Saint James's Catapult:
The Life and Times of Diego Gelmírez
of Santiago de Compostela

R. A. Fletcher
© R.A. Fletcher 1984
Used with permission of Oxford University Press



[1] Cape Finisterre is a mighty spur of granite which juts out from the mainland into the Atlantic in the extreme northwesterly quarter of the Iberian peninsula. What the earliest inhabitants of that area called it we do not know. As so often elsewhere within the territories which once were part of the Roman empire, it is the Latin name that has stuck: Finis Terrae, the end of the world. Even though we have long known that it is not accurate, the name remains fitting. The visitor to Cape Finisterre will find it so, as he contemplates, usually between the mists which sweep in on the western wind, the awful vastness of the waters which extend before him. At Cape Finisterre you are a long way from anywhere else. This is where the world stops. Out there the sea and the sky, earth and heaven, meet.

Santiago de Compostela lies a little inland, about forty miles due east of Finisterre. It too bears a Latin name, though not one which dates from imperial times: Sanctus Iacobus, Sant' Iago, Saint James. For over a thousand years devotion has drawn men and women to the apostle's shrine, to a tomb beneath a church in a holy city. Here miracles have been worked, the supernatural order has intermingled with the merely human. Here too, then, earth and heaven meet. Here at the margin of Christendom may be apprehended 'the point of intersection of the timeless with time'.

Galicia is cut off from the central tableland, or meseta, of the Iberian peninsula by a chain of mountains. The range known as the Cordillera Cantábrica, which runs along the north coast of Spain westwards from the fringes of the Basque country, inclines to the south shortly after passing the region of Oviedo and continues in a south-westerly direction towards the present northern frontier of Portugal. This south-westerly spur -- the Montes de León and the Sierra de la Culebra -- rarely drops below 4,000 feet, and [2] some of its peaks approach 7,000 feet in height. Its line continues through the mountainous provinces of northern Portugal, Tras os Montes, Alto Douro and Beira Alta, throwing up a last bastion in the Serra da Estrêla to the east of Coimbra, before dying into the coastal plain that leads down to Santarem and Lisbon. It is with the region on the Atlantic side of this formidable barrier that this chapter will be concerned, an area which now embraces the four Galician provinces of Spain -- Lugo, La Coruña, Pontevedra and Orense.

'Hard going, wearisome mountains.' In these words did a writer of the early twelfth century describe a journey made by the royal court of queen Urraca from Galicia to the meseta.(1) Communication between Galicia and the central tableland has always been difficult. In the north the foothills of the Cordillera Cantábrica run down close to the sea, no Roman road was driven along the coast, and we have little evidence of the use of the narrow coastal strip as a means of communication during the early medieval period. There is only one possible crossing-point of the mountains further south, on the line drawn between the towns of Astorga and Lugo. The road crosses two high passes, the Puerto de Manzanal between Astorga and Ponferrada, and the Puerto de Piedrafita del Cebrero between Ponferrada and Lugo. Ponferrada itself is set in the fertile depression known as El Bierzo. From here a route following the upper waters of the river Sil leads towards Orense. Further to the south again, the road from Zamora skirts the fringes of the Sierra de la Culebra, passes through Puebla de Sanabria and over the Padornela pass, forming the principal means of access to southern Galicia. Yet further to the south the valley of the Duero (Port. Douro) forms a corridor from the meseta into northern Portugal.

It is much easier to travel from Galicia southwards into Portugal than it is to cross the mountains on to the meseta. In the west the flood-plain of the lower reaches of the river Miño (Port. Minho) opens an easy access to Braga, Coimbra and the coastal plain of Portugal. The river valleys of the [3] Limia and the Tamega, draining in a south-westerly direction from the highlands of southern Galicia, form natural routes respectively from Orense through Celanova to Braga and from Verin through Chaves towards Viseu and Coimbra. Further east the upper waters of the river system which drains into the Sabor provide means of access only a little more difficult from Sanabria to Bragança.

The mountain barrier between Galicia and the meseta sharply differentiates the climates of the two regions. On the meseta winters are long and harsh, summers fiercely hot; rainfall is slight; woodland is little more than scrub. The aspect of the country is monotonous. Galicia's climate is temperate, Atlantic. Winters are generally mild, summers agreeably warm; rain is frequent, usually in the form of light showers; woodland is dense and lush. The countryside is easy on the eye, broken up and varied by outcrops of granite, rolling hills, abundance of rivers and streams. The scale of things is somehow comforting, manageable, human. There is not the desolation of the meseta, induced by an awareness of that brown, baked land stretching unchanging for miles and miles in every direction. These differences have not been without their effect on the inhabitants. The Galicians are friendly and cheerful; the people of the meseta are dour, sullen and charmless.

The meseta is the land of the classic staples of the Spanish economy: corn and sheep, the olive and the vine. The economy of Galicia is more varied. The Galician coastline is deeply indented with estuarine inlets known as rías These provide some superlative natural harbours, as at Vigo and Corunna, and we shall see that a lively trade was developing along these coasts during the twelfth century. For much of the earlier middle ages the opportunities presented by the long, serrated, island-studded Atlantic littoral were exploited by seaborne enemies -- pirates from the Muslim south and the Anglo-Scandinavian north. This never deterred the fishermen of Galicia from following a calling which has always been one of the mainstays of the Galician economy. When Diego Gelmírez issued an edict fixing prices in the market of Santiago de Compostela he listed octopus, lobster, oysters, cod, sea-bream, conger eel, pilchards and lampreys among [4] the fish for sale there. Fishermen returning up the Ría de Arosa paid toll on their catch at Torres del Oeste.(2) There were freshwater fish to be had in abundance too. When the bishopric of Tuy was re-established in 1071, for example, it was endowed with fisheries in the river Miño.(3)

The wild produce of the mainland was as varied as that of the sea and the rivers which fed it. Its gathering might be hazardous, for wolves were common -- the inhabitants of the district round Compostela were bound to turn out every Saturday to hunt them -- and even the occasional bear might be seen.(4) Some at least of the wax and honey of which we hear would have been made by wild bees.(5) Hares and partridges could be bought in the market at Compostela, rabbit-skin cloaks in that of Guimarães. It was for the pursuit of bigger game, wild boar or deer, that two Galician noblemen gave their king a hound called Ulgar and a hunting-spear, together worth 500 solidi, in 1118.(6) Our sources frequently refer to nuts, and it is no surprise that this is often in association with pigs. When Pedro Gundesíndez, canon of Compostela, granted a property to the monastery of Caabeiro he asked the monks to grant it in life-tenancy to his brother-in-law, and suggested that the rent should consist of suitable quantities of nuts, corn, beans, segunda and a pig.(7) The masons who worked to build Lugo cathedral about the year 1130 were supplied with hams, along with chickens, sheep, butter and bread from one of the bishop and chapter's estates.(8)

Innumerable charters refer to other domesticated livestock -- cows, sheep and goats, chickens, ducks and geese. The Galician economy was never a 'pastoral economy' plain and simple, but there can be no doubt that grazing played a very [5] significant part in it. Oxen were presumably the principal draught animals, as they still are in Galicia. The carts they drew a thousand years ago probably looked much like the carts they draw today; narrow and boat-shaped, the sides that run towards the 'prow' not meeting but leaving a gap of a few inches between them, the solid wooden wheels giving off a high-pitched squealing when in motion. We hear much of horses and of mules, the latter often more valuable than the former.(9) It is unlikely that these were draught animals, more probable that they were the mounts and pack-animals of warriors, merchants and the higher clergy. It is in such social contexts that we hear most of them. When the lady Ermesenda Núñez disposed of her dead brother's possessions in 1073 there were numbered among them three horses, a mule, two coats of mail and two swords; and it was horses, mules, coats of mail and helmets that Osorio Bermúdez seized from his nephew Bermudo Pérez in a family quarrel.(10) It is likely that many of the greater religious communities, such as the Compostelan monastery of San Martín Pinario, kept stud-farms.(11)

Vineyards are of common occurrence in our sources, but we hear much of orchards too, and it would seem that a good deal of cider was drunk. The French author of the guidebook for pilgrims which forms a part of the so-called Liber Sancti Jacobi commented that it was more often to be encountered in Galicia than wine. Cider as well as wine was drunk at a king's coronation in 1111, and a render of cider was stipulated as part of the rent in a lease of 1116.(12) The same author noted that wheaten bread was rare, rye bread being more commonly consumed. Wheat was certainly grown, though references to it are infrequent. We hear much more often of barley and rye, and very occasionally of millet.(13) The drying and storage of grain and hay, always difficult in the damp climate of Galicia, many then as now have been done in the [6] structures known locally as cabazos, which are one of the most striking features of the Galician countryside: narrow, gabled sheds of stone raised on piers to some three feet above the ground to keep them out of the reach of vermin; but no certain reference to them in these early documents has yet been found, and it may be that they were developed for the storage of maize only after its introduction in the sixteenth century.

The Romans had mined gold, tin and iron in the northwest of Spain. Only the last of these, it would seem, was still an active industry during our period. Place-names such as Ferreira furnish one indication of it, charters another: a charter from the monastery of Celanova, for example, refers to 'the mine where men make iron'.(14) The extraction of salt was another essential mineral industry. All our references are to coastal salt-pans. Ardiu Vimáraz, for instance, left to her son by her will in 1116 'the saltworks (salinas)which I constructed with him' on one of the islands off the Atlantic coast. Salt transported up the Ría de Arosa towards Padrón was liable to toll at Torres del Oeste.(15) There must have been a local ceramic industry, though the archaeologists have yet to reveal its products to us. Although foreign textiles did reach Galicia, as we shall see, there was doubtless some domestic production of cloth. The measures of cloth (pannos) which were exchanged in 1042, at a lowly social level, for a small plot of land near Sobrado, sound like local woollens. Some flax was grown -- we meet it in a Lugo charter of 1030, for instance -- and this helps to explain the fine bed- and table-linen to which some of our documents refer.(16) Leather must have been worked too. Footwear was a commodity perennially in demand among the pilgrims to Santiago, to say nothing of native users; and waterproof capes were an essential item of daily equipment in Galicia before the coming of the umbrella.

Tools and techniques were rudimentary. In the market of Compostela in the 1130s you might buy axes, mattocks, [7] ploughshares, billhooks, sickles -- all of iron -- and implements of uncertain function set with stone cutting edges (perhaps threshing sledges).(17) This is simple equipment, but adequate for working the light, rich soil of Galicia. The visitor will not see anything more complicated today. Ploughs are still light scratch ploughs, an iron share fastened to a wooden beam; harrows are wooden frames with iron spikes protruding beneath, pressed to the surface of the soil by the weight of rocks or children, drawn by oxen. Water-mills there were, and their use was spreading: the 'new mill' makes a frequent appearance in our charters.(18)

The latest-and indeed almost the only-historian to have studied the demography of early medieval Galicia has estimated her population about the year 1000 at the audaciously precise figure of 232,400. Dr García Alvarez's calculations, though ingenious, are highly speculative.(19) It is notoriously difficult, as every historian knows, to estimate the size of the population in any given area of medieval Europe. While it would be injudicious for one who can lay claim to no demographic expertise whatsoever to offer any but the most tentative of opinions on this topic, my own impression is that this figure is a good deal too low. I would guess that we should multiply it by at least five, conceivably by eight, to approximate to the demographic realities of the eleventh century. Be this as it may, it is more important for our purposes to estimate rather the density and distribution of the population than its aggregate size. The conformation of the terrain has always dictated that in Galicia the spread of the population is uneven. Much of the region is unsuitable for tillage, some of it even for grazing: the bleak moorland round Lugo for example, or the highlands in the province of Orense, or the rocky coastal areas of the province of La Coruña. In districts such as these settlement must have been extremely sparse. As against this, it can be demonstrated [8] that other areas were densely settled. Let one example stand for many. There survives in the cartulary of the Benedictine monastery of Celanova a series of charters and inventories, thirty-four documents in all, relating to the village of Bobadela, a little to the north of Celanova itself, between the years 988 and 1040. With the aid of these it can be shown that during this period the population of the village numbered at least 186, distributed in thirty-nine different properties. (In 1753 the population was 240 in sixty properties, in 1950 was 176 in thirty-eight properties.) Assuming that the present parish boundaries correspond roughly with the village territory of c. 1000, there is no escaping the conclusion that the density of the population was then something higher than the present-day density in the province of Orense, in which Bobadela lies, namely fifty-six persons per square kilometre.(20) Bobadela may have been a special case -- not least in the survival of abundant documentation relating to it -- but it can be shown that many other villages at least approached it in the density of their populations. Such figures are startling, even alarming, especially when set against the figure of eighteen persons per square kilometre postulated for the most densely settled parts of the populous county of Catalonia at the same date.(21) There can be no doubt that parts of Galicia were experiencing demographic saturation in the tenth and eleventh centuries.

On the central Spanish meseta the unit of settlement was and is the pueblo; that is to say, the large nucleated village surrounded by its own fields, with no outlying farms, separated from its neighbours by some considerable distance, sometimes as much as ten miles or so. The demands of agrarian routine and the need for defence, the simple desire for human society in the vast solitude of the plains, together dictated that it should be so. Nowadays the pueblo might have a population running into thousands. Doubtless they were smaller in the early middle ages, but we should probably not be far wrong if we think of them as having had populations [9] of some hundreds. In Galicia the word most commonly used to describe a unit of settlement is logar (This is the rendering in Gallego. The Castilian term is lugar. The words are of course derived from Latin locus.) The logar is a smallish village, it might today have a population of anything between fifty and three hundred people. The homesteads are normally scattered, there is not the tight bunching together that one finds in the typical pueblo. Each of them is surrounded by a complex of little yards, middens, orchards, vineyards and diminutive fields. Haphazardly strung out along a stream or road, on either side of a valley or near a suitable anchorage for fishing boats, the settlement may be distributed over a surprisingly large area. The logar does not have a centre, a focal point-again in contrast to the pueblo, which invariably has a central plaza, or square. It is characteristic that the road leading to such a settlement, which the visitor might expect to form a principal artery, dissolves within the logar into a confusing network of narrow lanes, with sharp twists and turns, unexpected dead-ends, and a disconcerting way of suddenly turning into a farmyard or of simply petering out on the edge of woodland or marsh: a maze where only the livestock appear to know their way about. All the evidence at our disposal suggests that the morphology of the Galician logar was much the same a millennium ago as it is today. That evidence is furnished by hundreds of charters recording donations, sales and leases of tiny parcels of land. The terms in which and the detail with which the bounds are described in these documents instantly call to mind the layout of the modern Galician logar.(22)

Larger units of settlement were almost entirely lacking. Lugo presents us with the most instructive example. Lucus Augusti had been the most important town in Galicia during the Roman period, though not a big one; her third-century walls, exceptionally well preserved, describe a circuit of only about 2,300 yards. The seat of a bishopric by (at latest) the latter part of the fifth century, Lugo was an administrative centre of some significance throughout the Suevic and Visigothic periods. Thereafter she went into a [10] long decline. In the middle of the eighth century bishop Odoario found the town deserted and set about reviving it. Yet in 910 a number of Galician aristocrats promised their king, Ordoño II, that they would rebuild the 'abandoned tenements' of Lugo (casas destructas) and themselves take up residence in them with their families. That this second resettlement had only limited success is suggested by a document of some sixty years later which reveals contemporary anxiety to repopulate the town so that it might serve as a defensive point against Viking attack.(23) This is not to suggest that the town was literally deserted. It remained the seat of a bishopric, royal charters were issued there, and it was administered by a count from at least the middle of the ninth century. But the documents do suggest that Lugo was not attracting and holding any considerable urban population. Its commercial and industrial role was insignificant.

The shrine of St. James attracted settlers to serve the needs of clergy and pilgrims. These may have included some immigrants from outside Spain from an early date: Bertenandus francus was a property-owner in Compostela about the middle of the tenth century.(24) The town was walled about a century later. But the smallness of the area enclosed, and the rarity of references to trade or merchants before the latter part of the eleventh century alike suggest that the town was a modest affair. The same may be said of Tuy, Braga, Orense and Chaves. This almost non-urban state of affairs in tenth- and eleventh-century Galicia contrasts with the situation on the meseta. León was not only a focus for government, the seat of the bishopric and of several other religious establishments, but also a centre of trade and industry which was attracting immigration and sustaining a growing population, many of whose members probably owed their livelihood exclusively to commerce or industry.(25)

The lack of a native merchant class in Galicia may have been supplied by Jewish traders. Surviving references to them, [11] though tantalizingly few, are suggestive. In 1044, for example, a certain Menedo González 'was entertaining Jews in his house who were transacting business there' (tenebat suos hebroes in sua casa qui faciebant suo mercatum); Arias Oduárez attacked them and despoiled them of 700 lengths of silk, 30 serges and 40 linens together valued at a thousand pounds -- an amazing haul.(26) The incident suggests one explanation of the presence of luxury goods in aristocratic wills and church inventories, to which we shall return. It occurred in southern Galicia, not far from Celanova; perhaps the merchants who suffered had travelled up from Portugal. In whose Muslim cities we know that there existed Jewish communities. But how frequent their journeys into Christian Galicia may have been, and what merchandise they carried back with them, we have no means of telling.

Texts such as this raise questions also about the presence of coinage and the means of exchange in Galicia. Remarkable as it may seem, itis tolerably clear that no coin was minted in Galicia after the end of the Visigothic period in the early eighth century for a very long time. The earliest reference to a royal mint which we possess seems to have occurred in the lost charter of Alfonso VI which we only know through a charter of his great-grandson Fernando II dated 19 February 1158. Alfonso's charter had granted one-third of the profits of the royal mint at Lugo to the bishop and chapter of that see; it was probably issued between 1088 and 1109. What did the Galicians do for money before the existence of a native currency? Suevic and Visigothic coins remained in circulation long after the Visigothic state had perished. Some of the gold coins of Muslim Spain found their way to the north-west, but one might guess that the intrinsic value of these coins was to high to articulate a brisk exchange in Galicia's agrarian economy.(27) It is evident that there was much barter. Land was sold in 1078 for 'a mare and an excellent cloak and a cow', exchanged in 1093 for the life tenancy of a church; a share in a church was sold in about [12] 1087 for 'a splendid roan stallion and a good cloak'.(28) We should beware of assuming that the prevalence of barter indicates a sluggish economy of exchange. Scores of documents from the tenth and eleventh centuries suggest rather that at a local level exchange was lively. It seems to have been articulated by the widespread acceptance of the equivalence of one solidus to either a measure (modius) of corn or a sheep. Thus we hear, for instance, of a sow worth two modii, and of an ox worth seven.(29)

Generally speaking, however, it appears that while a local trade based on barter was thriving, long-distance trade was rare and the regular use of coin unknown. In these respects the economy of Galicia stood in marked contrast to the regional economies of other, parts of the peninsula which were in Christian hands. León and Castile, and even the Pyrenean economies of Navarre and Aragon, were accustomed to the traffic of merchants who used coin: while the economy of Catalonia in the tenth and eleventh centuries was among the most advanced in Latin Christendom.

Notable by its absence from our documentation is any reference to the buying and selling of slaves. Yet slaves certainly existed in Galicia, just as they did in other parts of Christian Spain. The principal source of Moorish slaves was capture in war. This probably meant that they were in short supply in tenth- and eleventh-century Galicia, a period when it was generally true that the Muslims of Al-Andalus held the military and naval initiatives in Spain. Those who could find a use for slaves depended, probably, on occasional windfalls. Diego Gelmírez was presented with some captured Moorish pirates in 1115, and set them to work on the rebuilding of the cathedral of Santiago.(30) Later, with new Christian initiatives under Alfonso VII (1126-57), this source of supply may have quickened. The lady Guntroda Suárez included six Moorish slaves, three men and three women, among the endowments of her church of San Pedro de Villanova in about 1145.(31)

[13] Slaves were not used only for menial tasks. Two references indicate that great men particularly valued their culinary skills. (Arab cooking has no great gastronomic reputation: but those acquainted with the cuisine of Galicia will need no telling that any change could only be for the better.) St. Rosendo, the founder of Celanova, installed there as chef one Fez, described as 'a Moor from Córdoba'; and nearly two centuries later count Pedro Froílaz de Traba, the greatest lay magnate of Galicia in the early twelfth century, presented his son Fernando with a Moorish cook.(32)

Pedro Froílaz's cook was called Martín, and this Christian name presumably indicates that he had been baptized. This occurrence, if not uncommon, must make it well-nigh impossible to detect Moorish slaves in any list of names. When Rodrigo Gutiérrez married in 1037 he settled on his wife seventy-eight separate properties, eighty-one male and forty-three female slaves (mancipios et mancipias), flocks of sheep, a stud of mares and twenty stallions, two hundred head of other livestock, furniture, clothes and jewellery.(33) The names of some of the slaves were rather odd. Amicus is a curious name for a man, Nomenbonus a still more curious name for a woman. Were these perhaps baptismal names given to slaves who had been converted? If so, one may wonder how many other converted Moors lurk behind the impeccably Christian names -- Pedro, María, García, Guntroda, etc. -- of this slave list and others similar to it.

Slaves may have been integrated into more than the religion of their masters. A corrupt thirteenth-century copy of a document of 1110-28 from the monastery of Caabeiro seems to refer to the presence of a subject peasantry, now Christian but descended from Moors, on the monastic estates in this remote corner of Galicia.(34) Since the Islamic armies never occupied the extreme north-west it is out of the question -- quite apart from being inherently improbable -- that these people should have been downgraded descendants of [14] eighth-century Berber settlers; more likely that they were the descendants of slaves captured in battle or acquired by other means who had been settled on the land and had accepted Christianity. Our Caabeiro charter might therefore be said to bear out the contention of a distinguished historian of medieval slavery in the Iberian peninsula that in Galicia Moorish slaves were somehow assumed into the ranks of the serfs.(35) Perhaps the mancipios of Rodrigo Gutiérrez's marriage settlement should be termed not 'slaves' but 'serfs'.

We waver on the brink of a terminological morass into which historians are often sucked, there to flounder helplessly, when they try to find modern equivalents for words used in early medieval documents to describe social groupings. Who were the people whom I have described as 'serfs', and what do we know of them? The key word is criatio.(36) The criatio -- people occupied the lowest social rung on the ladder of the free. They were bound to the land in service to a lord. They could be sold, exchanged, or given away. Thus in 1087 a vendor disposed of 'my estates and my criazone' to a purchaser. The see of Tuy was endowed in 1071 with the church of Paderne 'and all its criazone'. In 1102 the abbot of Jubia and Pelayo Vélaz came to an agreement over three criatio -- men, probably brothers, disputed between them; two of them went to the abbot and Pelayo got the third. Diego Gelmírez acquired an estate 'with all its creatione' by exchange in 1124.(37) However, such people could own property. Three criatos were left a house in Compostela by a testator in 1116.(38) As property-owners they must have had access to the public courts of the land; and as law-worthy men they were free men.

For many years it was believed that Galician rural society below the élite levels consisted entirely of such criatio-people. Recent research has established that there existed an intermediate and very sizeable class of free property-owners -- allodialists they would be called elsewhere -- who unlike [15] the criatio-men acknowledged no lord below the king.(39) This rural middle class was densely distributed over the whole face of Galicia. The evidence for its existence is furnished by innumerable charters recording sales and donations of land which survive in the cartularies of monastic houses such as Celanova and Sobrado. These alienations were invariably made without the sanction of any lord. Such restraints on alienation as existed were exerted by the kinsfolk of the contracting parties whose names may be found in countless witness-lists. Relatives therefore exercised what might be called a horizontal pull upon the desire of a landowner to do what he pleased with his own; but of the vertical pressure of a superior lordship there is not a sign.

It would be rash to hazard any generalizations about these people. Doubtless they included, then as at any other time, both the fortunate and the unlucky, the contented and the ambitious, the careful and the improvident. It was an amorphous social group, as middle classes tend to be, within which it was possible both to rise and to fall. We have already seen evidence that parts of Galicia were densely populated. We know too that by Visigothic legal custom, which was still observed (as we shall see') in Galicia during this period, inheritances were divided equally among heirs at each generation. Under a customary system of partible inheritance there is a tendency for holdings to become ever smaller -- offset though this may be by other factors such as a high rate of both infant and adult mortality. 'There occurred a famine and great loss of life' we read in a charter of 1001.(40) In a society thus constituted crop-failure and cattle disease bring immediate distress for which borrowing is the initial remedy. Two very interesting charters, of 1023 and 1042, record surrenders of land to creditors by debtors who could not pay the interest (renovo) on their loans.(41) We shall never know how widespread indebtedness was among the lesser rural allodialists, but it may have been an important factor in [16] making possible the concentration of property in the hands of some of their neighbours. Take the case of Cresconio of Bobadela.(42) We know about him because in later life he entered the monastery of Celanova, bringing with him his considerable landed wealth -- and his title-deeds, preserved for us in the Celanova cartulary. Between 989 and 1010 he had assembled his estate by means of numerous separate transactions, most of them involving tiny parcels of land. Fortunately for us his charters usually tell us how it was done. We hear of debtors who defaulted on interest payments; of loans for the payment of compensation to other parties which could not then be repaid; of bribes to Cresconio, a prominent man locally, to secure his advocacy in lawsuits; and, of course, of straightforward purchase. Others could prosper by hitching their fortunes to men yet higher in the social scale: allodialists formed, in Galicia as elsewhere, a ministerial class. The fortunes of one Suero Pérez, for instance, seem to have been founded upon his services to Rodrigo Ovéquiz, count of Lugo, who rewarded him with a grant of land in 1084. By ingenuity or good fortune he survived the eclipse of count Rodrigo after his unsuccessful rebellion against Alfonso VI in 1085, and during the remainder of his life -- we lose sight of him after 1103 -- we can trace his acquisitions of land and men in a block of territory in southern Galicia which eventually came into the possession of the Cistercian abbey of Osera.(43) Some rose even higher. Diego Gelmírez himself was one of these, and he rose highest of all.

The topmost rung on the social ladder was occupied by the Galician aristocracy. Throughout the tenth century this was a very small group. To say that its members were drawn from a dozen or so families is a somewhat desperate attempt to give a rough indication of its size, but in itself a concession to twentieth-century understanding which may mislead: these aristocratic clans were all interconnected, and they had different ideas about what constituted a 'family' from ourselves. Its quality is perhaps better indicated by saying that in any one generation in tenth-century Galicia [17] there were perhaps a couple of hundred people, men and women, who mattered. They were linked by ties of blood and marriage both amongst themselves and with the royal house of León; they possessed enormous landed wealth; and they monopolized all the important positions in royal and ecclesiastical government. The inmates of the monastic houses which they founded and governed were keenly interested in genealogy -- probably reflecting in this the taste of patrons to whom, indeed, they were often related. This, and the abundant charters of the tenth century, enable us to discover a good deal about them.

The most prominent figure in the religious life of Galicia in the tenth century was Rosendo, leader of a monastic revival, founder of the abbey of Celanova, bishop successively of Mondoñedo and Compostela. Rosendo was to Galicia what Abbo of Fleury, Dunstan of Glastonbury or Oliba of Ripoll were to central France, southern England or Catalonia respectively. If we look to his nearer relatives we may hope to catch some flavour of the potentes of Galicia.(44) Rosendo was descended from the royal family, for his great-grandfather Gatón, count of Astorga in the 850s, was probably a younger brother of king Ordoño I. Rosendo's aunt Elvira married Ordoño II, his niece Gotona married Sancho Ordóñez king of Galicia, his cousin Adosinda married Ramiro II, another cousin Teresa married Sancho I, another niece Velasquita married Bermudo II, and his great-niece Elvira married Alfonso V. His male relatives occupied important positions in the royal government. His grandfather Hermenegild, hero of the conquest of Coimbra at the head of a royal army in 878, had been count of Tuy and Porto and the mayordomo of king Alfonso III. His father Gutierre and his uncle Arias were prominent at the courts of Ordoño II (their brother-in-law) and Sancho Ordóñez (their nephew). The bishopric of Mondoñedo was occupied successively by his great-uncle Savarico (907-25), by himself (925-47) and by his nephew Arias (947-58). The see of Compostela was held by three of [18] his relatives in the course of the century -- Gundesindo (920-4), Sisnando (952-68) and Pelayo (971-85) -- as well as by himself. The greatest monastic houses of Galicia were founded or restored by his family. Samos was restored by his uncle Arias in about 922; Rosendo himself founded Celanova in 934; Hermenegild, founder of Sobrado in 952, was the son of his great-uncle Abito; Osorio Gutiérrez, the founder of Lorenzana in 969, was a first cousin.

These religious houses were lavishly endowed. Let us take the last-named, Lorenzana, by way of example.(45) Osorio Gutiérrez founded it on a central nucleus of land to which he added a number of other estates. He gave to the monks he established there no less than eighteen proprietary churches. In addition to the stock already on the land he provided them with 90 mares and 2 stallions, 150 cows and 3 bulls, 150 yoke of oxen, 1,000 sheep, 500 pigs and 300 geese. He gave them church furnishings in abundance: four big bells and four small, three crosses, four chalices and patens, three crowns, three bronze thuribles, eight altar frontals, twenty-five priestly vestments and ten harps; the three capsas might have been caskets suitable for use as reliquaries. He also passed on to the monks a collection of liturgical and devotional books: that he had managed to assemble a large library is indicated by the fact that prayer-books alone ran to seven volumes For domestic use in the monastery he provided thirty-nine beds and sets of bedclothes (of differing degrees of comfort -- twelve were 'beds for the poor', presumably for use in the guest-house or for the monastic servants); seven benches with their cushions; mugs, dishes, bowls and jugs of silver; 6 candlesticks, 14 spoons and 184 table-napkins. Count Osorio himself entered his new foundation; we may suppose that his declining years were spent in some comfort.

[19] This is about as close as the surviving sources permit us to get to these noblemen. We have no Thietmar of Merseburg, no Orderic Vitalis, to illumine the aristocracy of Galicia as their writings so vividly light up the Saxon aristocracy of the tenth century and the Norman of the eleventh. But we may guess that the Galician nobility shared certain common features with their Saxon and Norman counterparts. It is reasonable to suppose that partition of inheritances between co-heirs and heiresses gave rise to frequent dissension among kinsfolk. The expanding kingships of Ordoño I and Alfonso III in the ninth century had offered new rewards -- in land, booty, lordships, military command, ecclesiastical office -- to enterprising and reliable king's friends. This would have driven a wedge of status and power between the new king's men and the rest. The forebears of St. Rosendo and count Osorio may have been 'made' by the ninth-century Asturian kings as consciously as the aristocracy which conquered England was 'made' by duke William of Normandy. New opportunities would have increased competition for rewards within the artistocratic group and threatened such fragile harmony as existed. The halt to expansion imposed upon the Christians of the north by the caliph Abd al-Rahman III in the early tenth century would have laid such tensions under further stress by stemming any increase in rewards and thus making more intense the competition for what remained. It is not unlikely that it was factors such as this which lay behind the internal strife which afflicted León, Galicia and Castile in the middle and later years of the tenth century. It is not impossible that such factors had something to do with the widespread foundation of monasteries during the same period.

We may be confident that this aristocratic society was frequently disrupted by feud. This would have been one of a number of reasons for the influential position occupied by its women. Feuds could be calmed by marriage-alliances; Galician like Anglo-Saxon brides could be 'weavers of peace'. Women could inherit land. If they survived the hazards of child-bearing and if their menfolk predeceased them, they might assemble under their control spectacular concentrations of property. This, for example, was the good fortune [20] (and the dilemma) of Ermesenda Núñez, herself a descendant of count Osorio Gutiérrez, who died at an advanced age in about 1084.(46) One might have expected that monastic foundations for women would have been frequent, as in tenth-century Saxony and to a lesser degree in Anglo-Saxon England. Women were certainly influential in monastic foundation; and, as we shall see in a later chapter, double houses for men and women did exist in Galicia: but no nunnery was established in the course of the Galician monastic revival of the tenth century. This was a lack which Diego Gelmírez's patronage was to supply in the twelfth century.

Rich, well-connected, powerful though these men and women were, their horizons were circumscribed. Tenth-century Galicia was not entirely isolated, but it is not going too far to say that her contact with the rest of Latin Christendom was fitful. It had not always been so. Late antique Galicia had been very much a part of the Roman and Mediterranean world. The Galician lady Egeria had made a pilgrimage to the Holy Places in the years 381-4, and has left us her account of it.(47) The chronicler Hydatius, a bishop in Galicia from 427 until his death in about 470, conscious though he was of living his life 'at the uttermost limit of the world', was yet a travelled man whose horizons were far from modest. He had journeyed to the east as a young man and met its learned men, including the great Jerome. In later life he was interested in the affairs of the lands bordering the eastern Mediterranean, and his references to travellers from the east coming to Galicia show us something of the means by which his curiosity was satisfied. As a bishop he travelled to Gaul on an embassy to Aetius in the years 431-2. We have to wait for over six centuries before we meet another native of Galicia whose vision ranged so far.(48) The barbarian period whose beginnings were chronicled by Hydatius, little [21] though we know of it, does not seriously impair our sense that Galicia continued to be a part of some larger cultural whole. Miro, king of the Suevi who settled in Galicia, had diplomatic relations not only with fellow barbarian kings in Neustria and Burgundy but also with the emperors in Constantinople. It was perhaps he who offered to the shrine of St. Martin of Tours the weight of his son in gold as the price for a miraculous cure. The most distinguished churchman of Galicia in the sixth century, Martin of Braga, was a native of Pannonia. The Visigothic king Leovigild could impound the ships of Gaulish merchants in Galicia. The fine sarcophagus at Lorenzana -- reputed to have received at a later date the mortal remains of count Osorio Gutiérrez -- was probably an import from southern Gaul in the seventh century. One of the coins in the Bordeaux hoard deposited about 700 was struck at a Galician mint; it may have made its way to Aquitaine through trade. British, possibly Breton, monks settled at Bretoña, near Mondoñedo, in the sixth century. Direct connections between western Spain and Ireland may have existed in the seventh. Certain of Isidore's works appear to have reached Irish centres of learning with remarkable speed, and it is possible that the monastic customs of Fructuosus of Braga owed something to Celtic usages.(49)

It was from the eighth century onward that Galicia slipped out of the mainstream of western Christian culture. It is not easy to determine why this should have happened. The Islamic invasion and conquest of Spain is only a partial explanation. True, it made communication by land through [22] the Pyrenees to Gaul far more difficult. A Muslim presence in north-central Spain, about the upper waters of the Ebro, endured for three centuries. The traveller from Galicia into Francia overland had to run the gauntlet of passing either through Islamic territory or -- perhaps still more alarming -- through the territory of pagan Vasconia; the latter a larger region now after the expansion of the Gascon tribesmen in the seventh century, people on whom Christian missionaries such as Amandus had made little, if any, impression. Yet the sixth - and seventh-century evidence suggests that Galician communication with the outside world had been made largely by sea. If traffic on the western seaways faltered and died, we may ask why and look in vain for an answer. Galicia is very vulnerable to attack from the sea. Hydatius tells of Vandal raids from Africa in 445, of Herul raids from the North Sea in 455 and 459.(50) But the Arabs and Berbers who invaded Spain in the eighth century were not great sea-goers at that early date -- though they were later on -- and it has been shown that the Vikings did not attack Galicia before the middle of the ninth century.(51) Our sources from Galicia, Aquitaine, western Francia and the British Isles are desperately meagre for the eighth century, and this should caution us to be wary in our search for any explanation. Perhaps the truth is that as the focuses of economic life, political power and intellectual activity moved gradually eastwards -- from Neustria to Austrasia, from Aquitaine to Lombardy, from Ireland to Anglo-Saxon England -- so Galicia was left out in the cold.

The Asturian kingdom, it is true, had relations with the court of Charlemagne and, perhaps, his successors. Alfonso III (866-910), in particular, was a ruler whose kingship may fruitfully be interpreted as having been cast in a sub-Carolingian mould -- not unlike that of his contemporary Alfred of Wessex. As we shall later see, he was in direct communication with the clergy of Tours, albeit anxiously and uncertainly. But it was in Oviedo and León and the [23] monasteries of the northern meseta, not in Galicia, that the heart of the imperium of the Leonese kings pulsed most vigorously; such at least is suggested by the works of art, the buildings and the texts which have come down to us.

By Alfonso III's day we do seem to be in an age when the Vikings were stifling such sea-borne communications as still existed. We know of raids on the Galician coast in 844 and 858; there may have been others of which we know nothing. Alfonso III was sufficiently worried by the threat of Viking attack to establish fortified strong points near his coastline, as other rulers were doing elsewhere. Perhaps the 'heathen men' against whom he fought (as his charters proudly tell us) were not always Muslims. The next big raid that we hear of occurred in 968: bishop Sisnando of Compostela was killed, the monastery of Curtis was sacked, and panicky measures were ordered for the defence of the inland town of Lugo.(52) At some point early in the eleventh century Tuy was sacked; its bishopric remained vacant for the next half-century. A pathetic piece of family history recorded in a Portuguese charter of 1018 lifts for a moment the curtain which normally obscures the more humble human consequences of the Viking raids, Amarelo Mestáliz was forced to raise money on the security of his land in order to ransom his daughters who had been captured by the Vikings in 10l5.(53) Bishop Cresconio of Compostela (c.1036-66) repulsed a Viking descent and built the fortress intended to protect the approach to the town of Compostela from the Atlantic which may still be seen by the water's edge at Torres del Oeste. A charter of 1086 refers to this or another raid in the Nendos district.(54)

Enemies were also coming by sea from the south. The Andalusian navy was called into being after the humiliating [24] Viking descent on the Guadalquivir valley in 844. It proved itself for the first time in the repulse of the Vikings in 859. Greatly enlarged, and its dockyards at Seville extended, it was employed to patrol the Iberian coastline under the caliphs Abd al-Rahman III (912-61) and Al-Hakam (961-76). Piracy by freebooting Saracens up and down the long Galician coastline became endemic. Those who lived by the sea from Seville to Coimbra, according to the Historia Compostellana, were accustomed to take to the ocean and raid the Christian coasts from the mouth of the Duero right round to the Pyrenees.(55) Later on we shall see what measures Diego Gelmírez took against them.

The Saracens also held the military initiative by land. The heady days of Christian expansion under Alfonso III were soon over. The roughly even balance of power in the first half of the tenth century was tipped decisively in Córdoba's favour by the massive growth in the caliphate's resources under Abd al-Rahman III. The Leonese rois fainéants of the third quarter of the century were the caliph's clients. And then, towards the end of the century there fell upon all of Christian Spain the, scourge of Almanzor, vizir of Al-Andalus and its ruler in all but name from 981. After fifty-seven victorious campaigns against the Christians Almanzor died in 1002, 'and was buried in Hell' as an eleventh-century chronicler noted with satisfaction; but not before he had, in 997, penetrated to the heart of Galicia, sacked the town of Compostela and carried off in triumph to Córdoba the bells from St. James's church. It was a calculated affront to one of Christian Spain's most holy places, and a demonstration of the vulnerability of Galicia.

Sporadic attack by land and sea, from the south and from the north, was powerful inducement to the Galicians to stay at home. This reinforced the tendency of a naturally conservative agricultural people to look backward to a reassuring past, to be content with what they knew rather than to reach out to embrace a world beyond their own. The law which they observed, the Lex Visigothorum, was an old law. It was the code of king Reccaswinth, enlarged by the Novels of Wamba, Erwig and Egica, that still governed the [25] vital transactions of family and therefore of political life -- the marrying and the giving in marriage, the transmission of property to heirs, the safeguarding of the rights of widows and orphans.(56) The names they gave their children were old names, drawn from a distant Suevic and Visigothic past: names such as Miro and Leovigild, Fromaric and Fagildo, Tudemir and Attila, abound in the charters of the tenth and eleventh centuries. The books which they copied and read have the same sort of story to tell. Ambrose, Augustine, Benedict, Cassiodorus, Eusebius, Fructuosus, Gregory the Great, Jerome, Ildefonsus, Isidore, John Cassian, Julian, Pachomius, Prosper: this is a roll-call of the learning available in about the year 700 -- but these are the authors recorded in Galician documents of the ninth, tenth and eleventh centuries.(57) Only two authors of a later date found a place. Beatus of Liébana's commentary on the Apocalypse, always a popular work in Christian Spain, was presented to the monastery of Guimarães by its founder in 959, and a copy of Smaragdus on the Rule of St. Benedict was given by no less a person than Rosendo of Celanova to the monastery of Caabeiro in 936. The presence of this Smaragdus is the only evidence we possess for knowledge of the Carolingian literary renaissance in western Spain.(58) The canon law which Galician churchmen observed was likewise the product of seventh-century codification. Pseudo-Isidore and the collections edited by Regino of Prüm, Abbo of Fleury and Burchard of Worms are conspicuous by their absence.(59) The liturgy used in Spain was the so-called Mozarabic liturgy, unwashed by the Franco-Roman liturgical currents which had flowed over the remainder of the Christian west. Their books were copied in the 'Visigothic' script peculiar to [26] Spain, its forms betraying no influence of the Carolingian minuscule. The very few churches which survive, such as Rosendo's chapel at Celanova, are markedly conservative in style, recalling the architecture of the Visigothic age. The clergy of Galicia were not travellers. They had no dealings with the popes, and we hear of no pilgrimages to Rome or further afield.

The organization of the Galician church was also an old one. As was the case elsewhere in Christendom, the structure of ecclesiastical administration had been grafted on to the administrative framework of the later Roman empire. The earliest bishoprics in Galicia were probably situated in the towns which gave their names to the three conventus juridici of the Roman province of Gallaecia -- Braga, Lugo and Astorga. The surviving correspondence generated by the Priscillianist controversy and the chronicle of Hydatius make it clear that there were bishops of these places by the middle years of the fifth century at the latest. Three precious documents from about a century later cast a shaft of light upon the ecclesiastical organisation of Galicia in the last days of the Suevic kingdom.(60) These are the records of the first (561) and second (572) councils of Braga, and the exceptionally interesting document known as the Suevic Parochiale, a list of the principal churches of each diocese in the metropolitanate of Braga drawn up about 580. Their testimony shows that there were by then thirteen episcopal sees in north-western Spain: Braga, Coimbra, Idanha, Viseu, Lamego, Porto, Dume, Lugo, Tuy, Orense, Astorga, Iria and the ecclesia Britonensis.(61) Two of these sees were of an unusual type. Dume, or Dumio, just outside the city of [27] Braga, was the site of a monastery founded by St. Martin of Braga. The Parochiale says of it simply, Ad Dumio familia servorum. This is probably to be interpreted, as Pierre David argued, as indicating that its abbot enjoyed the dignity of a bishop and exercised jurisdiction over the monks and the tenantry (familia) of the monastic estates. The ecclesia Britonensis, now Bretoña, was the seat of a bishop who ministered to the spiritual needs of the British immigrants to north-western Spain: in 572 its bishop, Mailoc, had a Celtic name.

The sixth-century evidence conjures up before us an active church, engaged in the foundation of rural churches and the slow christianization of the countryside. So much is suggested by the Parochiale, by the evidence for the foundation of proprietary churches, by Martin Of Braga's tract De correctione rusticorum and by conciliar concern for the pastoral duties of a bishop. It is also clear that monasticism was spreading in Galicia during the Suevic and Visigothic periods. Hydatius referred to a community of nuns (virgines Dei) at Braga, there was at least one monastery among the British settlements, Martin of Braga founded more than one, nearly all the houses in the monastic 'connection' of St. Fructuosus were in Galicia, and a seventh-century inscription attests a monastic establishment at Samos.

The structure of the Visigothic church was disrupted by the Islamic conquest and short-lived occupation of Galicia. But its memory survived. The restoration of ecclesiastical life in lands reconquered during the ninth, tenth and eleventh centuries was powered by the desire to resurrect the arrangements of the Visigothic period. In Galicia at least -- for matters took a somewhat different course elsewhere in Spain -- this desire was fulfilled. The diocesan structure with which the young Diego Gelmírez grew up in the 1070s and 1080s was that of the seventh century. (There is one important qualification to be made. After the Islamic conquest the bishop of Dume resided at Mondoñedo (Mindunietum), near Bretoña, and the ecclesia Britonensis ceased to exist as a separate see. The bishopric of Mondoñedo emerged in impenetrably obscure circumstances from a fusion of the two authorities. It was for this reason that its bishops styled [28] themselves variously Minduniensis, Dumiensis or Britoniensis during the tenth and eleventh centuries.)

Monastic life also was disrupted, and probably disappeared altogether from Galicia in the eighth century. A convincing case for continuity of regular religious life at any site has yet to be made. The ninth century saw several refoundations, notably at Samos, and some new foundations, often under the influence of clergy who had migrated from the Muslim south of Spain; and the tenth century, as we have seen, was the great age of monastic expansion in Galicia. Here again the emphasis was on restoration, on return to a revered and cherished past. The tenth-century revival had little if any contact with similar contemporary movements elsewhere in Latin Christendom. Despite Rosendo's copy of Smaragdus, the degree to which the revival diffused observance of the Benedictine Rule in north-western Spain seems to have been' small. Foreign influences on Galician monastic life were barely felt before the arrival of the Cluniacs towards - the end of the eleventh century.

The evidence, sparse though it is, is all of a piece. From the eighth century until the eleventh Galicia was effectively cut off from the community of western Christendom. Poised on the margin of the Christian world, hers was a world apart. Matters were not to remain thus for much longer.

Notes for Chapter One

1. HC, p. 128.

2. HC, p. 534; AHN 1749/21.

3. R. 13 June 1071.

4. HG, pp. 179-80; DMP, no. 27. Cold winter weather sometimes brings reports of sightings of wolves into the Galician local press even today.

5. AHN 491/13; Sobrado Cart. I, no. 131.

6. HC, p.534; DMP, no. 1; R. 29 July 1118. Might the dog Ulgar have been English? English dogs were much sought after abroad, and Ulgar is a possible rendering of some such name as Wulfgar. We shall have something to say of Anglo-Galician trade in a later chapter.

7. AHN 491/5. Segunda may have been inferior grain not good enough for milling or seed, but suitable as animal-fodder.

8. AHN 1325C/19 (2,3).

9. e.g. AHN 1325D/8 bis.

10. AHN 1067/2; A. López Ferreiro, Don Alfonso rey de Galicia y su ayo el conde de Trava (Santiago de Compostela, 1885), apéndice x.

11. AHN 512/10.

12. LSJ, p. 359; HC, p. 121; Jubia Cart., no. xxiv.

13. AHN 5 12/6; 1325C/13; Sobrado Cart. II, nos. 101, 132; Carboeiro Cart., no. xxii; AHN cód. 986B, fo. 132v-133r.

14. A. Rodríguez Colmenero, Galicia meridional romana (Bilbao, 1977), pp. 197-214; AHN cód. 986B, fo. 151v; compare also R. 28 March 959.

15. AHN 512/6, 1749/21.

16. Sobrado Cart., I, no. 198; AHN cód. 1043B, fo. 55v.

17. HG, pp. 534-5. Wooden sledges set with flints and drawn by mules were used for this purpose in Spain until recently. I have seen them for sale in the market of León, and in use on the meseta,though not in Galicia -- but I have never been in Galicia at harvest-time.

18. e.g. AHN 1325D/7 (of 1152).

19. M. R. García Alvarez, Galicia y los gallegos en la alta edad media (Santiago de Compostela, 1975), I, ch. y, especially pp. 180-2.

20. García Alvarez, op. cit., pp. 88, 237-57.

21. P. Bonnassie, La Catalogne du milieu du Xe à la fin du XIe siècle (Toulouse, 1975), I. 86-91. Professor Bonnassie describes the figure of 18/km2 as 'presque effarante'.

22. García Alvarez, op. cit., pp. 114-20, is good on this.

23. L. Vázquez de Parga, 'Los documentos sobre las presuras del obispo Odoario', Hispania 10 (1950), 635-80, doc. no. I; R. 7 June 910; AHN cód. 1043B, fo. 38v (pd. ES XL, ap. xxiii, pp. 403-4).

24. Sobrado Cart. I, no. 2.

25. C. Sánchez Albornoz, Una ciudad de la España cristiana hace mil años. Estampas de la vida en León (5th edn., Madrid, 1966), passim.

26. F. Fita, 'Los judíos gallegos en el siglo XI', BRAH 22 (1893), 172-7.

27. Many references are collected in M. R. García Alvarez, 'Moneda y precios del ganado en la Galicia altomedieval,' Cuadernos de Estudios Gallegos 24 (1969), 363-94.

28. AHN 1081/17; Jubia Cart., nos. x, xii.

29. The use of this equivalence was first established by L. G. de Valdeavellano, 'Economía natural y monetaria en León y Castilla durante los siglos IX, X y XI', Moneda y Crédito 10 (1944), 28-46.

30. HG, p. 199.

31. AHN 1509/14.

32. AHN cód. 986B, fo. 57r;Madrid, IVDJ, C.9/25.

33. Sobrado Cart. I, no. 127.

34. AHN 491/4. The crucial words refer to peasants de maladia sive christianos quomodo progenie de paganos. An emendation from maladia to muladia helps a little, and is justified by the consideration that copyists of the thirteenth and later centuries frequently misread Visigothic a for u and vice versa.

35. This was the suggestion of C. Verlinden, L'esclavage dans l'Europe médiévale: I. Péninsule ibérique (Bruges, 1955).

36. Spelt variously creatio, criacio, criazio, etc.

37. AHN 1508/8; R. 13 June 1071; Jubia Cart., no. v; HC, p. 419.

38. AHN 512/6.

39. C. Sánchez Albornoz, 'Pequeños proprietarios libres en el reino asturleonés', in his Investigaciones y documentos sobre las instituciones hispanas (Santiago de Chile, 1970), pp. 178-201; M. R. García Alvarez, Galicia y los gallegos en la alta edad media (Santiago de Compostela, 1975), I. 72-111.

40. Sobrado Cart. I, no. 132.

41. Sobrado Cart. I, nos. 39, 40.

42. Discussed by Sánchez Albornoz, art. cit.

43. AHN 1508/6, 8, 10, 16, 17.

44. In what follows I am indebted to the remarkable researches of Dr Emilio Sáez, especially his 'Notas al episcopologio minduniense del siglo X', Hispania 6 (1946), 1-79, and 'Los ascendientes de San Rosendo', ibid., 8 (1948), 3-76, 179-233.

45. The Lorenzana foundation charter is printed in ES XVIII, ap. xvii, pp. 332-40. This and other foundation charters of the period cry out for critical editing. The process might reveal the charter to be not altogether trustworthy, but I believe it to be sufficiently reliable to sustain the general point made here. Count Osorio's charter for Lorenzana survives in two different versions, AHN cód. 1044B, fo. 1r-8v (a separate quire bound into the cartulary) and 9r-14v. A future editor will need to take account also of a further single-sheet copy in AC Mondoñedo.

46. AHN 1067/1, 2.

47. For the context of Egeria's pilgrimage see now E. D. Hunt, Holy Land pilgrimage in the later Roman empire AD 312-460 (Oxford, 1982). The reader should bear in mind that the Roman province of Gallaecia was larger than the Galicia of today: to the east it included Astorga on the meseta and to the south stretched as far as the river Duero.

48. For Hydatius see most recently E. A. Thompson, 'The end of Roman Spain, I', Nottingham Medieval Studies 20 (1976), 4-18 (who to my mind exaggerates the remoteness of late antique Galicia from Roman culture). Hydatius has most recently been edited by A. Tranoy in the series Sources Chrétiennes (Paris, 1974).

49. For the foregoing see J. N. Hillgarth, 'Visigothic Spain and early Christian Ireland', Proceedings of the Royal Irish Academy 62 (1962), section C, pp. 167-94; E. A. Thompson, The Goths in Spain (Oxford, 1969), pp. 23-4, 68, 87-8; X. Barral i Altet, La Circulation des monnaies suèves et visigotiques (Munich, 1976), pp. 125-30; E. James, The Merovingian archaeology of southwest Gaul (British Archaeological Reports, Supplementary Series, no. 25: Oxford, 1977), pp. 221-5; M. Herren, 'On the earliest Irish acquaintance with Isidore of Seville', in Visigothic Spain. New Approaches, ed. E. James (Oxford, 1980), pp. 243-50; E. James, 'Ireland and western Gaul in the Merovingian period', in Ireland in early medieval Europe, ed. D. Whitelock, D. Dumville, and R. McKitterick (Cambridge, 1982), pp. 362-86, at pp. 373-4.

50. Ed. Tranoy, cc. 131, 171, 194.

51. C. Sánchez Albornoz, '¿Normandos en España durante el siglo VIII?,' in his Orígenes de la nacíon española. Estudios críticos sobre la historia del reino de Asturias, II (Oviedo, 1974), pp. 309-21.

52. Sampiro, Cronica, in J. Pérez de Urbel, Sampiro, su crónica y la monarquía leonesa en el siglo X (Madrid, 1952), at pp. 340-1; Cronicon Iriense, ed. M. R. García Alvarez, Memorial Histórico Español 50 (1963), pp. 1-240, c. 11; Sobrado Cart. I, no. 137; AHN cód. 1043B, fo. 38v.

53. Printed and discussed by R. Pinto de Azevedo, 'A expedição de Almanzor a Santiago de Compostela em 997, e a de piratas normandos a Galiza em 1015-16', Revista Portuguesa da História 14 (1974), 73-93. It may have been in the course of this raid, which lasted nine months, that Tuy was sacked.

54. HC, p. 15, Jubia Cart., no. ix.

55. HG, p. 197.

56. Many citations of Lex Visigothorum occur in charters of this period: see for example AHN cód. 986B, fo. 26v-27r (of 1052), and Sobrado Cart. II, no. 392 (of 1086).

57. M. R. García Alvarez, 'Los libros en la documentación gallega de la alta edad media', Cuadernos de Estudios Gallegos 20 (1965), 292-329.

58. On the diffusion of Smaragdus in the Peninsula see now A. Linage Conde, Los orígenes del monacato benedictino en la península ibérica (León, 1973), II. 794-801. All surviving MSS from Spanish scriptoria before c. 1100 come from Castile or further east.

59. A. García y García, Historia del derecho canónico: I. El primer milenio (Salamanca, 1967), pp. 176-84.

60. See the important study of P. David, 'L'organisation ecclésiastique du royaume suève au temps de Saint Martin de Braga', in his Études historiques sur la Galice et le Portugal du VIe au XIIe siècle (Lisbon-Paris, 1947), pp. 1-82.

61. Coimbra, Idanha, Viseu and Lamego were situated in the civil province not of Gallaecia but of Lusitania, whose administrative capital was at Mérida (Emerita). Because these areas were politically subject to the Suevic kings until the conquest of their kingdom by the Visigoths in the 580s, so their bishops had become ecclesiastically subject to the metropolitan see of Braga. At some point between 653 and 666 they were returned to the province of Lusitania at the instance of the metropolitan of Mérida. Disputes between the metropolitans of Galicia and Lusitania over their respective suffragan sees were to persist for many centuries.