INDIAN OCEAN. This subject will deal with the role of Indian Ocean in international trade in the following periods:
ii. Islamic Period. See Supplement.
The Indian Ocean is the third largest ocean in the world, covering some 75,000,000 sq. km. It unites culturally and environmentally diverse regions, such as Arabia and Antarctica, East Africa and Indonesia, Iran and Southeast Asia, and India and Madagascar. Offshoots such as the Red Sea in the west and the Straits of Malacca in the east give access to other marine systems, including the Mediterranean (and by extension the Atlantic Ocean) and the South China Sea (and by extension the Pacific Ocean). With the spread of Islam in the early centuries A.H. maritime connections through and beyond the Indian Ocean were to alter the course of world history dramatically, but it is now clear that navigation and trade in the pre-Islamic era were significant as well.
For many years the existence of maritime trade between southern Mesopotamia, southern Iran, the Persian Gulf, Bahrain, eastern Arabia, and the Indus valley during the late 3rd and early 2nd millennium B.C.E. has been well documented (Gadd, 1932, pp. 191-210; Ratnagar, 1981; Possehl, 1996, pp. 133-208; Possehl, 2002, pp. 325-42). This research is relevant in the present instance, because it provides clear evidence of the knowledge, ability, and experience in the use of the monsoon winds that were necessary for sailing between India and the West on the part of sailors in the region long before the Greek mariner Hippalus. The latter made the supposed ‘discovery’, in the early Roman imperial era, of the periodicity of the monsoons and how to use them for Indian Ocean seafaring (i.e., the southwest wind, May-October, and the northeast wind, December-March; see Kulke, 1998, p. 3; Wiesehöfer, 1998, p. 11). (On Hippalus see Pliny, Natural History 6.100, 104, 106; cf. Rouge, 1988, pp. 59-74; Casson, 1989; Periplus Maris Erythraei, sec. 57; Salles, 1998, p. 51; Schuol, 2000, p. 428.) While the Ptolemies and Romans may have become aware of this seasonal key to trans-Indian Ocean travel towards the end of the pre-Christian era, archeological and cuneiform evidence suggests that Indian, Arabian, and probably Mesopotamian traders had been covering long distances by alternately hugging the coasts and crossing open water (Arabian Sea) for several millennia.
Iranian evidence from the later 1st millennium B.C.E. is scarce. Herodotus says (4.44) that Darius I sent a fleet under the command of Scylax of Caryanda from the mouth of the Indus River to “that place whence the Egyptian king sent the Phoenicians … to sail round Libya,” after which, Herodotus tells us, Darius “subdued the Indians and opened their sea to his ships” (Salles, 1988, pp. 79-81; Karttunen, 1989, pp. 65-68). As the Indians are not mentioned in the Bisotun inscription, it was probably this event which first created the Achaemenid dahyu of Hinduš (DPe 17 ff., DNa 25, DSe 24, DSm 10, XPh 25; cf. Vogelsang, 1990, p. 104). On Darius’s Suez stela (DZc) the Great King records an order to cut a canal from the banks of the Nile “vers la mer qui vient de la Perse … et les navires allaient d’Egypte, par ce canal, vers la Perse, selon mon bon plaisir” (Lecoq, 1997, p. 248; cf. Salles, 1988, pp. 82-83; Tuplin, 1991, pp. 237-283). Leaving aside the many questions raised by these sources, neither in all likelihood reflects commercial or economic concerns, but rather political and military exigencies (cf. the Indian forces in Xerxes’ army; Karttunen, 1989, pp. 49-50). The payment of taxes, particularly in gold dust (Herodotus, 3.102), by the Indians of the twentieth nomos obviously implies great wealth but not necessarily active commerce with Achaemenid Iran and may, in any case, be a Greek interpolation on the part of Herodotus into an Achaemenid tax roll (Vogelsang, 1990, p. 105). However, if the testimony of Ctesias’s Indika, excerpted by Photius (Henry, 1947, pp. 62-63, secs. 4-5), is correct, then there was certainly knowledge at the Persian court in the late 5th century B.C.E. of gold, iron, onyx, and other precious stones in India (Vogelsang, 1990, p. 106). Whether messengers between Persepolis and Susa, and India (e.g., Hallock, 1969: PF 1318 8-9; PF 1383 9-11; PF 1397 1-4; PF 1556 5-7) during the period covered by the Persepolis fortification texts (509-496 B.C.E.) went overland, or by sea, is uncertain (Salles, 1996, p. 257).
Nor should one suppose that the opening of a sea-route between Iran and Egypt necessarily resulted in trade with the incense-rich kingdoms of South Arabia, for when Herodotus says that the “Arabs” delivered 1,000 talents of incense per year to the Persian king (3.7), this must refer to the inhabitants of north Arabia who controlled the overland incense route between Gaza and Egypt (Briant, 1996, pp. 736-37). On the other hand, Darius says that silver and ebony from Egypt were brought to Susa for the construction of his palace there, along with ivory from Ethiopia (Kush) and India (DSe 40-44). One can probably assume that some of the cargo from Africa and India traveled by sea, although royal procurement of this type does not automatically imply the existence of private commercial traffic as well.
During the Parthian and Sasanian periods we have some of the first archeological evidence of ties between Iran and the countries bordering the Indian Ocean and its appendages. Sherds of Indian Red Polished ware, probably dating to the first three centuries C.E., have been found on at least ten coastal sites between Bušehr and Jāsk, including Sirāf (Whitehouse and Williamson, 1973, Fig. SD-F; Kervran, 1996, p. 38 and Fig. 10A). Glazed sherds of both Parthian and Sasanian type have been recovered at Khor Rori on the coast of Dhofar in southern Oman (Sedov and Benvenuti, 2002, p. 189 and Pls. 13.5, 24.1) and at Ras Hafun in the Horn of Africa (Somalia; Chittick, 1979, p. 276; Smith and Wright, 1988, p. 121 and Fig. 7; Horton, 1996, p. 449). Late Parthian and early Sasanian coins (41-241 C.E.) have been reported from the Zanzibar Museum (Allibert, 1988, p. 119; Chami and Msemwa, 1997, p. 673), but these may have arrived in modern times (Horton, 1996, p. 447). Late Sasanian-type glazed pottery is reported from Aksum in Ethiopia (Munro-Hay, 1989; Munro-Hay, 1996, pp. 412-413; Phillipson and Reynolds, 1996, p. 135). Other glazed pottery, which could be either Sasanian or early Islamic, has been recovered on the Comores islands in the Mozambique Channel (Allibert and Vérin, 1996, pp. 466-67); at Unguja Ukuu on Zanzibar (Chittick, 1965, pp. 275-94); and at Mantai on the north coast of Sri Lanka (Wijayapala and Prickett, 1986, no. 14; Carswell, 1991, pp. 197-203; Charvát, 1993, pp. 13-29; Gollwitzer, 1998, p. 47-48). Ratto Kot, the site of a square fortress with round corner towers of Sasanian type on an island at the mouth of the Indus (Gharo) near Karachi, has been identified as a Sasanian foundation (Kervran, 1994, pp. 335-338) which may have been erected in the time of Bahrām V or one of his successors. Several late sources (e.g., Ṯaʿālebi, Ṭabari) recount variants of a story about Bahrām Gur and India, according to which the Sasanian king received al-Daybul (the port of Sind), Makrān, and adjacent regions as a gift from the king of Sind (Whitehouse and Williamson, 1973, p. 43; Kervran, 1994, p. 336; Bosworth, 1999, p. 102). It has been suggested that the Sasanian-type fort at Ratto Kot, near Karachi, may be a tangible reflection of the otherwise fabulous tales of Bahram Gur’s Indian adventures (Ṭabari, tr. Bosworth, V, pp. 100-102). Certainly it occupied a position which would have had strategic value in ensuring the safe passage of commercial vessels from Iran to the Indus. Sasanian administration in Hindestān (Huyse, 1999, I, p. 23 = Shapur KZ, sec. 3.11) or Sind (Gyselen, 2002, p. 168), attested in the time of Shapur I and Narseh, and again in the late Sasanian period, must have had economic as well as political implications.
Unmistakable fragments of late “Namord” ware, a type of fine, black-on-orange ware named after the type site of Tomb-e Namord in the Rudbar valley of southern Kermān (Sajjadi, 1989), have been found on at least four coastal sites in Iran and Arabia. These range from Rēšahr on the coast of Fārs near Bušehr (Whitehouse and Williamson, 1973, Fig. 5A) to ed-Dur on the Persian Gulf coast of the United Arab Emirates (Potts, 1998, Figs. 2-3); Jazirat al-Ghanam, on the coast of Ra’s Musandam in Omani territory (de Cardi, 1972, Fig. 2.1-15); and Qane, the ancient port of Hadramawt in Yemen (Sedov, 1996, Fig. 6.2-7).
Literary evidence for sea travel between Mesopotamia, Iran, and India in the early Sasanian period comes from several of the sources for the life of Mani (216-76 C.E.). The Mani Codex (Codex Manichaicus Coloniensis, secs. 142-44) recounts a discussion between Mani and Oggas, a Palmyrene merchant at Pharat, the main port of Mesene, in which Mani states that he saw merchants about to set sail for Persia and India (Koenen and Römer, 1993, p. 102; Wiesehöfer, 1998, pp. 19-20). In the Kephalaia Mani confirms that he himself sailed to India near the end of the reign of Ardashir, returning in the year of Shapur’ s coronation (Tardieu, 1994, p. 63). According to the Mani Codex, Iranian ships set sail from Rēv-Ardashir (mod. Rēšahr, near Bušehr), at the mouth of the river Tab; coasted along the littoral of Fārs and Makrān (including the islands); and from there proceeded to Deb (dyb; Gr. Dibous; Ar. al-Daybul), at the mouth of the Indus (Kervran, 1996/99, p. 52), possibly modern Banbhore (Kervran, 1994, p. 335).
Some of the most important literary evidence available for Sasanian trade with the east comes from the 6th century C.E. A statement made by Procopius of Caesarea confirms the presence of Persian traders in India who, the Byzantine historian states, made it impossible for the Axumites [Ethiopians] to buy silk at Indian ports by locating themselves “at the very harbours where the Indian ships [delivering silk from China] put in (since they inhabit the adjoining country) and are accustomed to buy the whole cargoes” (History of the Wars 1.20.12; cf. Liu, 1994, pp. 66-67). In the Christian Topography, the monk Cosmas Indicopleustes (“Indian navigator”) attests to considerable contact with Sasanian Iran at the island of Seledibal (Gk. Taprobane), that is, Sri Lanka (Weerak-kody, 1997). Cosmas mentions maritime trade in silk between Sri Lanka and Iran; and “though by sea it is a very long way from Persia” (Christian Topography 2.45-46), he says, “from the whole of India, Persia and Ethiopia the island [Sri Lanka], acting as intermediary, welcomes many ships, and likewise dispatches them” (Christian Topography 11.15). Nor was the trade one-way, for “in return it [Sri Lanka] gets the produce of each of the aforementioned markets, and passes them on to the peoples of the interior, and at the same time exports its own native products to each of these markets” (Christian Topography 11.15). In particular, the king of Sielediba (Sri Lanka) bought horses from Iran, “and he buys them and honours the importers with freedom from tax” (Christian Topography 11.22). The fact that there was a Christian community in Sri Lanka with a bishop, “ordained in Persia” (cf. Fiey, 1969, pp. 207-8, on the Christian community on Soqotra (ancient Dioskurides) with “clergy ordained in Persia and sent to those parts” (both Christian Topography 3.65) suggests that Cosmas was describing a trading colony of Nestorians (cf. Gropp, 1991, pp. 83-88) from Khuzestan and/or Fārs, where large numbers of Nestorian Christians lived (Fiey, 1969, pp. 176-205). Cosmas also related a story about a “Roman” merchant and a Persian envoy, which, while no doubt apocryphal, suggests that both Roman gold and Sasanian silver coinage circulated in Sri Lanka (Christian Topography 11.18-19), a point confirmed by numismatic findings on the island (Bopearachi, 1993, pp. 71-81).
The prevalence of precious and semi-precious stones in royal Sasanian dress and jewelry, such as rock crystal, carnelian (etched and plain), and rubies, described by Theophylactus Simocatta as “Indian gems,” strongly suggests that India and Sri Lanka were an important source of these commodities (Simpson, 2003, p. 63). While the sources cited above clearly show that Persian merchants traveled east, there is less evidence available to suggest that Indian merchants traveled to or set themselves up in the ports of southern Iran (Kröger, 1979, pp. 441-48). Typical “Indo-Arabian” stone anchors from Sirāf (Vosmer, 1999, p. 259) belong to a type which was widespread throughout the Indian Ocean in the pre-Islamic and early Islamic era, from East Africa to Iran and from Oman to Sri Lanka. As such, they cannot be used as evidence of Indian merchant vessels at Sirāf. Yet, just as the spread of Nestorian Christianity from west to east is likely to have been bound up with commercial enterprise, it is tempting to suggest that, in the opposite direction, the westward expansion of Buddhism into southern Iran, suggested by a growing body of as yet equivocal evidence (Ball and Whitehouse, 1976, p. 147-150; Ball, 1987, pp. 95-115; Ray, 1994, p. 138), may have been a by-product of mercantile contacts. Thus, the commerce of the Indian Ocean may have exerted influences far beyond the purely economic domain.
C. Allibert, “Les contacts entres l’Arabie, le Golfe persique, l’Afrique orientale et Madagascar: Confrontation des documents écrits, des traditions orales et des données archéologiques récentes,” in L’Arabie et ses mers bordiéres I. Itinéraires et voisinages, ed. J.-F. Salles, Travaux de la Maison de l’Orient 16, Lyon, 1988, pp. 111-26.
C. Allibert and P. Vérin, “The Early pre-Islamic History of the Comores Islands: Links with Madagascar and Africa,” in The Indian Ocean in Antiquity, ed. J. Reade, London, 1996, pp. 461-70.
W. Ball, “Some Rock-Cut Monuments in Southern Iran,” Iran 24, 1987, pp. 95-115.
W. Ball and D. Whitehouse, “Qal’at-i Haidari,” Iran 14, 1976, pp. 147-50.
C. E. Bosworth, tr., The History of al-Ṭabarī Vol. V. The Sāsānids, the Byzantines, the Lakhmids, and Yemen, Albany, 1999.
O. Bopearachi, “La circulation des monnaies d’origine étrangére dans l’antique Sri Lanka,” in Circulation des monnaies, des marchandises et des biens, ed. R. Gyselen, Res Orientales 5, Bures-sur-Yvette, 1993, pp. 63-87.
P. Briant, Histoire de l’empire perse, Paris, 1996.
B. de Cardi, “A Sasanian Outpost in Northern Oman,” Antiquity 46, 1972, pp. 305-10.
J. Carswell, “The Port of Mantai, Sri Lanka,” in Rome and India: The Ancient Sea Trade, ed. V. Begley and R. D. De Puma, Madison, 1991, pp. 197-203.
L. Casson, The Periplus Maris Erythraei, Princeton, 1989.
F. A. Chami and P. J. Msemwa, “A New Look at Culture and Trade on the Azanian Coast,” Current Anthropology 38, 1997, pp. 673-77.
P. Charvát, “External contacts of Sri Lanka in the 1st millennium A.D. (Archaeological evidence from Mantai),” Archív orientální, 1993, pp. 13-29.”
N. Chittick, “The ‘Shirazi’ Colonization of East Africa,” Journal of African History 6, 1965, pp. 275-94.
Idem, “Early Ports in the Horn of Africa,” International Journal of Nautical Archaeology and Underwater Exploration 8, 1979, pp. 273-77.
Cosmas Indicopleustes: J. W. McCrindle, The Christian Topography of Cosmas, an Egyptian Monk, New York, 1887; E. O. Winstedt, The Christian Topography of Cosmas Indicopleustes, Cambridge, 1909; Wanda Wolska, La Topographie Chretienne de Cosmas Indicopleustes, Paris, 1962.
J. M. Fiey, “Dioceses Syriens Orientaux du Golfe Persique,” in Mémorial Mgr Gabriel Khouri-Sarkis (1898-1968), Louvain, 1969, pp. 177-219.
C. J. Gadd, “Seals of Ancient Indian Style Found at Ur,” Proceedings of the British Academy 18, 1932, pp. 191-210.
M. Gollwitzer, “Sri Lanka: Drehscheibe des Handels im Indischen Ozean von den Anflingen bis zur Kolonisierung,” in Der Indische Ozean in historischer Perspektive, ed. S. Conermann, Asien und Afrika 1, Hamburg, 1998, pp. 37-56.
G. Gropp, “Christian Maritime Trade of Sasanian Age in the Persian Gulf,” in Golf-Archäologie: Mesopotamien, Iran, Kuwait, Bahrain, Vereinigte Arabische Emirate und Oman, ed. K. Schippmann, A. Herling, and J.-F. Salles, Internationale Archäologie 6, Buch am Erlbach, 1991, pp. 83-88.
R. Gyselen, Nouveaux matériaux pour la géographie historique de l’empire sassanide: Sceaux administratifs de la collection Ahmad Saeedi, Studia Iranica Cahier 24, Paris, 2002.
R. T. Hallock, Persepolis Fortification Tablets [PF], Oriental Institute Publications 92, Chicago, 1969.
R. Henry, Ctésias: La Perse, l’Inde, les sommaires de Photius, Paris, 1947.
M. C. Horton, “Early Maritime Trade and Settlement of Eastern Africa,” in The Indian Ocean in Antiquity, ed. J. Reade, London, 1996, pp. 439-59.
P. Huyse, Die dreisprachige Inschrift Šābuhrs I. an der Kaʿba-i Zardust (ŠKZ), Corpus Inscr. Iran. III/I/I, 2 vols., London, 1999.
K. Karttunen, India in Early Greek Literature, Studia Orientalia 65, Helsinki, 1989.
M. Kervran, “Forteresses, entrepôts et commerce: Une histoire à suirvre depuis les rois sassanides jusqu’aux princes d’Ormuz,” in Itinéraires d’Orient: Hommages á Claude Cahen, ed. R. Curiel and R. Gyselen, Res Orientales 6, Bures-sur-Yvette, 1994, pp. 325-51.
Idem, “Indian ceramics in Southern Iran and Eastern Arabia: Repertory, classification and chronology,” in Tradition and Archaeology: Early Maritime Contacts in the Indian Ocean, ed. H.P. Ray and J.-F. Salles, New Delhi and Lyon, 1996, pp. 37-58.
Idem, “Le port multiple de l’Indus: Barbariké, Deb, Daybul, Lahori Bandar, Diul Sinde,” Res Orientales 8, 1996, pp. 45-92; = “Multiple Ports at the Mouth of the River Indus: Barbarike, Deb, Daybul, Lahori Bandar, Diul Sinde,” in Archaeology of Seafaring: The Indian Ocean in the Ancient Period, ed. H. P. Ray, Indian Council of Historical Research Monograph Series 1, Delhi, 1999, pp. 70-153.
L. Koenen and C. Romer, “Der Kölner Mani-Kodex,” in Mani: Auf der Spur einer verschollenen Religion, ed. L. Koenen and C. Römer, Freiburg, Basel, and Vienna, 1993, pp. 43-103.
J. Kröger, “Sasanian Iran and India: Questions of Interaction,” in South Asian Archaeology 1979, ed. H. Härtel, Berlin, 1979, pp. 441-48.
H. Kulke, “Die Geschichte maritimer Beziehungen im Indischen Ozean. Eine Einführung in das Thema,” in Der Indische Ozean in historischer Perspektive, ed. S. Conermann, Asien und Afrika 1, Hamburg, 1998, pp. 1-8.
P. Lecoq, Les inscriptions de la Perse achéménide, Paris, 1997.
X. Liu, Ancient India and Ancient China: Trade and Religious Exchanges, AD 1-600, Delhi, 1994.
S. C. H. Munro-Hay, Excavations at Aksum, British Institute in Eastern Africa, London, 1989.
Idem, “Aksumite Overseas Interests,” in The Indian Ocean in Antiquity, ed. J. Reade, London, 1996, pp. 403-16.
D. W. Phillipson and A. Reynolds, “B.I.E.A. Excavations at Aksum, Northern Ethiopia, 1995,” Azania 31, 1996, pp. 98-147.
G. L. Possehl, “Meluhha,” in The Indian Ocean in Antiquity, ed. J. Reade, London, 1996, pp. 133-208.
Idem, “Indus-Mesopotamian Trade: The Record in the Indus,” Iranica Antiqua 37, 2002, pp. 325-42.
D. T. Potts, “Namord ware in southeastern Arabia,” in Arabia and its Neighbours: Essays on Prehistorical and Historical Developments Presented in Honour of Beatrice de Cardi, Abiel 2, Turnhout, 1998, pp. 207-20.
S. Ratnagar, Encounters: The Westerly Trade of the Harappa Civilization, Delhi, 1981.
H. P. Ray, The Winds of Change: Buddhism and the Maritime Links of Early South Asia, Delhi, 1994.
J. Rougé, “La navigation en mer Érythrée dans l’Antiquité,” in L’Arabie et ses mers bordiéres I. Itinéraires et voisinages, ed. J.-F. Salles, Travaux de la Maison de l’Orient 16, Lyon, 1988, pp. 59-74.
M. Sajjadi, “A Class of Sasanian Ceramics from Southeastern Iran,” Rivista di Archeologia 13, 1989, pp. 31-40.
J.-F. Salles, “La circumnavigation de l’Arabie dans l’Antiquité classique,” in L’Arabie et ses mers bordiéres I. Itinéraires et voisinages, ed. J.-F. Salles, Travaux de la Maison de l’Orient 16, Lyon, 1988, pp. 75-102.
Idem, “Achaemenid and Hellenistic Trade in the Indian Ocean,” in The Indian Ocean in Antiquity, ed. J. Reade, London, 1996, pp. 251-68.
Idem, “Antique Maritime Channels from the Mediterranean to the Indian Ocean,” in From the Mediterranean to the China Sea. Miscellaneous Notes, ed. C. Guillot, D. Lombard, and R. Ptak, Wiesbaden, 1998, pp. 45-68.
M. Schuol, Die Charakene: Em mesopotamisches Kiinigreich in hellenistisch-parthischer Zeit, Oriens et Occidens 1, Stuttgart, 2000.
A. V. Sedov, “Qana’ (Yemen) and the Indian Ocean: The archaeological evidence,” in Tradition and Archaeology: Early Maritime Contacts in the Indian Ocean, ed. H. P. Ray and J.-F. Salles, New Delhi and Lyon, 1996, pp. 11-35.
A. V. Sedov and C. Benvenuti, “The pottery of Sumhuram: General typology,” in Khor Ron Report 1, ed. A. Avanzini, Arabia Antica 1, Pisa, 2002, pp. 177-248.
St. J. Simpson, “Sasanian Beads: The Evidence of Art, Texts and Archaeology,” in Ornaments from the Past. Bead studies after Beck, ed. I. C. Glover, H. Hughes-Brock, and J. Henderson, London, 2003, pp. 59-78.
M. C. Smith and H. T. Wright, “The Ceramics from Ras Hafun in Somalia: Notes on a Classical Maritime Site,” Azania 23, 1988, pp. 115-41.
M. Tardieu, “L’Arabie du nord-est d’aprés les documents manichéens,” Studia Iranica 23, 1994, pp. 59-75.
C. Tuplin, “Darius’ Suez Canal and Persian Imperialism,” in Achaemenid History VI. Asia Minor and Egypt: Old Cultures in a New Empire, ed. H. Sancisi-Weerdenburg and A. Kuhrt, Leiden, 1991, pp. 237-83.
W. Vogelsang, “The Achaemenids and India,” in Achaemenid History IV. Centre and Periphery, ed. H. Sancisi-Weerdenburg and A. Kuhrt, Leiden, 1990, pp. 93-110.
T. Vosmer, “Indo-Arabian Stone Anchors in the Western Indian Ocean and Arabian Sea,” Arabian Archaeology & Epigraphy 10, 1999, pp. 248-63.
D. P. M. Weerakkody, Taprobanê: Ancient Sri Lanka as Known to Greeks and Romans, Turnhout, 1997.
D. Whitehourse and A. Williamson, “Sasanian Maritime Trade,” Iran 11, pp. 29-49.
J. Wiesehöfer, “Mare Erythraeum, Sinus Persicus und Fines Indiae: Der Indische Ozean in hellenistischer und römischer Zeit,” in Der Indische Ozean in historischer Perspektive, ed. S. Conermann, Asien und Afrika 1, Hamburg, 1998, pp. 9-36.
W. Wijayapala and M. E. Prickett, Sri Lanka and the International Trade: An exhibition of ancient imported ceramics found in Sri Lanka’s archaeological sites, Dept. of Archaeology, Colombo, 1986.
(D. T. Potts)
Originally Published: December 15, 2004
Last Updated: March 27, 2012
This article is available in print.
Vol. XIII, Fasc. 1, pp. 87-91