Navigation – Plan du site
Connexions de basse intensité et symboliques partagées

The Indian Ocean and Swahili Coast coins, international networks and local developments

Les monnaies de l'océan Indien et de la côte swahili, réseaux internationaux et développement local
John Perkins

Résumés

Les monnaies caractéristiques de la côte swahili (vers 800-1500) ont longtemps été considérées comme un monnayage dégénéré de la périphérie du monde islamique, produit des implantations coloniales arabes et persanes dans la région. Grâce à l’exploitation de nouveaux matériaux, il a été démontré que les villes de la côte étaient en réalité des ports de commerce africains prospères, qui s’étaient développés de façon homogène à partir de villages préexistants. L'un des vestiges les plus durables de la culture matérielle de ces villes, les monnaies, n'a pas reçu une attention suffisante, en dépit de l'utilité de son étude pour mieux comprendre les villes de pierre swahili. En déconstruisant les travaux antérieurs, qui ont essentiellement cherché des modèles, il est avancé que les monnaies sont de remarquables marqueurs de la culture des villes de pierre swahili et non simplement de piètres copies d'autres monnaies islamiques de l'océan Indien. Toutefois, leur rôle à la fois sur la côte swahili et dans le monde de l'océan Indien est encore mal connu. Afin de se faire une meilleure idée de leur usage, cet article s'intéresse aux monnaies non seulement depuis une perspective locale, mais aussi à une échelle plus large, celle de l’usage des pièces de monnaies dans le commerce de l'océan Indien.

Haut de page

Notes de l’auteur

Acknowledgements
The results of this paper are based on parts of my PhD research. This was part of a Collaborative Doctoral Award between the University of Bristol and the British Museum, which was funded by the Arts and Humanities Research Council, UK. I wish to acknowledge and thank these institutions for their support. Further, the material from Songo Mnara was made available to me by Dr Jeffrey Fleisher and Dr Stephanie Wynne-Jones; without it, much of this work would have been impossible and I thank them for this. Also, the support of the British Institute in Eastern Africa, as well as the Tanzanian Department of Antiquities and the National Museum of Tanzania, has been invaluable. I would also like to thank my PhD supervisors Prof. Mark Horton and Dr Catherine Eagleton for their advice and guidance. Finally, I would like to thank Dr J.D. Hill for his advice in writing this paper.

Texte intégral

Introduction

1Coins and coinage are an important class of evidence for understanding the economies, societies, and political histories of the Indian Ocean pre-European arrival—if a little utilized one. This paper uses the results of a recent reassessment of the coinage from the Swahili Coast c. 800–1500 to contribute to debates about the origins of the Swahili culture and its place within the wider Indian Ocean world before the incursion of Europeans. In so doing, it aims to raise the profile of coins as evidence for this time period, but also highlight the related difficulties. The paper provides a brief summary of the key aspects of the coinage from the Swahili Coast to ask what circumstances led to the rise of the coinage and what inferences can be drawn about the economy of these coastal settlements and their relationship with the wider Indian Ocean world. In particular, it will be argued that the evidence of the coinage shows that for monetary systems at least, the peoples of the Swahili Coast were not slavishly copying other parts of the Islamic world but developed a distinctly regional monetary system that lasted for more than 600 years until the arrival of the Portuguese, although at that point it was already in decline.

  • 1  This definition follows G.S.P. Freeman-Grenville, 1988, p. ix.
  • 2  See for example J. Kirkman, 1954, L. Harries, 1964, H.N. Chittick, 1965a for earlier approaches; a (...)

2The area covered by this article ranges approximately from modern Tanzania’s border with Mozambique in the south to the Bajun Islands in southern Somalia in the north, including the offshore islands like Zanzibar and Mafia1 (see map, Figure 1). It is part of the wider Swahili world of East Africa, which includes the Comoros Islands and Madagascar. From c. 800 this coastal strip, along with other parts of the Swahili cultural sphere, developed a distinctive Islamic culture of which towns such as Shanga, Manda, Kisimani Mafia, and Kilwa Kisiwani are reminders. These were the first urban settlements in this part of Africa and they were closely connected to trade and exchange across the Indian Ocean. Much has been written about the origins of this culture in terms of the impact of perceived Arab or Persian colonization, the spread of Islam, and the contribution of indigenous languages and cultures.2 The shift has taken place from the idea of an implanted culture to that of a homogenously grown, but essentially African culture influenced by the Indian Ocean world.

Figure 1: Map of the Indian Ocean with relevant places

Figure 1: Map of the Indian Ocean with relevant places

After G.F.P. Freeman-Grenville, 1962, Map 1.

  • 3  A detailed discussion on this problem is provided by D.M. Varisco, 2007; also see S.D. Goitein, 19 (...)
  • 4  This term was suggested to me by C.E. Bosworth through V. Curtis of the British Museum and was use (...)

3There remains the question of periodization. How will one address this period? No consensus on what terms to use in the broader periodization of Islam currently seems to exist.3 The term medieval, with all its attached European, and primarily Christian, connotations seems inappropriate here in this Islamic sphere during a period where no European contact existed. Here, this East African period from approximately 800–1500 will be referred to as the Islamic Middle Period (IMP),4 with both the beginning and end dates not being overly strict cut-off points.

  • 5 C. Eagleton, J. Williams, 2007, p. 111-115 and p. 258
  • 6  See H.N. Chittick, 1966a. The coins were seen by this author in the Ethnographisches Museum Berlin (...)
  • 7  G.S.P. Freeman-Grenville, 1962a, p. 1-4; L. Casson, 1980, p. 22-23; M.C. Horton, J. Middleton, 200 (...)

4A distinctive feature of these towns along the Swahili Coast was that they minted and used their own coins, probably from as early as the mid-9th century and in some places possibly up to the 15th century. To understand the distinctive nature of this coinage, it is important to place it within the wider context of the coins used around the Indian Ocean world. Making coins came comparatively late to the east coast of Africa. Coins were being issued around other parts of the western Indian Ocean from circa the 5th and 3rd centuries BC.5 Some Roman, Byzantine, and Sasanian coins are reported from the East African coast; however, none of these come from excavations, and the surrounding evidence suggests that they probably did not reach the Swahili Coast in antiquity.6 Evidence for contacts and trade between this part of Africa and the Roman and Persian worlds is mainly recorded in the limited written records.7 At the time when the Swahili Coast started minting its own coinage in the 9th century, as will be discussed below, the use of coinage was already well established in the Indian Ocean world and there were clear models for people on the Swahili Coast to adopt or adapt.

  • 8  M. Broome, 1985, p. 21. The singular of fulus is fals.
  • 9  M.L. Bates, 1978, p. 8; S. Album, 1998, p. 5-6; and M. Broome, 1985, p. 21-22
  • 10  N.D. Nicol, 2012, p. 9
  • 11  M. Broome, 1985, p. 29-33.
  • 12 Ibid., p. 21-22.
  • 13  P. Grierson, 1960, p. 246-247.
  • 14  M. Broome, 1985, p. 33-34; see also P. Grierson, 1960, p. 254.
  • 15  S. Album, 1998, p. 5-6.
  • 16  M. Broome, 1985, p. 33-34.
  • 17 Ibid., p. 33-34.

5However, a key point this paper intends to make is that the Swahili Coast did not simply adopt an existing coinage system. In particular, Swahili Coast coins are not simply East African versions of the Islamic coinage of the Middle East, North Africa, and Spain. These areas followed a common model from the Umayyad period with gold dinars and silver dirhams, and after an initial transitional period from pre-Islamic to Islamic coinage they usually carried just inscriptions on them and no images. Typical of these coins are those of the Abbasids, who minted gold dinars, silver dirhams, and copper fulus8 (Figures 2–4). The vast majority of the Abbasid coinage was in silver dirhams and they were minted throughout that period, with a weight standard of about 2.97 grams and a diameter of c. 25 mm,9 while the dinars, in design virtually identical to the dirhams,10 weighed c. 4.2 grams with a diameter of 17–18 mm and later 19–21 mm.11 Both the gold and silver coins from the 8th century onward carry the name of the caliph and the date and the place where they were minted. Generally, they also used the denomination dinar or dirham and various references to Allah.12 At first the copper coins, to which less importance was attached,13 were smaller than the silver coins but weighed more, between 3–5 grams. Later coins are thinner and carry the name of the caliph; they weighed only 1.5–2 grams.14 However, the weight was never standardized, and not much research has been carried out on these coins.15 Generally, the legends consist of mint, date, and issuer, often with any one of these missing. The coins later copy the dirham design, but replace dirham with fals in the legend.16 Unlike the gold and silver coins, these were a coinage local to the individual cities and are known only over a period of about 80 years, from c. 815 to 897.17

Figure 2: Abbasid copper coin of al-Mansur. Baghdad mint 773

Figure 2: Abbasid copper coin of al-Mansur. Baghdad mint 773

Copyright British Museum. Coins and Medals IOLC.6044.

Figure 3: Abbasid silver coin of Harun al-Rashid. Basra mint 797

Figure 3: Abbasid silver coin of Harun al-Rashid. Basra mint 797

Copyright British Museum. Coins and Medals 1847.0401.59.

Figure 4: Abbasid gold coin of al-Mutamid, 4.12 grams. Baghdad mint, 875

Figure 4: Abbasid gold coin of al-Mutamid, 4.12 grams. Baghdad mint, 875

Copyright British Museum. British Museum Coins and Medals 1936,0605.3.

  • 18 Ibid., p. 52-55.

6Other Middle Eastern, North African, and Spanish coinages followed this similar model, including that of the Fatimids, although the Fatimids’ main coinage was in gold, also with an average weight of 4.2 grams. The silver coins are much rarer than the gold coins and vary greatly in weight, with the half dirham weighing about 1.4 grams.18 At first their coin designs resembled those of the other Islamic coinages, thus following the Abbasid example, but the Fatimid Caliph al-Mansur (c. 945–952) introduced a new type. It featured concentric circles around a central legend with a circular margin legend. Variations of this design were used by the following rulers. In particular, the “bulls-eye” type coinage issued by al-Mu’izz (c. 953–975) was a much used variation (Figure 5). This latter design possibly was copied by the coins from Manda, mentioned below.

Figure 5: Fatimid gold coin with circular design of al-Muizz. Misr (Egypt) mint, date ?

Figure 5: Fatimid gold coin with circular design of al-Muizz. Misr (Egypt) mint, date ?

Copyright British Museum. Coins and Medals 1853,0406.65.

  • 19  M. Mitchiner, 1979, p. 17.
  • 20  J.S. Deyell, 1990, p. 26.
  • 21 Ibid., p. 21.
  • 22 Ibid., p. 26-32.
  • 23  See M. Mitchiner, 1979, in particular p. 17-195, and J.S. Deyell, 1990, in particular p. 21-43.
  • 24  See for example M.C. Horton, 2004a.

7In contrast, if we look at the non-Islamic Indian coinage of this time, it is very diverse, with many different systems. Mitchiner says that these systems were mostly based on the Indo-Sasanian coinage19 and ultimately, like the Islamic coinage, evolved from Sasanian models.20 As Deyell points out, these were aimed mainly at local distribution.21 In northern India, for example, a silver coinage with a weight standard of around 3.8 grams was used by the Pratiharas.22 The coins mostly show either symbols, images, or writing, or any combination of these and use local scripts depending on the region.23 It seems highly unlikely that these played a part in the inception of the Swahili Coast coinage, despite the obvious links between the East Coast of Africa and the Indian subcontinent in other archaeological material.24

  • 25  J. Cribb, pers. comm. 2011 for the earliest date, and J.S. Deyell, 1990, p. 44-46 for the latest, (...)
  • 26  S. Goron, J.P. Goenka, 2001, p. xxiii-xxiv and J.S. Deyell, 1990, p. 47.

8A special mention needs to be made of the local coinage of Sindh in the north-western part of the Indian subcontinent. These Islamic coins are generally referred to as the coins of the Amirs of Sindh, and their dating is somewhat problematic, with dates from c. 810 to post-900 being suggested.25 They are made of silver and are minute in size and weight, consistently averaging at around 0.5 grams26 (Figures 6–7). The similarities and possible links, which have been suggested in the past, to the coinage of the Swahili Coast will be discussed further on in this paper.

Figure 6: Coin of the Amirs of Sindh. Coin of Abdallah

Figure 6: Coin of the Amirs of Sindh. Coin of Abdallah

Copyright British Museum. British Museum Coins and Medals 1996,0609.07.

Figure 7: Coin of the Amirs of Sindh. Coin of Muhammad

Figure 7: Coin of the Amirs of Sindh. Coin of Muhammad

Copyright British Museum. Coins and Medals 1996,0609.02.

The coinage of the Swahili Coast and its chronology

  • 27  G.S.P. Freeman-Grenville, 1962(b), p. 45-57 and H.N. Chittick, 1965(a), p. 276.
  • 28  J.d.V. Allen, 1982.
  • 29  R.L. Pouwels, 2002, p. 405.
  • 30  L. Harries, 1964, p. 225.
  • 31  J.E.G. Sutton, 1990, p. 57-59 and M.C. Horton, J. Middleton, 2000, p. 27-28, among others.
  • 32  M.C. Horton, 1996.
  • 33 M.C. Horton, H.W. Brown, W.A. Oddy, 1986.
  • 34  M.C. Horton, 1986.

9Up to the end of the 1970s the study of these coins had been linked with attempts to reconstruct and/or confirm the political history of the towns and dynasties of the Swahili Coast, comparing readings of the coin inscriptions with the very limited historical sources there are for this region. A large part of this is due to the two versions of the history of Kilwa, which state, in various fashions, that people had come from Shiraz in Persia and settled on the Swahili Coast, where they founded towns like Kilwa Kisiwani in the 10th century.27 Thus they would be seen as the founding fathers of what now is referred to as the Swahili. James de Vere Allen has since suggested a much older mythical base of this founding story,28 and this notion is also seen by Randall Pouwels as the most likely explanation.29 It is, then, only more recently that there has been a drive away from this view of a “foreign élite”,30 with a recognition of the inherently African nature of Swahili Coast settlements.31 This, naturally, does not mean that they were populated only by Africans, in the same way other Indian Ocean ports were not populated solely by an indigenous population. Although the views among historians seemed to have slowly shifted, it was not until the 1980s that through excavations on the Kenyan coast32 and a hoard find on Pemba,33 as well as a re-evaluation of the evidence from Chittick’s excavations at Manda34 that also the archaeology showed that this theory of an East Africa colonized by Arabs or Persians finally had to be put to rest. At the same time, the Swahili Coast was also not isolated from the Indian Ocean world but actively part of it.

  • 35  G.S.P. Freeman-Grenville, 1988, IX, p. 1.
  • 36 The dating of al-Hasan ibn Sulaiman is probably the most secure of all rulers of Kilwa and this gen (...)
  • 37  The three coins were published by H.W. Brown, 1991. H.W. Brown, 1993, p. 13 speaks of four coins, (...)
  • 38  M.C. Horton, J. Middleton, 2000, p. 93.
  • 39  There now can be little doubt about this, as a deposit from Songo Mnara shows—J. Perkins, J. Fleis (...)

10The Swahili Coast coinage is, with the exception of the classically inspired coinage of Aksum, the earliest coinage of Sub-Saharan Africa.35 All coins bear an Arabic inscription, and it is important to stress that no coins are known that use the vernacular KiSwahili or any predecessor thereof. They are known in three types of metal: silver, copper, and gold. The silver and copper coins all share common core traits. Three gold coins are known, and all carry the name of al-Hasan ibn Sulaiman, who probably ruled in the 1330s.36 Apart from the ruler’s name they appear to carry a date, which Brown has tentatively read as 72- or 73- AH, and cover a range from 1320–1338 AD. They also name the mint as Kilwa. They weigh 3.58, 4.33, and 4.71 grams respectively.37 While this weight averages precisely 4.2 grams, as pointed out by Horton and Middleton,38 it is clear that the spread is significant and no inferences about a possible desired weight for these coins can be drawn from such a small sample. They are, maybe not surprisingly, very different in appearance from both the silver and copper coins. However, their find-circumstances are unclear and the author would like to reserve judgement on these coins until further evidence emerges. As they are all of one ruler—and therefore might have been only an isolated appearance—and little more is known either regarding their use, circulation, or purpose, combined with their appearance towards the end of the coinage and different design from the coinage tradition in silver and copper, they are not relevant to this current discussion. The silver coins are very small, and in comparison the copper coins are much larger. The silver coins range from 7 to 12 mm in size and 0.02 to 0.27 grams in weight, while the copper coins range between 19 to 25 mm in size and roughly 0.5 to 3 grams in weight. They carry an Arabic inscription which continues from obverse to reverse, the name of what is believed to be the ruler, his belief in Allah, and usually the rhyming of the obverse and reverse inscriptions. In nearly all cases, the obverse was the side which carried the ruler’s name, while the reverse carried the reference to Allah. All these traits remained relatively unchanged from the 9th century up until the coinage ended—probably in the late 14th century.39

  • 40  H.W. Brown, 1996, p. 369.
  • 41 Ibid., p. 369.

11What is clear when one compares the earliest Swahili Coast coins and contemporary dirhams is how different these East African coins are. The only thing they have in common is the silver they are made of and the Arabic language and script. The earliest Swahili Coast coins are far smaller and lighter than an Abbasid dirham or a fraction thereof, 8–9 mm compared with 25 mm and between 0.11–0.27 grams compared with 2.97 grams. In fact, all 24 coins of group A and B at Shanga together weigh only 3.82 grams.40 In this respect it should also be mentioned that there was no known local East African source of silver41 and the silver needed to be imported, possibly by melting down foreign coins. However, as the example shows, it did not require many dirhams to create a large number of Swahili silver coins. Further, the Swahili Coast coins throughout their existence, with the exception of the three gold coins, did not conform to the conventions of Islamic coinage—they did not carry either date, mint, or a reference to the caliph. Added to this are the rhyming inscriptions which continue from one side to the other, as mentioned.

  • 42  J. Perkins, J. Fleisher, S. Wynne-Jones, 2014.
  • 43  J. Walker, 1936 and 1939.

12A detailed re-examination for my PhD thesis has sought to refine the dating and typology of these coins, in particular the copper coins. The initial stumbling block is that no more information can be gained from the archaeological publications, owing to lack of details and contexts relating the coins directly to the rest of the archaeology. It is further hindered by the long life of the various copper coin types; for example, Ali ibn al-Hasan coins, believed to be the earliest copper coins, are still found along with all other types in large numbers in contexts of the 14th and 15th centuries, as the excavations at Songo Mnara (discussed below) have shown.42 At the same time, the Shanga finds are well published but form only a small corpus; and the Pemba finds, being a hoard and being scarce at other sites, allow only limited conclusions. It has become clear that the typology presented by Walker, published in 1936 and 1939,43 is surprisingly complete, while the approach of stylistic re-evaluation has to be treated with great caution, as pointed out by Brown:

  • 44  H.W. Brown, 1996, p. 368.

Dating on the basis of style is subjective and unreliable; this is especially true for coins minted in an area so remote from the production area of those with which they might reasonably be compared.44

13This is even more pertinent to the poor condition of a large portion of these coins.

  • 45  The full results can be found in my PhD thesis, J. Perkins, 2014, Chapters 6 and 7 in particular, (...)

14However, a possible starting point is the use of XRF analysis. In conjunction with Dr. D. Hook and Dr. I. Young of the British Museum, I carried out XRF analysis of the coins from the Songo Mnara excavations for my PhD thesis.45 The results suggest that there are two distinctive groups, separated by their metal composition, with gradual changes from ruler to ruler within each group. This led me to postulate two separate minting periods for the coins of Kilwa, possibly separated by 100 years or more, but with the continued use of the earlier coins until the end of the coinage. Further, the Nasir ad-Dunya-type coins were surprisingly coherent as a group of coins. Only further archaeological investigations will be able to give a clearer picture, however.

  • 46  F. Stuhlmann, 1909, p. 854-861.
  • 47  A list of all publications here would be too numerous. The following is a selection: G.S.P. Freema (...)

15Although there were earlier German reports on the ruins of the Swahili Coast, which also mentioned the coins,46 the earliest published detailed reports on coins from East Africa are those by Walker mentioned above. Since then Freeman-Grenville, Chittick, Brown, Horton, Album, and Fleisher and Wynne-Jones have discussed new coin finds and various aspects of these coins.47 The most recent general publication was by Album in 1999; however, Brown 1993 includes dates and coins not mentioned by Album.

The coins

  • 48  This includes around 11,000 published by G.S.P. Freeman-Grenville, 1957, p. 179, Table VI; 4,500 c (...)
  • 49  H.W. Brown, 1992, p. 9.

16In total, at least 20,000 Swahili Coast coins are known.48 Helen Brown divided the coins into five groups in chronological, sometimes overlapping, order. New research conducted by this author has not changed this grouping or chronology, although the term Tanzanian-type silver is chosen over Pemba silver. The five groups are as follows: Shanga silver, Tanzanian-type silver, Kilwa copper, Zanzibar copper, and Kilwa gold.49

Shanga silver

  • 50  M.C. Horton, 1996, p. 377.
  • 51  H.W. Brown, 1996, p. 369-372.
  • 52 Ibid., p. 373.
  • 53  H.N. Chittick, 1984, p. 213 lists only 6 coins. I discovered a further coin from the Manda excavat (...)
  • 54  See H.N. Chittick, 1984, p. 213.
  • 55  H.W. Brown, 1996, p. 373.
  • 56  See N.D. Nicol, 2006.
  • 57  N.D. Nicol pers. comm. 2012.

17Two rulers are known from the Shanga coins: Muhammad and Abdallah (Figures 8–9). Further coins are known, which are later, crude, and illegible; they too are small silver coins but will not be discussed here in more detail. All coins taken together suggest a minting tradition from the second half of the 9th century up until the 12th century.50 The coins are minute, between 8–9 mm in diameter, while their weights, ranging from 0.11–0.27 grams, vary greatly. Both types have a regular die-axis, with those of Muhammad lying at 12, 3, and 6, and those of Abdallah at 10, 1, and 7;51 9 and 4 were not recorded, but this is probably by chance. All 24 coins of group A and B together weighed only 3.82 grams, as noted above. Another type coin, of which two specimens were found at Shanga,52 is not listed by Brown, possibly because of their very small number. They were first discovered at Manda (Figure 10), where seven coins were found,53 and no name for the issuer can be discerned. These coins were originally described by Brown as Fatimid coins from Sicily and copies of these;54 however, it seems that all of these too were local issues, albeit possibly Fatimid-inspired.55 Dr. N.D. Nicol, a leading scholar of Islamic and in particular Fatimid coins,56 kindly looked at some images of these coins and says they do not match any Fatimid coins of Sicily or North Africa he is aware of.57 Despite their minuteness, they seem to follow the Fatimid round inscription design mentioned above.

18Names known from the coins:

  • Abdallah

  • Muhammad

Figure 8: Shanga group A type coin, Muhammad

Figure 8: Shanga group A type coin, Muhammad

Image drawn from M.C. Horton, 1996, Pl. 131, 9.

Figure 9: Shanga group B type coin, Abd Allah

Figure 9: Shanga group B type coin, Abd Allah

Image drawn from M.C. Horton, 1996, Pl. 131, 15.

Figure 10: Manda coin 1 with round design (coin in poor condition)

Figure 10: Manda coin 1 with round design (coin in poor condition)

British Institute in Eastern Africa

Tanzanian-type silver

  • 58  H.N. Chittick, 1966(b), p. 11-12 and, 1974, p. 270.
  • 59  A. Juma, 2004.

19All factors taken together suggest a date of around 10th–11th century for these coins. Although these coins are most prominently known from a hoard of 2,060 coins at Mtambwe Mkuu on Pemba, individual specimens, all of one particular ruler, were excavated earlier by Chittick at Mafia and Kilwa58 but had been largely ignored by scholars. Since, four more were found by Juma at Unguja Ukuu on Zanzibar.59 These coins were previously described as Pemba-type by Brown; however, all coins found on Pemba come from the same hoard, while none were found elsewhere on the island. At the same time, they have also been found at excavations elsewhere in Tanzania and their place of mint is far from certain. Thus, a more neutral “Tanzanian-type silver” is suggested here.

  • 60  M.C. Horton, 1986
  • 61  As Philippe Beaujard pointed out to me in pers. comm. 2012, the names Ali and Bahram, which are li (...)

20The coins from the Mtambwe Mkuu hoard range from 7–12 mm in diameter and 0.02–0.20 grams in weight. Only around half of these coins were legible, and on these ten rulers’ names appear,60 with Ali61 ibn al-Hasan (Figure 11) being the name on the coins also found by Chittick and existing in a very similar type among the Kilwa copper coins (Figure 12).

21Names known from the coins:

  • Ali ibn al-Hasan

  • Bahram ibn Ali

  • Said ibn Ishaq

  • Muhammad ibn Ishaq

  • Ibrahim ibn Ismail

  • Khalid ibn Ahmad

  • Ahmad ibn Khalid

  • Muhammad ibn Abdullah

  • Muhammad ibn ?

  • Muhammad ibn Sulaiman

Figure 11: Ali ibn al-Hasan coin of Mtambwe Mkuu

Figure 11: Ali ibn al-Hasan coin of Mtambwe Mkuu

Drawn from M.C. Horton, H.W. Brown, W.A. Oddy, 1986, Plate 1.1.

Figure 12: Kilwa coin of Ali ibn al-Hasan—two-line type—similar to silver coins

Figure 12: Kilwa coin of Ali ibn al-Hasan—two-line type—similar to silver coins

Tanzania Coin Collection 36.

Kilwa copper

  • 62  This rough figure is reached by subtracting the 2,000 coins of Mtambwe Mkuu and the c. 4,500 Zanzi (...)
  • 63  N.M. Lowick, 1985, p. 1.

22By far the largest corpus of coins, with more than 13,000 known,62 is that which has been ascribed to the sultans of Kilwa (Figure 13). They were likely minted between the late 11th or early 12th century up until the 14th century, probably with a break of about 100 years without the minting of new coins. They probably stayed in use after that. These copper die-struck coins are usually in poor condition; and, as has been stated for the coins found at Siraf,63 this is probably a consequence of the high salt content of the earth/sand they were found in. Current evidence suggests that the Kilwa coins did not aim for a precise weight standard, but this might be due to the far from perfect condition of the coins.

Figure 13: Kilwa coin of Sulaiman ibn al-Hasan

Figure 13: Kilwa coin of Sulaiman ibn al-Hasan

Tanzania Coin Collection 670.

  • 64  H.W. Brown, 1993, p. 11-12.

23Of all the coins, approximately 2,000 were analysed by me. The relatively well-preserved ones weigh anywhere from 2–3 grams; the vast majority, however, weigh well below that, largely owing to their condition. The only exceptions to this rule seem to be the coins of Daud ibn al-Hasan, which generally weigh less than 2 grams. Again, relatively well-preserved coins range from 19–22 mm in diameter, but a few examples have been measured up to 24 or even 25 mm. The Nasir ad-Dunya-type coins, which are probably of Kilwa origin, weigh roughly from 1.0–1.7 grams with a diameter from 20–22 mm, but are even more rarely in good condition. The vast majority of coins have a regular die-axis at one of four positions: 12, 3, 6, or 9 o’clock. Of 1,130 coins, including those of Nasir ad-Dunya, where the die-axis could be determined with a high degree of certainty, 980 (87%) had one of these four alignments and there do not appear to be any major differences between the different rulers. This would suggest a continued striking tradition involving pegs or a similar type of system. Eight names are known for these coins—although Brown lists only seven,64 omitting those of Nasir ad-Dunya type.

  • 65  J. Perkins, 2014, p. 219-233.

24Names known from the coins, following my recent re-evaluation of the chronology:65

  • Ali ibn al-Hasan

  • Daud ibn al-Hasan
    [Break in minting]

  • al-Hasan ibn Talut

  • Sulaiman ibn al-Hasan

  • al-Hasan ibn Sulaiman

  • Daud ibn Sulaiman

  • Sulaiman ibn al-Hasan

    • 66  H.W. Brown, 1970.

    Nasir ad-Dunya (not listed by Brown, but discussed by her in detail elsewhere.66 Uncertain origin, but due to their number at sites such as Kilwa, Mafia, and Songo Mnara, it seems conceivable that they come from this area)

Zanzibar copper

  • 67  H.W. Brown, 1993, p. 12.

25Brown ascribes these coins to the 12th–14th centuries. Coins of three rulers are ascribed to Zanzibar, owing to the prevalence of the finds there. These weigh from 2.5–3.0 grams and have a diameter of 20–23 mm.67

26Names known from coins:

  • al-Hasan ibn Ali

  • al-Husain ibn Ahmad

  • Ishaq ibn al-Hasan

27No mint has been found so far for any of the coinages of the Swahili Coast, and also no dies. This makes it difficult to ascribe coins to a certain place, apart from find frequency and the Kilwa king lists. For example, the coins tentatively ascribed to Pemba by Brown are known only from one hoard there; they might originate elsewhere along the coast.

28To reiterate, one of the most remarkable things about the silver and copper coinage is that it does not at any point conform to the traditions of Islamic minting. The coins do not bear the name of the caliph or the name of the mint and do not carry a date. Also, apart from the anonymous Nasir ad-Dunya coins, they avoid using any form of titulature.

Distribution of Swahili Coast coins and foreign imports

  • 68  See H.N. Chittick, 1961, 1974, and 1984; M.C. Horton, 1996; A. Juma, 2004; S. Wynne-Jones, J. Flei (...)
  • 69  J.S. Kirkman, 1954, p. 149.

29From the earliest times onward the Swahili coinage, though large, seems to have been very limited in its distribution. The coins are found at various sites on the Swahili Coast: the silver coins are found nearly exclusively on the northern Kenyan coast, Zanzibar, and Pemba, with only a few examples having been found at Kilwa and Mafia; the copper coins are found virtually exclusively on the central Tanzanian coast and on Zanzibar, with only a few examples being found at Shanga and Manda.68 It is interesting to note that apart from two Chinese coins, no coins were found at the important Kenyan site of Gedi.69

  • 70  This is shown in the archaeological evidence at Shanga; see M.C. Horton, 1996, p. 376.
  • 71  A hoard found at this site, which the author has been analysing, was sealed in the foundations of (...)
  • 72  G.S.P. Freeman-Grenville, 1962b, p. 103-108.

30The above suggests that coinage on the Swahili Coast was first minted on the northern Kenyan coast at sites such as Shanga and Manda from the second half of the 9th century. A silver coinage clearly inspired by, or related to, the original coinage was then minted further south, perhaps on Pemba and/or Zanzibar, probably in the 10th and 11th centuries, while at the same time it appears that the coinage at Shanga became cruder and probably stopped being minted in the 12th century.70 The second silver coinage was in turn replaced by a copper coinage around the end of the 11th or in the early 12th century at Kilwa and/or Mafia, which in turn inspired a minor copper coinage of Zanzibar from around the 12th century. The end date of the copper coinage remains unclear. The evidence from Songo Mnara would suggest that all known copper coins believed to come from Kilwa, including the Nasir ad-Dunya coins, were already in circulation towards the end of the 14th century.71 It needs to be clearly stated that it was, thus, not the arrival of the Portuguese which put an end to the minting of this coinage; and it seems likely that this coinage remained in circulation at least until then, as they report the use of a copper coinage at Kilwa in 1505.72

  • 73  H.N. Chittick, 1982, p. 54.
  • 74  T. Huffman, 1972, p. 362 and Plate I.
  • 75  G.S.P. Freeman-Grenville, 1988, IX, p. 2 and p. 6. Also, I have seen some letters at the British M (...)

31While the distribution on the Swahili Coast is very limited, the moment we leave the Swahili Coast these coins become exceptionally rare finds. Unsurprisingly, some examples come from Mogadishu73 and one from Great Zimbabwe;74 but outside East Africa, there are only two known reports. There is one Kilwa-type coin, which was found along with some Nasir ad-Dunya coins (found so numerously at Kilwa and Mafia) at al-Balid, ancient al-Mansura, in Dhufar province in Oman. However, these coins have, to this author’s knowledge, never been published and the archaeological context is unknown.75

32The only other find of Swahili Coast coins outside of East Africa comes, astonishingly, from Australia. Five Kilwa-type coins, two of Ali ibn al-Hasan, two of Sulaiman ibn al-Hasan, and one of al-Hasan ibn Sulaiman, were found on Marchinbar Island, part of the Wessel Islands off the coast of the Northern Territory. The find was claimed to have been made during World War II by an Australian soldier stationed there. It is not entirely clear how these coins came to rest on these shores. The finder claimed to have also found Dutch coins dating from the late 17th through to the late 18th centuries on the same beach.76 Freeman-Grenville believes it unlikely that these coins left the East African coast after the 16th century, their period of circulation.77 However, they would have been readily available long after they ceased to be used. They are even today found on beaches, and it appears just as likely that they were picked up at a later date as a curiosity on the East African coast and taken on a voyage much further east. It seems that initially no further investigation of these extraordinary finds took place—they can be found with some information in Sydney’s Powerhouse Museum’s online database.78 A recent expedition to the island did not produce any further related finds or contexts.79 Therefore, again, we are left with this little evidence and only some very speculative answers and doubts about these coins remain.

33Other than these two finds, neither with detailed contextual information, it is worth reiterating that no Swahili Coast coins have been found outside the Swahili Coast—a pattern that deserves more consideration, given the important role ascribed to the Swahili Coast in regards to wider Indian Ocean trade and contacts.

  • 80  Two of these are now in the National Museum of Tanzania in Dar es Salaam. One was published by H.N (...)
  • 81  Mostly published by G.S.P. Freeman-Grenville, 1960, p. 42. Further additions are H.N. Chittick, 19 (...)
  • 82  M.C. Horton, forthcoming.
  • 83  This is further confirmed by M.C. Horton, forthcoming.

34In this respect it is interesting to look at what foreign coins were found on the Swahili Coast and, in particular, at a comparison with other regions around the Indian Ocean. For the IMP, the author is aware only of just over 300 foreign coins from Kenya and Tanzania compared with the at least 20,000 local Swahili Coast coins. These foreign coins consist of seven Indian Chola coins,80 approximately 70 non-East African Islamic copper, silver, and gold coins, and about 250 Chinese coins81—although Horton’s forthcoming publication, which deals with the archaeological sites of Pemba and Zanzibar, adds small numbers of confirmed and unconfirmed coin finds to this list.82 It should be noted that the majority of Chinese coins come from hoards, usually without any local coins.83

  • 84  D. Whitehouse, 1970.
  • 85  N.M. Lowick, 1985, p. 1.
  • 86  See N.M. Lowick, 1985, p. 11-63.
  • 87 Ibid., p. 1 and 6.

35Unfortunately, there are relatively few published coin assemblages from excavations bordering the Indian Ocean, related to the IMP. One of the few Indian Ocean ports of this period to have been excavated and published is Siraf, in southern Iran. It was excavated between 1966 and 1973 and dates from the 9th century. It already started to fall into decline towards the end of the 10th century, but seems to have survived until the 16th century as a small town.84 It therefore can serve only as a very limited site for comparison. During the excavations, 949 coins were found, four gold, 74 silver, 436 bronze, and 435 lead. Of these, 460 coins could not be identified. The majority of the coins come from the Great Mosque and its underlying structures.85 Interestingly, the majority of Shanga coins were also found in the platform of the mosque. Of the 489 legible coins, 418 are of interest here.86 The remaining 71 coins either post-date the period under discussion or pre-date the Islamic period. Of the ones of interest, 335 coins were Islamic coins either of local mint or those of the dynasties that ruled Persia at one point. Only one coin of the relevant period comes from Umayyad Spain, while 13 are uncertain Mongol-period coins and could possibly be counted towards the other number. A total of 69 Chinese coins were found, but 60 of these were in a single hoard. This was deposited in a layer of the 14th century.87

36More work is required in looking at the distributions of coins around the Indian Ocean during the IMP, to gain a fuller picture of the monetary history of the Indian Ocean. If Siraf and the evidence for East Africa are typical, then it is likely that few coins from one part of the Indian Ocean moved to other parts. Research would include looking at which local and regional coinages stayed in the areas of their local use, as is the case for Swahili Coast coins, and which had a greater tendency to travel—and then to ask why and what their use was in facilitating trade or moving as bullion and raw metal.

The origins of the coinage

37As it currently stands, the history of coinage on the Swahili Coast begins with the tiny silver coins probably minted at Shanga from the mid-9th century. The earliest evidence for Islam there dates to 780, while the earliest archaeological evidence for the coins dates to the late 9th to early 10th century. Therefore, it seems likely that the coinage in Shanga started some time during the second half of the 9th century, when Islam had become more widely spread. The coins are a phenomenon that is also inseparably linked to the rise of the Swahili towns.

  • 88  See M.C. Horton, 1996 and M.C. Horton, 2004b.

38What, though, were the reasons for these towns to issue coins? With the current state of evidence, it is very difficult to say what exactly gave rise to the use of coinage on the Swahili Coast. What is clear, however, is that it happened at the same time as an increase in trade with the Indian Ocean world took place and the East African coastal towns started to grow and also to adopt Islam.88 To theorize about the reasons behind their introduction, the function of the coins has to be better understood first.

  • 89  H. Brown, 1993; M.C. Horton, 1996, p. 399-400; M.C. Horton, J. Middleton, 2000, p. 92-94; and J. M(...)
  • 90  J. Perkins, 2014; most of Chapter 8 is dedicated to this discussion.
  • 91  The gold coins are too singular in their appearance and issue, and combined with their unclear fin (...)
  • 92  S. Wynne-Jones, J. Fleisher 2010 and 2011, as well as pers. comm. with S. Wynne-Jones and J. Fleis (...)
  • 93  S. Wynne-Jones, J. Fleisher, 2012.

39It has been suggested that the coins were used in the international trade as local trade tokens or even as part of an interchangeable currency with a fixed exchange rate for the Indian Ocean trade.89 I have argued against both these notions.90 On the one hand, the very low intrinsic metal value and non-conformity to Islamic minting traditions of both the silver and copper coins would render them worthless outside of the sphere of authority of the issuer, even if they were deemed of high value within this sphere.91 Thus, it must be assumed that they would have been of little use in international trade. The other aspect, that of a trade token, also seems unlikely in light of the distribution pattern emerging from the excavations at Songo Mnara,92 where many were found in the wealthy houses of traders, but also in open spaces and in poorer dwellings. This suggests that they were carried around and lost in local everyday transactions. However, at the same time it is perfectly possible that this local value might, indeed, have been of token value, as initially suggested by S. Wynne-Jones and J. Fleisher.93 Therefore, an individual coin’s value would not be bound to its metal contents.

40Returning to the initial question of why the Swahili Coast started to use coinage, one possible explanation is that with the rise in international trade interactions, both with foreign traders arriving on the Swahili Coast and Swahili traders travelling abroad and the increasing size of the settlements, the use of coinage was seen as a practical step forward. More importantly, they might have been issued as a sign of authority, or to establish authority, as well as providing a public proclamation of faith by the rulers, with both messages aimed at the local inhabitants and possibly the foreign—often Muslim—traders who would have spent considerable time in the towns waiting for the monsoon winds to turn. That the Swahili were aware of the use and purpose of coinage in the rest of the Indian Ocean world is a given, and by issuing coins the local rulers might have not only sought to affirm their authority in this form, but also to aid trade with Muslim trade contacts.

  • 94  See J. Perkins, 2014, in particular p. 259-262.
  • 95  M.C. Horton, 1996, p. 97 and p. 101, Plate 49.

41When looking at the monetary system of the Swahili coast and the origin of this coinage, one factor has usually been omitted from the discussion. Although this is beyond the scope of this paper, other means of currency might have been in circulation at the same time as the coins.94 At Shanga two hoards of cowrie shells were found next to each other in layers of the first half of the 14th century.95 This might suggest that cowries were used alongside the coins as a local means of exchange. It might be considered whether the coins, which hardly left the towns in which they were minted, were a local currency as opposed to non-coin money, such as beads and cowries, forming a more inter-regional East African/African currency. However, the evidence in this case is even less informative than with the coinage, and cowries and beads were hardly recorded, or recorded with little detail, in the archaeological record prior to Horton’s excavations at Shanga.

  • 96  G.S.P. Freeman-Grenville, 1960, p.37.
  • 97  J. Walker, 1936, p. 74-75.
  • 98  G.S.P. Freeman-Grenville, 1978, p. 194.
  • 99  H.W. Brown, 1996, p. 371.
  • 100  M. Broome, 1985, p. 57-60.

42These coins in many ways are a microcosm for how other features of the culture, history, and archaeology of the Swahili Coast have been interpreted. For every stand-out trait of these coins, like the rhyming of obverse and reverse, and certain formulas used in the inscription, scholars have tried to find a precedent. It seemed inconceivable that the traits that would occur on what was seen as a minor local coinage could not have appeared elsewhere first. Much of this is linked to the problems of dating the coins, and many of these interpretations must be considered as outdated by the discovery of the silver coinage. For example, Freeman-Grenville argued that the Kilwa coins could not date before a certain time owing to the fact that the other earliest known coins with rhyming inscriptions were Fatimid copper coins of the end of the 11th and 12th centuries.96 This and other examples, such as the use of the phrase ‘aza nasir (“may his victory be glorious”) also show more generally the dominant notions about the Swahili culture—that they were purely the receptacle of ideas, thus incapable of innovation, and all innovations had to be either implanted or borrowed. Walker had argued that ‘aza nazir, which appears on the coins of al-Hasan ibn Sulaiman, could only have appeared on Kilwa coins after its first appearance on the coins of the Mamluk al-Mansur in 1377; thus he ascribed the coins to a 15th-century ruler of the same name.97 Since then, Freeman-Grenville’s research showed that the phrase appeared earlier, on the coin of the Mamluk ruler al-Nasir Muhammad in 1317, but he too used this as a terminus post quem for any Kilwa issues.98 This date fits much better with the dates for the al-Hasan ibn Sulaiman described by Ibn Battuta who is now thought to have issued these coins. While there is no doubt that the features on the Kilwa coins will in some cases have precedents in the world of Islamic coinage, it is the basic assumption that all features have to have had a precedent which grates. We now know that these interpretations can be flawed. For example, as seen above, rhyming inscriptions occur on some of the earliest known silver Swahili coins,99 predating the Fatimid examples by some 200 years; and while it is possible that the latter example might indeed give us a terminus post quem, it has to enter the thought process that it is possible that it and other features might have appeared independently on the coins of the East African coast. In fact, their non-conformity might make them more likely to be innovative in both design and phrasing. However, having said this, their basic design remains relatively constant throughout their history. Just to reiterate the point, neither from the heartlands of the Abbasids nor the Fatimids, nor the geographically closer Yemen and Oman are any coins known which could be considered as prototypes to the early coinage of the Swahili Coast. The known Yemeni coins, for example, followed the design of the Abbasids and later the Fatimids but used a lighter weight standard100—though nowhere near as light as the coins from Shanga and Manda.

  • 101  R.L. Pouwels, 2002, p. 424-425.
  • 102  M. Walsh, 2010, p. 469.
  • 103  S. Goron, J.P. Goenka, 2001, p. xxii.
  • 104  Although there appear to be doubts regarding the identification of the ancient Debal/Daybul with t (...)
  • 105  H.P. Ray, 2004, p. 45.
  • 106  J.S. Deyell, 1990, p. 44-45 and p. 47.
  • 107  S. Goron, J.P. Goenka, 2001, p. xxii. While S. Goron and J.P. Goenka do not give a reason for the (...)
  • 108  Cribb, pers. comm.
  • 109  E. de Zambaur, 1927, p. 279. Daud ibn Yazid and Bashir ibn Daud.

43What does this mean for the early silver coinage and the adoption of coin making on the East African coast? There has been a natural tendency to look for a single external origin and inspiration for these coins—a single prototype. As this coinage is not a simple adoption of Middle Eastern dirhams, other prototypes were looked for. A possible prototype has been identified in the silver coinage of the Islamic Amirs of Sindh. That there was trade between South Asia and East Africa early in this period is shown by finds at Shanga and also a claim found along the Swahili Coast in oral traditions that Wadibuli, people of what is variously known as Daybul, Debal, or Banbhore in Sindh, founded towns here. As Pouwels points out, this claim has to be seen in the same light as the claims to Persian or Arab origin,101 and more recently Walsh has argued that all the oral traditions referring to the Wadibuli in various terms do not allow for the identification of one specific group—or in other words, place of origin—but rather represent “the telescoping and conflation of once separate histories”.102 It might, though, suggest a frequent contact at some point. The coins of Sindh do share some of the attributes of the coins of the Swahili Coast. They are very small silver coins which do not refer to the caliph; some use the phrase billah yathiqu (“in god trusts”),103 while yathiqu is frequently found on Swahili Coast coins from the earliest times onward; they have a legend which seems to continue from obverse to reverse; and a few have a star and/or crescent design, which features prominently on some coins from Mtambwe Mkuu and later Kilwa-type copper coins. However, the coins from Sindh do not have rhyming inscriptions, are by comparison heavier, use different inscriptions, and do not appear to be of the same “family” stylistically. In fact, in appearance they largely seem to be closer to the more traditional Abbasid coinage. Then there is the problem of dating. Deyell points out that not one local silver coin was recovered at the excavations of Debal/Banbhore,104 which dates from the 8th century onward.105 Instead, a copper coinage was issued by the local Arab governors. He concludes, based on archaeological evidence, that the local silver coins started to be issued only post-900106 —although Goron gives a date of 870 without further explanation,107 while Cribb believes that they date as early as 810.108 Cribb’s notion is based on the occurrence of two coins which carry names that are also known in that sequence from the list of Amirs.109

  • 110  M.C. Horton, J. Middleton, 2000, p. 48; R.L. Pouwels, 2000, p. 251.

44Nonetheless, at this point it cannot be said for certain that they actually pre-date the mid-9th century coins of Shanga. Also, the two coins mentioned by Cribb do not use the word yathiqu, and at least one of them has a star and crescent. The star and crescent does not feature on the coins of Shanga, while yathiqu appears here. As such, there is no evidence for a close copying of a Sindh prototype for the coins that began to be produced on the Swahili Coast, even if there are similarities in form. Of course, no one has suggested the coins of Sindh might have borrowed from the coins of the Swahili Coast. Rather than seek simple, single prototypes that travelled across the Indian Ocean, perhaps we need to recognize that East African coin makers were aware of the use of coins in different parts of the Indian Ocean; and when the need for coin ‘money’ arose in the new trading settlements along the coast, they took ideas of coin use but created a distinctly regional type of coin for use in local exchanges. Along with non-coin money, they might have formed a complex local, regional, and inter-regional Swahili or East African currency system. Finally, when discussing the coinage of the Swahili Coast, one point has so far never been considered. Unlike nearly all other parts of the Islamic world at this time, Islam was not spread here by conquest.110 It was a voluntary development, in the same way that the choice to use coinage in the first place was voluntary; and as the coinage was intended for local, and possibly regional, use, there would have been no necessity to conform to traditional Islamic minting traditions. This might be a further explanation for the unique nature of this coinage on the Swahili Coast.

Conclusion

45When we look at the evidence presented, it becomes clear that many aspects of the rise and indeed development of Swahili Coast coinage currently remain unexplained. Many of the theories put forward over the years have been shown not to have held up to the test of time, leading more recent publications to be void of speculation and err on the side of caution.

  • 111  R.L. Pouwels, 2001, p. 644.

46By no means should this paper be understood to suggest that the wider Indian Ocean did not lend ideas to the Swahili Coast. As Pouwels put it so aptly:111 the Swahili Coast did not live in a vacuum. The general idea of a coinage itself, as we have seen, is without a doubt such a loaned idea. The voice of the ruler is clear on these coins; it was important for him to express clearly his faith in Allah. Whether this was genuinely out of piousness or for political purposes remains unclear. For whatever reason this coinage was initially adopted—possibly as a sign of power, or religion, or both—it is one of the most durable remains of Swahili material culture that, with its rhyming inscriptions, still speaks of a distinctive culture.

47While the coinage is the first recognizable monetary system on the Swahili Coast, it is possible that other systems, such as cowrie shells, were used concurrently and maybe even preceded the coins. The coinage, right from the beginning, seems to have been a very local affair and perhaps not universally used even in the areas where it originated. There are sites where no coins have been found at all, or in such minute quantities that they must be seen as foreign to the site, despite the sites being flourishing centres and sometimes close to sites where coins were in use. Both these factors suggest a very limited area of use, possibly linked to the influence of the issuing authority, and most likely also that the coinage was not used in international trading activities. In fact, if we look at the scarcity of foreign coins, both on the Swahili Coast and at Shiraz, it might have serious implications regarding the use of coinage, or more precisely lack thereof, in the Indian Ocean trade during the IMP. More comparative material is required to allow any clear-cut conclusions.

48As we have seen, the only coinage that shares some aspects with the Swahili Coast coinage was issued by the Amirs of Sindh. Again, the evidence is weak, but why should the earliest coinage of the Swahili Coast not have arisen without a direct precedent? At the same time, it is possible that the similarities are indeed due to the interaction between these two regions of the Indian Ocean—but the emphasis has to be on interaction. The Indian Ocean is described as an exchange network, but even now the Swahili Coast is usually seen purely as a receptacle in this exchange. Why should the Swahili Coast not have actively taken part in this exchange? Few scholars look for, or even suggest, ideas which might have been exported from the Swahili Coast. Ultimately, the evidence at the moment does not permit any final or clear conclusions, but the coins at least allow the suggestion that the Swahili Coast did not only actively take part in trade but also in the exchange of ideas.

Haut de page

Bibliographie

Album, S., 1998, A checklist of Islamic coins, 2nd ed., Santa Rosa, S. Album.

Album, S., 1999, Sylloge of Islamic coins in the Ashmolean, vol. 10, Arabia and East Africa, Oxford, Ashmolean Museum.

Allen, J.d.V., 1982, “The ‘Shirazi’ problem in East African coastal history”, Paideuma, 28, p. 9-27.

Bates, M. L., 1978, “Islamic numismatics. Section 3”, Middle East Studies Association Bulletin, 12 (3), p. 2-18.

Broome, M., 1985, A handbook of Islamic coins, London, Seaby Press.

Brown, H.W., 1970, “Fakhur-al Dunya and Nasir al-Dunya: Notes on two East African topics”, Numismatic Chronicle, p. 253.

Brown, H.W., 1991, “Three Kilwa gold coins”, Azania, 26, p. 1-4.

Brown, H.W., 1992, “Early Muslim coinage in East Africa: The evidence from Shanga”, Numismatic Chronicle, p. 83-87.

Brown, H.W., 1993, “Coins of East Africa: An introductory survey”, Yarmouk Numismatics, 5, p. 9-16.

Brown, H.W., 1996, “The coins”, in M.C. Horton, H.W. Brown, N. Mudida, Shanga. The archaeology of a Muslim trading community on the coast of East Africa, London, British Institute in Eastern Africa Memoir, n° 14, p. 368-375.

Casson, L., 1980, “Rome’s trade with the East: The sea voyage to Africa and India”, Transactions of the American Philological Association, 110, p. 21-36.

Chittick, H.N., 1961, Kisimani Mafia. Excavations at an Islamic settlement on the East African coast, Dar Es Salaam, Antiquities Division, Occasional Paper n° 1.

Chittick, H.N., 1965a, “The ‘Shiarzi’ colonization of East Africa”, The Journal of African History, 6 (3), p. 275-294.

Chittick, H.N., 1965b, “Director’s Report”, London, British Institute of History and Archaeology in East Africa, Report 1964–1965, p. 3-10.

Chittick, H.N., 1966a, “Kilwa: A preliminary report”, Azania, 1, p. 1-36.

Chittick, H.N., 1966b, “Six early coins from near Tanga”, Azania, 1, p. 156-157.

Chittick, H.N., 1969, “An archaeological reconnaissance of the southern Somali Coast”, Azania, 4, p. 110-130.

Chittick, H.N., 1973, “On the chronology and coinage of the Sultans of Kilwa”, Numismatic Chronicle, p. 192-200.

Chittick, H.N., 1974, Kilwa. An Islamic trading city on the East African coast, 2 vols, Nairobi, British Institute in East Africa Memoir, n° 5.

Chittick, H.N., 1982, “Medieval Mogadishu”, Paideuma, 28, p. 45-62.

Chittick, H.N., 1984, Manda. Excavations at an island port on the Kenyan coast, Nairobi, British Institute in Eastern Africa Memoir, n° 9.

Deyell, J.S., 1990, Living without silver. The monetary history of Early Medieval north India, Delhi, Oxford University Press.

Eagleton, C., Williams, J., 2007, Money: A history, London, British Museum Press.

Fleisher, J., Wynne-Jones, S., 2010, “Kilwa-type coins from Songo Mnara, Tanzania: New finds and chronological implications”, Numismatic Chronicle, p. 494-506.

Freeman-Grenville, G.S.P., 1957, “Coinage in East Africa before Portuguese times”, Numismatic Chronicle, p. 151-175.

Freeman-Grenville, G.S.P., 1960, “East African Coin finds and their historical significance”, The Journal of African History, 1 (1), p. 31-43.

Freeman-Grenville, G.S.P., 1962a, The Medieval history of the coast of Tanganyika with special reference to recent archaeological discoveries, Berlin, Akademie-Verlag.

Freeman-Grenville, G.S.P., 1962b, The East African coast: Select documents from the first to the earlier nineteenth century, Oxford, Clarendon Press.

Freeman-Grenville, G.S.P., 1971, “Coin finds and their significance for East African chronology”, Numismatic Chronicle, p. 283-301.

Freeman-Grenville, G.S.P., 1978, “Numismatic evidence for the chronology at Kilwa”, Numismatic Chronicle, p. 191-196.

Freeman-Grenville, G.S.P., 1988, The Swahili Coast, 2nd to 19th centuries, London, Variorum Reprints.

Goitein, S.D., 1968, “A plea for the periodization of Islamic history”, Journal of the American Oriental Society, 88 (2), p. 224-228.

Goron, S., Goenka, J.P., 2001, The coins of the Indian Sultanates, Delhi, Munshiram Manoharlal.

Grierson, P., 1960, “The monetary reforms of ‘Abd al-Malik: Their Metrological basis and their financial repercussions”, Journal of Economic and Social History of the Orient, 3 (3), p. 241-264.

Harries, L., 1964, “The Arabs and Swahili culture. Africa”, Journal of the International African Institute, 34 (3), p. 224-229.

Horton, M.C., 1986, “Asiatic colonization of the East African coast: The Manda evidence”, Journal of the Royal Asiatic Society of Great Britain and Ireland, 2, p. 201-213.

Horton, M.C., 2004a, “Artisans, communities, and commodities: Medieval exchanges between northwestern India and East Africa”, Ars Orientalis, 34, p. 62-80.

Horton, M.C., 2004b, “Islam, archaeology, and Swahili identity”, in D. Whitcomb (ed.), Changing social identity with the spread of Islam. Archaeological perspectives, Chicago, The Oriental Institute of the University of Chicago, p. 67-88.

Horton, M.C., forthcoming, Zanzibar and Pemba, archaeological investigations of an Indian Ocean Archipelago, London, British Institute in Eastern Africa.

Horton, M.C., Brown, H.W., Oddy, W.A., 1986, “The Mtambwe hoard”, Azania, 21, p. 115-123.

Horton, M.C., Brown, H.W., Mudida, N., 1996, Shanga. The archaeology of a Muslim trading community on the coast of East Africa, London, British Institute in Eastern Africa Memoir, n° 14.

Horton, M.C., Middleton,J., 2000, The Swahili. The social landscape of a mercantile society, Oxford, Blackwell Publishers.

Huffman, T., 1972, “The rise and fall of Zimbabwe”, The Journal of African History, 13 (3), p. 353-336.

Juma, A., 2004, Unguja Ukuu on Zanzibar. An archaeological study of early urbanism, Doctoral thesis, Uppsala University. URL:http://uu.diva-portal.org/smash/record.jsf?pid=diva2%3A164832&dswid=2470. Accessed 21 October 2011.

Kirkman, J.S., 1954, The Arab city of Gedi. Excavations at the Great Mosque, Architecture and Finds, London, Oxford University Press.

Kusimba, C.M., 1999, The rise and fall of the Swahili states, Walnut Creek, AltaMira Press.

Lowick, N.M., 1985, Siraf XV. The coins and monumental inscriptions, London, The British Institute of Persian Studies.

Mattingley, H, 1932, “Coins from a site-find in British East Africa”, Numismatic Chronicle, p. 175.

Middleton, J., 2003, “Merchants: An essay in historical ethnography”, The Journal of the Royal Anthropological Institute, 9 (3), p. 509-526.

Mitchiner, M., 1979, Oriental coins and their values, vol. 3, Non-Islamic states and Western colonies, London, Hawkins Publications.

Nicol, N.D., 2006, A corpus of Fatimid coins, Trieste, G. Bernardi.

Nicol, N.D., 2012, Sylloge of Islamic coins in the Ashmolean, vol. 3, Early Abbasid precious metal coinage (to 218 AH), Oxford, Ashmolean Museum.

Perkins, J., 2014, The coins of the Swahili Coast c. 800–1500, PhD Dissertation, University of Bristol.

Perkins, J., Fleisher, J., Wynne-Jones, S., 2014, “A deposit of Kilwa-type coins from Songo Mnara, Tanzania”, Azania, 49 (1), p. 102-116.

Pouwels, R.L., 2000, “The East African coast from c. 780 to 1900 CE”, in N. Levitzon, R.L. Pouwels (eds), The history of Islam in Africa, Athens, Ohio University Press, Oxford, James Currey Ltd, p. 251-272.

Pouwels, R.L., 2001, “A reply to spear on early Swahili history”, The International Journal of African Historical Studies, 34 (3), p. 639-646.

Pouwels, R.L., 2002, “Eastern Africa and the Indian Ocean to 1800: Reviewing relations in historical perspective”, The International Journal of African Historical Studies, 35 (2/3), p. 385-425.

Ray, H.P., 2004, “The beginnings: The artisan and the merchant in early Gujarat, sixth–eleventh centuries”, Ars Orientalis, 34, p. 39-61.

Stuhlmann, F., 1909, Beiträge zur Kulturgeschichte von Ostafrika, Allgemeine Betrachtungen und Studien über die Einführung und wirtschaftliche Bedeutung der Nutzpflanzen und Haustiere, mit besonderer Berücksichtigung von Deutsch-Ostafrika, Berlin, D. Reimer (E. Vohsen).

Sutton, J.E.G., 1990, A thousand years of East Africa, Nairobi, British Intitute in Eastern Africa.

Varisco, D.M., 2007, “Making “Medieval” Islam meaningful”, Medieval Encounters, 13 (3), p. 385-412.

Walker, J., 1936, “The history and coinage of the Sultans of Kilwa”, Numismatic Chronicle, p. 43-81 (and plates VIII-IX).

Walker, J., 1939, “Some new coins from Kilwa”, Numismatic Chronicle, p. 223-227.

Walsh, M., 2010, “Deep memories or symbolic statements? The Diba, Debuli and related traditions of the East African coast”, in C. Radimilahy, N. Rajaonarimanana (eds.), Civilisations des mondes insulaires (Madagascar, îles du canal de Mozambique, Mascareignes, Polynésie, Guyanes): Mélanges en l’honneur du Professeur Claude Allibert, Paris, Éditions Karthala, p. 453-476.

Whitehouse, D., 1970, “Siraf: A Medieval port on the Persian Gulf”, World Archaeology, 2 (2), p. 141-158.

Wynne-Jones, S., Fleisher, J., 2010, “Archaeological investigations at Songo Mnara, Tanzania, 2009”, Nyame Akuma, 73, p. 2-9.

Wynne-Jones, S., Fleisher, J., 2011, “Archaeological investigations at Songo Mnara, Tanzania, 2011,” Nyame Akuma, 76, p. 3-8.

Wynne-Jones, S., Fleisher, J., 2012, “Coins in context: Local economy, value and practice on the East African Swahili Coast”, Cambridge Archaeological Journal, 22 (1), p. 19-36.

Zambaur, E. de, 1927, Manuel de généalogie et de chronologie pour l’histoire de l’Islam, Hannover, Heinz Lafaire.

Haut de page

Notes

1  This definition follows G.S.P. Freeman-Grenville, 1988, p. ix.

2  See for example J. Kirkman, 1954, L. Harries, 1964, H.N. Chittick, 1965a for earlier approaches; and J.E.G. Sutton, 1990, p. 57-88, C.M. Kusimba, 1999, in particular p. 89-154, and M.C. Horton, J. Middleton, 2000, for more up-to-date views. These also give an overview of the development of thought on the matter.

3  A detailed discussion on this problem is provided by D.M. Varisco, 2007; also see S.D. Goitein, 1968.

4  This term was suggested to me by C.E. Bosworth through V. Curtis of the British Museum and was used throughout my PhD thesis.

5 C. Eagleton, J. Williams, 2007, p. 111-115 and p. 258

6  See H.N. Chittick, 1966a. The coins were seen by this author in the Ethnographisches Museum Berlin, H. Mattingley, 1932; H.N. Chittick, 1965a, p.283, footnote 18; G.S.P. Freeman-Grenville, 1960, p.32-33; and H.N. Chittick, 1969, p. 124-130

7  G.S.P. Freeman-Grenville, 1962a, p. 1-4; L. Casson, 1980, p. 22-23; M.C. Horton, J. Middleton, 2000, p. 32-33.

8  M. Broome, 1985, p. 21. The singular of fulus is fals.

9  M.L. Bates, 1978, p. 8; S. Album, 1998, p. 5-6; and M. Broome, 1985, p. 21-22

10  N.D. Nicol, 2012, p. 9

11  M. Broome, 1985, p. 29-33.

12 Ibid., p. 21-22.

13  P. Grierson, 1960, p. 246-247.

14  M. Broome, 1985, p. 33-34; see also P. Grierson, 1960, p. 254.

15  S. Album, 1998, p. 5-6.

16  M. Broome, 1985, p. 33-34.

17 Ibid., p. 33-34.

18 Ibid., p. 52-55.

19  M. Mitchiner, 1979, p. 17.

20  J.S. Deyell, 1990, p. 26.

21 Ibid., p. 21.

22 Ibid., p. 26-32.

23  See M. Mitchiner, 1979, in particular p. 17-195, and J.S. Deyell, 1990, in particular p. 21-43.

24  See for example M.C. Horton, 2004a.

25  J. Cribb, pers. comm. 2011 for the earliest date, and J.S. Deyell, 1990, p. 44-46 for the latest, with S. Goron, J.P. Goenka, 2001, p. xxii suggesting around 870.

26  S. Goron, J.P. Goenka, 2001, p. xxiii-xxiv and J.S. Deyell, 1990, p. 47.

27  G.S.P. Freeman-Grenville, 1962(b), p. 45-57 and H.N. Chittick, 1965(a), p. 276.

28  J.d.V. Allen, 1982.

29  R.L. Pouwels, 2002, p. 405.

30  L. Harries, 1964, p. 225.

31  J.E.G. Sutton, 1990, p. 57-59 and M.C. Horton, J. Middleton, 2000, p. 27-28, among others.

32  M.C. Horton, 1996.

33 M.C. Horton, H.W. Brown, W.A. Oddy, 1986.

34  M.C. Horton, 1986.

35  G.S.P. Freeman-Grenville, 1988, IX, p. 1.

36 The dating of al-Hasan ibn Sulaiman is probably the most secure of all rulers of Kilwa and this general date is based on the re-evaluation I carried out for my PhD thesis.

37  The three coins were published by H.W. Brown, 1991. H.W. Brown, 1993, p. 13 speaks of four coins, which presumably is a mistake.

38  M.C. Horton, J. Middleton, 2000, p. 93.

39  There now can be little doubt about this, as a deposit from Songo Mnara shows—J. Perkins, J. Fleisher, S. Wynne-Jones, 2014.

40  H.W. Brown, 1996, p. 369.

41 Ibid., p. 369.

42  J. Perkins, J. Fleisher, S. Wynne-Jones, 2014.

43  J. Walker, 1936 and 1939.

44  H.W. Brown, 1996, p. 368.

45  The full results can be found in my PhD thesis, J. Perkins, 2014, Chapters 6 and 7 in particular, and will be published separately.

46  F. Stuhlmann, 1909, p. 854-861.

47  A list of all publications here would be too numerous. The following is a selection: G.S.P. Freeman-Grenville, 1957, 1960, 1971 and 1978; H.N. Chittick, 1973, 1974 and 1984; H.W. Brown, 1970, 1991, 1993 and 1996; M.C. Horton, H.W. Brown, W.A. Oddy, 1986; S. Album, 1999; J. Fleisher, S. Wynne-Jones, 2010; and S. Wynne-Jones, J. Fleisher, 2012.

48  This includes around 11,000 published by G.S.P. Freeman-Grenville, 1957, p. 179, Table VI; 4,500 coins found and excavated by H.N. Chittick; 1,700 coins collected by Perrot; the 2,000 coins of Mtambwe Mkuu; the c. 500 coins that have been found at Songo Mnara so far; and various smaller publications.

49  H.W. Brown, 1992, p. 9.

50  M.C. Horton, 1996, p. 377.

51  H.W. Brown, 1996, p. 369-372.

52 Ibid., p. 373.

53  H.N. Chittick, 1984, p. 213 lists only 6 coins. I discovered a further coin from the Manda excavation while working on these coins at the British Institute in Eastern Africa in 2010. In style it looks more like the Shanga coins of group A and B, but carries the name “Yusuf”.

54  See H.N. Chittick, 1984, p. 213.

55  H.W. Brown, 1996, p. 373.

56  See N.D. Nicol, 2006.

57  N.D. Nicol pers. comm. 2012.

58  H.N. Chittick, 1966(b), p. 11-12 and, 1974, p. 270.

59  A. Juma, 2004.

60  M.C. Horton, 1986

61  As Philippe Beaujard pointed out to me in pers. comm. 2012, the names Ali and Bahram, which are listed below, might suggest Shii and Persian influences.

62  This rough figure is reached by subtracting the 2,000 coins of Mtambwe Mkuu and the c. 4,500 Zanzibar coppers from the total of 20,000 mentioned above.

63  N.M. Lowick, 1985, p. 1.

64  H.W. Brown, 1993, p. 11-12.

65  J. Perkins, 2014, p. 219-233.

66  H.W. Brown, 1970.

67  H.W. Brown, 1993, p. 12.

68  See H.N. Chittick, 1961, 1974, and 1984; M.C. Horton, 1996; A. Juma, 2004; S. Wynne-Jones, J. Fleisher, 2010 and 2011.

69  J.S. Kirkman, 1954, p. 149.

70  This is shown in the archaeological evidence at Shanga; see M.C. Horton, 1996, p. 376.

71  A hoard found at this site, which the author has been analysing, was sealed in the foundations of a house and probably dates to the late 14th or early 15th century; Wynne-Jones pers. comm. 2012. Nearly all known coin types were contained within this hoard.

72  G.S.P. Freeman-Grenville, 1962b, p. 103-108.

73  H.N. Chittick, 1982, p. 54.

74  T. Huffman, 1972, p. 362 and Plate I.

75  G.S.P. Freeman-Grenville, 1988, IX, p. 2 and p. 6. Also, I have seen some letters at the British Museum from N. Lowick to the excavators identifying some of the coins.

76  G.S.P. Freeman-Grenville, 1988, IX.

77 Ibid., IX, p. 9.

78 http://www.powerhousemuseum.com/collection/database/?irn=305381. The records are, however, outdated and the images do not match the registered objects.

79  Ian McIntosh, pers. comm. 2013.

80  Two of these are now in the National Museum of Tanzania in Dar es Salaam. One was published by H.N. Chittick, 1961, p. 12; the other is probably the one mentioned by H.N. Chittick, 1965b, p. 4; both come from Mafia. A further five South Indian Chola coins are mentioned by M.C. Horton, forthcoming and pers. comm, to have come from Mkokotoni on Zanzibar.

81  Mostly published by G.S.P. Freeman-Grenville, 1960, p. 42. Further additions are H.N. Chittick, 1974, p. 273-301 and M.C. Horton, H.W. Brown, W.A. Oddy, 1986, p. 117, Table 1. Even with these additions, as far as I am aware, the numbers have hardly changed since G.S.P. Freeman-Grenville’s publication.

82  M.C. Horton, forthcoming.

83  This is further confirmed by M.C. Horton, forthcoming.

84  D. Whitehouse, 1970.

85  N.M. Lowick, 1985, p. 1.

86  See N.M. Lowick, 1985, p. 11-63.

87 Ibid., p. 1 and 6.

88  See M.C. Horton, 1996 and M.C. Horton, 2004b.

89  H. Brown, 1993; M.C. Horton, 1996, p. 399-400; M.C. Horton, J. Middleton, 2000, p. 92-94; and J. Middelton, 2003, p. 514-516.

90  J. Perkins, 2014; most of Chapter 8 is dedicated to this discussion.

91  The gold coins are too singular in their appearance and issue, and combined with their unclear find-circumstances they cannot be considered in this discussion at this stage.

92  S. Wynne-Jones, J. Fleisher 2010 and 2011, as well as pers. comm. with S. Wynne-Jones and J. Fleisher.

93  S. Wynne-Jones, J. Fleisher, 2012.

94  See J. Perkins, 2014, in particular p. 259-262.

95  M.C. Horton, 1996, p. 97 and p. 101, Plate 49.

96  G.S.P. Freeman-Grenville, 1960, p.37.

97  J. Walker, 1936, p. 74-75.

98  G.S.P. Freeman-Grenville, 1978, p. 194.

99  H.W. Brown, 1996, p. 371.

100  M. Broome, 1985, p. 57-60.

101  R.L. Pouwels, 2002, p. 424-425.

102  M. Walsh, 2010, p. 469.

103  S. Goron, J.P. Goenka, 2001, p. xxii.

104  Although there appear to be doubts regarding the identification of the ancient Debal/Daybul with the modern site of Banbhore, see: M. Kervran: http://www.orient-mediterranee.com/spip.php?article1243&lang=fr. This, however, has no bearing on the current discussion.

105  H.P. Ray, 2004, p. 45.

106  J.S. Deyell, 1990, p. 44-45 and p. 47.

107  S. Goron, J.P. Goenka, 2001, p. xxii. While S. Goron and J.P. Goenka do not give a reason for the dating, it is possibly based on the demise of the caliphal rule and the rise of the Qurashites and subsequently the Qarmatians here from c. 880 onward. This is briefly described by J.S. Deyell, 1990, p. 46-47.

108  Cribb, pers. comm.

109  E. de Zambaur, 1927, p. 279. Daud ibn Yazid and Bashir ibn Daud.

110  M.C. Horton, J. Middleton, 2000, p. 48; R.L. Pouwels, 2000, p. 251.

111  R.L. Pouwels, 2001, p. 644.

Haut de page

Table des illustrations

Titre Figure 1: Map of the Indian Ocean with relevant places
Crédits After G.F.P. Freeman-Grenville, 1962, Map 1.
URL http://afriques.revues.org/docannexe/image/1769/img-1.jpg
Fichier image/jpeg, 144k
Titre Figure 2: Abbasid copper coin of al-Mansur. Baghdad mint 773
Crédits Copyright British Museum. Coins and Medals IOLC.6044.
URL http://afriques.revues.org/docannexe/image/1769/img-2.jpg
Fichier image/jpeg, 64k
Titre Figure 3: Abbasid silver coin of Harun al-Rashid. Basra mint 797
Crédits Copyright British Museum. Coins and Medals 1847.0401.59.
URL http://afriques.revues.org/docannexe/image/1769/img-3.jpg
Fichier image/jpeg, 44k
Titre Figure 4: Abbasid gold coin of al-Mutamid, 4.12 grams. Baghdad mint, 875
Crédits Copyright British Museum. British Museum Coins and Medals 1936,0605.3.
URL http://afriques.revues.org/docannexe/image/1769/img-4.jpg
Fichier image/jpeg, 44k
Titre Figure 5: Fatimid gold coin with circular design of al-Muizz. Misr (Egypt) mint, date ?
Crédits Copyright British Museum. Coins and Medals 1853,0406.65.
URL http://afriques.revues.org/docannexe/image/1769/img-5.jpg
Fichier image/jpeg, 56k
Titre Figure 6: Coin of the Amirs of Sindh. Coin of Abdallah
Crédits Copyright British Museum. British Museum Coins and Medals 1996,0609.07.
URL http://afriques.revues.org/docannexe/image/1769/img-6.jpg
Fichier image/jpeg, 12k
Titre Figure 7: Coin of the Amirs of Sindh. Coin of Muhammad
Crédits Copyright British Museum. Coins and Medals 1996,0609.02.
URL http://afriques.revues.org/docannexe/image/1769/img-7.jpg
Fichier image/jpeg, 12k
Titre Figure 8: Shanga group A type coin, Muhammad
Crédits Image drawn from M.C. Horton, 1996, Pl. 131, 9.
URL http://afriques.revues.org/docannexe/image/1769/img-8.jpg
Fichier image/jpeg, 20k
Titre Figure 9: Shanga group B type coin, Abd Allah
Crédits Image drawn from M.C. Horton, 1996, Pl. 131, 15.
URL http://afriques.revues.org/docannexe/image/1769/img-9.jpg
Fichier image/jpeg, 20k
Titre Figure 10: Manda coin 1 with round design (coin in poor condition)
Crédits British Institute in Eastern Africa
URL http://afriques.revues.org/docannexe/image/1769/img-10.jpg
Fichier image/jpeg, 8,0k
Titre Figure 11: Ali ibn al-Hasan coin of Mtambwe Mkuu
Crédits Drawn from M.C. Horton, H.W. Brown, W.A. Oddy, 1986, Plate 1.1.
URL http://afriques.revues.org/docannexe/image/1769/img-11.jpg
Fichier image/jpeg, 28k
Titre Figure 12: Kilwa coin of Ali ibn al-Hasan—two-line type—similar to silver coins
Crédits Tanzania Coin Collection 36.
URL http://afriques.revues.org/docannexe/image/1769/img-12.jpg
Fichier image/jpeg, 40k
Titre Figure 13: Kilwa coin of Sulaiman ibn al-Hasan
Crédits Tanzania Coin Collection 670.
URL http://afriques.revues.org/docannexe/image/1769/img-13.jpg
Fichier image/jpeg, 36k
Haut de page

Pour citer cet article

Référence électronique

John Perkins, « The Indian Ocean and Swahili Coast coins, international networks and local developments », Afriques [En ligne], 06 | 2015, mis en ligne le 21 décembre 2015, consulté le 21 août 2017. URL : http://afriques.revues.org/1769 ; DOI : 10.4000/afriques.1769

Haut de page

Auteur

John Perkins

Senior Research Associate, Flinders University

Haut de page
  • Logo Institut des mondes africains
  • Logo DOAJ – Directory of Open Access Journals
  • Les cahiers de Revues.org