t threaten to strip his honores, unlike Erchengar and unlike Otto would do with Conrad the Red a few years later. All of this suggests not that West Frankish kings couldn’t as a matter of fact strip honores, but that as a matter of principle they didn’t.  

A contemporary image of Louis IV and Hugh the Great from the margins of Vatican MS Reg. Lat. 991. I actually want to dwell on this a little more, because this image is underneath the chapter titles for the Lex Bavariorum, specifically the bit that goes ‘If a proud and rebel duke spurns the king’s decrees’. The punishment for this is being stripped of ducal office, and – given the provenance of this manuscript is probably Lotharingian, the association suggests that it was on people’s minds in the Lotharingian circles that were dealing with the problem. That it wasn’t mooted at Ingelheim suggests deliberate acclimatisation to West Frankish political culture on the part of the Ottonian elites.

So what’s the timing here? The answer to this question is also obscure, for a couple of reasons. First, there’s a persistent shibboleth that the 877 Capitulary of Quierzy is the smoking gun which made counties hereditable. Most historians wouldn’t subscribe to that anymore – there are provisions about the inheritance of counties in it, but they’re temporary provisions for while Charles the Bald is away in Italy – but, in general, a tendency to ascribe ‘loss of control over the distribution of honores’ to ‘royal weakness’ leads to a vague ‘late ninth century’ date. We also have another problem, which is less recognised: the generation who came to prominence in the 880s and 890s lived forever. I tried to count up every West Frankish count during Charles the Simple’s reign (there aren’t a lot! There’s like 30-50 at any one time!), and the coincidence in death dates is striking: the map I drew on my pad of paper is covered with dates from 916 to 923 and even later. The genealogical threads of the central medieval aristocracy generally begin with this ‘Charles the Fat’ generation, but men like Odo, Robert of Neustria, Ebalus Manzer, William the Pious, were generally appointed. There was a lot of controversy about some of them, for certain, and a fair amount of self-help; but they look like ninth-century counts, not eleventh-century ones.

Hereditary succession, though, is clearly in place by the time the ‘Charles the Fat’ generation dies, and we even have some discussion of it. In 914, a charter of Robert of Neustria introduces Hugh the Great as heir to all his honores by royal concession; in 918, a dispute settlement charter names Raymond I of Toulouse as count by appointment of his father Odo during Odo’s lifetime; at the death of Baldwin the Bald of Flanders in 918, Folcuin of Saint-Bertin fifty years later describes his sons Arnulf the Great and Adalolf partitioning his honores between them; and around 1000 Dudo of Saint-Quentin claims that the Rollonid duchy in Rouen was always intended to be hereditary by royal grant. In this latter case, we might raise our eyebrows; but so much of Dudo’s narrative involves complicated reasons why the Norman rulers can keep the ‘hereditary’ bit without the ‘obligations to the king’ bit that one gets the definite sense that the royal aspect was generally known and had to be manoeuvred around.

This evidence, then, would seem to suggest a date in the 910s or 920s, but we can probably push it a bit earlier: not everyone was quite as long-lived as Robert of Neustria or Richard the Justiciar, and in Maine and the Spanish March (where the counts die much earlier) insofar as we can make deductions from practice, the direct transmission of honores within the family appears to have been in place from c. 900.

This is interesting, because the question of appointment to and deposition from comital honores was extremely important during the reign of King Odo (888-898). This was true basically everywhere – Odo was able (for obvious reasons) to appoint his brother Robert of Neustria to the position he had vacated, but not to the countship of Poitiers, which was disputed between Robert, Ebalus Manzer, and one Adhemar. Equally, the countship of Maine was disputed between one Gozlin (whom Odo and Robert backed) and one Roger (who may have been backed by Charles the Simple and was certainly married to his aunt). In Clermont, Odo appointed Hugh of Berry to replace William the Pious; again, without success as William was able to defeat Hugh in battle. At Laon (probably), Odo was able to free up the countship by executing Count Walker – drastic, but the most successful short-term strategy… In any case, all of these issues caused Odo immense trouble. The interrelated complaints of the Poitevin and Maine countships were fundamental in allowing Charles the Simple’s rebellion in 893 to gain initial traction. Odo eventually had to capitulate to all of the people who he had tried to dismiss from their countships and buy them off. 

It is perhaps also relevant here that Charles the Simple’s claim to the throne was explicitly about patrilineal inheritance of office. Yes, kingship was supposed to be different to countship; and as Stuart Airlie has argued, the Carolingians expended a lot of effort to make themselves special. This included Charles the Simple and his backers, although as I’ve argued before, it seems clear that this effort was found generally unconvincing. Nonetheless, Charles did stick with it: David Pratt, in a very smart bit of work, has argued that an early version of the Ordo of Seven Forms called the Leiden Ordo (which included all the bits about ‘paternal thrones’ and so on) was probably used at Charles’ coronation. The middle ground between a nobility more angered than usual at attempts to directly control the transmission of honores and a king whose claim to the throne rests on direct inheritance rather than (say) acclamation by the magnates would definitely include the extension of direct inheritance to other offices.

Now, this already fails the Bachrach test, because there isn’t a smoking gun. We know that Charles’ accession was negotiated but we have no texts equivalent to the election of Louis the Blind as king of Provence in 890. However, we know what a slow shift in mentalités looks like, because that is more-or-less what happens in the Ottonian kingdom (as well as elsewhere, such as Provence and Transjurane Burgundy).  The West Frankish case is too complete, quick, and extreme to be anything other than a principled policy decision with a wide base of political support. If so, it was one of the most important decisions in French history: the new comital and vicecomital dynasties would ultimately provide some of the silver threads of continuities around which regional elites, and ultimately regional polities, could coalesce.