TARSHISH-TARTESSOS, THE EMPORIUM REACHED BY KOLAIOS OF SAMOS. By F. González de Canales.

Fernando GONZÁLEZ DE CANALES CERISOLA “Tarshish-Tartessos, the Emporium Reached by Kolaios of Samos” CIPOA 2, 559-576. PHÉNICIENS D’ORIENT ET D’OCCIDENT. MÉLANGES JOSETTE ELAYI. Cahiers de l’Institut du Proche-Orient Ancien du Collège de France (CIPOA) II. Édité por André Lemaire avec la collaboration de Bertrand Dufour and Fabian Pfitzmann. Editions Jean Maisonneuve. Paris. 2014 TARSHISH-TARTESSOS, THE EMPORIUM REACHED BY KOLAIOS OF SAMOS* Fernando GONZÁLEZ DE CANALES CERISOLA Centro de Estudios Fenicios y Púnicos Abstract: Documentary evidence contributed by archaeological finds over the last decades in the city of Huelva (southwestern Spain) tends to converge with the references transmitted by written sources, if adequately interpreted, to the point of confirming the identity of Phoenician/Hebrew Tarshish with Greek Tartessos. Keywords: Tarshish, Tartessos, Huelva, Ora Maritima. 1. TARSHISH AND THE EMPORIUM OF HUELVA The toponym Tarshish, of disputed etymology, is the only reference to the Far Occident in the Old Testament. Since this issue has been broadly dealt with in other publications1, as far as Huelva is concerned, I will now limit myself to outline some slight considerations. The exhumation of a wide Phoenician-indigenous pottery assemblage in Huelva, dated between ca. 900 B.C. and ca. 770 B.C.2 according to tradition* I would like to thank Prof. ANDRÉ LEMAIRE, for his invitation to participate in this well deserved homage to Mme JOSETTE ELAYI, and AURELIO MONTAÑO’s help in preparing the English version of this paper. 1 F. GONZÁLEZ DE CANALES, Del Occidente mítico griego a Tarsis-Tarteso. Fuentes escritas y documentación arqueológica, Madrid 2004, p. 169-270; F. GONZÁLEZ DE CANALES – L. SERRANO – J. LLOMPART, El emporio fenicio precolonial de Huelva, ca. 900-770 a.C., Madrid 2004, p. 209-210; idem, “The earliest Phoenician, Greeks and Sardinian ceramics found in Huelva: a support for Tarshish in 1 Kings 10.22”, in VI International Congress of Phoenician and Punic Studies, Lisbon 2005 (to be published); idem, “The Pre-colonial Phoenician Emporium of Huelva ca 900770 BC”, Babesch 81, 2006, p. 13-29; idem, “The Emporium of Huelva and Phoenician Chronology: Present and future possibilities”, in C. SAGONA (ed.), Beyond the Homeland: Markers in Phoenician Chronology, ANES Sup. 28, 2008, p. 631-655; idem, “The Two Phases of Western Phoenician Expansion beyond the Huelva Finds. An Interpretation”, Ancient West & East 8, 2009, p. 1-20; idem, “Tarshish and the United Monarchy of Israel”, ANES 47, 2010, p. 136-163. 2 Understanding by ca. 900 B.C. to the end of the 10th century B.C. until the first half of the 9th century B.C. (F. GONZÁLEZ DE CANALES et alii, El emporio…, op. cit. [n. 1], p. 199), and not excluding, in a wider sense, the second half of the 10th century B.C. (F. GONZÁLEZ DE CANALES, op. cit. [n. 1], p. 242). Radiocarbon records tended to elevate this ceramic dating (A.J. NIJBOER – J. VAN DER PLICHT, “An Interpretation of the Radiocarbon Determinations of the Oldest Indigenous-Phoenician Stratum thus CIPOA 2 p. 559-576 © Maisonneuve, Paris, 2014 560 FERNANDO GONZÁLEZ DE CANALES CERISOLA al ceramic chronology, together with Phoenician inscriptions, numerous remains of industrial, handicraft, agricultural, fishing and commercial activities3, and some elements previously documented, such as the Phoenician wall of San Pedro Hill4, ensures that we are facing an initial emporial stage of western Phoenician expansion. During the following period the Huelva habitat grew to an extension of around twenty densely urbanized hectares, with dwellings of oriental type, and at the same time, the sumptuousness of La Joya necropolis denoted the enrichment of its elites. Relative to these finds, regardless of the first origin of the biblical verse I Kings 10, 22, the problems associated to its transmission, and the margin of historicity granted to Solomon, it is hard to assume that several centuries after the kingdom of Hiram I of Tyre a Hebrew writer would have attributed to him for the first time the provisioning from a distant place (“…once every three years…”) of certain products5, some of which so exclusive, like silver and ivory, whose development by the Phoenicians is attested in Huelva in a period long before the Deuteronomy. As a result, these findings represent an argument in favor of the historicity of I Kings 10, 22 as regards Hiram I and the Phoenicians. The identification of Tarshish with Huelva is strengthened by the facts that the working of all products obtained in Tarshish by Tyre, according to Ez. 27, 12 (silver, iron, lead and tin) and Jer. 10, 9 (silver), are confirmed, and by the position of Tarshish in a Far Occident in other Biblical verses (Jon. 1, 3; implicitly, Ps. 72, 10 and Ez. 38, 13), the Assyrian inscription of Esharhaddon at the beginning of the 7th century B.C. (where Tarshish appeared as Tar-si-si6) and, plausibly, the Phoenician stele of Nora, ca. 800 B.C., provided b-tršš can be read7 on its first line. Anyway, the fact that far, excavated at Huelva, Tartessos [South-West Spain]”, Babesch 81, 2006, p. 3136). 3 F. GONZÁLEZ DE CANALES et alii, El emporio…, op. cit. (n. 1), p. 137-176, 228237 and pl. LXIII-LXXII. 4 This wall, sitting in a “Final Bronze” context according to D. RUIZ MATA – J.M. BLÁZQUEZ MARTÍNEZ – J.C. MARTÍN DE LA CRUZ, “Excavaciones en el Cabezo de San Pedro (Huelva): Campaña de 1978”, Huelva Arqueológica 5, 1981, p. 149-316, esp. 259, was typified following J. ELAYI, “Remarques sur un type de mur phénicien”, RSF 8/2, 1980, p. 165-180. 5 Gold, silver, and ivory, with serious trouble to interpret the Hebrew terms q(w)pym and t(w)kim. 6 R. BORGER, Die Inschriften Asarhaddons, Königs von Assyrien, AfO, Beih. 9, Graz 1956, p. 86. 7 If it is so, the emergence in Huelva of an assemblage of Sardinian vessels (F. GONZÁLEZ DE CANALES et alii, El emporio..., op. cit. [n. 1], p. 100-106 and pl. XXI and LX), and a Phoenician inscription on a Sardinian amphora with an epigraphic analysis by M. HELTZER (ibidem, p. 133, No. 2, and pl. XXXV, 2 and LXI, 2), favor the possibility that the term Tarshish in the Nora stele could refer to Huelva. E. LIPIŃSKI, Itineraria Phoenicia, Studia Phoenicia 18, OLA 127, Leuven 2004, p. 244 TARSHISH-TARTESSOS 561 Tarshish is mentioned in extra-Biblical documents is of paramount importance. 2. GREEKS IN HUELVA With the precedent of Tarshish, I will briefly explain the Greek finds in Huelva, and later I will try to analyze whether this habitat properly answers the information of ancient Greek written sources about Tartessos. 2.1. Greek pottery Since the convincing periodization of B.B. Shefton in 19828 and the synthesis of P. Cabrera in 19909, works to which I shall preferably attend, Greek pottery found in Huelva has experienced a notable increase. In accordance with thousands of shards, mostly unpublished, and taking into account the terrains of the city actually excavated, the Greek vases reaching Huelva could be estimated in hundreds of thousands. The first instant is marked by sporadic quality imports attributed to Phoenician trade. The beginning of this phase, represented by a Middle Geometric II Attic krater10, two Late Geometric Euboean skyphoi and an Early Protocorinthian cotyle, has been enhanced by other 33 vessels of similar or former chronology, among which are to be remarked several Middle Geometric II Attic kantharoi, skyphoi and a jug, as well as some Subprotogeometric I-III Euboeo-Cycladic pendant semicircle plates and skyphoi11. From the end of the 7th century B.C. until ca. 540 B.C. imports, already dependent on Ionian trade, were predominantly eastern Greek ceramics, in which Samian productions have a prominent role. Among other luxurious Attic vessels, some pieces of the Gorgon Painter Circle12, a dynos attributed and 246, recall the old connection between Sardinia and southwestern Iberia, including the similarities between Monte Sa Idda bronze hoard weapons and those found in the Odiel Estuary. 8 B.B. SHEFTON, “Greeks and Greek Imports in the South of the Iberian Peninsula. The archaeological evidence”, in H.G. NIEMEYER (ed.), Phönizier im Westen, Madrider Beiträge 8, Mainz a/R 1982, p. 337–370, pl. 30–32. 9 P. CABRERA BONET, “El comercio foceo en Huelva: cronología y fisionomía”, Huelva Arqueológica 10-11/3, 1990, p. 41-100. 10 B.B. SHEFTON, op. cit. (n. 8), p. 342-343. 11 F. GONZALEZ DE CANALES et alii, El emporio…, op. cit. (n. 1), p. 82-94, pl. XVIII-XIX, LV-LVIII. 12 One of which recently released by M. GARCÍA FERNÁNDEZ – M.C. MAESTRE RUIZ, “Esfinges, Leones y Górgonas en la Huelva Tartésica del siglo VI a.C. La excavación arqueológica del solar C/ Concepción nº 3”, in XII Jornadas de Arqueología y Patrimonio de Aljaraque, Huelva 2009. The finds of this excavation are still under study. 562 FERNANDO GONZÁLEZ DE CANALES CERISOLA to Sophilos or his circle13 and komasts’ cups and skyphoi14, mostly from painters KX and KY, can be accounted for ; an olpe from Kleitias with a dipinto reading “Athenea”15; three Gordion cups, one of which signed by Kleitias16 and another most likely decorated by this same painter17; a blackfigure horse-head amphora18; another black-figure amphora from a possible pupil of the Amasis Painter; Siana cups and band cups from the Little Masters. Also lavish are several Laconian cups19, including one from the Boreads painter20; an Early Corinthian cup21 and a Middle Corinthian cup; two Samian Siana cups with astragalus and ram protome plastic representations22. Notably, none of Eolian, Quios, Miletus and Massalian vessels are missing. Finally, a group has been differentiated characterized by a very clear green yellowish paste and low adherent black decoration, whose absence in the Mediterranean and high representation in Huelva leads us to suspect that it was made in this city. Perhaps other Greek vessels of doubtful provenance also were manufactured in Huelva. From ca. 540 B.C. only some Attic vessels are documented, preferably cups. Historical facts indicate that this type of trade was now dependent on Carthaginian intermediaries. 2.2. Greek inscriptions Some Archaic Greek inscriptions incised after firing on Greek import pottery are documented. On a bowl, perhaps Miletian, the term Niethoi (ΝΙΗΘΩΙ) in archaic Ionian of mid-6th century B.C. has been interpreted as a local anthroponym23, and, alternatively, as a theonym24. On a cup, another 13 E. GARCÍA ALFONSO, “Fragmento de dinos ático de figuras negras”, in M.D. LÓPEZ DE LA ORDEN – E. GARCÍA ALFONSO (ed.), Catalogue of the Exhibition, Cádiz y Huelva: Puertos fenicios del Atlántico, Sevilla 2010, p. 174-175, No. 39. 14 J. FERNÁNDEZ JURADO, “La presencia griega arcaica en Huelva”, Monografías Arqueológicas, Colección Excavaciones en Huelva 1, 2nd ed., Huelva 1985, p. 26-28 and Fig. 8; F. GONZÁLEZ DE CANALES – L. SERRANO – J.P. GARRIDO – J. ORTEGA, “Nuevos comastas en Tarteso”, Revista de Arqueología 120, 1991, p. 14-17. 15 R. OLMOS ROMERA – P. CABRERA BONET, “Un nuevo fragmento de Clitias en Huelva”, AEArq 53, 1980, p. 5-14. 16 M. GARCÍA FERNÁNDEZ – M.C. MAESTRE RUIZ, report cit. (n. 12). 17 J. FERNÁNDEZ JURADO, op. cit. (n. 14), p. 20-23 and Fig. 6, 9. 18 J. FERNÁNDEZ JURADO, op. cit. (n. 14), p. 36 and Fig. 13. 19 J. FERNÁNDEZ JURADO, op. cit. (n. 14), p. 18-20 and Fig. 5; F. GONZÁLEZ DE CANALES, op. cit. (n. 1), p. 321. 20 M. GARCÍA FERNÁNDEZ – M.C. MAESTRE RUIZ, report cit. (n. 12). 21 F. GONZÁLEZ DE CANALES – L. SERRANO, “Consideraciones en torno al Tarteso griego y al Tarsis de Salomón con motivo de unos grafitos hallados en Huelva”, Revista de Arqueología 175, 1995, p. 8-17, esp. 10-11. 22 F. GONZÁLEZ DE CANALES, op. cit. (n. 1), p. 323-325. 23 J. FERNÁNDEZ JURADO – R. OLMOS ROMERA, “Una inscripción jonia arcaica en Huelva”, Lucentum 4, 1985, p. 107-114, esp. 110-111. TARSHISH-TARTESSOS 563 one read ΡΑΚΛΕΟΣΗΜΙ in Cnidian alphabet, which can be transcribed as [‛Η]ρακλέος ήμί (“I am of Heracles”)25. Out of a photograph M. Kerschner considers that this cup could be a “Knickrandschale” (Ionian cup) type 5 or 6 of Schlotzhauer, dated between 670 and 610 B.C.26 The presence in Huelva of people writing in Greek is confirmed by three other inscriptions on local bowls of orientalizing grey ceramic: the first one composed of a sigma and an iota27; the second one, reading IΣΤIAI Δ and including numeral 10, has been interpreted28 as a dedication to Hestia; the third one comprises the reading NIKHΣEI, interpreted as a verbal form for “to win”, as a dedication to Nike or, else, to a local divinity assimilable by the Greeks to their idea of Nike29. The multiracial and emporial character of Huelva is reaffirmed by these Greek epigraphs together with other indigenous or Phoenicians epigraphs on supports of diverse adscription. 3. GREEK WRITTEN SOURCES ABOUT TARTESSOS The confusion around Tarshish amongst the interpreters of Biblical texts since ancient times keeps a curious parallel with similar contradictions over Tartessos. The cause cannot be other than the loss of information about the distant occident owing to the disconnection between the southwest of the Iberian Peninsula beyond Cadiz and the Greek and oriental Phoenician worlds in the second half of the 6th century B.C. In the case of Tartessos its memory remained alive among writers of the Classical, Hellenistic and 24 M. ALMAGRO GORBEA, “Una probable divinidad tartésica identificada: Niethos/Netos”, Palaeohispanica 2, 2002, p. 37-70; idem, “NIETHOS – Néit: The earliest documented Celtic God (c. 575 BC) and the Atlantic relationships between Iberia and Ireland”, in H. ROCHE – E. GROGAN – J. BRADLEY – J. COLES – B. RAFTERY (ed.), From megaliths to metals: Essays in honour of George Eogan, Oxford 2004, p. 200-208. 25 A.J. DOMÍNGUEZ MONEDERO, “Fragmento de copa con inscripción griega”, in M.D. LÓPEZ DE LA ORDEN – E. GARCÍA ALFONSO (ed.), op. cit. (n. 13), p. 60-61, No. 10. 26 I appreciate this information from A.J. DOMÍNGUEZ MONEDERO. 27 F. GONZÁLEZ DE CANALES – L. SERRANO, art. cit. (n. 21), p. 10; F. GONZÁLEZ DE CANALES – L. SERRANO PICHARDO – J.P. GARRIDO ROIZ, “Nuevas inscripciones fenicias en Tarteso: su contexto histórico”, in M.E. AUBET – M. BARTHÉLEMY, Actas del IV Congreso Internacional de Estudios Fenicios y Púnicos, Cádiz 2000, Vol. I, p. 227-238, esp. 230, fig. 2, 5 and pl. 5A-5B. 28 Independently, by R. PEDRERO and L. DUBOIS, in J. LLOMPART – E. M. ORTA – J.P. GARRIDO – F. GONZÁLEZ DE CANALES – L. SERRANO, “Discusión en torno a la lectura y soporte de una inscripción griega arcaica con dedicatoria a la diosa Hi/estia hallada en Huelva”, Huelva en su Historia 13, 2ª época, 2010, p. 3-14. 29 M. GARCÍA FERNÁNDEZ – A.J. DOMÍNGUEZ MONEDERO – F. GONZÁLEZ DE CANALES – L. SERRANO – J. LLOMPART, “Una inscripción griega arcaica hallada en el Cabezo de San Pedro (Huelva)”, SPAL: Revista de prehistoria y arqueología 18, 2009, 93-103. 564 FERNANDO GONZÁLEZ DE CANALES CERISOLA Roman epochs, to whom we owe a good number of speculations arising from their absolute ignorance about the location of the once famous city. During the past century the city-emporium of Tartessos, categorically outlined in the written sources, was replaced by the idea of a culture/civilization of autochthonous character whose contents had to be filled. Under this premise, when research could prove a deep oriental acculturation, a great deal of “the orientalized” was qualified as “Tartessian”, even if links to the Phoenician were undoubted, as in the case of some luxurious objects of bronze metalwork and refined gold jewellery. Thus, the concept of “Tartessian culture” was subject to a permanent state of reinterpretation. Whatever the case, in the written sources the Greek term “Ταρτησσός” does not mean a culture or civilization but a city-emporium and its namesake river. As these sources continue to generate uncertainty, it seems appropriate to inquire whether they have any fundaments. 3.1. Dating of the Greek sources 1. The first references to Tartessos come from the Archaic Greek Period. Without ruling out any dose of fantasy in some instances, like Arganthonios’s longevity according to Anacreon30 and Herodotus31 or, partly, in the Kolaios voyage told by the latter32, and acknowledging the bridge role of Stesichorus between the old myths and reality, when he links Geryon to the geography of the Tartessos River33, we get a series of references offering some very concrete data. 2. Sources born after the Archaic period and until the Roman conquest reflect utter ignorance of southwestern Iberia beyond Cádiz (considered to be unreachable) or revive old western myths in a scenario that again becomes confused34. 3. The last period starts with the reopening of the territories west of Cádiz by the Roman legions. Greek and Roman writers and geographers showed a living interest for the Tartessos of the old sources, which they continued to 30 Anacreon of Teos, in D.L. PAGE, Poetae Melici Graeci, Oxford 1962, frg. 361 (apud Strabo, Geography III, 2, 14). 31 Herodotus, History I, 163, 2. 32 Herodotus, History IV, 152, 1-5. 33 Stesichorus, Gerioneys, quoted by Strabo, Geography III, 2, 11. 34 Such do we appreciate in authors like Herodotus, History III, 115, 1, referring to his time; Pindar, Olympian Odes III, 43-45; Nemean Odes III, 21-22; Nemean Odes IV, 69; Isthmian Odes IV, 12-13; Euripides, Heracles 234; Isocrates, Philip 112; Panathenaicus 250; or Euctemon, quoted by Avienus, Ora Maritima 350-369. TARSHISH-TARTESSOS 565 consider a city, but the many identifications proposed35 prove that the memory of its location had been lost. Consequently, with the tinges expressed above, only the sources originating in the Archaic Period are apt to transmit trustworthy information about Tartessos. 3.2. Interpretation and verification of Greek archaic sources Through Herodotus we know that coinciding with the foundation of Cyrene in northern Africa, ca. 630 B.C., Kolaios of Samos crossed the Pillars of Heracles and reached the commercial emporium of Tartessos36; we also know about the arrival of Phocaeans and their friendship with the Tartessians king (βασιλέι) Arganthonios37. Later on, when their city was threatened by the Persians, the Phocaeans could not go back to Tartessos because Arganthonios had died38. As this fact took place five years before the naval battle of Alalia, ca. 535 B.C., the Ionians must have remained in touch with Tartessos between ca. 630 and ca. 540 B.C. The reliability of these dates is supported by the knowledge of Greeks about events that were not far from their time39. Owing to their activities, the emporium west of the Pillars of Heracles mentioned by Herodotus can only find an answer in Huelva, which three hundred years after the arrival of the first Phoenicians had reached a surprising expansion and affluence. Ionian movements toward this city are confirmed by the emergence of numberless pieces of eastern Greek ceramic between the end of the 7th century B.C. (arrival of Kolaios) and the beginning 35 Between the two mouths of the Baetis River mistaken for the Tartessos River (Posidonius of Apamea, On the Ocean, quoted by Strabo, Geography III, 2, 11; Eustathius of Thessalonica, Paraphrase of Dionysius Periegetes 337, in GGM II, p. 276: 45 - 277: 1); next to the Ocean (Scholia in Iliadem VIII, 479); in an island close to the Pillars of Heracles (Scholia in Lycophronem, Alexandra 643); next to Calpe (Gibraltar) as an ambiguous Tartessis (Eratosthenes, whom Artemidorus contradicts, quoted by Strabo, Geography III, 2, 11), perhaps Tartessos and its immediate territory. It would also be identified with Carteia (Pliny the Elder, Natural History III, 7; Pomponius Mela, Chorography II, 96; Pausanias, Description of Greece VI, 19, 3, who names Carpia as Carteia; Appian, Roman History VI. The Wars in Spain 2,63, who calls it Carpessus) and, frequently, with Gades/Gadir (Pliny the Elder, Natural History IV, 120; VII, 156; Silius Italicus, Punica XVI, 465-467; Arrian, Anabasis of Alexander II, 16, 4; Valerius Maximus, Memorable Doings and Sayings VIII, 13, 4; Sallust, Historiae, Frg. II, 5 [B. MAURENBRECHER]; Cicero, Letters to Atticus VII, 3, 11; Avienus, Ora Maritima 85 and 269-270; Lydus, On signs in the heavens XXXVIII, 1-8). 36 Herodotus, History IV, 152, 1-2. 37 Herodotus, History I, 163, 1-3. 38 Herodotus, History I, 165, 2. 39 About this knowledge, A.J. DOMÍNGUEZ MONEDERO in F. GONZÁLEZ DE CANALES et alii, “The Emporium..., art. cit. (n. 1), p. 647-648. 566 FERNANDO GONZÁLEZ DE CANALES CERISOLA of the second half of the 6th century B.C. (incapability of the Phocaeans to go back to Tartessos) and the Greek inscriptions expounded. With higher precision than Herodotus, Pseudo-Scymnus40 locates the emporium of Tartessos at two days of daylight westward sail from Gadira. This course is very adequate for the 54 nautical miles (100 km), generally against sea currents and westerly winds, separating Cádiz Bay from Huelva (fig. 1), where archaeology proves the existence of a protohistoric city of extraordinary extension, which in no way could be merely imagined by PseudoScymnus in the 2nd century B.C. Pseudo-Scymnus source is 4th century B.C. geographer Ephorus of Cyme. However, the origin of these news may not have come from Ephorus, due to the lack of connection in his time between the Hellenic world and the west of Cádiz, but, as can be deducted from Pseudo-Scymnus’s text, from the period in which Tartessos was in full activity and in touch with the Ionians, whose old records and information Ephorus compiled. May we remark that the punctual location assigned by Pseudo-Scymnus to Tartessos is consistent with that of a city-emporium, as he himself qualifies it, instead of a wide territory like Turdetania became later, when Tartessos was already eclipsed. Undoubtedly, the most detailed information is provided by Avienus’s poem Ora Maritima41. We are obviously interested in the passage dealing with the geography of Tartessos, whose original source does not have to coincide with other passages since Avienus points out (41-50) that he compiled authors from different epochs. Ora Maritima establishes an on foot route between the Gulf of Lisbon and the Tartessian shore (Tartessiorum litus) of scarcely four days, and another one between the Tartessian shore and Malaca feasible in five days (174-182). These distances, whereas the steep final stretch of the second journey is rather harder, take us strictly to the Huelva coast (fig. 2). The Ana River, natural border between the Cynetes of the Portuguese Algarve and the Tartessian territory (Tartesius ager) (222-225), has preserved its name: Guadi-Ana (fig. 1). Between the Ana River and the Tartessians, other verses locate a territory called Hiberia with its homonymous Hiberus River, for which the Hiberians (Hiberos) were named (248-255). As, for reasons that I will justify, the Tartessos River is the Tinto River, there are two chances for the identification 40 Pseudo-scymnus, Periplus 161-164, in GGM I, p. 201. I have considered the following editions of Ora Maritima: P. VILLALBA I VARNEDA, Ruf Fest Aviè. Periple (Ora Maritima), Barcelona 1986, followed in J. MANGAS – D. PLÁCIDO (ed.), Avieno (Testimonia Hispaniae Antiqua I), Madrid 2000, p. 31-169, and J.P. MURPHY, Rufus Festus Avienus. Ora Maritima (Description of the Seacoast from Brittany to Marseilles [Massilia]), Chicago, Illinois 1977. 41 TARSHISH-TARTESSOS 567 of the Hiberus River: either the Piedras42 or the Odiel43 (fig. 1). In favor of the latter, Posidonius, who quotes Onoba (Huelva) as a city of Iberia44, characterizes the Iber River45 as such: “...this phenomenon (oceanic tides)… is general in the whole context of the coastline… Iber River –says Posidonius-… sometimes raises its level… when the northern winds grow strong, blaming the lagoon it crosses, for under the impetus of the winds the lagoon waters would flow towards it” It is not necessary to elucidate that this Iber is the same aforementioned Hiberus River in Ora Maritima instead of the extant Ebro in northeastern Spain. Of late, downtown Huelva underwent severe floods from the Odiel River, which merges with the Tinto in a large estuary (Ligustine Lake: vide infra and fig. 3) before flowing into the Atlantic Ocean. Next to the Hiberus River the Cartare Island (255) is mentioned, plausible location of the city of Tartessos (Qrt = city). This island or peninsula (the Greek suffix –nessos is common to both geographical features) fits the old Peninsula of Huelva, almost an island bounded by the Tinto on the east, the Odiel on the west and the Nicoba Stream on the northeast (fig. 2). After Cartare, the Cassius Mountain appears (259), doubtless identified with the fossil dunes composing the Asperillo Hill on the coastline (fig. 1)46. East of the Tartessians were the Cilbiceni (255), a people related to the Cilbus River, for which there is firm evidence in favor of the Guadalete47. Before the Cilbus River the Besilus is cited (320), from which the Baetis (Guadalquivir) can be derived without much ado (fig. 1)48. At one day’s sail from a river, most likely the Hiberus (Odiel), since it is the last one cited (248), both the city of Gadir and the Tartessian Gulf (sinus Tartesii) shore can be found (265-270). One day’s sail between the Odiel River and the Cádiz area is acceptable if we count upon the favorable and 42 According to J.M. LUZÓN, “Tartessos y la Ría de Huelva”, Zephyrus 13, 1962, p. 97-104, esp. 103. 43 According to A. ARENAS LÓPEZ, El verdadero Tarteso, Valencia 1927, p. 400. 44 Posidonius of Apamea, On the Ocean, quoted by Strabo, Geography III, 5, 5. Plutarch, Sertorius VIII, 1, also mentions imprecisely a region of Iberia in the southwest of the Peninsula in relation to the Betis (Βαίτιος) River. 45 Posidonius of Apamea On the Ocean, quoted by Strabo, Geography III, 5, 9. 46 Besides its location in the passage, this identification is based on the existence of another Cassius Mountain configured by coastal dunes east of Pelusium in Egypt. A third Cassius Mountain is found in Hatay, Turkey, north of ancient Phoenicia. 47 The hydronym is convenient for the Lacibis (Λαχιβίς) of Ptolemy, Geography II, 7, 9, ordo lacilbulensium of CIL II, 1342 and the Lacidula of supl. 5409 at the foothill of the Audita Rock, in present Grazalema and the fountainheads of the Guadalete River as C. PEMÁN claimed, El pasaje tartéssico de Avieno a la luz de las últimas investigaciones, Madrid 1941, p. 80. 48 J.M. LUZÓN, art. cit. (n. 42), p. 103. 568 FERNANDO GONZÁLEZ DE CANALES CERISOLA predominant westerly winds and sea currents. As to the approximation of the Tartessian Gulf to Gadir, incompatible with the two days’ sail between Gadira and Tartessos in Pseudo-Scymnus, it can be attributed to an erroneous identification of Avienus between Gadir and Tartessos in the same verses, proving that he copied from a source with whose geographical contents he was unfamiliar. In another reference (423-424) the Tartessians (Tartesii) extend to the Calactican Gulf (Calacticum sinum), that might belong to an unnamed gulf in verses 205-210 wherein the Ana River flowed; a second option is the gulf where the Guadalquivir mouth was, since the Doñana Spit was still far from embaying it49 (fig. 4) and this fact is reflected in the configuration of the protohistoric sites50. Another problematic location is the mountain of the Tartessians (Tartesiorum mons), dark with forests, near Erythia Island (308-312), i.e.: in the area of Cádiz. However, just like in the case of the Tartessian Gulf shore, this location is nullified because Avienus erroneously identified Tartessos with Gadir. Finally, the mountain of the Tartessians is mentioned before the Besilus and Cilbus rivers (317-321); i.e.: within the Huelva area. Ignoring landscape changes by deforestation, one possibility is provided by the hillocks (“cabezos”) north of the old habitat of Huelva where sector C of La Joya necropolis is located. Nevertheless, the detailed description of the course and features of the Tartessos River (283-298) provides a real clue to the location of the city, as follows: “… The Tartessos River, however, flowing through open fields from the Ligustine Lake (Ligustino lacu), girds the island on both sides with its current. And it does not run through a single bed, neither does it cut through the underlying soil, for, on the side where dawn breaks, it projects three branches (tria ora) over the fields, with a mouth twice double (a precise translation of the original ore bis gemino: ore = ablative singular), also bathes the southern parts of the city. But over the marsh the Argentarius Mountain is projected, so called by the ancients due to its appearance, for its hills shine from afar when the sun hurts its lofty heights with beams of fire. The same river with its waters, in turn, rolls filings of heavy tin and drags the precious metals near its walls”. The only river in the region responding to this description is the Tinto. The Ligustine Lake, where the Tartessos River flowed (into the Ocean), corresponds to the broad Tinto-Odiel Estuary at that time (fig. 3). 49 A. RODRÍGUEZ et alii, “Evolución costera de la desembocadura del Guadalquivir en los últimos 6.000 años (SW de España)”, Geogaceta 20/5, 1966, p. 1086-1088. 50 J.L. ESCACENA, “Fenicios a las puertas de Tartessos”, Complutum 12, 2001, p. 73-96, esp. 75. TARSHISH-TARTESSOS 569 The fluvial island surrounded by the Tartessos River is Saltés (fig. 3), where archaeological sounding has yielded 7th century B.C. Phoenician pottery51. Avienus’s scholars have expressed a great deal of difficulty in identifying the “Tartessos River mouths”. The three mouths (tria ora) it brought to the fields from the eastern side do not fit the interpretation of three river mouths open to the Ocean, but they do fit the Tinto sources52 which so much captivated later Arab authors53: the fountainhead or Lahšar, today called Jarrama Creek, the Alum Fountain, today Peña del Hierro Creek, and the Vitriol (copperas or copper vitriol) Fountain, today Agrio Creek flowing from the Riotinto Mines (fig. 5). This geographic connection with the extraordinary argentiferous deposits justifies the attention paid to the Tinto/Tartessos River and explains Stesichorus’s54 mention of “...the abundant springs of the silverbedded Tartessos River”. As to the twice double mouth (ore bis gemino) with which the river bathed the south (of the territory) of the city (neither “ore” fits here with river mouth, nor any river in the region ever opened to the Ocean through four mouths) coincides with a big inlet 1100 meters wide, today called the Rincón Marsh, which penetrated into the Peninsula of Huelva from the Tinto River by the southeast and, once inside it, divided itself into two new inlets, each one undergoing an additional subdivision (fig. 3). In accord with the poem, the Tinto River, just like the Tartessos River as regards the namesake city, bordered the south of the protohistoric habitat of Huelva from the east. Although the reference appears in other authors (Pseudo-Scymnus55, Stephen of Byzantium56, Eustathius of Tessalonica57), instead of carrying tin perhaps we should infer, somewhat symbolically, that what the Tartessos River dragged was silver, since tin is related to the Argentarius (of the silver) 51 J. BEDIA GARCÍA, “El mobiliario protohistórico y antiguo”, in A. BAZZANA - J. BEDIA GARCÍA (ed.), Excavaciones en la Isla de Saltés (Huelva), 1988-2001, Monografías de Arqueología 22 , Sevilla 2005, p. 230-259, esp. 240-241, 244-246. 52 Virgil, Aeneid I, 245, mentions Ora as a synonym of “tributaries” of the Timavus River in Slovenia. His works caused a great deal of influence over later authors: Avienus, Ora Maritima 220: castrorum in usum et nauticis uelamina, even reproduced the same idea in Georgics III, 313: usum in castrorum et miseris uelamina nautis (J.P. MURPHY, op. cit. [n. 41], p. 56). 53 Al-‛Udrī, al-Himyarī, al-Qazwīnī, Ibn Gālib, in D. CATALÁN – M.S. DE ANDRÉS, Crónica del moro Rasis, Madrid 1975, n. 11-14 in p. 90-91. 54 Stesichorus, Gerioneys, quoted by Strabo, Geography III, 2, 11. 55 Pseudo-Scymus, Periplus 165, in GGM I, p. 201. 56 Stephen of Byzantium, Ethnica, s.v. Ταρτησσός. 57 Eustathius of Thessalonica, Paraphrase of Dionysius Periegetes 337, in GGM II, p. 277: 5-6. 570 FERNANDO GONZÁLEZ DE CANALES CERISOLA Mountain identifiable with the Solomon Hill at Riotinto58, whose rich argentiferous ores were intensely exploited in the past. Another possibility is a tin supply from Extremadura and Portugal to the city of Tartessos, because in Pseudo-Scymnus59 this metal was transported through the Tartessos River from the Celtica region. Perhaps this tin was traded through a route reaching Riotinto, then conveyed to the intermediate site of Niebla and later, by the Tinto/Tartessos River, to Huelva60 to be used by its industrial infrastructure in order to produce bronze or be exported abroad. The appearance of tin61 and Portuguese-Extremadurian pottery62 in the context of Huelva before 770 B.C. shows that this trade had been long established. In conclusion, land distances from the Gulf of Lisbon to the Tartessian shore and between the latter and Malaca in Ora Maritima, and the two nautical days from Gadira to Tartessos indicated by Pseudo-Scymnus, take us to a specific place matching the Huelva habitat. Equally important is the description of the surroundings of the Tartessos city, its namesake river and its “mouths” which fit the particular geography of the Huelva Estuary and the course of the Tinto River. Additional comment deserves the dating from the literary analysis applicable to the source of the Tartessian passage in Ora Maritima. Setting aside the rest of the poem, this passage, like the periegetic genre, practiced in the 6th century B.C., describes land routes, mainland villages and some customs and historical events. In any case, it is hard to accept that amongst the Ionians there could be no geographical records about an emporium like Tartessos after some ninety years of relationships, especially when Greek writing is fully attested in the emporium proper. 4. THE “TARTESSIAN TERRITORY” Besides Tartessos, two other cities appear as Tartessian in Stephen of Byzantium: Hecataeus’s Elibirge63 and Ibila, with gold and silver mines, which 58 A. ARENAS LÓPEZ, op. cit. (n. 43), p. 416; J.M. LUZÓN, op. cit. (n. 42), p. 100- 101. 59 Pseudo-Scymus, Periplus 165, in GGM I, p. 201. A Roman way in the Ravena Anonymous went from Onoba to the Celtic Beturia and Portugal passing through Urion, usually identified with the Riotinto old settlement. 61 F. GONZÁLEZ DE CANALES et alii, El emporio…, op. cit. (n. 1), p. 150-151 and pl. XXXVIII, 9 and LXIV, 20. 62 F. GONZÁLEZ DE CANALES et alii, El emporio…, op. cit. (n. 1), p. 108, 191-192 and pl. XXII, 12-24. 63 Hecataeus, in FGH 1, Frg. 38 (apud Stephen of Byzantium, Ethnika, s.v. Έλιβύργη). 60 TARSHISH-TARTESSOS 571 G. Nenci estimates that Stephen of Byzantium took also from Hecataeus64. Regarding Elibirge, one could think about some habitat near the city of Tartessos, like Niebla, the Roman Ilipla, halfway between Huelva and the Riotinto Mines. Whereas Ibila reminisces the protohistoric mining and metallurgical habitat of Solomon Hill, at Riotinto65, the Roman Urion. Perhaps they should be understood as a mining-metallurgical axis ending in Huelva. Obviously, the boundary of the Tartessians in the city of Herna referred to in Ora Maritima (462-463) and identified with Peña Negra de Crevillente (Alicante)66, should not be understood as ethnographic, something absurd, but instead, as an influence of the great city-emporium of Tartessos. Similar interpretation should be given to the island devoted to Noctiluca off Menace/Malaca under sway of Tartessos (426-430). The Tartessian territory has gone through various attempts of delimitation giving credit to toponym markers with -ipo and –uba/oba, to pottery with burnished geometric decoration and, occasionally, to “Warriors’ Stelae” from the southwest of the Iberian Peninsula, but all of these proposals face serious difficulties whose exposure exceeds the limits of this work. Consequently, it seems adequate to restrict the Tartessian territory to the one determined by ancient Greek sources, since any enhancement would prompt us to consider Tartessian other peoples (Cilbiceni, Ileates, Etmaneum, Cempsi) that are well differentiated in those same sources. In any case, though in the presence of a language and material culture exceeding the territory assigned by the sources to the city of Tartessos and its immediate environs, the term “Tartessian culture” would continue being conventional and should not be mistaken for the Greek concept of Tartessos. 5. CONCLUSIONS The city-emporium that the Ionians reached when they sailed beyond the Pillars of Heracles was none other than Tarshish. From this toponym the Greek name Tartessos would derive, sharing the same root and adding the suffix –ssos, so common in Asia Minor. This identification counts today with the necessary archaeological support. 64 Hecataeus, in G. NENCI, Hecataei Milesii fragmenta, Florence 1954, frg. 45 (apud Stephen of Byzantium, Ethnika, s.v. ῎Ιβυλλα). 65 A. BLANCO FREJEIRO – J.M. LUZÓN NOGUÉ – D. RUIZ MATA, Excavaciones arqueológicas en el Cerro Salomón (Riotinto, Huelva), Anales de la Universidad Hispalense: Serie Filosofía y Letras 4, Sevilla 1970. 66 A. GONZÁLEZ PRATS, “Las importaciones fenicias en la Sierra de Crevillente”, in G. DEL OLMO - M.E. AUBET (ed.), Los fenicios en la Península Ibérica, Sabadell 1986, Vol. II, p. 279-302, esp. 280. 572 FERNANDO GONZÁLEZ DE CANALES CERISOLA Ruling out the notion of Tarshish/Tartessos, besides denying any hint of credibility to numerous written sources, would leave without response many issues affecting the nature, objectives and dating of the first contacts between the Phoenicians, the Greeks and the Iberian Peninsula, the mechanism of commercial exchanges, the socioeconomic transformations experienced by the indigenous population, the introduction of the city as a concept, and the origin of paleo-Iberian writing, that is to say: it would hamper any viable proposal for historical reconstruction. TARSHISH-TARTESSOS Fig. 1 Southwest of Iberian Peninsula according to Avienus. 573 574 FERNANDO GONZÁLEZ DE CANALES CERISOLA Fig. 2 Roman routes in the Antonini Itinerarium and the Ravenna Anonymous. TARSHISH-TARTESSOS Fig. 3 Tinto-Odiel Estuary and Huelva Peninsula ca. 6th century B.C. 575 576 FERNANDO GONZÁLEZ DE CANALES CERISOLA Fig. 4 Guadalquivir Gulf ca. 2600-2300 B.C. according to A. RODRÍGUEZ et alii, art. cit. (n. 49), fig. 1-C. Fig. 5 The three “mouth” (tria ora) of the Tinto/Tartessos River.