A Shipwreck on the Dalmatian Coast and Some Gold Coins of Romanus III Argyrus Metcalf, Michael Greek, Roman and Byzantine Studies; Jan 1, 1960; 3, 2; ProQuest pg. 101 A Shipwreck on the Dalmatian Coast and Some Gold Coins of Romanus III Argyrus Michael Metcalf G OLD COINS OF ROMANUS III have frequently been found in cen­ tral Dalmatia. They have come particularly from the upland districts of that province, rather than from the coasts or islands. The abbe Bulic, who directed the archaeological museum at Split from 1884 until 1926, twice commented on the fact that pieces of this particular emperor were often found. Single finds and small hoards, he said, had been discovered throughout the whole of central Dalmatia, but particularly in the uplands. l The gold of Romanus III (1028-34) is not common; the only provenance re­ corded by Mosser in his Bibliography is the important Sofia hoard of 1897, deposited in the reign of Alexius I. Byzantine gold of whatever period is far from plentiful in Dalmatian finds. The only eleventh century hoard is an unpublished one from the time of Alexius I (1081-1118), represented by a few coins recently acquired by the Split museum. Such Byzantine hoards as are known are characteristically either from the islands or from the little ports along the coast. The Dubrovnik find of 1892 and that from the site of ancient Narona in the lower Neretva valley, discovered about 1901, belong to the group associated with the ports2 ; a Beneventan solidus struck in the name of Charlemagne, IF. Bulie, in Bttllettino di Archeologia e Storia Dalmata III (1895) 78, where he says, 'E da notari che in varii incontri furono trovati in tutta la Dalmazia media monete d'oro di questo imperatore, rna special mente nella parte montana della provincia'; and the same, 'Un piccolo ripostiglio di monete bizantine', BASD 36 (1913) 60, where three further find spots are listed. 2See S. MeA. Mosser, A Bibliography of Byzantine Coin Hoards (New York 1935), for references. (Dubrovnik under Ragusa.) The list of ports may be extended by adding the bronze coins found at Durres (Durazzo) in 1924 (attributed to Basil; Albania [1925] 28£.) and at Senj before 1863 (John I [or successors?]; Archiv xxix, 336). 101 Copyright (c) 2003 ProQuest Information and Learning Company Copyright (c) Duke University, Department of Classical Studies 102 MICHAEL METCALF [GRBS 3 found at Trogir in 1937,3 may be classed with them. Two island finds of sixth and seventh century gold have been recorded from Krk and Brac.4 The parcel recently acquired at Split comes from Korcula.o Finds of gold from the hinterland are rare. The only such eleventh century Byzantine coin from Jugoslavia of which I know is a nomisma of Alexius I, discovered at Trebinje in the Herzego­ vina.6 Valentini remarked that stray finds of nomismata in northern Albania were not as commonly known as one might expect, con­ sidering the esteem in which the coins were held by peasants in modern times.7 Some explanation for the finds of Romanus III was obviously required. Bulie suggested that the interior of the province of Dalmatia in the eleventh century was more prosperous and was the scene of greater commercial activity than the coast. This explan­ ation does not seem very probable but it shows that Bulie felt clear about the facts. Cedrenus chronicles an incident which affords a much more likely explanation. In the year 1040, when the emperor Michael IV was at Saloncia, John "Orphanotrophos", his brother and chief minister of state, sent him ten centenaria of gold by ship. The action sufficiently characterizes the two men: while Michael visited the tomb of St. Demetrius and made widespread gifts of money in the hope of a providential improvement in his health, John was concerning himself energetically with the administration of the em­ pire and its finances. The gold very possibly would have been used to defray the expenses of Michael's Bulgarian campaign; but the ship 3The coin is now in the Split museum. I am indebted to the Director for the courtesy with which the accession registers were made available to me. 4Por the Krk hoard, see Viestnik Hrvatskoga Arheoloskega Druztva 4 (1882) 124f.; and for the Brae hoard, J. Klemenc, 'Nalazi novaca u Jugoslaviji, 1910-1936', Numizmatika 2-4 (1934-6) 124ff. 5There are four coins, of Romanus IV, Michael VII and Alexius 1. It seems likely that most if not all of the sixteen solidi of Theophilus described by I. Marovic in 'Zlatnici cara Theofila u numizmatiekoj zbirci arheoloskog muzeja u Splitu', Vjesnik ;::;a Ariz. i Hist. Dalmatinsku 55 (1953) 213ff., are from a single discovery. 6The coin was acquired by the Zagreb archaeological museum. I am indebted to Professor M. Gorenc for the courtesy with which he made the numismatic accession registers available to me for study. 7G. Valentini, 'La numismatica in Albania (esperienze di un collexionista)" Numismat­ ica 5 (1939) 122ff. Copyright (c) 2003 ProQuest Information and Learning Company Copyright (c) Duke liniversity, Department of Classical Studies 1960J GOLD COINS OF ROMANUS III ARGYRUS 103 was caught in a storm and was wrecked on the Illyrian coast. Stefan Voislav, the ruler of Zeta, is said to have appropriated the treasure and ignored a letter from the emperor threatening war if the gold was not returned. In doing so, he was taking the oppor.;. tunity to join in a more general movement of revolt among the Slavonic peoples of the western part of the Balkan peninsula. The object of the uprising was no doubt to take advantage of the death of Basil II ("the Bulgar-slayer") in 1025 and of the weakness of the rulers who succeeded him. The mountains which lie between the Vardar valley and the Adriatic are a difficult territory in which to campaign. An army sent against Stefan Voislav under the com­ mand of Georgios Provatas, Cedrenus reports, was lost without hav­ ing achieved its aim.s The treasure doubtless remained in the hands of the Slavs. It seems more than likely that the Dalmatian finds of coins of Romanus III once formed a part of it. Five of that emperor's nomismata were discovered at Gornie Ogorje in or shortly before 1895 in the neighbourhood of the Stari Gaj (the Old Wood), and 13 more were subsequently reported to have been found nearby. Ogorje is a small village in the hills twenty miles north of Split, not easily accessible from the coast.9 Another hoard was found in 1912 in the Sinjskopolje (where the Split airport now is). The exact locality was given as J abuka, on the hill Osoj. Jabuka and Osoj are both common placenames in Dalmatia, so that it is difficult to be certain of the spot to which Bulie was referring. It may have been the hill Osoje, near Gala.10 Eight nomismata were found; they all passed into private collec­ tions.ll Single finds are recorded from Dugo Polje, a village about ten miles northeast of Split, from the Poljica, a hilly district between Split and the lower Cetina valley, and from the little island of Bisevo.12 All five localities are within the orbit SCedrenus II, 527. 9Its exact location is 43.44 N., 16.28 E. The best detailed gazetteer of Jugoslavia is Yugoslavia. Index gazetteer showing place names in 1:100,000 map series, 7 vols. (Cairo 1944) . 101£ so, the location is 43.43 N., 16.44 E. llSee J. Klemenc, op.cit. 12The fimlspot is given as Dugopolje by Bulie. Once again, it is a common name; the reference would have been much more satisfactory if the position had been more adequately described. The village of Dugo Polje ten miles from Split is located 43.35 N., 16.37 E. The Poljica is presumably the district at about 43.30 N., 16.40 E. Bisevo, which was formerly called Busi, is at 42.58 N., 16.0 E. Copyright (c) 2003 ProQuest Information and Learning Company Copyright (c) Duke University, Department of Classical Studies 104 MICHAEL METCALF [GRBS 3 of Split, so that at first glance one might think that the shipwreck was probably on the coast nearby. But they are likewise within the orbit of the Split museum, which would be more likely to acquire finds which came from the neighbourhood than those made at a greater distance. If similar coins had been found further to the south, they might well have escaped the record. One other coin of which the provenance may be significant is mentioned below. The distribution of the finds cannot be said, then, to offer clear evidence of the scene of the shipwreck. It is perhaps just worth pointing out that a ship making the direct crossing of the Adriatic from Pescara might be expected to pass close to Bisevo or Vis.18 There are specimens of the gold of Romanus III in most national collections. The Vienna cabinet has six, for example, all acquired before 1875, and there are nine at Budapest, which are also early accessions. Few if any have provenances?· Under the ticket of one of the coins in the British Museum (BMC 2) there is, however, a note which reads, "One similar specimen found in Servia shown at B.M. 1904." The boundary of Serbia in 1904 lay a good hundred miles from Split at the nearest, and in any case the exact find-spot is not recorded, so that there are insufficient grounds for seeing any connection with the treasure-ship. One coin of Romanus III has been published from Transylvania. It was found at Caransebes, in the Timisoara district, before the first world war.Hi As is usual in monetary studies of the earlier middle ages, many more provenances would be welcome. Michael IV had been emperor for six years at the time of the Dalmatian shipwreck and yet none of his coins have been reported from Dalmatia, either in the hoards of Ogorje and J abuka or separately. His nomismata are scarcer than those of Romanus; there 13Cf. John's remark about the Roman republic, Cedrenus, loc.cit. The Via Valeria, running east from Rome, reached the Adriatic at Pescara. It is a little surprising that Voislav should have controlled, as apparently he was thought to do, territory so far afield as Dalmatia: we have only the Byzantine statement that the gold fell into his hands, and he may have been willing to take any pretext for an attack on the army sent against him. The few details which are known about eleventh century Serbian history have recently been discussed by V. Laurent, in 'Le theme byzantin de Serbie au XIe siecle', Revue des etudes Byzantines 15 (1957) 185ft 141 must thank Dr. Koch and Dr. Fillep for their help. 15Numizmatikai KozlOny 13 (1914) 26; and see 1. Sabau, 'Circulatia monetara in Transilvania secolelor XI-XIII, in lumina izovoarelor numismatice', Studii fi Cercetari de Numismatica 2 (1958) 269ff. Copyright (c) 2003 ProQuest Information and Learning Company Copyright (c) Duke University, Department of Classical Studies 1960] GOLD COINS OF ROMANUS III ARGYRUS 105 were none in the Sofia hoard.16 Gold may have been struck only for Michael's official and ceremonial requirements and not in quantity for general circulation. In any case, the preservation of a large store of gold in the imperial treasury must often have re­ sulted in the drawing of sums in old coin to meet some heavy expense such as that of a campaign. The single pieces and small hoards from Dalmatia, where there was no regular monetary economy employing coinage (as is shown by the absence of finds of petty coins) and where the gold of other emperors does not occur, suggest that these nomismata, acquired by some particular chance, were treasured by the people living in the hinterland as medals, or for the social esteem which their possession conferred, or at any rate for reasons not connected with the commercial use of coinage. Novak has discussed in detail the way in which Slavs and Latins lived side by side in Dalmatia.1'f He argues that from early times the Latin element in the popula­ tion was small, and that Split, among other towns, was "slavicised" well before the time with which we are here concerned. The dis­ tribution of coin finds of Romanus III, taken together with the find spots of other Byzantine hoards, shows how closely the trade and monetary affairs of Dalmatia were confined to the little com­ munities of the seaboard and to the islands, and offers a reminder that it was not until the late middle ages that there were many commercial contacts with the hinterland.18 It is characteristic of the extent of Byzantine Dalmatia that when the inhabitants of Salona abandoned their city and fled to the safety afforded by the 16The best account of the hoard is in Comptes rendus de l'academie des inscriptions d belles lettres, 4th Series, 25 (1897) 303f, but it awaits proper publication. Note that P. Grierson's analyses of fineness ('The debasement of the bezant in the eleventh century', Byzantinische Zeitschrijt 47 [1954] 379ff) confirm the traditional attributions to Romanus III and Michael IV and add the tetarteron, formerly given to Romanus IV, to Romanus III's issues. 17V. Novak, 'The Slavonic·Latin symbiosis in Dalmatia during the middle ages', Slavonic and East European Review 32 (1954) Iff. 18But the peace treaty of 1186 between Ragusa and the Grand Zupan suggests that trade may have reached further than the circulation of coinage. The middle ages afford other examples of the restriction of monetary transactions to the ports where trader and producer met. The peace treaty is printed in S. Ljubic, Listine 0 odnolajih izmedju juznoga slavenstva i mletacke republika vol. i (1868) No. XVII, and in Monumenta Hun­ gariae Historica, Diplomataria vol. xi, 105. The latter version includes the agreement that at Portus Narentis 'denarios toJIatur, per quos antea acceptum fuit.' Copyright (c) 2003 ProQuest Information and Learning Company Copyright (c) Duke University, Department of Classical Studies 106 MICHAEL METCALF [GRES 3 walls of Dioc1etian's palace, the site which was menaced by the Slavs was still within their view. The numismatist cannot often show connections between de­ posits of coins and specific events, although he may suspect them.19 It is gratifying, therefore, that a story of shipwreck can be attached to the coins of Romanus III from the Split museum. In 1177 there was another instance of misadventure on the Dalmatian coast about which interesting numismatic evidence has survived to our own day, although this time by means of a document. Some pirates from the neighborhood of Sibenik attacked a ship which was sailing down the Adriatic and overreached themselves so far as to rob a cardinal and apostolic legate. They called down on the heads of the Archbishop of Split and the Bishop of Trogir the displeasure of Alexander III, who threatened the whole district of Sibenik with a general interdict, excluding only penance and infant baptism, if the cardinal's belongings were not returned. A clerk drew up a list of the missing goods, among which various kinds of coin were named: sterlingi, tarreni, and oboli masmutini as well as perperi were among them.20 If one day an English '·Cross-and-Crosslets" penny is dug up in, for example, Rogoznica, we shall know that the cardinal did not recover all his goods. ST. JOHN'S COLLEGE, CAMBRIDGE UNIVERSITY February 1960 19There is bound to be a certain amount of inference in any such connection; the degree of uncertainty depends on the accuracy with which the coins can be dated. A. Blanchet (,Les rapports entre les depots monetaires et les evenements militaires, politiques et economiques. III. IV.', Revue Numismatique 4, xxxix [1936] 205ff) in writing about the French feudal coinages, where chronology is notoriously difficult, in­ dulges in flagrant circular argument to arrive at connections which no doubt are correct. 20See Monumenta Hungariae Historica, Diplomataria xi, nos. 77-8, pp. 124ff. Copyright (c) 2003 ProQuest Information and Learning Company Copyright (c) Duke University, Department of Classical Studies