1Northern Iberia has left no early medieval polyptyques, which means that we do not have the kinds of written source about land-use and land management that Jean-Pierre Devroey has done so much to illuminate1. Indeed, thinking about northern Iberia in the early middle ages is in any case problematic because there is an acute hiatus in available source material for the eighth and ninth centuries. That hiatus occurs in the first instance because of the Muslim conquest of the early eighth century and the creation of the Emirate of al-Andalus in mid-century, encompassing more than half of the peninsula2. However, there is some written material that can shed light on land-use and even some that can throw light on land management before the development of the great monastic estates of the central middle ages3.

2The monastery of San Millán de la Cogolla lies in the Rioja in the hills on the south-western side of the Ebro valley. Its origins may well lie in the late sixth century but its growth into a powerful institution dates from the tenth century, following extension of political control of the Ebro valley by kings of Pamplona (Navarre) from round about 923, kings whose patronage of San Millán becomes clear from the 970s4. Documentation of the monastery’s interests survives from the mid-tenth century and by the early eleventh century the foundation was receiving major gifts from aristocratic patrons5. During the eleventh century it established a substantial portfolio of property, by gift and purchase, becoming one of the major landowners of the region, a process thoroughly analysed by Professor García de Cortázar6. The Becerro Galicano is a late twelfth-century cartulary including records of these properties, arranged to meet the demands of eleventh-and twelfth-century political interests7; it includes twelfth-century forgeries, designed to emphasize the patronage of the kings of Navarre and counts of Castile, especially in the early to midtenth century; but it also includes material from a lost early twelfth-century cartulary, the Becerro Gótico, into which were copied records acquired by San Millán in the eleventh century as it enveloped many smaller monasteries of the region, together with their properties and their archives8. It is these earlier archives, and the information they reveal about the property interests of the monasteries of Old Castile, that form the subject of this paper.

The Texts

3Early medieval records of property conveyance and acquisition are for the most part charters, of standard format. Throughout western Europe charterwriting was heavily formulaic: whether a longer or a shorter text, a charter was largely composed of a great number of standard formulas, the texts only varying in the specifics of persons and of property conveyed. Across northern Iberia the same formulas were used over and over again in the ninth and tenth centuries, from central Portugal to Asturias and from Galicia to Cataluña, formulas which have analogues in the practice of many places beyond the Pyrenees. Although there are plenty of variants, the level of standardization overall is very striking9. Remarkably, most of these very common formulas do not occur in the pre-eleventh-century charters of Becerro Galicano at all and those that do feature occur very rarely. So, for example, the exceptionally common disposition formula, which announces that a donor or vendor is making the transaction of his or her own free will, does not occur at all: placuit nobis atque conuenit nullius cogentis imperio nec suadentis articulo sed propria nobis accessit uoluntas ut…; nor does the formal handover formula, de meo iure sit abrasa et in tuo iure sit tradita. The almost invariable liberam habeatis potestatem only occurs four times and all of those occurrences are in fabricated charters10. The common sanction clause que in iudicio/concilio uindicare non ualuero ([for which] I cannot provide a valid confirmation in court) does not occur at all; the very common disclaimer at the start of a sanction, quod fieri minime credimus (which we scarcely believe could happen), only occurs twice; and the sanction formula listing those prohibited from interfering in a transaction, an ego an quemlibet subrogata persona (whether I myself or any substitute), only occurs twice. The latter exceptions are in royal charters of the 990s and one 959 record of a gift to San Millán11. By contrast, the extremely common meseta sanction Si quis contra hanc factum meum ad disrumpendum uenerit uel uenero is exceptionally common throughout the Becerro Galicano, occurring 260 times in the infinitive version, and looks as if it was added at the time of compilation of this cartulary12; for example, the fabricated charter of 929 attributed to King García Sánchez has Si quis homo, ex meis propinquis aut extraneis, hoc meum factum disrumpere voluerit, fiat a Domino Deo maledictus et confusus13. The absence of many standard formulas is both striking and remarkable, given their wide-spread occurrence elsewhere.

4If standard formulas are absent, what can we say about the pre-eleventhcentury charters of this cartulary? While the forgeries certainly do have charter form, as do the credible royal grants of the late tenth century, with dispositions, dates, sanctions, witness lists and usually some initial protocol, many of the remaining pre-eleventh-century texts of the Becerro Galicano cannot be called charters. They do not even approximate to charter form and lack its essential elements. Some, however, do and there is a small group of texts that we could reasonably call charters. For the most part they record gifts in favour of monasteries and churches absorbed by San Millán in the eleventh century, and they date from 867—to Orbañanos, Acosta, Obarenes, Quijera, Hiniestra, Barticare, Herrán, Arce14. The content of these unusual charters is intrinsically credible, partly because they have a very individual diplomatic but also because, with one exception, they come from a limited area lying to the north west of San Millán, from the Sierra de la Demanda to the valley of Tobalina, largely within the modern province of Burgos and the primitive county of Castile15. This distribution appears to represent a coherent zone of scribal practice distinct from the standard tradition. Given these two attributes, it seems reasonable to propose that these charters derive from credible products of the ninth and tenth centuries.

5However, much of the pre-eleventh-century material does not have charter form. This material consists of notes of acquisitions made by the beneficiaries of a property transfer. These notes are inconsistent in form: occasionally they have an associated sanction or a date and sometimes a few witness names, but most do not have witness lists; they lack formal protocol; they have very few formulas and they are—in particular—short on expressions conveying the freedom to control the property acquired. So, for example, there are two sets of bare tenth-and early eleventh-century notes from Salcedo, the first set with dates and the second set without, neither having witnesses; or there is a list of individual sales to Hiniestra, and three gifts, with an attached sanction; or a set of three notes in favour of Santiago de Villapún-Maurdones, each note without witnesses, date or sanction16. So, for example, the Hiniestra list of sales takes this form:

Ego Nunnu Durgania dono pro me anime in molina de Aslanzone, ad Sancti Emiliani, una vice.
Ego Gonzalvo Nunniz vendo in iii solidos alia vice et iii argenzos.
Ego Argelo et meo filio vendimus in ii solidos duas vices.
Ego Alla vendo in xi argenteos una vice.
Ego Feles Nunniz dono vice.
Ego Didaco Albura vendo tres vices in iiii solidos.
Ego Alvaro Zema dono tres vices.
Ego Alvaro Ovecoz et Beila Didaz vendimus duas vices in solido et vi argenteos.
Ego Blasco Gomiz et mea soror vendimus media vice in una novellam.
Ego Munnio Vacoda vendo una vice in vi argenteos.
Abite vendidit una vice in vi argenteos.
Ego Monnio Ferrero et meo germano vendimus una vice in vi argenteos.
Ego Beila Monnioz, i vice in vi argenteos.
Ego Godemiro de Aslanzone, una vice in vi argenteos et duos solidos.
Ego Endura, una vice in i solido.
Ego Lili, i vice in vi argentos.
Ego Monnio Nunniz, una vice in v aranzatas de cera, et tres vices sunt de Salitus abba.
Ego Nunnu Durgania, cum filiis meis, qui sumus venditores atque, pro animabus, donatores firmamus perhenniter ad tibi, Salitus abba.
Et sunt xxv vices.
Nos, omnes supra scriptos, unicuique singillatim ad honorem Sancti Emiliani de Fenestra roboramus supra scripta hereditatem.
Si quis homo ex nostris propinquis vel extraneis hanc nostram offertionem in aliquo disrumpere voluerit, sit a Deo omnipotentem maledictus, et a parte comitis exsolvat lxa solidos argenti; et ad regula parte, duplet hereditatem17.

6While one of the Villapún-Maurdones notes has this form:

In Dei nomine. Ego Nunu placuit mihi, pro salute anime mee, trado ad regula Sancti Iacobi me ipsum cum facultate mea, id est: uno orto et una ferragine in loco qui dicitur Rateziella, iuxta aqua qui currit ad Sancti Martini: de alia pars, penniella de eras super villa; et in illa tova de parras, mea tercia parte. Et in molino de Robuela, ii feria, die et nocte, de viiio in viii dies. Et una vinea in foio de Robuela, iuxta rivo maiore Flumenziello. Et uno agro latus Rivo Maiore. Et in molino de Biggas, iii feria, die et nocte, de octo in octo dies. Et una faza deorsum Calzata de Mirone. Et agrum de Dolquiti, latus villa: de alia pars, Beila. Isti ab omni integritate serviant ad Sancti Iacobi18.

7Whereas charters record transactions, these notes are different for they record acquisitions: the principal interest of the recorder was in the goods acquired rather than in the moment of acquisition.

8All of this suggests that it was the standard practice of monasteries in this area to keep records of acquisitions. There is nothing intrinsically incredible about the practice nor about the content of the notes, particularly since their subjects are often of very small scale. Take the example of six notes made in favour of the monastery of Taranco, running from the early ninth to the early tenth century: they include a gift of an apple orchard and of a small time-share in a mill19. The circumstantial detail is so particular, and the properties so small, it is extremely unlikely that such content could have been made up from nothing. There is of course no corroborative evidence, and there could well be errors here and there, but the notes offer credible texts. If monasteries were making notes of their acquisitions, in effect they were keeping registers—a practice similar in principle though not in scale to compiling a polyptyque. There seem to have been several such monasteries already in the ninth century—Taranco and San Miguel de Pedroso from early in the century, Salcedo, Oca, Acosta, and perhaps Herrán and Hiniestra, as well as those which are not evidenced in text before the tenth century but which, like San Millán, may have pre-tenth-century origins; most of these places were keeping records into the early eleventh century20. They lie in a coherent block, north west of San Millán, in the modern provinces of Burgos and Álava.

9From a purely diplomatic point of view, it is evident that there was a further stage to record keeping before much of this material was copied into either Becerro: different monasteries gathered separate notes together into composite documents, recording the acquisition of a range of different properties, at different times. From their appearance in the Becerro Gótico this often seems to have been an early eleventh-century stage, as is in any case implied by the dates of the final items of some lists—like the long list from Obarenes, ending with a date of 7 May 1009, or the two Salcedo composite documents, ending with a date between 1004 and 1035, and the Hiniestra composite, which has entries to 103521. One of the Salcedo lists is rounded off with a confirmation clause, naming virtually all the donors mentioned, and a sanction; and the Obarenes list is rounded off with sanction, date and confirmation list. The three Villapún-Maurdones notes were put together and completed as a composite with date, sanction and list of witnesses22. What the redactors seem to have been trying to do in constructing these composites was to create something that looked like a charter rather than a set of notes, using confirmation clauses which looked like witness lists.

10These monastic houses were therefore developing their own recording practice, using lists and notes to compile an overview of their holdings. As an aspect of Iberian recording and archiving in the early middle ages, this tells us that there was a lot more to record-making than creating charters. But it also has an economic significance because the purpose of making and keeping such lists and notes was to know about each monastery’s property. The scale was of course much smaller than that of the spread of properties of the great monasteries of northern France and the Low Countries, but the kind of interest was comparable, as was the intention to manage and profit from the acquisitions. These records were far from being polyptyques but their construction into composite documents functioned as a kind of landlord survey.

Early monasteries and their estates

11It is quite clear that the early ninth-century monastery of Taranco, situated 100 km north west of San Millán in the Valle de Mena, had landed possessions, just as other ninth-century monasteries did. It is also clear that the land was cultivated; references to planting, barns, mills and planted pasture imply that cereals and green fodder were grown23. The properties specified in the ninth-century notes relating to Taranco also include apple orchards and a stand of enclosed woodland for pannage, presumably for pigs24; and there is one unusually explicit reference to a time-share in a mill of one day and night a week, every Tuesday25. The so-called foundation charter of 800, which may well be a retrospective construction, adds horses, cattle, sheep, goats, pigs; some vineyards and wine presses; and pasture and meadows26. All of this emphasizes a mixed but planned agricultural regime. Since the arable and orchards were in scattered locations, and since Taranco absorbed several churches together with their appurtenant properties during the ninth century, it is clear that the totality of this endowment was considerably more than could be worked by the eight persons of the foundation charter and indeed more than they needed to sustain themselves27.

12The Becerro Galicano includes a number of would-be “foundation charters”, like that of Taranco. Whether or not they actually reflect a single moment of foundation we cannot possible know, but they certainly offer a constructed view of property holdings, presumably from a member of the community, at a significant stage in a monastery’s early history. One view of an initial endowment has been plausibly reconstructed by Juan José Larrea for another monastery in this region, San Román de Tobillas, from a later charter with questionable elements28; Abbot Avito’s supposed bequest in 822 shows the monastery’s interests stretching some 10 km to the north, with arable plots in a series of villages, woodland, sernas and also some salt pans farther to the south east. Another view comes from the less questionable “foundation charter” of Orbañanos, which offers an unusually precise description of a ninthcentury endowment; it is worth consideration in some detail29.

13The village of Orbañanos sits on the south bank of the river Ebro, in the Merindades region of Burgos province, where the Ebro cuts canyons through the limestone of a landscape reminiscent of the Dordogne. Protected from both the harsher climate and Arab raiding parties of more exposed areas to the south, these fertile, orographically complex valleys consistently offer the earliest testimonies of post-711 communities in northern Iberia. In Orbañanos the valley floor has been submerged as a result of the modern damming of the Ebro in the nearby Sobrón gorge, flooding some of the most valuable land and with it in all probability some of the place-names preserved in the 867 text which forms the basis of this section. Nonetheless, modern cartography allows a remarkably high proportion of the early endowment of the monastery of San Juan, by the “founding” abbot Guisandus, to be identified, revealing a coherent and compact estate in Orbañanos itself (at 600 m above sea-level), complemented by a second endowment on a similar scale in the more arid Obarenes (at 750 m), a now abandoned village 8 km to the south east.

14The topography of the Orbañanos estate is marked by the proximity of the Ebro: cannare, irrigation (?) channels; rivo, the river bank itself; saucto, a riverside wood, “soto” in modern Spanish; Aqua Fierco; nave, a boat and presumably, in a topographical context, metonymically a landing. The properties gifted by Guisandus consist of one ager, five terrae, the channel mentioned above, a vegetable garden (hortus), a ferragine (field sown with green forage), and four vineyards. All lie within a radius of one kilometre from the church, except, rather incongruously, for a single field some 6 km due east in the orographically very aptly named Recuenco (bowl) valley. In what would appear to be a separate gift, Guisandus then added further properties (4 terrae, 1 ager and 1 vineyard) in Orbañanos and again a few kilometres east, this time also encroaching across the Ebro to Sobrón (Villa Semprun).

15To the south east in Obarenes there are four more terrae, and only one vineyard, but several references to apple-trees amidst a variety of other possessions: another ferragine (containing apple-trees), another hortus (also with apple-trees), an apple-orchard in its own right, and a flax-field. Finally, a mill (with its own hortus) is also included, though the phrase “four turns, every eight days” indicates it was shared, perhaps with the rest of the village community, many of them named as neighbours of the various possessions being gifted30. Time-sharing in mills will recur in the next section on the similarly variegated though more abundant, and documented slightly later, possessions of the monastery of Hiniestra.

The topography and geography of Hiniestra

16A particularly good example of a monastery with a range of economic interests is provided by that of Hiniestra. Hiniestra lies at 1,000 metres above sea level and is today a near deserted hamlet situated on the northern flank of the Montes de Oca, a relatively lightly inhabited but heavily wooded area of rolling hills straddling the watershed between the Duero and Ebro river valleys; it lies between the fertile Bureba basin to the north east and the alluvial slopes of the Arlanzón valley to the south west and is some 20 km east of the city of Burgos. The Montes de Oca is an agriculturally marginalized area, valued for grazing and timber, and notoriously dangerous for the pilgrims passing through on the route to Santiago31. Another indicator of the marginality of these hills is the fact that Basque immigrants settled there around this time, as reflected in place-names such as Galarde and Zalduendo.

17What little we know of the monastery of Hiniestra, dedicated jointly to the saints John and (confusingly) Millán, results from its incorporation into the domain of San Millán de la Cogolla in 1052—from this point “San Millán” in this paper will refer exclusively to San Millán de la Cogolla, while we will refer to San Juan and San Millán de Hiniestra simply as “Hiniestra”32. According to San Millán’s Becerro Galicano, Hiniestra was founded shortly before 947 under the protection of the prestigious count Fernán González, although some of the Hiniestra texts imply an earlier history, such that we might think of the 947 event as something of a prestigious restart33. By the early eleventh century Hiniestra had acquired numerous possessions in a number of rural communities through dozens of donations and purchases34. Taken together the Hiniestra material, as yet unstudied as a block, must constitute one of the largest and earliest dossiers for the study of rural Castilian society in the tenth century. In what follows we will focus on three contrasting and complementary aspects of Hiniestra’s possessions: the monastery’s core estates in the Montes de Oca hills; the acquisition of milling rights on the Arlanzón river, 8 km to the south; and the acquisition of arable land and vineyards around the town of Briviesca, some 20 km to the north east (see fig. 1).

18The monastery of Hiniestra itself should be associated with the microtoponym San Millán situated a kilometre south of the village of Hiniestra, from which it is clearly distinguished in the charter BG377 (via qui discurrit ad Fenestra villa). In this initial record of endowment the monastery received a mixture of grazing rights, vineyards, orchards, meadows and mills35, but the only element described in any detail was a rhomboid area of approximately 250 hectares around the monastery, termed its dehesa. This is delimited with a fair degree of precision, the terms of reference being paths (represented by arrows on fig. 2) heading towards a series of identifiable cardinal points: Milanes, circled; Hiniestra, circled; and Oca, a much more important centre and reference point 10 km due east and thus beyond the map36. At the northwestern extreme of the area demarcated the microtoponym La Dehesa occurs, seemingly marking the limit between the monastic dehesa and the village of Barrios de Colina to the north. Similarly, the Alto de la Ermita may well have marked the limit between the village of Hiniestra and monastery’s core estate. The conditions set out for the use of this dehesa refer exclusively to logging, setting severe penalties of 5 solidi per tree illegally felled. Rather more problematically, a second dehesa with the same conditions applying is then delimited, seemingly some 3 km farther east but rather harder to locate37. Regardless of the precise geography of this second area, what is clear is that the main resource being protected are the trees themselves; grazing rights, for example, go unmentioned. Elsewhere too the dehesa was primarily an instrument for the management of woodland. The reference to glandiferos in the early Taranco text mentioned above suggests pannage, but aside from references in texts which are almost certainly later forgeries, the first mention of grazing rights in dehesas in this material dates from 100638. Timber and firewood were undoubtedly important resources, and indeed within the same Montes de Oca we will later see San Millán coming into conflict with neighbouring village communities over such matters39.

19Beyond the dehesa, in the surrounding constellation of villages, though markedly only those to the north (Hiniestra 6 times, Milanes 3 times, [Barrios de] Colina and Arraya), sporadic acquisitions by the monastery are recorded during the century of documented independent existence, at the rate of one a decade, with a concentration in the early eleventh century (see table 1). From such notices it is dificult to make out any systematic process of management or expansion, except perhaps a tendency to consolidate the monastery’s holdings with purchases on the southern edges of these villages and/or just outside (auditus) the villages, and on one occasion explicitly next to one of Hiniestra’s existing possessions (see table 1)40. The purchases (as also most of the donations) are all small scale, and generally of terrae or agri, with two mentions of apple trees, and one turn in a mill.

20Hiniestra also received donations in another dozen locations which are either more scattered (Rábanos, Alarcia (Falariza)?, Cerezo) or now impossible to identify. More significantly, there are also separate occasions on which it is recorded that individuals made donations to Hiniestra of possessions in different villages of the Montes de Oca (Colina, Atapuerca and Hiniestra itself) and simultaneously of other holdings in or near Briviesca, some 20 km to the north east, mirroring Hiniestra’s own split holdings (as we will see shortly) between these two non-contiguous and ecosystemically complementary areas. It is striking that a number of lay people had similarly distributed properties, and these appear to have been wealthier individuals making donations on a grander scale than the modest ones so far discussed. This pattern is repeated four times, and there is not a single case of a hybrid donation involving other locations (see table 2).

21Hiniestra itself had very extensive interests around Briviesca, traditional capital of the agriculturally rich Bureba basin, situated 725 m above sea level41. Hiniestra would acquire here four different privately owned churches, on each occasion by donation from individuals of no stated social rank: St Clement’s, St Tirso’s, St Sebastian’s and St Leocadia’s42. There are references to other neighbouring monastic holdings in these brief donation texts, giving the impression of a densely populated and perhaps highly coveted area: one of Hiniestra’s terrae was next to another owned by Froncea, a monastery some 30 km to the south west on the Arlanzón river; another borders possessions of the brothers of Bovatella, an otherwise unidentified monastery43. Both the count of Castile and the king (of Pamplona?) also had property here44.

22There are at least a dozen further acquisitions, as many as in the immediate surroundings of the monastery (see table 3). Around Briviesca two things are striking. Firstly, the concentration of Hiniestra’s possessions, to the extent that on three different occasions the newly acquired land bordered on a piece already owned by Hiniestra, suggests both very extensive and highly concentrated holdings. Secondly, the majority of these acquisitions (and it is true of all the early ones) are from donations rather than purchases. In other words, it is not a case so much of Hiniestra buying into the area, as that the monastery seems to have had significant influence around Briviesca, despite the distance between the two and the fierce competition from other churches. We can also note the much higher proportion of references here to vineyards when compared to the Montes de Oca, which is in accordance with the ecosystemic contrast between the two areas already noted.

23According to the 947 “foundation text”, the abbot Salitus endowed Hiniestra with all his possessions, including very specifically—in an otherwise rather general list—the mills at Arlanzón45. This is complemented in a separate undated text by a list of 18 transactions detailing Salitus’s acquisition for Hiniestra of turns in the Arlanzón mill (see table 4)46. In contrast to Hiniestra’s acquisitions in both Briviesca and the Montes de Oca, the mill list mainly comprises purchases, though three donations are included as well Salitus’s personal holding (and donation) of three turns. Two different units of account are mentioned (solidi and argenteos), and in at least two cases Salitus paid in kind (a quantity of beeswax and a calf). A turn in a mill tended to cost between 1 solidus and 6 argenteos, although those of Gonzalo Núñez and Gudmir of Arlanzón seem to have been significantly more valuable; the latter’s exceptional usage of the locative surname perhaps suggests a superior social status. Similarly, two of the four donations (those of Alvaro Zema and Salitus) were of multiple shares, again suggesting more prosperous members of the community. Another multiple share-holder, Diego Albura, is perhaps remembered in the name of the nearby hamlet of Villalbura. From their names just two of the sellers, Argelo and Alla, appear to have been women (though in the latter case it is unclear), in addition to the anonymous sister of Blasco Gómez.

24The list is laconic and contains no topographical information other than the reference to the mill and to Arlanzón itself, the name common to both village and river. An attempt has been made, presumably retrospectively, to turn the list into something resembling a charter, repeating the name of the first donor, Nuño Durgania (whose surname is unusual), and presenting the others as if they were his children.

25An even more laconic list from the same period insists on the interests of Salitus and Hiniestra in acquiring mill turns in Arlanzón, with 30 turns being bought from Nuño, Alvaro, Oveco and Gudmir of Arlanzón for 100 solidi in 951, a significantly higher price, three times the rate paid in the undated tenth-century record cited earlier47. Gudmir was clearly the same man as in the earlier text, in which his selling price was exceptionally high. If we similarly regard the Nuño of 951 as the same as Nuño Durgania, here Alvaro could well be Alvaro Zema, the donor of three turns in the earlier record. In other words, these may have been the more prosperous and prominent members of the community, selling larger shares, unless the price differential simply relates to a different and more valuable mill. At the same time, Hiniestra was also acquiring milling rights in Briviesca, at the unidentified Tovas mill. There are records of two separate donations, the second of which, from 947, outlines a steep penalty for appropriating water from the dam during the winter months48.

26A couple of generations later, it is evident that Hiniestra’s interest in milling rights on the Arlanzón had not diminished for it purchased two thirds of the mill of Villabáscones (today Castañares), just outside Burgos, for 100 solidi, in 1017. The price is no greater than the earlier prices, but now the resource is concentrated in the hands of just three parties, where previously it had been shared much more evenly within a group: the sellers themselves, brother and sister Enneco and Totadonna, the monks of Froncea, and a Jew called Citiello49. Indeed, just two years previously Enneco and Totadonna had acquired a significant two-day time-share for 20 solidi50.

Concluding thoughts

27At least from the early ninth century some little-known monasteries were managing substantial properties, such that one has to ask whether, even in the ninth century, they were producing a surplus. The detailed study of Hiniestra suggests that by the mid-tenth century monasteries of this kind could have been producing for distribution. In that case timber from the core estate was clearly highly valued and the monastery accumulated arable plots in different locations within a 2–3 km radius, as well as some vines and orchards. Also essential to its monastic economy were more distant properties 20 km away near Briviesca, where the interest in vineyards is notable, although not exclusive. Many other monasteries of this region had interests in vineyards too, interests which are evidenced from as early as the earliest records, with the rate of acquisition increasing across the tenth century. Hiniestra, like some larger monasteries, was establishing an interest in salt pans by the mid-tenth century51; and acquiring mills also appears to have been deliberate policy from the mid-tenth century, with the monastery taking the significant initiative of buying multiple turns of mill time52. It is very difficult to see this exceptional interest in mill rights as anything other than a deliberate attempt to develop capacity to process cereals on some scale53. Despite the fragmented nature of the source material, what emerges is intensive use of resources and a high degree of spatial specialization, as illustrated by some repetitive but illuminating passages54. This was also the case with monasteries that do not directly feature in the Becerro Galicano, like that of Froncea, which had possessions ranging from Burgos to Briviesca55. With but one significant exception, the Ebro itself, everywhere (Obarenes, Briviesca (Tovas), Milanes, Arlanzón, Burgos) water was being harnessed for mills, and access to it regulated, protected, traded and shared.

28All of this raises critical questions about labour. The number of properties in a monastery’s portfolio, and their distance from the monastic centre, must mean that some of the essential agricultural labour came from beyond the monastery itself, although it does not exclude the possibility that monks themselves participated. It is also clear that some lay donors, without being great magnates, had multiple properties, like Gómez who gave Hiniestra arable and vineyards in two separate areas in 99156. The texts provide no explicit detail of working arrangements and offer very few clues. However, we might note that Diego Beilaz’s gifts to San Millán in 952—a man who was a very wealthy owner with many scattered properties—included people called meos homines and at least three casatos57. Casatos here must mean dependent workers; homines, since they are differentiated, were perhaps free tenants58. Reference to casati is exceptionally rare in this material but BG505 (913) lists four casas and BG342 (971) names seven collazos. All three donors of these properties were evidently aristocrats and one might suppose dependent labour more likely on large aristocratic estates. (There are also references to sernas in this material, especially in association with the owners of multiple properties, but although some sernas had explicit arable use, there is no suggestion in this material that they were necessarily worked by dependent labour59.) By contrast, the many ninth-and tenth-century gifts of single parcels of arable (often bounded by those of other named individuals), as also of fruit trees and patches of vines, look as if they came from peasant proprietors; where the properties were distant from the monastery it must be likely that the donor continued to work the land given, making a regular return to the new owner, in effect becoming a (free) tenant in respect of the donation.

29Overall it is likely that in the ninth and tenth centuries there was a mixture of arrangements for working the land: there could have been agricultural workers who were members of the monastic household (although there is no evidence of this); there appear to have been some dependent (i.e. unfree) workers, living on or near the lands they worked, especially in association with large aristocratic estates; and there appear to have been free tenants, as well as free proprietors. On balance, since there are so few references to dependents and so many small donations and some small sales, it looks as if free tenants dominated the monastic landscape. This area, like so much of northern Iberia pre-1000, seems to have shared the system of exploitation of wealthy resources through free and tied peasant labour, based on the peasant working unit, managed by peasant households paying rent, landlords having little involvement with the management of the farming regime60.

30In the end this is debatable and there can be no certainty about working arrangements at this date. What we can be sure of, however, is that long before the creation of the great monastic estates that are so well known, many monasteries were managing the mixed agricultural regimes of their property portfolios, with an eye on specialized land-use for grape production and on controlling cereal production. By 950 many of these monasteries must have been producing for the market—which presumably explains why San Millán de la Cogolla was so interested in acquiring them across the following century.