Seminary LXII: from these hilltops we can see for centuries

Here is a much-delayed seminar report for you. On 9th March, already, Damián Fernández of NYU came to speak to the Cambridge Byzantine Seminar. Since his topic was “Hilltop Settlement and Economic Change in Late-Antique Northern Iberia”, which isn’t Byzantine at all, it’s not entirely clear to me why that was, but it was of obvious interest to me (you’ve heard me mention hilltops here before, right?) and there are people in Cambridge I only see at the Byzantine seminar, so I happened along.

The basic question Fernández was setting out to answer came out of a couple of quotes from Hydatius’s Chronicle, of which one goes like this:

The Sueves under King Hermeric pillaged the central areas of Gallæcia, but when some of their men were slaughtered and others captured by the people who remained in possession of the more secure fortified sites [castella tutiora], they restored the peace treaty.¹

The question that comes out of this is, what exactly were these castella? This treads lightly into some very tangled questions, about the degree of Romanisation in the north of Spain—Fernández thinks that recent work that has found seals, ceramics, buildings, walls to towns and so on demonstrates that it was more considerable than hard-line ethno-continuity theories would accept, but I know there are those reading who would say that this is what the authorities of the area want archæologists to find.² However, what Fernández was mainly attacking was a historiography in which the period after the arrival of the barbarians in Spain and before the eventual attempted extension of the consolidated Visigothic kingdom into the north is a time in which all that the locals could do was, quite literally, run to the hills. As Fernández pointed out, however, a lot of the hilltop forts they supposedly found seem to have been there under the Romans, Gijón for example being a third- or fourth-century foundation (which was still going for the Muslims to try and run Asturias from when they arrived, of course). Some are more ancient than that, even, but are refurbished during the Roman period (castro ventosa, seen below). I suppose the question then becomes, is that Roman occupation or local resistance to the Romans? It’s all very well to say that Roman material culture indicates this area was part of the Empire, but we know from the Scots and German borders that the peoples on the outside of the limes are very often keen buyers of Roman gear, and even happy to join the army (and serve in far-away places) without that necessarily meaning that they’re now cives romani.³

Current state of Castro Ventosa, near Bierzo

Fernández’s answer to this was that this sort of question can’t be addressed from the archæology. The material culture doesn’t differ between areas that may have been outside the frontier and areas everyone is sure were in it; it’s not ethnicity, it’s just poverty. Okay, fair point, but we still don’t know what was going on. One thing that was going on, however, was wall-building, in the third century right through to the early fifth, not because of any particular threat but because walls are a prestigious thing to have round your late Roman settlement. They associate not with decline, but with wealth. He suggested therefore that fortified hilltop settlement (and indeed fortified lowland settlement, of which there is also lots contemporaneously) was not an aberration caused by military, economic or demographic crises but the new mode of settlement for the period, a cultural shift not a strategic one. He saw a state-driven change in the settlement network caused by, well, fashion as much as economy, though that too. This seemed somewhat circular to me, the state encouraging change in settlement morphology because lots of people have changed the morphology of their settlements because the state… I wanted to know whether these sites have rôles as burial centres, as my pet ones from later certainly do, but this didn’t appear to work here: apparently some do and some don’t, and almost none have churches. I don’t think I’d expect churches, actually, I think those would be more local until later, so this didn’t really get me anything.

Remains of the walls of the ancient fortification at Viladonga

So, okay, there were a lot of small things here, and some quite big things, where I think alternative theses might be arguable or even preferable, but what I did like about this paper was his overall argument that these castella need to be seen not as crisis symptoms but as part of the same growth that is, at the same time that many of them are being refurbished (or even built, like Muelas below), causing the sprouting of new villas in the lowlands from third right through to sixth and in some cases seventh centuries. (Visigothic Spain was, after all, not apparently short of wealth.) Where the land is good for large-scale agrarian agriculture, you get villas; where it’s better for pastoralism and living on hilltops, you get castella—it’s environmental not military. That makes a lot of sense to me. The other important thing that he stressed is that, unlike the situation with hillforts like Dinas Powys in Britain, these are not aristocratic centres.⁴ There’s no evidence of resource redistribution, of patronage of craftsmen or of accumulation or special treatment of food animals (such as Alcock found at Dinas Powys); they’re just where people live, villages with walls. That’s so in the south, anyway; in the far north we know (from the slates!) that rents were collected at these places. Here a situation where control over transhumance routes is a source of power becomes more likely, even if the material culture, as said, is no different.

The (fifth-century) settlement of Muelas del Pan, near Zamora

So, plenty to chew on, and much that I thought other scholars would have disagreed with, not necessarily correctly but they would have. But my last paragraph of notes reads:

Settlements are a decision taken by social actors: response to soc.-econ. change by e. g. state, moving to a more dynamic organisation of a local kind under a hands-off barbarian k’dom; aristocracies, intensifying local and decentralised econ. Good places to put walls!

and I hope that shows that this is a guy with some big and powerfully explanatory ideas, which I’m sure I’ll meet again and which perhaps the readership might also find useful.

---

1. R. W. Burgess (ed./transl.), The Chronicle of Hydatius and the Consularia Constantinopolitana (Oxford 1993), Hydatius cap. 81 rev. D. Fernández.

2. The most relevant reference that the handouts provide seems to be Carmen Fernández-Ochoa [& Ángel Morillo], “Walls in the Urban Landscape of Late Roman Spain: Defense and Imperial Strategy” in Kim Bowes & Michael Kulikowski (edd.), Hispania in Late Antiquity: current perspectives, The Medieval and Early Modern Iberian World 24 (Leiden 2005), pp. 208-340, but Dra Fernández-Ochoa’s webpages would seem to be a good place to find more. That does of course rely on the assumption that she is not, contrary to what some people think, involved in the suppression of pre-Roman evidence from these areas so as to promote, “un pasado romano hipertrofiado por cuestiones políticas”. If you are concerned by that possibility you probably ought to follow the link; I’m not in any position to judge from here.

3. If you want an actual academic reference here rather than links to other blogs, no matter how authoritative they be, I offer you Karl Hauck, “Der Missionsauftrag Christi und das Kaisertum Ludwigs des Frommen” in Peter Godman & Roger Collins (edd.), Charlemagne’s Heir: new perspectives on the reign of Louis the Pious (Oxford 1990), pp. 275-296, which manages with startling ease to be more relevant to this question than you would imagine from its title.

4. I’ve given all these references before, but because it was a dig worth reading, I’ll do so again: Leslie Alcock, Dinas Powys (Cardiff 1963), rev. in idem, Economy, Society and Warfare among the Britons and Saxons (Cardiff 1987); see also idem, Kings & Warriors, Craftsmen & Priests in Northern Britain AD 550-850 (Edinburgh 2003), pp. 209-210.