Version classiqueVersion mobile
OpenEdition Books

Penser la paysannerie médiévale, un défi impossible ?

 | 
Alain Dierkens
, 
Nicolas Schroeder
, 
Alexis Wilkin

Les polyptyques et au-delà. Produire, échanger, administrer et consommer au miroir des documents médiévaux

The Tivoli breve of 945

Chris Wickham

Texte intégral

  • 1 A. Castagnetti et al. (ed.), Inventari altomedievali di terre, coloni e redditi, Rome, Istituto st (...)
  • 2 G. Luzzatto, Dai servi della gleba agli albori del capitalismo, Bari, Laterza, 1966, p. 13–14, lit (...)

1The definitive edition of the Italian polyptychs, in an edition of 1979 by four experienced Italian historians, published almost every ninth-and tenth-century text which can be compared to the polyptychs of Francia to which Jean-Pierre Devroey has dedicated so much of his academic career—only one has been found since, a third text listing the rights of the bishop of Lucca1. The last of the documents in the 1979 volume, dated to 945 and concerning the lands of the bishop of Tivoli, is by far the least studied; in fact, it is barely referred to in any previous or subsequent work on Italian economic history, or, indeed, in discussions of the history of early medieval Tivoli (few as these are)2. It is, however, an interesting text in its own right, and my tribute to Jean-Pierre is thus an analysis of what it tells us about the economy and socio-political structures of Tivoli in the ninth and tenth centuries. It has to be recognised at the outset that there are not so many peasants in this article; the evidence we have is mostly—as it is also (in fact, still more) for neighbouring Rome in the same period—for élite tenants of the bishop. But what the implications of this are for peasant society we shall see as we proceed.

  • 3 L. Bruzza, Regesto della chiesa di Tivoli, Rome, Della Pace, 1880 [henceforth RT]. The cartulary i (...)
  • 4 I will Italianise all names except those of popes and emperors.

2The Tivoli text is entitled a breve recordationis de casalibus et rebus Tyburtine ecclesiae, and it survives in a short late eleventh-century cartulary of the church of Tivoli, now in the Vatican Archive, which contains 57 folia and eighteen documents, some added in the later twelfth. The whole cartulary was edited in 1880 by Luigi Bruzza, pretty well by the standards of the period; I have checked his readings for the other documents in the volume, although for the 945 text itself I have been happy to rely on Augusto Vasina’s excellent 1979 edition3. Despite its title, which resembles those of other polyptychs and quasi-polyptychs in Italy, this is not actually a list of “villages/estates and properties of the church of Tivoli”. A preface to the document makes its status clearer: it is an exemplar collectionum et brevium iam antea digestorum, “a copy of compilations and notices already summarised/enumerated”, from the times of popes Nicholas, John and Leo, already vetustate consumptos, “consumed by age”, and copied so that their memory not be lost, in Rome at S. Pietro [Vaticano] in the third year of Pope Marinus II (hence the date of 945), at the suggestion of Bishop Hucbertus, Uberto4, of Tivoli. It is not immediately clear why the bishop needed to go to Rome, 25 km away, to copy out his own document, even given that the pope was sovereign over Tivoli (I will come back to the point); but the author of the manuscript had no doubt about it, for he included a picture of Marinus giving the bishop a parchment roll with a third title above the image, cartula pensionum, a “charter of rents”.

  • 5 J.-O. Tjäder (ed.), Chartae latinae antiquiores, Dietikon/Zürich, Urs Graf, 1987, t. 26, n. 808.
  • 6 See above all C. Carbonetti Vendittelli, “Sicut inveni in thomo carticineo iam ex magna parte vetu (...)

3What this text is, in fact, is a summary list of leases from the bishops of Tivoli. It is not a “real” polyptych, but rather resembles the “Alahis list” from shortly before 800, a list of documents for a family in Pisa collected across the eighth century, although the information contained in the Tivoli list is more substantial5. Each one appears in the text in a standardised format: per person X (or his/her heirs, heredes), property Y, at rent Z—as in the case of the first of all, per Ioh(ann) i Pertu, clusura vineata que ponitur in Orgiale, den(arium) i. There are 256 such abbreviated leases, for up to 180 separate tenants, for many appear more than once, or else their heirs do. The list was not finished, since the text ends Per… and there are some lines blank in the manuscript. Whether this was the choice of the copyist around 1100, or of that of 945, or simply a sign of the poor preservation of the 945 text by 1100, is not possible to say. But it is at least likely that the 945 text was itself a copy, perhaps made more authoritative by papal ratification, of an already-abbreviated list of leases, made initially because the leases were “consumed by age”, a phrase usually used for papyrus documents which are being copied onto parchment to make them last longer6.

  • 7 L. Duchesne (ed.), Le Liber pontificalis, Paris, De Boccard, 1981, t. 2, p. 232.

4When do these leases date from? Our first clue lies in the list of popes. Nicholas can only be Nicholas I (858-867). We might assume that John was John VIII (872-882), as John VII (705-707) was probably too early, and later Johns (there were three between 898 and 936) might have been too late for all this copying and consumption by age to be done, although it is worth remarking that John IX (898-900) was the only early medieval pope who was actually from Tivoli7. Leo could well be Leo IV (847-855), but again there were three other, if short-lived, Leos between 903 and 939, and Leo III (795-816) was not so far back in the past.

  • 8 The complete set, excluding the papal documents for Subiaco, consists of L. Allodi, G. Levi (ed.),(...)
  • 9 Inv., p. 259, 269 (Pasquale), 257, 259, 261, 264, 269 (Grimone), 258, 260, 263, 269, 274 (Adriano) (...)

5These observations might make the leases seem to be mostly ninth-century; but we should not take the formulaic vetustas phrase too literally, and certainly not the naming of these three popes, for it is highly unlikely that all the leases derived from the spans of only three papacies. We have two other clues, in fact. The first is that up to five tenants appear elsewhere—not many out of 180-odd, but we are lucky to have even that many, given the small number of Tivoli documents surviving before 945 (only five, in fact, excluding papal grants to the mountain monastery of Subiaco at the other end of the diocese)8. Pasquale primicerius, who appears once, and his heirs once as well, renting the same vineyard in Silicata for one tremissis rent, is possibly a papal official, and the only primicerius of that name known from our (admittedly incomplete) lists is documented between 798 and 801, under Leo III. Grimaldo or Grimone (consul et) dux, whose heirs appear on five separate occasions, must certainly be the Grimone magister militum who is recorded in a now-lost inscription as rebuilding the urban church of S. Paolo in 840, for two of the 945 leases are for that church, and also for the fundus of Balviniano, which in the inscription he gave to the church. These correlations again put their leases, and those of their heirs, squarely into the ninth century; but a Tivoli document of 911 allows us to suppose that some were later too, for in that text, a court-case, the judges were headed by Adriano clarissimus comes and included as iudices Petrunaci and Talaro. All three appear as tenants in the 945 text: Adriano comes does so on several occasions, as might befit a clarissimus, a high-status title; his name is common, which makes the identity less certain, but the other two names are rare. At least some of these leases, I conclude, were only a generation old in 945, notwithstanding the claims for age in the text—or indeed less old than that, for the heirs of Talaro also appear in the 945 text, in a lease which ought to postdate 924, when Talaro is again attested9.

  • 10 Last gold: RS, n. 116; first silver: RS, n. 112.
  • 11 Toubert, Les structures…, op. cit., p. 561–576, 621n; A. Rovelli, Coinage and Coin Use in Medieval (...)
  • 12 Inv., p. 256, 265 (Agnello), 264, 265 (Arnone), 260, 263, 265, 266, 268 (Spasiano), 264, 267 (Leon (...)
  • 13 Rovelli, Coinage…, op. cit., art. iv, p. 8–9, 17.

6This is reinforced by a third clue, the rents. Of these, 102 are in gold coins (tremisses, aurei, solidi aurei, siliquae), whereas 81 are in silver (usually denarii; also some siliquae)— with 16 others in solidi without any indication of metal, which might be added to the solidi aurei but cannot be assumed to be gold automatically. (The others are in agricultural products, especially pigs and wine: see below.) In Roman documents of the period, even if leases are not so numerous before 945, rents are calculated in gold up to the end of the ninth century (to 897), and in the tenth (from 919) are always in silver10. This does not, sadly, mean that we can simply assume that the rents in gold are ninthcentury and those in silver are from the tenth. For a start, the coins actually minted in Rome were all in silver after the 770s, even if gold coins, sometimes called mancosi, a term which does not appear in the 945 text, also seem to have been available in the city in the ninth century. The use of a gold terminology in our text might thus in part have been a hangover, helped by the traditionalism of formularies, as well as reflecting a perhaps intermittent availability of gold in Tivoli; but there was certainly never a period after 800 in which only gold was available11. Furthermore, some of the tenants in the text appear giving rents in both gold and silver in different entries, as with Agnello comes, the heirs of Arnone dux, Spasiano comes and his heirs (significantly, the heirs pay in gold, Spasiano himself in silver), and the heirs of Leonino comes, to name only a few—indeed, one entry asks for both a gold tremissis and a silver denarius12. What objects the Tiburtini (the inhabitants of Tivoli) were actually giving in these rents cannot be known, indeed; they might sometimes have been agricultural rents or unminted metal valued in money, as was common in the nearby Sabina13. But the least we can say is that if 40-45% of money rents were valued in silver in these leases, it is highly unlikely that they all predated 900. I conclude that the leases date from across a substantial period, between ca 800 and ca 930, and that the abbreviation and copying processes took place in a relatively short period of time.

*

7Having established the status of the evidence which the 945 text provides, we can better understand what it tells us about the society and economy of Tivoli: about the agricultural environment of the city, the status of the tenants of the bishop, episcopal estate management, and the nature of agrarian society, insofar as it can be perceived. These will be the core foci of the rest of the article. But before we do this, we need to situate Tivoli briefly in the sociopolitical history of the ninth and tenth centuries.

Fig. 4 – Tivoli and its region.

  • 14 The texts for the 1001 siege are set out conveniently in J. F. Böhmer, Regesta imperii, 2/3, Graz/ (...)
  • 15 For counts, see Pacifici, “Tivoli…”, op. cit., p. 205–206, 219–222, with caution ; RS, n. 154, 186 (...)

8Tivoli lies on the edge of the Campagna Romana, the plain around Rome, on the first line of hills to its east, in a striking location with cliffs around it on two sides (fig. 4). It was part of the papal Patrimonium, the core of the later papal states, from its formation when the popes replaced Byzantine imperial rule in the second half of the eighth century, right up to the Italian conquest of Rome in 1870; its bishop was prominent in papal politics. But Tivoli was not as dominated by Rome as were the other sees immediately around the latter; unlike them, or indeed any other city for some 80 km around Rome, it was a real political and urban centre, with a substantial population and its own local élite, which is very visible in our documents for the city. As a result, its relations with Rome were often very bad. Its revolt against Otto III in 1001, which Otto quelled, according to one source moved the Romans to revolt and expel Otto, because he had been too sympathetic to the defeated Tiburtini; and, later, the similarly generous treatment of Tivoli by Pope Innocent II after a fierce war in 1142-1143 was one of the reasons why the Romans established their own city commune against him14. Tivoli had a count; Adriano in 911, already cited, appears to have been local given his leases, and Berardo in 983, given his unusual ascription as comes civium of Tivoli, may therefore have been the choice of the Tiburtini, but Graziano in 971 is explicitly dux et comes of Tivoli as the representative of (in vice) Pope John XIII. All the same, these are the only texts which cite a count of Tivoli across the tenth century, or indeed the eleventh, and he does not appear in some of the most important ones, such as a remarkable document from 1000 (two months before Otto’s siege) in which 62 nobilissimi viri, acting in the name of all the habitatores of the city, a maiore usque ad minore, ceremonially agreed an annual money rent to the bishopric, seniorem nostrum et defensor, in a clear and momentous act of civic solidarity in the face of danger15.

  • 16 P. Delogu, “Territorio e cultura fra Tivoli e Subiaco nell’alto medioevo”, AMST, 52, 1979, p. 25–5 (...)
  • 17 H. Zimmermann (ed.), Papsturkunden 896-1046, 1-2, Vienna, Österreichischen Akademie der Wissenscha (...)
  • 18 But we might not ; although this claim is unlikely to be an interpolation, topographical overinclu (...)

9Tivoli was thus a serious protagonist, with an active élite, keen on its own autonomy; no wonder the Romans hated it so. It had its own territory or comitatus, the Tiburtino, which was defined by the borders of its diocese; this was largely mountainous, and in the eastern mountains the monastery of Subiaco was in the tenth century extending its power, partly through papal generosity, partly through its own aggrandisement and castle-building; relations between Subiaco and the bishop of Tivoli, who saw his own authority as being undermined, were as a result also poor16. But the bishopric was itself remarkably rich in land, as an enormous papal confirmation from 1029 attests17. Some of that was in the mountains or up the Aniene river valley east of Tivoli; some was in the city itself, which, if we were to take the confirmation literally, the bishop owned the whole of18; and some was on the rich hill slopes and the plain just below the city to the west. This document shows that, even if the leases recorded in the 945 text are very numerous indeed, they by no means covered all his lands. All the same, most of the place names in the leases match up with the bishop’s lands a century later, and make up around half of these; they thus give us a fairly good guide to the nature of his resources, and what he got from them.

  • 19 Inv., p. 255, 257, 266, 271 (Baruniano, divided into unciae [twelfths], half-unciae and scripuli)  (...)

10One thing is evident: the agricultural environment of Tivoli was clearly demarcated. 43 of the leases are for rents in animals, usually pigs but sometimes goats or sheep or chickens; eight are in wine, and one in olives; many of the money rents ask for animals as well. (There are no rents in grain.) And, most importantly, although many of the leases are for estates, whether large or (often) small, called (as was normal in Lazio in this period) by the classical term fundi, many are for still smaller lands, including some individual fields (terrae), olive-groves, and above all vineyards, vineae and clusurae. The fact that some fundi had many leases associated with them indicates that they were often in reality also broken up. One explicit example is Baruniano upstream from the city, with 8 leases of carefully-calculated percentages of it (two of them are doublets, repeating the same information; they are probably lease renewals)—although the bishop claimed it as a single block in 1029, so it was still conceptually a whole, however much divided. Another, less explicit, is Ancariano, for which 4 leases are for the fundus— in one case called in integrum, the whole estate—but two (they are again doublets of each other) are for a vineyard and two future vineyards, plus 5 modia of arable land, a meadow and two houses, i.e. one or two tenant holdings19. This was, plausibly, how many or most of these fundi were in reality managed tenurially.

  • 20 S. Carocci, Tivoli nel basso medioevo, Rome, Istituto storico italiano per il medio evo, 1988, map (...)
  • 21 Wickham, Roma medievale…, op. cit., p. 123–132.
  • 22 For the mills, see Inv., p. 263, 265–266 (a set listed together, probably reflecting an archival c (...)
  • 23 RS, n. 152 ; I. Lori Sanfilippo, “Le più antiche carte del monastero di S. Agnese sulla Via Noment (...)

11In the framework of these divisions, we find some clear specialisations. This is especially the case in the area inside 3 km of the city, which provides over half the leases, as we can see by comparing the placenames of 945 with the casual spatial indications in other early Tivoli documents, plus the identifications on the map provided by Sandro Carocci in his book on fifteenthcentury Tivoli—for many of them, even when not anchored by being names of stable settlements, had not changed in 500 years and more20. The areas of Silicata and Caccabelli, for example, which were both on the hill-slope below Tivoli, and furnish large numbers of leases (36 and 11 respectively), were exclusively dedicated to wine, as was Bisciano (11 leases) on the other side of the Aniene, and Valeria (3 leases) and Orgiale just up the Aniene from the city. This was common in the immediate hinterland of cities: Albano, a smaller city in Rome’s hinterland, shows the phenomenon even more strongly slightly later in the tenth century, and so would Rome in the eleventh21. Less common was the specialisation in olive-cultivation shown in Pesoni (8 leases) immediately to the south of Caccabelli; and it is also interesting that two smallish fundi, Paterno (5 leases), beside the Aniene on the plain just below Caccabelli, and Baruniano, already mentioned, although not described in more detail in terms of their terrain, were characterised above all by rents in pigs (they provide a fifth of all our references to them in the 945 text). Above Silicata and Bisciano, in the narrow valley just under the city, where the Aniene ran off the hills into the plain, we also find areas devoted to mills: Griptulas (7 leases) to the south of the river, Puzalia and above all Trullia (7 leases) to the north. Fifteen of the sixteen mills leased in the 945 list, none of which are obviously doublets, are located here, a large number for the period, which compares well even with the major concentration of urban mills attested along the Tiber in Rome in the tenth century and the two following22. And nor were they the only ones there, for, even if the three mills in Trullia referred to in a document of 984, one of which had been given to a city notable by the bishop, were perhaps some of those referred to in 945, the molas de Besta in a text of 982, in the valley of a torrent running down to the Aniene from the city, seem to have been different again23. There may have been no grain rents in the 945 text, but there was evidently plenty of grain grown in the Tiburtino: further from the city than the vineyard areas just mentioned, but necessarily for the urban market, for grain is milled only shortly before it is used.

  • 24 Wickham, Roma medievale…, op. cit., p. 129.
  • 25 Carocci, Tivoli…, op. cit., map facing p. 524 ; the main element missing in the tenth century is o (...)

12I have just noted that there are parallels in the distribution of vineyards with Albano and Rome, and, later, northern cities too; but it must be said as well that in no other case does one proprietor seem to own the local terrain to such an extent. In Albano, which was dependent on and linked to Rome in a way that Tivoli certainly was not, fifteen churches, mostly Roman, are documented as owning land on the hill-slope below the town in the tenth century and another eleven in the eleventh24; below Tivoli, however, the bishop alone seems to have been in control. Doubtless other owners held here too, as we can see, at least sketchily, for the land inside the city, but the bishop must have dominated over them. We cannot conclude that these specialisations were his choice, and indeed it is not likely, but he was certainly the main person to benefit from them, apart, of course, from his tenants who paid money rents to him and made that money back and more by selling the produce, probably above all to the Tiburtini themselves. What is also striking, then, is the degree to which the Tivoli market was strong enough to allow such specialisations, all of which, not just the vines, indeed resemble the specialisations mapped by Carocci for the fifteenth century25. This is very early—only Rome itself surpassed it in central Italy as yet—and it reinforces our sense of Tivoli as a substantial population focus.

  • 26 Tabelliones : Inv., p. 268, 269 ; other offices in 945, of obscure function (see Pacifici, “Tivoli (...)
  • 27 Examples to 1000: L. M. Hartmann (ed.), Ecclesiae S. Maria in Via Lata tabularium, Vienna, Caroli (...)
  • 28 RS, n. 153-154, 156 (?) ; RT, n. 3. Comites and duces recorded as living outside the city : Inv., (...)

13Most of the bishop’s tenants were, as already implied, the city’s élite. Including references to their heirs, 7 of them are called dux, and 15 comes (plus two comitissa); 26 are called miles or mili (a local Italianisation; we find comi too). Furthermore, almost all the 17 tenants who lease land more than twice in the text (again including their heirs) are duces, comites and milites— all except two, a rural monastery, SS. Cosma e Damiano, up the Aniene at Vicovaro, and Benedetto calciolarius, the shoemaker, to whom we shall return; nearly a third of the total set of leases were to this élite group. Except in a few examples (as with Adriano comes), these titles are almost certainly not offices, but generic status markers; they were also local ones, for most of the names of these figures do not match the naming system of Rome, the only plausible alternative base for such tenants, and the term miles is unknown in Rome in this period. Elegiadorus comi, who appears as a mili later in the text paying the same rent for the same land, shows the generic nature of the terminology well enough, for the two terms here appear interchangeable; the phrase consul et dux, which appears once for Grimone dux, makes the point too, for this is the standard word for “aristocrat” in ninth-and early tenth-century Rome. (Grimone’s 840 inscription, by contrast, gave him a real office, magister militum, a phrase absent in 945. The local offices which appear in other texts, notably iudex and tabellio, are also much less visible: there are no iudices in 945, and only two tabelliones26.) Miles is found in other Lazio cities too, Nepi and Sutri; the only missing title of a similar sort is tribunus, which can be found in Nepi, Orte and (especially) Sutri, but not here27. It can be added that the élite of Tivoli was also urban, not rural, for the most part; the 945 list has very few references to comites or duces living in the countryside. In the city, we have some references to their residences; interestingly, they were often towers in the city’s walls. These were not yet the free-standing tower-houses of the future, but they show an evident involvement in housing of a certain solidity, and maybe also defensibility, which in itself implies an interest in self-projection28.

  • 29 Leases : RT, n. 3, 4 ; for the titles, Wickham, Roma medievale…, op. cit., p. 236.

14It is thus clear that the bishop maintained a clientelar relationship with prominent Tiburtini, from at least the early ninth century, through leasing land to them. This continued after 945 as well, as two leases from the 950s also show (both for locations mentioned in the 945 list)—although by now élite terminology, among lessees and witnesses alike, has shifted, and we find viri magnifici and nobiles viri instead of comites and milites. This may show influence from Rome, as we find the same titles there29. But that did not affect the political attachment of the leaders of Tivoli to their city and their bishop; the actions of the nobilissimi viri of the 1000 document, already cited, who were also a very numerous group, show the point quite clearly. The bishop dominated the city’s substantial élite politically as well as tenurially, and the two processes were evidently related; they probably needed him more than he needed them.

  • 30 M. Lenzi, La terra e il potere, Rome, Società romana di storia patria (Miscellanea della società r (...)
  • 31 Exact doublets might just be double registrations of the same document ; but the smaller number of (...)

15In tenth-century Rome, and in other places in ex-Byzantine Italy (notably Ravenna), the basic lease between a landowner and an aristocratic tenant was the three-generation emphyteusis, which gave extensive rights of possession to the lessee. Such leases were generally for very low rents, a few denarii each; if there was also an entry-fine, to make the transaction less uneconomical (as I think very likely), it was not included in the hardly-changing formulary used for such documents. Three of the four later tenth-century leases of the bishop of Tivoli were indeed such emphyteuses, and the small sums of money asked for most leases in the 945 list fit the scale of Roman emphyteutic rents30. I do not think there can be much doubt that most of the 945 leases were indeed of this type. If so, it is worth wondering whether they were really all for three generations, given that the doublets in which x and/or his/her heirs lease the same land twice are not uncommon31—for, even though the lease list covers more than a century in all probability, many single leases across three lives will have lasted seventy-plus years. This detail might show that Tivoli’s practices were not identical to those of Rome. But this can only be a suggestion, for our indications here are sketchy.

  • 32 Inv., p. 265–267.

16Excluding hypothetical entry-fines, the total sums gained from this entire set of leases were not huge. If we make the (risky) assumption that a gold and a silver solidus had the same value, and if we make rough calculations of the value of rents purely in kind, we only reach the figure of between £6 and £7 per annum, for over a century’s-worth of leases; this seems accurate at least as a ball-park figure, and it is not a large sum for the period. But it is not the case that the bishop simply handed out his land for recognitive sums to gain political influence. For a start, a substantial proportion of the leases went to people who were by no means clearly members of the élite. Furthermore, sometimes we can see gradations of rent which reflect the value of the land more clearly. Examples include the gold solidus paid for a quarter of an olive-grove by the heirs of Spasiano comes, or the 17 denarii paid in two transactions by Theodoro and Anualdo comites and by Spasiano comes, for different sets of mills (mills seem indeed to have provided a fairly standard rent of around 5 denarii each; Benedetto calciolarius had one for that sum, for example), or the 30 denarii paid by the daughters of Leonino comes for a house and garden in the city32. Even then, in each case except probably the last, the lessees will have made rather more out of the land than they paid; but we can see here an attention to valuation which may also extend to some of the vineyard leases for low rents as well, given that some of them were certainly small plots. Both élites and probable non-élites, it can be added, can be seen paying both apparently recognitive and apparently non-recognitive rents; there are no gradations of status here.

  • 33 Inv., p. 260, 264, 273, 275. Cecci is on the Carocci map, cited above, n. 20 ; for the rough locat (...)
  • 34 Inv., p. 254–255, 258, 266 (Maurica), 268, 273.

17In some cases, too, the bishop had a more direct relationship with the exploitation of his lands. This is visible above all in two estates, fundus Cecci in the hills 4 km south-east of Tivoli and the paired fundi of Afloru and Piciano, touching Cecci but further out, around 8 km from the city. One of the leases for Cecci was in money, but the other three were for 30 decimatae of wine each, which in one case had to be brought to the dominatio— a local word for estatecentre, as we know because one of the Afloru leases was for its dominatio33. For Afloru, that latter lease was in money, one tremissis, but a vineyard in the fundus paid 15 decimatae, another portion of the estate paid 130 decimatae and 2 tremisses, and a further portion paid 150 decimatae of olives (in both cases with the obligation to take the rent to the dominatio). Finally, one section of Afloru was leased to the heirs of Maurica de Afloru, evidently a local inhabitant, for 2 gold tremisses, 110 decimatae of wine (to be taken to the dominatio), 8 pairs of doves (tortili), a sheep, 4 pairs of chickens and 16 eggs, 2 modia of pannage (esca), and 8 decimatae more of wine pro viveratica, for drinking, presumably (judging by parallel cases across Italy) by the bishop’s representatives34. This large rent is shot through with extra signs of subjection. Some of the other substantial rents in kind just mentioned were paid by milites and comites, but it is safe to conclude that Maurica and her family were peasant cultivators.

  • 35 Inv., p. 257 (Leone), 260, 274 (Bassulo).
  • 36 Inv., p. 266, 267, 268 (Benedetto), 260, 262, 268, 269 (ferrarii).

18The 945 list of leases does, then, show that the bishop did indeed have direct relationships, not just with local estate-management on at least two estates, but also with peasants. That is backed up by at least one other directly-evidenced case, Leone colonus, the only colonus in the document, who owed 10 decimatae of wine from a portion of Falbiano, and probably also by the parvulus Bassulo, who paid one of the wine rents from Cecci, as well as a money rent for a portion of another estate35. It is also very likely that the numerous people who paid money-rents for vineyards of unspecified sizes cultivated them directly on at least some occasions. When we come to Benedetto calciolarius, who leased a mill for 5 denarii as we have seen, a vineyard in Silicata for one denarius and two others in Valeria for 2 denarii, or else the three examples of ferrarii, smiths, who rented respectively a house in the city and two vineyards, also for money, we cannot be sure whether these artisans exploited their suburban properties themselves, seasonally, or whether (perhaps more likely in Benedetto’s case) they sub-let36. All the same, we are here, too, clearly looking at the strategies of non-élites, in each case in the framework of leases held directly from the bishop.

  • 37 For libelli, Toubert, Les structures…, op. cit., p. 450–493 and Lenzi, La terra e il potere…, op. (...)
  • 38 R. Meneghini, R. Santangeli Valenzani, I fori imperiali, Rome, Viviani, 2007, p. 150 et 153 ; RS, (...)
  • 39 Toubert, Les structures…, op. cit., p. 303–447.

19There are some parallels to this in our other documents. The Afloru leases are very likely to be libelli, not emphyteuses, short-term leases for, often, more substantial rents, and indeed a later bishop issued one such libellus in 990, a 19-year lease of a terra sementaria, arable land, to five laboratores (a standard word for cultivator) in two family groups, who owed 8 denarii and an eighth of the grain and beans they produced: although a grain rent of one in eight is not very high, and each family was headed by a priest, these are most likely to be cultivators37. And we also have one lease from a layman, Caloleo illustris viro, from 963, which backs this up. Caloleo was an interesting man; a member of Prince Alberico’s aristocratic entourage in Rome in the 940s, he can also be associated (in a convincing archaeological argument by Riccardo Santangeli) with the reconstruction of an urban quarter on top of the Foro di Traiano, around what slightly later texts call the Campo Carleo, presumably as a financial investment. But he also had interests in Tivoli, as we can see from his gift of S. Barbara in the city to Subiaco, already mentioned; and in the 963 text he can be seen leasing half of the fundus or casale of Papi, 4 km up the Aniene from the city, to ten laboratores, acting as socii or consortes, i. e. collectively, for an eighth once again of grain and beans, one in twenty of their pigs and sheep, and 50%, an unusually high percentage, of the wine38. These were inhabitants of Papi, because they undertook to sell their portions only to other inhabitants. They were clearly peasants, which further underlines the case made for the 990 text. And they show in this text something of the local collective solidarity which would from this point onwards, and for a century to come, underlie the process of incastellamento which would transform the landscape of the Tiburtino, as indeed farther afield39.

  • 40 Ibid., p. 516-533  ; for Rome, Wickham, Roma medievale…, op. cit., p. 100–103.

20The satisfaction of having got past the veils of our documentation to the peasantry in the Tiburtino on at least a few occasions soon wears off. It is true that this is almost impossible to achieve in our documentation for Rome, even if that is rather more extensive than for Tivoli. But Italy as a whole has hundreds of contracts with cultivators surviving from the ninth and tenth centuries; and for the lands around the monastery of Farfa in the Sabina, not far north of Tivoli, well over a hundred such contracts survive for the century and a half after 950 in that monastery’s cartularies40. All the same, we can at least now see how peasants dealt in ninth-and tenth-century Tivoli. Some of them, on certain episcopal estates, paid serious rents in kind (with some money) directly to the bishop; some of them, doubtless, are among those without titles who are documented paying money-rents for small plots to the bishop in 945; some will have paid kind-rents to the middlemen whose leases are most often attested in the breve, or to independent lay owners like Caloleo. Our main absence here is peasant landowning, so common in most of Italy, even if very rare around Rome. Whether there were owner-cultivators around Tivoli, located between the large blocks of episcopal land, we simply cannot say.

  • 41 Toubert, Les structures…, op. cit., p. 450–493 (in the Sabina, bipartite estates steadily became w (...)

21One more thing, however, is clear in our evidence: around Tivoli (like around Rome, but unlike around Farfa, at least at the beginning of our period), there was no articulated structure to estates; they were just locations for rentpaying41. There are explicit signs of estate-centres in the Tiburtino, unlike Rome in this respect, but there is no sign of demesnes; and, one can add, there is no significant sign of rural unfreedom (it only appears in the formularies of papal bulls, which were notoriously archaising). Quite the opposite: the few leases to cultivators which we do have show a certain level of protagonism and local association. This may be atypical (we cannot say what happened on the majority of estates, and cultivators without written agreements may well have had less protagonism); all the same, that sort of local association would, as just noted, be an important basis for the transformations of the future.

22The lack of organisation in estates brings us straight back to our startingpoint, the nature of the breve of 945. Written leases of portions of land both large and small were, essentially, the main way in which agricultural organisation—as also political organisation—was managed in Tivoli, from the top down, in the ninth and tenth centuries. That is why, I conclude, something as simply structured as the 945 list was worth getting papal ratification for: because it was through such sets of leases that power worked in the society of the diocese of Tivoli. The Tivoli list is in this context not a failed polyptych, but, rather, a guide to how estates worked in just the same way as polyptychs are, only in a different economic reality. Jean-Pierre Devroey has worked all his life on the difficult process of unpicking such rural realities, in their different articulations from place to place. Tivoli offers its own differences, and has its own fascination in precisely that articulation.

Notes

1 A. Castagnetti et al. (ed.), Inventari altomedievali di terre, coloni e redditi, Rome, Istituto storico italiano per il medio evo (Fonti per la storia d’Italia, 104), 1979 ; n. xii, p. 249-275, ed. A. Vasina, is the Tivoli text [henceforth cited as Inv.; the dating to 945 is customary, and I use it here, but it could also be 944. For the new Lucca text, see P. Tomei, “Un nuovo ‘polittico’ lucchese del ix secolo”, Studi medievali, 53, 2012, p. 567–602. I am grateful to Sandro Carocci for a critical reading of this text.

2 G. Luzzatto, Dai servi della gleba agli albori del capitalismo, Bari, Laterza, 1966, p. 13–14, little more than a citation, but giving a useful list of parallel (and not so parallel) texts ; P. S. Leicht, “L’ordinamento feudale nel Regesto di Tivoli”, Atti e memorie della Società tiburtina di storia e d’arte [henceforth AMST], 27, 1954, p. 209–219, a focussed analysis but wholly outdated. P. Toubert, Les structures du Latium médiéval, Rome, École française de Rome, 1973, the basic guide to the economic structures of Lazio, hardly mentions the text. G. Pasquali, “L’azienda curtense e l’economia rurale dei secoli vi-xi”, in A. Cortonesi (ed.), Uomini e campagne nell’Italia medievale, Rome/Bari, Laterza, 2002, p. 5–71, cites it more recently at p. 30–31, making some of the points about the text which are developed here ; see also below, n. 11. Basic overviews of Tivoli : V. Pacifici, “Tivoli nel medio-evo”, AMST, 5-6, 1925–1926, a classic of romantic nineteenth-century scholarship (p. 255–258 for 945) ; for overviews of the city, C. F. Giuliani, Tibur, 1, Rome, De Luca (Forma Italiae, 1/7), 1970, p. 32–40 and passim, problematised and developed by I. Belli Barsali, “Problemi dell’abitato di Tivoli nell’alto medioevo”, AMST, 52, 1979, p. 127–147.

3 L. Bruzza, Regesto della chiesa di Tivoli, Rome, Della Pace, 1880 [henceforth RT]. The cartulary is in Archivio Segreto Vaticano, Archivum Arcis, Armari i-xviii, n. 3658, and the 945 text is at fols. 25v-33r. The “restoration” of the MS in the 1970s has left it in a terrible state, with fol. 56 misplaced and fol. 51 actually lost; much of the text is now too blackened to be readable even with the UV lamp. But Bruzza’s readings can be seen to be for the most part accurate, fortunately.

4 I will Italianise all names except those of popes and emperors.

5 J.-O. Tjäder (ed.), Chartae latinae antiquiores, Dietikon/Zürich, Urs Graf, 1987, t. 26, n. 808.

6 See above all C. Carbonetti Vendittelli, “Sicut inveni in thomo carticineo iam ex magna parte vetustate consumpto exemplavi et scripsi atque a tenebris ad lucem perduxi”, in C. Braidotti et al. (ed.), Scritti in memoria di Roberto Pretagostini, Rome, Quasar, 2009, t. 1, p. 47-69, who shows how papyrus was still a normal medium for documents in and around Rome, well into the tenth century. It is also conceivable that the word exemplar refers to the copying in the surviving manuscript, and that Uberto in 945 had brought the original leases to Rome to be summarised there—possibly by more expert palaeographers ?—but this seems too complicated a procedure.

7 L. Duchesne (ed.), Le Liber pontificalis, Paris, De Boccard, 1981, t. 2, p. 232.

8 The complete set, excluding the papal documents for Subiaco, consists of L. Allodi, G. Levi (ed.), Il Regesto Sublacense del secolo xi , Rome, Reale società romana di storia patria, 1885 [henceforth RS], n. 111, 60, 154-153 and 155—as can be seen from their location, they are all for properties which in the end came under Subiaco’s control (every document for Tivoli before 982 survives either in RT or RS).

9 Inv., p. 259, 269 (Pasquale), 257, 259, 261, 264, 269 (Grimone), 258, 260, 263, 269, 274 (Adriano), 258, 269 (Petrunaci), 265, 274 (Talaro). For Pasquale primicerius, see L. Halphen, Études sur l’administration de Rome au moyen âge (751-1252), Paris, Honoré Champion, 1907, p. 93 (although it must be noted that the church of Tivoli had primicerii too : RT, n. 6, 8, so the identification must remain hypothetical). For Grimone, see Pacifici, “Tivoli…”, op. cit., p. 189-192, for an edition of the seventeenth-century transcription (he reads Grimo for the Primo of the transcription, given the correlation with the 945 text, as would I) ; for Adriano etc., see RS, n. 154, 153.

10 Last gold: RS, n. 116; first silver: RS, n. 112.

11 Toubert, Les structures…, op. cit., p. 561–576, 621n; A. Rovelli, Coinage and Coin Use in Medieval Italy, Farnham, Ashgate, 2012, arts. iii (esp. p. 6n, 13-14, where she cites the 945 text), iv, v (p. 7-15), and see the “Addenda and corrigenda”, p. 1–3, for more recent bibliography. For where the gold came from, the most recent contributions (partially diverging) are P. Delogu, “Il mancoso è ancora un mito?”, in S. Gasparri (ed.), 774, Turnhout, Brepols, 2008, p. 141–159, and V. Prigent, “Le mythe du mancus et les origines de l’économie européenne”, Revue numismatique, 171, 2014, p. 701–728.

12 Inv., p. 256, 265 (Agnello), 264, 265 (Arnone), 260, 263, 265, 266, 268 (Spasiano), 264, 267 (Leonino) ; 267–268 (gold and silver).

13 Rovelli, Coinage…, op. cit., art. iv, p. 8–9, 17.

14 The texts for the 1001 siege are set out conveniently in J. F. Böhmer, Regesta imperii, 2/3, Graz/Cologne, Hermann Böhlau, 1956, p. 782–786 ; see K. Görich, Otto iii . romanus saxonicus et italicus, Sigmaringen, Thorbecke, 1995, p. 99–102 for doubts about the major source, the Vita Bernwardi (which is the one relating the reaction of the Romans) ; G. Althoff, Otto III., Darmstadt, Primus, 1997, p. 169–176, is the most up to date, but a less critical, account. For 1142-1143, see the bibliography in C. Wickham, Roma medievale, Rome, Viella, 2013, p. 495-496 and 502-505.

15 For counts, see Pacifici, “Tivoli…”, op. cit., p. 205–206, 219–222, with caution ; RS, n. 154, 186, 185. For 1000, RT, n. 9.

16 P. Delogu, “Territorio e cultura fra Tivoli e Subiaco nell’alto medioevo”, AMST, 52, 1979, p. 25–54 ; L. Travaini, “Rocche, castelli e viabilità fra Subiaco e Tivoli”, ibid., p. 65–97 ; more recently, P. Rosati, “I confini dei possessi del monastero sublacense nel medioevo (secoli x-xiii)”, Archivio della Società romana di storia patria, 135, 2012, p. 31-62, and L. de Lellis, “Lo sfruttamento agricolo della massa Iubenzana nel medioevo”, in G. Ghini, Z. Mari (ed.), Lazio e Sabina, Rome, Quasar, 2012, t. 8, p. 115–120, references kindly given to me by Camille Rouxpetel.

17 H. Zimmermann (ed.), Papsturkunden 896-1046, 1-2, Vienna, Österreichischen Akademie der Wissenschaften, 1988-1989, n. 585. Nn. 228 and 316, earlier confirmations from 973 and 993, are both forged in their present form, although their content is effectively the same as that of 1029; the 1029 text is itself quite possibly interpolated, but the additions, even if of uncertain scale, do not affect the general point about episcopal wealth in at least the western half of the diocese.

18 But we might not ; although this claim is unlikely to be an interpolation, topographical overinclusivity of this kind is very typical of papal bulls of the period. A gift to Subiaco by a Tivolese priest in 924 (RS, n. 153, cf. 154) of land in and around the city is explicitly in full property, and a diploma of Otto i from 967 claims that Subiaco’s church of S. Barbara, beside the cathedral of Tivoli, was a gift from the Roman aristocrat Caloleo (Die Urkunden Otto i. , ed. by T. Sickel, Hanover, Hahnsche Buchhandlung [Monumenta Germaniae Historica, Diplomatum regum et imperatorum Germaniae, 1], 1884, no 336). All the same, the bishop probably dominated the city tenurially, and both the 945 document and a later text (RT, n. 3) show him leasing out land there. Rome, it may be added, was also dominated tenurially by churches : Wickham, Roma medievale…, op. cit., p. 80–88, 157–158.

19 Inv., p. 255, 257, 266, 271 (Baruniano, divided into unciae [twelfths], half-unciae and scripuli) ; 257, 260, 264, 268, 272 (Ancariano). For doublets as lease renewals, see below.

20 S. Carocci, Tivoli nel basso medioevo, Rome, Istituto storico italiano per il medio evo, 1988, map at end of book ; as he notes on p. 6, much of this topographical work also depends on sixteenth-to nineteenth-century catasti.

21 Wickham, Roma medievale…, op. cit., p. 123–132.

22 For the mills, see Inv., p. 263, 265–266 (a set listed together, probably reflecting an archival collocation), 273, 274, 275 ; Wickham, Roma medievale…, op. cit., p. 187–189.

23 RS, n. 152 ; I. Lori Sanfilippo, “Le più antiche carte del monastero di S. Agnese sulla Via Nomentana”, Bullettino dell’“Archivio palaeografico italiano”, NS, 2-3, 1956-1957, p. 65–97, n. 2.

24 Wickham, Roma medievale…, op. cit., p. 129.

25 Carocci, Tivoli…, op. cit., map facing p. 524 ; the main element missing in the tenth century is orti, which lay immediately below the city in the fifteenth. (Three of the four placeable ortua in 945 were inside the city ; the other was in Trivio, which was indeed close to the city, but it is the only such example : see Inv., p. 257, 267, 269, 273, with RT, n. 3, for Trivio’s location.)

26 Tabelliones : Inv., p. 268, 269 ; other offices in 945, of obscure function (see Pacifici, “Tivoli…”, op. cit., p. 217–230 for over-precise definitions) are castaldi, actionarii, regionarii, vicarii, magistri : see Inv., p. 258, 263, 267, 269, 271, 273. See RS, n. 87, 83, 6, 115, 27, 38 for Roman examples of consul et dux up to 960.

27 Examples to 1000: L. M. Hartmann (ed.), Ecclesiae S. Maria in Via Lata tabularium, Vienna, Caroli Gerold filii, 1895, t. 1, n. 3, 5, 15, 18, 19, 23; P. Fedele (ed.), Carte del monastero dei SS. Cosma e Damiano in Mica Aurea, secoli x e xi , Rome, Società romana di storia patria (Codice diplomatico di Roma e della regione romana, 1), 1981, n. 2, 4, 7.

28 RS, n. 153-154, 156 (?) ; RT, n. 3. Comites and duces recorded as living outside the city : Inv., p. 268, cf. 265, for Spasiano comes, 272 for Maurizio dux. No milites are so recorded. For early exemples of the private use of towers in walls, see A. A. Settia, “Cerchie murarie e torri private : urbane”, Atti del XXI Convegno Internazionale di Studi, Pistoia, Centro italiano di studi di storia e d’arte, 2009, p. 45-66.

29 Leases : RT, n. 3, 4 ; for the titles, Wickham, Roma medievale…, op. cit., p. 236.

30 M. Lenzi, La terra e il potere, Rome, Società romana di storia patria (Miscellanea della società romana di storia patria, 40), 2000, passim ; F. Theisen, Studien zur Emphyteuse in ausgewählten italienischen Regionen des 12. Jahrhunderts, Frankfurt, Vittorio Klostermann, 2003, p. 220–273 ; Wickham, Roma medievale…, op. cit., p. 107–110 (including for entry-fines) ; RT, n. 3, 4, 8. The first Tivoli document of all, RS, n. 111 from c. 760 (the edition gives 758, but in fact it could be any date between 757 and 767 ; it is indeed earlier than any for Rome, although an isolated text), is an episcopal lease of a similar kind, but to another church, which slightly changed the rules.

31 Exact doublets might just be double registrations of the same document ; but the smaller number of cases where the first lease is to a tenant and the second, identical, is to his heirs, must be renewals. For the latter, see Inv., p. 258 and 265 (Agnello and heirs), 259 and 269 (Pasquale and heirs), 260, 268 (Spasiano and heirs).

32 Inv., p. 265–267.

33 Inv., p. 260, 264, 273, 275. Cecci is on the Carocci map, cited above, n. 20 ; for the rough location of Afloru, see RT, n. 4 (there, “Floru”), plus Zimmermann, Papsturkunden…, op. cit., n. 585, the 1029 bull.

34 Inv., p. 254–255, 258, 266 (Maurica), 268, 273.

35 Inv., p. 257 (Leone), 260, 274 (Bassulo).

36 Inv., p. 266, 267, 268 (Benedetto), 260, 262, 268, 269 (ferrarii).

37 For libelli, Toubert, Les structures…, op. cit., p. 450–493 and Lenzi, La terra e il potere…, op. cit., p. 31–40. For 990, RT, n. 6 ; although a cultivator lease, it also included a church.

38 R. Meneghini, R. Santangeli Valenzani, I fori imperiali, Rome, Viviani, 2007, p. 150 et 153 ; RS, n. 145 (942), 93 (963). Caloleo gave this half-estate to Subiaco too ; by 973 (Zimmermann, Papsturkunden…, op. cit., n. 226) half of the casale Papi appears in lists of Subiaco properties confirmed by popes.

39 Toubert, Les structures…, op. cit., p. 303–447.

40 Ibid., p. 516-533  ; for Rome, Wickham, Roma medievale…, op. cit., p. 100–103.

41 Toubert, Les structures…, op. cit., p. 450–493 (in the Sabina, bipartite estates steadily became weaker across the ninth century) ; cf. for Rome : Wickham, Roma medievale…, op. cit., p. 103 and 110–112.

Table des illustrations

Légende Fig. 4 – Tivoli and its region.
URL http://books.openedition.org/psorbonne/docannexe/image/28002/img-1.jpg
Fichier image/jpeg, 137k

© Éditions de la Sorbonne, 2017

Conditions d’utilisation : http://www.openedition.org/6540