This gallery contains 3 photos.
I am travelling this weekend and have time only for a short post, but happily one of the two final posts from my 2021 trip to North Wales is very short, and is therefore presented herewith. This was another bit … Continue reading
Sorry about last week; deadlines, is all I will say, and we’ll see what happens. However, also a factor was that I didn’t really want to write this post. I was telling myself that’s because it would be hard or long, but it’s not really. The actual problem is that it’s one of the posts where I have to admit I was wrong. But you gotta, when you realise; so here goes.
Entrance to the Archivo de la Corona d’Aragón in 2011, where if I’d been able to spend any time in my doctorate some things might have been different
This one goes back to my doctoral research, when while working through the charters from the comital archive of Barcelona that are now in the Archivo de la Corona de Aragón I noticed a man called Unifred turning up a lot, and once with a woman called Riquilda. Lots of women were called Riquilda, but once I also had the charters from Vic in play, I thought I saw more bits of what had been going on and it was pretty interesting. Unifred, or as we might now say it if anyone still used the name, Umfred or Onfret depending on what language you want, was the name of one of the sons of the vicar Sal·la, who founded the monastery of Sant Benet de Bages; FACT.1 Riquilda had been married to someone called Unifred, Unifred Amat in fact, and had a son called Guillem; also FACTS.2 But plot twist! Sal·la’s son Unifred also turns up with a woman called Sesnanda, who seems to have been running his foundation at Òdena for him and after his death, defended her claim to lands there on the basis of him having been her man, vir; also FACTS!3 And I put all the documents I could get together and concluded that what seemed to have happened here was that Unifred son of Sal·la had been married to Riquilda and had Guillem, but had then put Riquilda aside – and in the charter which mentions her and Guillem a priest called Seniol who is elsewhere shown to be her brother apparently felt it necessary to warn her of the dangers of an active libido, so I wondered if she’d been caught sleeping with someone else.4 Then once Unifred and Riquilda were separated, or even before who knows, he had taken up with Sesnanda and passed a lot of his property onto her, but without marrying her, so that her inheritance lay open to challenge. And in that latter phase of his life he seemed to be carrying this extra name, Amat, in the documents Amado, which literally just means "beloved", so I thought his romantic entanglements might actually have brought him a nickname. Not FACT, but a smart reading of limited evidence! And so I put this in the book.5
Remember the book? Still available, kids! Jonathan Jarrett, Rulers and Ruled in Frontier Catalonia, 880-1010: pathways of power (Woodbridge 2010)…
Nonetheless, I should have thought harder about that byname. In the period of these documents the folk of what would become Catalonia didn’t use surnames, which can be extremely frustrating when trying to distinguish them. If they use two names at all, then the second one is either their father’s, by way of identification, or a nickname or byname.6 I knew who Unifred’s father was and that he wasn’t called Amat; so I went for the latter explanation because my theory fitted nicely with it. Even then, I should have remembered that the main reason you ever get these extra names is because someone needed to distinguish their holders from someone else. A few charters ago we met Guitard Beraza, remember; he’s almost certainly recorded thus because he’s not the only Guitard following Count Borrell of Barcelona around at that time.7 So the fact that Unifred Amat had a second name should have made me think there were probably two contemporary Unifreds, and then maybe look at the two relationships and think: two people, two partners? But I had a theory, and the evidence didn’t outright contradict it, so I didn’t think any of these things, and as long as no new evidence came to light no-one was going to be able to say I was wrong. Oh well.
Home of the Amats? Read on. This is Castellví de Rosanes in deep Barcelona, image by Pere López Brosa – own work, licensed under CC BY-SA 4.0 via Wikimedia Commons
So, remember that back in late 2021 I was going through the Barcelona documents in the last volumes of the Catalunya Carolíngia? There’s the new evidence. Unifred Amat is in there and one document identifies him nice and clearly as the son of Sendred, vicar of Castellví de Rosanes.8 So, two people. At that point, of course, the whole thing comes crashing to bits. They’re still quite big bits: Unifred Amat was married to Riquilda, her priestly brother still smack-talks her urges and, note, Amat actually is a surname the family carries, as neither Unifred nor his son Guillem had fathers of that name. Also, Unifred son of Sal·la did apparently have a long but unofficial relationship with a lady castellan of his that nearly left her without means after his death. It’s just that these weren’t the same story. And I know it’s not changing the face of the earth, but I did put it in the book, and rested several blog posts on the story as well, because it was a good story. But my version, at least, was only a story. Sorry, sorry, I’ll try not to do it again…
1. He is identified as Sal·la’s son in the act of consecration of the church of Sant Benet de Bages, printed in Ramon Ordeig i Mata (ed.), Catalunya Carolíngia volum IV: els comtats d’Osona i Manresa, Memòries de la Secció històrico-arqueològico 53 (Barcelona 1999), 3 vols, doc. no. 1127, and in Sesnanda’s court case, mentioned below, ibid. doc. no. 1736, to name but two.
2. All visible in ibid. doc. no. 1564, with the family links set out in the earlier Ignasi J. Baiges i Jardí & Pere Puig i Ustrell (edd.), Catalunya Carolíngia volum VII: el comtat de Barcelona, Memòries de la Secció històrico-arqueològico 110 (Barcelona 2019), 3 vols, doc. no. 688.
3. She makes that claim in ibid. doc. no. 1736, as said above, and previously called Unifred "my lord", senior meus, in ibid. 1133; she otherwise occurs in conjunction with him in ibid. doc. no. 1263 during his life, and posthumously, executing his will, in doc. nos 1276 & 1283.
4. Ibid. doc. no. 1564, where the gift is made, "in such a way, namely, that while you shall live as long as you do not through any libidinous urge join yourself with another man you may hold and possess [this land] and provide for your sons and daughters from it and do as well as you can to do as much better as you can see with it, and after your death indeed let it revert freely and integrally and let them hold and possess it and divide it among themselves equally," in tale videlice racione ut dum ti vixeris si ad ullum ominem ten non coninuxeris per nullam libidinem teneas et possideas et filiis et filias tuas exinde nutrire facias and bene facere ut melius viderius vel potueris, post obitum vero tuum ad illis remaneat liberum vel integrum et illi teneant et possideant et inter se equaliter dividant. He has already specified that he means this gift to go to "your sons and daughters from one father", et filios et filias tuas ex uno patre genitis. I might be over-reading this, of course, but these phrases are unusual; I’ve never seen them elsewhere.
5. Jonathan Jarrett, Rulers and Ruled in Frontier Catalonia, 880‒1010: pathways of power (Woodbridge 2010), pp. 148‒150.
6. See, if you want, Josep Moran i Ocerinjauregui, "L'antroponímia catalana l'any mil" in Imma Ollich i Castanyer (ed.), Actes del Congrés Internacional Gerbert d'Orlhac i el seu temps: Catalunya i Europa a la fi del 1r. mil·leni, Vic-Ripoll, 10-13 de novembre de 1999 (Vic 1999), pp. 515–525, or the older but more accessible Jordi Bolòs, "Onomàstica i poblament a la Catalunya septentrional a l'alta edat mitjana" in Philippe Sénac (ed.), Histoire et archéologie des terres catalanes au moyen âge (Perpignan 1995), pp. 49–69, online here. In English there is the less directly relevant survey of Lluís To Figueras, "Personal naming and structures of kinship in the medieval Spanish peasantry" in George Beech, Monique Bourin and Pascal Chareille (edd.), Personal Names Studies of Medieval Europe: Social Identity and Familial Structures, Studies in Medieval Culture 43 (Kalamazoo MI 2002), pp. 53–66.
7. See, indeed, Jarrett, Rulers and Ruled, pp. 155‒159, for two more.
8. He therefore occurs in Baiges & Puig, Catalunya Carolíngia VII, doc. nos 324, 682, 688 as said above & 711, and also Ramon Ordeig i Mata (ed.), Catalunya Carolíngia volum VIII: els comtats d’Urgell, Cerdanya i Berga, Memòries de la Secció històrico-arqueològico 111 (Barcelona 2020), 2 vols, doc. no. 662. Of these, CC7 324 is a gift from his father in 951, and even his father calls him Unifred Amat.
Posted in Catalonia, Charters, Currently reading...
Tagged Òdena, Castellví de Rosanes, medieval marriage, prosopography, Unifred Amat
I think there are three more of these posts from my tour round North Wales in summer 2021, and this one was going to be another fairly simple photo-dump, until the first bit turned into an evening’s research by itself… So now you get to find out what I found out! This was one of those happenstance discoveries made by following a random sign, a practice I wholeheartedly recommend. So this is the story.
It was only the most incidentally personal history that had led us to Dolwyddelan, and nothing to do with the interests of either blog or blogger to be honest, but once our purpose was more or less accomplished, we found our map rather inadequate and had to pause for a while to try and work out our bearings.1 And when I strode up the road a bit to see if there were signs, I found one saying, “Roman Bridge”. To which I was, as it is classically put, like, "huh", not least because I didn’t think there was really much Roman in this bit of the country. So once we had decided how we were going to get out of where we were, we thought we’d have a look down that way first. And this is what we found.
This is the northwards face, zoomed in tight to show the architecture; it was about the clearest angle one could get without actually being in the river
This is a view from further to the north, showing some of what I mean about photography problems but also the mouth of a railway tunnel behind it, which has in the end been vital in working out where we actually were
Here’s a view along the piers, another struggle to get any camera angle on the structural elements. I couldn’t photograph the southern face at all, which would probably have told me something as, from the pictures in the archæological watching brief, that’s where the 1950s concrete is visible. But I couldn’t see it! And in any case I’m getting ahead of myself…
Just as I don’t like arguments which amount to, "it’s really old, it must be Roman/Muslim/whatever", or indeed ones which assume that the first time we have surviving writing about something is when it dates from, I am also not keen on ones that go, "it was built shoddily, it must be medieval." So I went digging a bit further. Coflein’s entry for the site is disappointingly short, and just gives the 18th-century date without further details; but it does link to a 96-page archæological watching brief document, honestly a bit more than I’d bargained for.3 And it turns out that there is a really obvious reason for that date, which is that according to maps older than 1794 there was a lake here until at least 1701. So 18th-century is really all it can be.4 And in fact it’s been mightily mucked about with since then, including the addition of a concrete core in 1953, and the archæological watching brief arose because the year after we were there and I took these photographs, they put even more concrete in it so it could take heavier vehicles to Blaenaeu Dolwyddelan which lies beyond, most importantly fire engines, which had had to be banned when it started to crack. The bridge is Grade II listed, so I assume it still looks roughly the same, but it’s barely even early modern inside by now.
So what’s the moral here? Perhaps that sometimes it takes archæologists to show so-called "historians" how to do archival work? But since they did, so can I and now so can you. Or perhaps it’s that judging a structure by standing fabric without opening it up is always risky; but again, in this case we can now do better. Quite what your chances are of running into someone from near Dolwyddelan who wants to tell you they have a medieval or even Roman bridge across the Lledr there, I don’t know; but now you’re ready for ’em!
1. We were running off an 2019 AA Road Atlas, which for England and even Scotland was fine, but in Wales’s case just missed out roads, and not just tiny lanes to nowhere but numbered B-roads sometimes (though granted often still quite tiny). Compounded with that, our Garmin satnav has a marked tendency to try to cut corners off A-roads with ‘short-cuts’ which are national speed limited, in theory, but actually hill tracks, and we needed to be able to second-guess her. In the end I bought a Landranger for North Wales out of sheer frustration. The 2023 AA Road Atlas we now have has not filled in all the gaps, either. Is Wales just not thought worth mapping or something?
2. David Evans, "Conwy bridge with medieval origins visibly cracking", in Nation.Cmyru (29th October 2021), online here.
3. Carol Ryan Young, "Pont Sarn Ddu, Dolwyddelan / Sarn Ddu Bridge, Dolwyddelan: Briff Gwylio Archeolegol / Archaeological Watching Brief / Cofnodi Adeiladau / Building Recording", Yr Amgylchedd Hanesyddol yn Cofnodi Prif Gyfeirnod / Historic Environment Record Event Primary Reference Number 46197, Prosiect Rhif / Project No. G2702, Adroddiad Rhif / Report No. 1610 (March 2021), online here.
4. Ibid. p. 10, arguing directly with the Cadw Historic Buildings listing. That apparently uses the same argument as reported by Evans, "Conwy bridge", suggesting he may at least have gone as far as looking up the listing, and to be fair, I’d have trusted that too.
Posted in archaeology, Celts
Tagged architecture, bridges, conservation, Dolywddelan, Medieval tourism pictures, Wales
I’m still on the road this weekend so will have to ask your forgiveness for another post written through the app, which I am definitely still learning. Anyway, the prompt for this one was reading for my 2022 piece on the numismatic term follis, which had me briefly looking through papyrological evidence for how people actually used the word, and this brought me to a very interesting paper by the late Leslie MacCoull, doing things with the numbers in a sixth-century Egyptian tax record.1 I have been taken to task before for hating on any maths done with historical numbers, and this paper was actually a really good example of both what I think can be done with such evidence and what I think can’t. So let me try and make that interesting!
I can’t find an image of the actual papyrus, and I’m not sure there is a digital one from the database record…
… so instead here is a Byzantine nomisma, relevant as you will see, and specifically a gold solidus of Justinian I struck at Constantinople in 539-65 CE, from the University of Leeds’s Brotherton Library. It was CC/WC/BYZ/001, but now they’ve renumbered all the coins and I don’t have a list, so that’s about as far as I can go!
The text in question is one that MacCoull wrote several pieces about (and I admit, I haven’t read the others).2 What it is is several sheets of tax register from the reign of Emperor Justinian I, giving dues from one village in the Hermoupolis area as equivalencies between three accounting units, the nomisma or solidus, the keration or carat (but not necessarily our modern carat) and the talent, by this time a local base-metal unit of very low value. Each render is given in either of the first two units, and sometimes both, and talents. Of these only the nomisma existed as real money, as you see above, but it was very high-value so most people were presumably paying in sums of talents, hence the need for the list, and this is sort of where we come in.
A copper-alloy 33-nummi coin of Emperor Justinian I struck at Alexandria in 539-65 CE, CNG Coins electronic auction 229 lot 461
MacCoull did three things with this. Firstly she observed a recurrent unit of 100 talents being used in calculations, which I think is indeed there and makes sense given the very low value of a notional talent. But we have no clear ratio of Roman nomisma to Egyptian talent, so she wondered if this 100-talent unit could relate to a coin of 33 Roman nummi that reforms of Justinian introduced in Egypt in 539 CE, as shown above, and she used the equivalencies in this papyrus for maths that showed it might. Then, with that giving her the extra steps needed for the sums, she worked out the whole tax render of the village as just over five pounds of gold and argued, reasonably, that if one sole village paid five pounds of gold a year in tax, the tax take for all Egypt must have been huge, tons of gold.3 But she did not try to calculate it, which for me saves the paper; the point can be made without the sum. And actually, we don’t even need the preceding maths to be bang-on to make that point safely; an order of magnitude answer, one village paid several pounds, still does the same job.
That may, however, be just as well, because I’m really not sure it’s possible to do the middle bit, that is, proving that 100 talents equalled 33 nummi. There are a number of obstacles, almost all of which Maccoull acknowledged. This is the list.
Copper-alloy 12-nummi coin of Emperor Justinian I struck in Alexandria in 539-65 CE, Savoca Numismatik, Online Auction 209, lot 586
So that means that quite a lot of bits of MacCoull’s maths have multiple possible input values. After grappling with this for a while I had to tabulate it (sorry). So what you have here is the starting sum of 5 pounds of gold, reckoned at 72 nomismata to the pound, so 360 nomismata all told for the village, but run through Maccoull’s arithmetic for each variable combination. That is, I’ve multiplied up to talents for each rate of the keration, then divided by Maccoull’s 100-talent unit which she thought was worth 33 nummi, then multiplied by that 33 for a value in nummi, and lastly divided down back into nomismata, one row for the low valuation of them and one for the high. And thus…
| 360 s. | 18 kar. (16,800 tal.) | 24 kar. (22,400 tal.) | 25,000 tal. |
|---|---|---|---|
| in nummi | 360 x 16,800 = 6,048,000 tal. ÷ 100 = 60,480 units x 33 = 1,995,840 |
360 x 22,400 = 8,064,000 tal. ÷ 100 = 80,640 units x 33 = 2,661,120 |
360 x 25,000 = 9,000,000 tal. ÷ 100 = 90,000 units x 33 = 2,970,000 |
| In nomismata @ 7,200 nummi each | 277.2 | 369.6 | 412.5 |
| In nomismata @ 8,400 nummi each | 237.6 | 316.8 | 353.6 |
So, what does all that tell us? Well, most obviously, none of the answers are exactly right so any one is a fudge. If the system ran this way at all, it ran through roundings up or down that might be tiny one-by-one but could multiply up to much larger differences. One might give up here and say that no-one could use such an inaccurate reckoning, or one might do what people, not least Michael Hendy whom I gainsay only with great caution, tend to do instead and say that these margins were where money-changers made their profit and everyone else needed the coin too badly to refuse.9 If we take that path, then there are clearly two fudges there that are closer: the maths comes closest to working if the nomisma was 7,200 nummi and 24 keratia or 8,400 nummi and 25,000 talents. The latter is closer and I like it better because both variables are high. Maccoull had noticed that if you calculated with the low value of the nomisma and the high value of the keration or vice versa the valuations nearly balance out, but of course the nomisma should have gained or lost value in both measures at once, and in any case the differences aren’t the same, a seventh against a quarter, so can’t have been meant to counter each other as she suggested.10 But does the fact that with just the right set of input figures her maths nearly works actually prove it was historical and she was right?
Well, the input figures themselves are not junk, they are all attested. But then again, they are attested alongside other figures which were obviously in use at the same time and must have made sense to the users, but with which MacCoull’s maths can’t work. And also, the error margin between the highest and lowest outputs of the maths is a whopping 149% (I think), which really ought to make anyone worry about choosing a result within it because that makes their numbers go. But most of all, even the one version which seems to me plausible rests on the premise that MacCoull was right about the 100-talent unit being 33 nummi. If she was wrong about that none of the above works at all, so you can hardly say that if it works with her premise her premise must be right; that’s just circular. At the end I’m happy to say that if Maccoull was right about that equivalence that maths is probably how it worked, and that whether she was right or not her point about the provincial tax take is valid and important. But that doesn’t mean we should accept everything that comes with it…
1. Leslie S. B. MacCoull, “Accounting in BM 1075” in Adam Bülow-Jacobsen (ed.), Proceedings of the 20th International Congress of Papyrologists, Copenhagen, 23-29 August, 1992 (Copenhagen 1994), pp. 484-487.
2. The others were MacCoull, “Money and People in the Late Antique Hermopolite: BM 1075 and Related Texts” in Tyche Vol. 2 (Vienna 1987), pp. 99-105; and MacCoull, “Bemerkungen zu Papyri IX” in Tyche Vol. 11 (Vienna 1996), pp. 243-254 (no. 221), as well as MacCoull, “Hermopolite Taxation in BM 1076” in Journal of Juristic Papyrology Vol. 23 (Warszawa 1993), pp. 119-124, online here, which takes her conclusions in our paper and applies them to the next papyrus in the batch.
3. MacCoull, “Accounting”, pp. 485-487.
4. Procopius, Anecdota, ed. & transl. H. B. Dewing, Loeb Classical Library 290 (Cambridge MA 1935; repr. 1969), XXV (pp. 295-297). For details on this, despite some reservations I express below, your best resort is Philip Grierson’s Byzantine Coins (London 1982), but I’m away from my copy so can’t give the page references I’m afraid.
5. MacCoull, “Accounting”, pp. 486-487.
6. Ibid., inc. the term ‘εὔσταθμα’, p. 486.
7. Jonathan Jarrett, “Follis or follaron? The name of the Byzantine coin of 40 nummi” in Acta numismàtica Vol. 52 (Barcelona 2022), pp. 225–248.
8. In that same book I can’t access right now as cited in n. 4 above.
9. Michael Hendy, Studies in the Byzantine Monetary Economy (Cambridge 1985), pp. 344-360, though I have to admit I don’t think that bit of the book makes a lot of sense.
Posted in Byzantium, Currently reading..., Now working on..., numismatics
Tagged bad mathematics, Egypt, Justinian I, Leslie MacCoull, tax
A Corner of Tenth-Century Europe 
