Asiatic influences on Somali culture 1

  • Published on

  • View

  • Download


This article was downloaded by: [University of North Carolina]On: 11 November 2014, At: 23:10Publisher: RoutledgeInforma Ltd Registered in England and Wales Registered Number:1072954 Registered office: Mortimer House, 37-41 Mortimer Street,London W1T 3JH, UKEthnos: Journal ofAnthropologyPublication details, including instructions forauthors and subscription information: influences on SomalicultureVinigi L. Grottanelli aa Museo Nazionale PreistoricoEtnografico , RomaPublished online: 20 Jul 2010.To cite this article: Vinigi L. Grottanelli (1947) Asiatic influences onSomali culture , Ethnos: Journal of Anthropology, 12:4, 153-181, DOI:10.1080/00141844.1947.9980670To link to this article: SCROLL DOWN FOR ARTICLETaylor & Francis makes every effort to ensure the accuracy of allthe information (the Content) contained in the publications on ourplatform. However, Taylor & Francis, our agents, and our licensorsmake no representations or warranties whatsoever as to the accuracy,completeness, or suitability for any purpose of the Content. Any opinionsand views expressed in this publication are the opinions and views ofthe authors, and are not the views of or endorsed by Taylor & Francis.The accuracy of the Content should not be relied upon and should beindependently verified with primary sources of information. Taylor andFrancis shall not be liable for any losses, actions, claims, proceedings,demands, costs, expenses, damages, and other liabilities whatsoeveror howsoever caused arising directly or indirectly in connection with, inrelation to or arising out of the use of the Content. article may be used for research, teaching, and private studypurposes. Any substantial or systematic reproduction, redistribution,reselling, loan, sub-licensing, systematic supply, or distribution in anyform to anyone is expressly forbidden. Terms & Conditions of accessand use can be found at by [University of North Carolina] at 23:10 11 November 2014 influences on Somali culture1VINIGI L. GROTTANELLIMuseo Nazionale Preistorico-Etnografico,RomaI. The influence of Asiatic cultures on the Bantu populations ofthe East-African coast has been studied by a number of authorsduring the last few decades and is now a well documented pheno-menon in some of its aspects.2 It is an accepted fact that severalelements both of the material and spiritual culture of these Bantucan be traced back to Asiatic sources, and these have in turn beenidentified with different geographical and cultural areas, viz. Arabia,Persia, India, and Indonesia. Influences from these areas, possiblywith exception of the last, often appear intermingled: this is due tothe actual admixture of Arab, Persian and Indian elements and totheir often parallel action in space and time in the millennial processof their commercial colonization of East-African coasts. Hirschbergis thus justified, up to a certain extent, in speaking of an arabisch-1 A short lecture on this subject was given by the author at a meeting of theInstitute of Anthropology, Rome University, on March 29th, 1947. The authorexpresses his thanks to Prof. C. Conti Rossini, Roma, Univ. Dozent S. Lagercrantz,Uppsala, Prof. K. G. Lindblom, Stockholm, and Dr. A. E. Steinmann, Zrich,who helped him with valuable suggestions and information while he was writingthis paper.2 See especially: M. Devis, Le pays des Zendjis ou la Cte Orientale d'Afriqueau Moyen-Age, Paris 1883; O. Baumann, Usambara und seine Nachbargebiete,Berlin 1891; F. Stuhlmann, Handwerk und Industrie in Ostafrika, Hamburg 1910;E. Werth, Das deutsch-ostafrikanische Kstenland, 2 Vols., Berlin 1915; W.Schilde, Die afrikanischen Hoheitszeichen, ZfE 61 (1930), 138; W. Hirschberg,Die arabisch-persisch-indische Kultur an der Ostkste Afrikas, MAG Wien, LXI(1931), 5. Heft; J. Hornell, Indonesian Culture in East Africa, Man XXVIII(1928), 1-4; the same, Indonesian Influence on East African Culture, JRAILXIV (1934), 305-332; U. Monneret de Villard, Note sulle influenze asiatichenell'Africa Orientale, RSO XVII, Roma 1938, 303-349; W. Hirschberg, Asia-tische" Kultureinflsse an der Ostkste Afrikas, Kol. Rdsch. XXXIV, Leipzig 1943,49-57. For fairly complete bibliography, see Monneret, op. cit. Further contri-butions by S. Lagercrantz are summed up by Hirschberg (1943) 53-54. Also seeG. Lindblom, Die Stosstrommel, insbesondere in Afrika, Ethnos 1945. 1, 37.153Downloaded by [University of North Carolina] at 23:10 11 November 2014 Ethnos No. 4. 1947indisch-persische Kultur, though other scholars, like H. Baumann,more correctly keep the indischen Einwirkungen distinct from thepersisch-arabische Mischkultur.Ethnologists, however, have not so far devoted the same attentionto the similar problems in the area immediately adjoining Bantu EastAfrica towards the north-east, viz. the Somali peninsula. Historianshave given due attention to ancient and modern Asiatic establishmentson this coast; the oriental character of the Somali coastal towns haslong been recognized; the well known Somali traditions referringto an Arab origin of their people have caused much speculation ontheir ethnical relation to the Semites; and it has always been obviousthat Islam has deeply affected their spiritual culture and, to a lesserdegree, their sociology. But, as I say, strictly ethnological enquiriesinto the Mischkultur's influences, especially in the field of materialculture, have not so far been extended to the Somali, with the fewexceptions we shall shortly see. Cerulli, as a matter of fact, explicitlyrecognized the presence of Arab, Persian, Indian and Indonesiancultural currents among the Somali, but merely from a historicalviewpoint.3 It is significant that the two main existing works onSomali ethnography, Paulitschke's and Puccioni's,4 contain no mentionof this problem.In 1946, I had the opportunity of dealing personally with a large,fairly complete ethnographical collection from southern Somalia,5 andwas thus induced to devote my attention to the Asiatic affinitiesshown by a number of these items. To a somewhat experienced eye,a non-Hamitic object is at once recognizable out of the monotonous3 E. Cerulli, L'Islam nell'Africa Orientale, in Aspetti e problemi attuali delmondo musulmano, 87-89, Roma, R. Accad. d'Italia, 1941.4 Ph. Paulitschke, Ethnographie Nordostafrikas, Berlin 1893-96; N. Puccioni,Antropologia e etnografia delle genti della Somalia, Vol. III, Bologna 1936.5 These collections, gathered in 1907 and 1908 by order of the Government ofItalian Somalia in five different districts (Merca, Brava, Gelib, Lugh, Bardera),were delivered to the Reale Museo Preistorico-Etnografico L. Pigorini, Rome(now Museo Nazionale, hereafter indicated as MPE) in 1911. They consist of over1000 items, and could not be exhibited, owing to lack of space, until 1946, whena new gallery was made available for them.154Downloaded by [University of North Carolina] at 23:10 11 November 2014 Vin igi L. G r ot t anelli: Asiatic influences on Somali culturemass of Hamitic ones, in which the uniform reddish-brown colouringsof hide and leather, crude pottery, wood, and plaited reeds, withtheir characteristic lack of phantasy and decoration, are remarkablypredominant. The few coarse implements of the Somali and otherCushitic tribes fully respond to the limited needs of a nomadic life, inwhich cattle alone matters; they are in harmony, as it were, with thebarren steppe and bush in which they wander. As soon as we come tomore elaborate types of implements, on practically any metal tool, oncotton and silk fabrics with brightly coloured patterns, on delicatelycarved wooden objects and so on, a foreign influence can at oncebe suspected. Even the rare horn and ivory ornaments, the beadsand cowries of doubtless ancient introduction, seem to partake inthe general dullness, confirming the common impression given by trueHamitic material culture, which always betrays indifference tocolours and shapes, and an almost complete absence of artistic tasteand imagination.In this respect, as well as from many other viewpoints, contrastwith the neighbouring Bantu is striking. Though they have nowbeen in contact with them for centuries, the Somali have copied andadopted very little from the latter, except a few reluctantly usedagricultural tools: the universal contempt in which Negroes are heldby Hamites well accounts for this fact. The same, though more forsocial-historical than for racial reasons, applies to the Galla, whomalso the Somali openly despise. "With the Abyssinians the latter havehad little intercourse as a whole; national and religious antagonism,born perhaps even before Gran's wars, has no doubt been embitteredby the frequent Amhara raids and razzias since the Abyssinianconquest of the Harar region and of southern Ethiopia. Practicallyall additions to Somali material culture come therefore from theeastern countries overseas. This is further proved by the differenceexisting between the Somali culture in the coastal area and that ofthe hinterland, where external influences have found it more difficultto penetrate.68 Not only did the Shirazi and their forefathers trade with the Somali coastI5SDownloaded by [University of North Carolina] at 23:10 11 November 2014 Ethnos No. 4.In some cases, a direct Arab (Mohamedan) origin is self evident:tisbas (islamitic rosaries), wooden koranic tablets, kohl bottles, horsetrappings, most of the silverwork, such as bracelets, anklets, amulet-cases etc., are not only of Arab type,7 but are often produced byArab (or Persian) craftsmen living in the coastal towns of Somalia.The same applies to arms of various oriental designs: I myself boughtin Mogadiscio, in 1933, a handsome ivory-handled, silver-sheatheddagger of Persian model, manufactured by an Arab silversmith ofthat town. Such fancy arms are however produced for the use offoreign minorities, and are not used by the Somali proper, exceptpossibly by a few orientalized chiefs. Arabo-Persian influenceson the stone-and-mortar architecture of the coastal towns, as wellas on the manufacture of wooden doors and windows, have easilybeen recognized. Monneret, for one, has recently pointed out somePersian characteristics of the Mogadiscio minarets8; and if a thoroughstudy of structural and decorative elements in Somali coastal archi-tecture were made, it is highly probable that direct relationship tocorresponding Arab, Persian, and also Indian elements would beascertained on a more general scale, as it was not difficult for vonLuschan and others to do with regard to the Swahili area furthersouth.9In spite of the fact that weaving has been known in the Hamiticworld for a very long time, the advanced character of Somali weav-ing technique is doubtless due to Asiatic influence. The fine, multi-coloured cotton textiles produced in the coastal towns, and thencefor times immemorial, and settle down there at least ten centuries ago; and notonly do their descendants live there to this day; but some Somali themselvesbecame sea-faring people. The Mijurtin used to sail on their small crafts as faras Zanzibar and Bombay (G. Toni, II paese somali, in L'Esploratore, IX, feb.1885, 58.7 Revoil (La valle du Darror, 316, Paris 1888) thought that Somali ornaments,though imported by the Arabs, rather had a Jewish character.8 U. Monneret de Villard, I minareti di Mogadiscio, RSE III (Rome 1943),127-130.9 F. V. Luschan, Fremder Einfluss in Afrika, Westerm. Monatsh., Sept. 1898,quoted by Stuhlmann, Handwerk etc., 94-97; W. H. Ingrams, Zanzibar, itsHistory and its People, 217 ff., London 1931.xS6Downloaded by [University of North Carolina] at 23:10 11 November 2014 Vinigi L. Grottanelli: Asiatic influences on Somali culturelargely traded overseas, have little in common with the strong butrough fabrics made by Hamitic weavers of the interior.Indonesian influences, as I said, tend to have a place of their ownin the general picture of this Mischkultur: they are no doubt lessdiffused, and therefore less known. Among the Bajuni of Chisimaioand of the islands off the Juba river mouth (who, properly speaking,are not Somali), Puccioni noted Indo-Melanesian cultural elementssuch as the use of outrigger canoes and the system of sea-turtle fishingwith the aid of the sucking-fish,10 but neither of these elements isfoun,d further north. Among those which we do find, I will mentionhere coconut scrapers (called mbusi in Brava), coconut vessels, andthe plangi system for dyeing hagog kerchiefs etc. I do not think Ican agree with Stuhlmann's cautious, and Hornell's rather sweepinginference, of the Indonesian origin of rectangular huts with mudwalls and palm-leaf (or thatched) roof11: a type of hut, anyhow,which I have also seen along the Somali coast, from Jubalnd rightup to Hafun in the north. Apart from these, there are a few morecultural elements, which have hitherto escaped attention, and most ofwhich show an Indonesian connection of some sort. To these Ishall dedicate a more detailed examination in the course of thefollowing- pages. But of course the present article must only beregarded as a preliminary essay, and as a very inadequate attemptto deal with so vast a subject as the title suggests.I shall avoid entering upon the strictly historical questions arisingin connection with Asiatic influences on the Somali area. In thebest paper we possess on this problem as a whole, Monneret has10 N. Puccioni, Le popolazioni della Somalia Italiana, 69-71 , Bologna 1937;of the same author, see also Beobachtungen bei den Badschuni, Ztschr. f. Rassenk.III (1931), and Gli Etiopici meridionali, in Biasutti's Razze e Popoli della Terra,II, Torino 1941, 175-176; D. R. Parenti, I Bagiuni, RSE V (1946), 156-190.11 Stuhlmann, op. cit., p. 85 and 104; Hornell (1934). 327. Quite apart fromthe question of the west- and central-African distribution of rectangular huts,these exist in other parts of inner East Africa (V. L. Grottanelli, I Mao, 180-181,Roma 1940) where they are built by primitive tribes only. Here,. as well as inWest Africa, they can hardly be regarded as a relatively recent element of foreignintroduction. If any migration theory must be followed with regard to this cul-tural element, Frobenius' Melanesian hypothesis appears, at least, less unlikely.r*7Downloaded by [University of North Carolina] at 23:10 11 November 2014 E thn o s No. 4. 1947Fig. I. Somali dagger.already examined this aspects with rare erudition, and readers hadbetter be referred to his article12; much of the historical andarchaeological evidence he quotes has too often been overlookedby anthropologists dealing with this subject.II. The sword and sabre are unknown to the original Hamiticculture. In Abyssinia these weapons were certainly introduced bythe South-Arabs, perhaps two thousand years ago or more, but theynever became popular, Spear and javelin remaining the nationalarms." The Galla were using no swords when they first made their12 Note, etc., v. s.1 3 See C. Conti Rossini, Storia d'Etiopia, 205, Bergamo 1928; the same, Appuntie comenti, RSE III (1943), 103. Maqrz's testimony (see further) is very expliciton this point.158Downloaded by [University of North Carolina] at 23:10 11 November 2014 Vinigi L. Gr ot t anelli: Asiatic influences on Somali cultureFig. II. Bur Eibi rock-engravings (from a photograph in P. Graziosi, L'etdella pietra in Somalia, Tav. I, 2).appearance in history, as late as the XVI century; as a matter offact, they appeared to have no knowledge of metals whatsoever.The Somali, who as well as the Danakil had obtained iron fromthe Semites centuries before the Galla, have long had, and stillretain, a typical cutting weapon, a straight bladed dagger of adesign peculiar to them alone (fig. I). That this weapon is tradi-tionally old among them, is proved by the ancient Somali rock-engravings discovered by Graziosi at Bur Eibi (fig. II),14 whichfaithfully reproduce the unmistakable shape of their hilts andstraight doubleedged blades. The antiquity of this type of daggerwould prove even greater if it were possible to identify it, as14 P. Graziosi, L'et della pietra in Somalia, Firenze 1940.I I IJ9Downloaded by [University of North Carolina] at 23:10 11 November 2014 Ethnos No. 4. 1947mFig. III. Somali dagger-sword (MPE 95774) andsabre (MPE 9J773).Rvoil maintained, with that worn at the belt by Punt chiefs asdepicted in the Deir-el-Bahri wall paintings.15 Be this as it may,it is a fact that the Somali departed so unwillingly from thisparticular type, that when they started forging swords for them-selves they remained faithful to the same model, merely makingthe blade longer (fig. III). They adopted the Semitic word forsword (syf or sf), but Reinisch's texts show that they oftencontinue to call this schwertartiges Dolchmesser bilawa or bilo,i. e. dagger.18 To this day, swords and sabres are very rare amongthem,17 as is the case with most Cushitic tribes.However, MPE collections contain three sabres of a quite different15 La valle etc., 299; also see G. Caniglia, Genti di Somalia, 29, Roma 1935.16 L. Reinisch, Die Somali Sprache, "Wien 1900-1903, I and II vol., passim.17 N. Phccioni, Antropologia etc., III, 76.160Downloaded by [University of North Carolina] at 23:10 11 November 2014 Vinigi L. Grottanelli: Asiatic influences on Somali cultureand striking type, which have little if anything in common eitherwith the original Somali dagger type, or with the well-knownAbyssinian and Galla sabres of Arab origin. Two of these sabres(nos. 41929 and 42262) were collected by Robecchi-Bricchetti in1890; they had belonged to two Aissa-Somali aban (caravan chiefs)of the Wardig and Abgal sub-tribes respectively. The blunt, heavy,one-way-cutting Hades show rude skill in iron workmanship, andleave little doubt as to their having been produced locally. Theircurve is so very slight, that the catalogue describes them as straightblades. The third sabre (no. 95774) is slightly more curved, butto an extent in no way comparable to the almost sickle-like crescentsof Abyssinian and Arab sabres. This last sabre (fig. I l l b) wascollected in Merca in 1892, but further details are lacking.The sheaths of the three sabres are composed of two thin woodstrips, cut to the blade's dimensions, and covered with brownleather. Here is a further difference from most Abyssinian andArabian sabres, whose scabbards are made simply of leather.But it is the hilts that attract our special attention), because theyhave no parallel, as far as I know, in trie whole of East Africaor in Arabia. In literature, I could only find a couple of itemswhich partly correspond to these types. One is shown in Robecchi-Bricchetti18 as belonging to a Mijurtin Somali of Alula; but in thiscase the hilt is more ornate, with decorative lozenges of probablyPersian pattern, and there is a straight wrist guard, again as inPersian swords. The other example is supplied by Paulitschke,19but apart from other differences in design this one has an iron handguard, not to be found with our Somali hilts. It is also possiblethat similar types might be encountered further south, among thecoast Bantu of Kenya and Tanganyika: Stuhlmann possibly referredto such types as these when he spoke of ein gebogenes Schwert,dessen Griff einen starken, zurckgebogenen Knauf, manchmal sogar18 L. Robecchi-Bricchetti, Somalia e Benadir, 216, Milano 1902.19 Ph. Paulitschke, Harar, Taf. S. 112, Leipzig 1888.161Downloaded by [University of North Carolina] at 23:10 11 November 2014 Ethnos No. 4. 1947Fig. IV. Sabre hilts. 1, Somali (MPE 42262). 2, Somali (MPE 9J774). 3, Galla(MPE 31069)- 4 Danakil (MPE 32777). 5, Morocco (MPE 95845). 6, Gelebes(MPE 95939)- 7. Persia (MPE 33262). 8, Java (MPE 1798).i6zDownloaded by [University of North Carolina] at 23:10 11 November 2014 Vinigi L. Grottanelli: Asiatic influences on Somali cultureFig. V.' Sabre hilt. Java (by kind permission, fromDr. A. Steinmann's personal collection, Zurich).einen Bgel hat,20 used by village chiefs of former Deutsch-Ost;but unfortunately he gives no illustration -of. them. Stuhlmannthought of a Persian origin for such swords (or sabres?), and asa rule it is quite true that the Persian type is the most widelydiffused in the Swahili area and elsewhere in East Africa,21 but inour case this view cannot be held owing to the many and widedifferences from Persian models.There are two origins I can suggest for this weapon. It is eithera simplified derivation from the Moroccan model (fig. IV, 5)," orit is an africanized Indonesian type. Though the two suggestedareas of origin are at an immense geographical distance from oneanother, there is no necessary contradiction in making reference20 Handwerk etc., 126.21 See, out of many illustrations, the Zanzibar Arab hilt shown in F. B. Pearce,Zanzibar, 217, London 1920, Merca's governor's one in Robecchi-Bricchetti, op.cit., 108, and the fine Ababde sword hilt in G. A. Hoskins, Travels to Ethiopia,London 1835. Also in Arabia, preference has long been given to Persian arms(C. M. Doughty, Travels in Arabia Deserta, I, 224 and 457, Cambridge 1888).As Zeller puts it, Persien war von jeher das Zentrum orientalischer Waffen-schmiedekunst (Fhrer ber die orientalische Sammlung etc., 10, Bern 1923).22 I have been unable to see Olon, Relation de l'Empire de Maroc, Paris 1645,where (p. 96) I am told a sabre-hilt similar to MPE Somali ones is reproduced.163Downloaded by [University of North Carolina] at 23:10 11 November 2014 Ethnos No. 4. 1947to both at the same time, as the Somali (presumably Tumal) crafts-men1 who produced the sabres can very well have been unconsciouslyinfluenced by both. No. 42262 (fig. IV, 1) could be regarded asa coarse imitation of a Moroccan hilt, deprived of the hand guard,but still retaining the wrist guard. No. 95774 (fig. IV, 2) could,in a sense, simply be a local variation of the previous, but evenso one cannot help being struck, so to speak, by the Indonesian airabout it. The guards have disappeared 'altogether; the curving ofthe hilt-top is more marked, the carvings deeper-set. In generalappearance, this hilt stands more than half-way, as it were, betweenthe north-African and the Malay types. I cannot remember anydirect parallel to this type among Indonesian hilts,23 but both fig.IV, 6 and 8 (Celebes, Java) [and fig. V (reproduced by courtesyof Dr. Steinmann from his private collection of Javanese swords)]show resemblance to our item. Significant comment by a Britishofficial with life-long experience of Borneo tribes was, that bothhilts were similar but not typical of the Malay sword pedanghilt, and also of the short cutting sword parang or doku incommon use among Mohamedans and aborigines of Borneo.It is in no way surprising that also in this field Indonesianinfluences should have reached Somaliland. The "possibility ofMoroccan influences needs however some explanation. As early asthe XV century Abyssinian kings had turned to North Africa intheir attempt to improve national armaments. Maqrz writes:Imperium eius [of king Yeshaq] robur accepit, quod Mamlucusquidam Circassus, de numro eorum qui loricas conficiunt, in regio-nibus Aegypti, ad eum venit, sedesque apud eum fixit. Hie extruxitipsi armamentaria magna, quae arma beilica, veluti gladios, hastas,loricas et plura huius generis continebant. Antea enim arma eorumerant hastae breves quas jaciebant. At that time, the Christianarmies of Abyssinia were at war with the Mohamedan emirates inthe east and south, whose armies included large numbers of Somali.23 Maqrizi, Historia regura islamiticorum in Abyssinia, ed. Rinck, 6, LueduniBatavorum 1790.164Downloaded by [University of North Carolina] at 23:10 11 November 2014 Vinigi L. G r ott a ne Hi: Asiatic influences on Somali cultureHere, then, was an easy opportunity for the latter to get acquaintedwith, or enter in possession of, North-African swords, which veryeasily may have included Moroccan ones. About a century later.and this is even more significant we learn from Chihab Al-DinAhmad24 that warriors from Maghreb were fighting in Gran'sarmies. The swords they were using may well have been taken asa model by their Somali allies. And of course even in more recenttimes occasional opportunities of adopting foreign types of armsmay have presented themselves, and have been made use of, evenby as conservative a people as the Somali. '*III. I here wish to make a passing reference to another peculiarMPE item, the spear shown in fig. VI, i. Information concerning itis incomplete, and though Giglioli's always accurate and reliablecatalogue puts it down to Abyssinia, I doubt whether we canaccept this statement otherwise than in the broader sense of North-East Africa. The description, given in Giglioli's work and repeatedon the label attached to the spear, runs: Lancia africana. Lo stelo stato tagliato, ma vi la ghiera e un ornamento di ferro a spiraleoriginale simile a queilo dei Wollo-Galla. Abissinia."5This type of spear is quite unparalleled, as far as I know, throug-out North-East Africa, in fact in Africa as a whole. The iron bladeitself, long and marrow, slightly triangular in section, is differentfrom the usual Hamitic spear-head. Further, the system of fasteningthe blade laterally to the shaft is altogether un-African. The bambooshaft is also unusual, though not unknown.26 The leather straps, with24 R. Basset, Histoire de la conqute de l'Abyssinie par Chihab ed-DirtAhmed ben 'Abd el Qder, trad., 331, 336, Paris 1897.25 La collezione etnografica del Prof. E. H. Giglioli geograficamente classificata,Vol. II, 93, Citt di Castello 1912.26 Iron headed spears with long and thick bamboo shafts, used by the Ingassanaof the eastern Sudan, are imported from Abyssinia, Prof. Evans Pritchard was told(A preliminary account etc., SNR X, 1927, 80). In Abyssinia, however, I do notknow them to be common; they occur more frequently among the Galla. Bambooshafts are common in India (W. Egerton, Handbook of Indian Arms, 123,London 1880), and we know that they used to be imported into Arabia (R. F.Burton, A pilgrimage to Mecca and Medina, II, 247, Leipzig 1874). B.'s de-scription is insufficient to identify their type.165Downloaded by [University of North Carolina] at 23:10 11 November 2014 Ethnos No. 4. 1947Fig. VI. Spear-heads. 1, EastAfrica (MPE 17024 G). 2, Galla(MPE 41979). 3, Dayak, Borneo(MPE 1579). 4, Borneo (MPE80222).which the blade is strongly secured, do suggest on fhe other handthe make by a cattle-breeding tribe, and the iron spiral is no doubtidentical in form and material to those mentioned by Giglioli (cp.fig. VI, 2), though similar spirals, fitted close to the blade, are alsoto be found on spears from Bantu Africa, e. g. among the Rotse,the Luena and several other tribes.The only parallel I can think of for this puzzling and ratherhybrid-looking weapon is the Indonesian sumpitan, in some cases ablowing-pipe and a spear at the same time. Here there is of coursea practical motive for fastening the spear-blade sideways, in ordernot to obstruct the shaft's hollow (fig. VI, 3); but the same methodis also applied, in Borneo, to common spears (fig. VI, 4), where theabove motive no longer holds. Rotang, or thin brass wire, neverleather straps, are naturally used in Indonesia.166Downloaded by [University of North Carolina] at 23:10 11 November 2014 Vinigi L. G r o 11 a ne Hi: Asiatic influences on Somali cultureIV. Now to return to our subject proper. The presence of thesame cultural item both in SomalilaBd and in Indonesia does notalways necessarily imply, of course, that the latter is to be regarded asthe area from which such items originate. This is the case of elementsintroduced by Mohamedan culture in both areas, as for instancemen's skull-caps made of straw or other vegetable fibres, and toe-peg sandals.The former have a fairly wide distribution in Africa, thoughthey appear to be typical of the Swahili area. On the other handthey are rare and particularly foreign-looking among the Hamiticpeople of East Africa, who nearly always go bare-headed and areproud especially the Somali of their fuzzy hairdressing.In the coastal area of Tanganyika, these caps called by the Arabname of kofia are made of fibres of different colours, and areworn by the so-called yumbe or village elders.27 Similar caps,though made of uniformly brown fibres, are also used in Somalia,where they are called kofi (fig. VII, i). Here too,, they are notworn by the average man, but only by sfaekhs or priests (wadadin),or by the hajji on their return from Mecca, as a variant of thekofiyyah proper, i. e. of the well-known embroidered white caps.28These plaited fibre are not altogether, as one might believe, a purelyAfrican form of the kofiyyah: Stuhlrnann states that the abovementioned Swahili caps are usually not of native make, being mainlyimported from Bombay, and Ferrandi informs us that the Somaliones are lavoro proveniente d'Arabia, e specialmente da Ged'da,I need not go into the African and extra-African distribution ofthis item, but it may be assumed that in Congo (fig. VII, 4) it wasintroduced by Arab merchants, and a Mohamedan origin can simi-larly be assigned to the caps we find in Madagascar (fig. VII, 3)27 F. Stuhlmann, Handwerk etc., 118-119. For a slightly different type ofSwahili kofiyyah, see Robecchi-Bricchettii op. cit., 579. Also note resemblancebetween the Zanzibar cap published by Baumann (Volk. v. Afr., 200) and theHova one reproduced above (fig. VII, 3), with the typical fringe.28 U. Ferrandi, Lugh, emporio commerciale sul Giuba, 264-265, footnote, Ro-ma 1903. MPE items are simply described in the catalogue as berrette da prete.167Downloaded by [University of North Carolina] at 23:10 11 November 2014 E thnos No. 4. 1947Fig. VII. Plaited skull-caps. 1, Somali (MPE 88957). 2> North Borneo (MPE80941). 3, Hova, Madagascar (MPE 31063). 4, Mayombe, Lower Congo(MPE 64020).and Indonesia (fig. VII, 2), though shape, plaiting technique andmaterial vary slightly in the different areas.V. The wooden sandals with toe-peg, used by men and womenalike, are again an uncommon item in North-East Africa, and onewhose foreign origin cannot be doubted. Hamites (including Abys-sinians) usually go bare-foot, but in the Somali steppe, where soilis often scorchingly hot and strewn with countless thorns, nativeswear sandals made of ox, giraffe, or buffalo hide, usually with onelarge leather strap across to hold the foot. "Wooden sandals, held168Downloaded by [University of North Carolina] at 23:10 11 November 2014 Vinigi L. Grottanelli: Asiatic influences on Somali cultureEg. VIII. Toe-peg sandals, i, i a, Swahili (MPE 37205). 2, 2 a, Somali, Harar '(MPE 42184). 34, Arabia (from C. Niebuhr).high from the soil by means of two heel-like salients under the sole,carved out of one single piece of wood, are sometimes used by thehighland populations during the rainy season, to prevent the footfrom sinking in the mud, and a variety of these is to be found alsoamong the Somali. But even in these types, the foot is held by straps.The Harar-Somali toe-peg specimen (fig. VIII, 2) scarcely has any169Downloaded by [University of North Carolina] at 23:10 11 November 2014 Ethnos No. 4. 1947parallels from the whole Hamitic area. Ferrandi mentions a similartype in Lugh,29 and a specimen or two from Eritrea can be seen inMC.30 On the other hand, this item appears to be quite commonalong the Swahili coast (fig. VIII, i),31 and in all likelihood it haspenetrated from there into the interior, very much through the samechannels as the skull-cap. Migeod, who carefully noted this item innorth Congo, puts it down to die Arabiss.32 MPE has a specimenof these sandals from the Acholi of Uganda,33 and MME has onefrom the Upper Nile.33a Toe-peg sandals are also to be found inSouth Africa, where, however, they are not worn by the Ban.a, butby the Malays of the Cape (specimens in PRM and SEM).The geographical distribution of this item outside Africa is wide,and provides us with more evidence that it typically belongs to theoriental Mischkultur. Toe-peg sandals existed in Arabia in the18th century, as Niebuhr's illustration shows (fig. VIII, 34),34 andthey are still worn to this day in the Persian Gulf region33; but Iam told' that in Arab' countries they are no longer common. Similarsandals are worn in many parts of India, where they acquire moreelaborate shapes and finer decoration: they can be lacquered andpainted, as the MPE specimen shown in fig. IX, 5, or minutely andartistically carved (see the fine PRM specimens in Oxford).38 Even29 Ferrandi, op. cit., 220.30 Catalogue Nos. 8828 to 8832.31 M. Guillain, Documents sur l'histoire . . . de l'Afrique Orientale (1846-1848),II, 84, Paris s. d.; Voeltzkow, Reise in Ostafrika, II. 1, 21, and I. 1, pl. 19;F. Stuhlmann, op. cit., 112; A. Juenger, Kleidung und Umwelt in Afrika, 140,Leipzig 1926; H. Baumann, Vlkerkunde v. A., fig. p. 200; Ingrams, op. cit., 310.32 F. W. H. Migeod, Across Equatorial Africa, 194, 264, London 1923.33 R. Boccassino, Una raccolta di oggetti etnografici degli Acioli dell'Uganda,Ann. R. 1st. Sup. Or., Napoli, IX, 1937, 22-23 and pl. VII, 3.33a Catalogue no. AF 54 b, c. Other MME toe-peg sandals are from Zanzibar(AF 4401 a) and Danakil (AF 4401 b). Of a further specimen (AF 2647 A) noexact catalogue indication is given (possibly Kenya or Uganda). I owe thisinformation to the courtesy of P. P. Maarschalkerweerd, O. F. M.34 C. Niebuhr, Description de l'Arabie, pl. II, Amsterdam 1774.35 F. Stark, Baghdad Sketches, 194, Leipzig 1938.36 Also see Sonnerat, Voyage aux Indes Orientales et la Chine, pl. 12 andtext vol. I, 50, Paris 1782. The Somali specimen also has a (quite un-African)turned and varnished red knob; but lacquering and the use of the turning lathe170Downloaded by [University of North Carolina] at 23:10 11 November 2014 Vinigi L. Grottanelli: Asiatic influences on Somali culture5 a7 8Fig. IX. Toe-peg sandals. 5, j a, India, Malabar (MPE 4470). 6, 6 a, Acholi,Uganda (MPE 960289). 78, Indonesia (from drawings kindly suppliedby Mr. G. L. Tichelman, Amsterdam).more elaborate forms of toe-peg sandals occur in Indonesia, particu-larly in Java, where they are called gamparan. The sole is some-times inlaid with tortoise-shell; in one specimen (fig. IX, 8) it ishave long been known in Harar. R. F. Burton, First Footsteps etc., London 1856,has a coloured table showing similar turned knobs on a red and green varnishedHarar divanI 7 iDownloaded by [University of North Carolina] at 23:10 11 November 2014 E thn o s No. 4. 1947carved to reproduce the shape of a fish. In some cases, the knobis made of horn or white bone with red ornamental patterns etc.,or shaped as a lotus flower. A rarer type, also found in Java, isentirely made of cast copper and richly decorated.87Tihe fact that these sandals were unknown in Indonesia duringthe Hinduistic period, and are undoubtedly not pre-Mohamedan,38seems to me to point out the probability that they were introducedalong with islamic culture from India to Indonesia, but that theywere not originally Indian. Islam was brought to Indonesia throughGujarat, a part of India particularly open to Arabo-Persian influ-ences. In India herself, localities where (PRM) specimens of thesesandals were brought from (North-Western Province, Amritsar inPunjab, Muzzaffarpur in Bengal) are either predominantly Mo-hamedan, or contain strong Mohamedan minorities. It thus seemslikely that the cradle of this culture element should lie further west,probably as I have said in Arabia.Cape specimens show, at any rate, that this item has in turn beenbrought westward to South Africa by (Mohamedan) Malays orIndonesians. On the other hand, a more direct Arab origin is almostcertain for the East- and Central-African specimens.VI. Influences of the oriental Mischkultur have also reachedSomaliland in the field of musical instruments, though it has oftenbeen said that the Somali have no musical instruments at all.39Paulitschke, having written that sie [die Nordostafrikaner] pflegenkeine Instrumentalmusik, goes on however to give a short descrip-tion of flutes, horns, bells, drums, castanets and wooden pipes asbelonging to Hamitic culture and known to the Somali.40 A more37 / . E. Jasper en Mas Pirngadie, D e inlandsche Kunstnijverheid in Neder -landsch Indie, Vol. V, De bewerking van niet-edele metalen, 1930, fig. 227; seealso L. Th. Mayer, Een blik in het Javaansche Volksleven, 471 ; H. H. Juinboll,Katalog des ethn. Reichsmuseums, I X , Leiden 1914, 1 0 9 n o .38 I am indebted for this information to Mr. G. L. Tichelman, Amsterdam.39 G. Revoit, op. cit., 335; G. Ferrand, Les omalis, 204, Paris 1903.40 Paulitschke, op. cit., I, 148.I72Downloaded by [University of North Carolina] at 23:10 11 November 2014 V in i g i L. Grottanelli: Asiatic influences on Somali culturecomplete list was given by Puccioni,41 who added to the previouslymentioned Somali instruments also another type of (wooden, cylin-drical) trumpet (as distinct from horn-trumpets), Triton shells, andlastly a so-called clarinet-trumpet.These last two items certainly have direct Asiatic connections.Triton shell trumpets are so common, that they deserve no more thana passing reference. Quoting Waitz' opinion, that the similar Hovaand Sakalava shell-horns should be considered Malayo-polynesian,and von Pappenheim's, that on account of their Madagascar namethey should be regarded as South-Arab, Schilde sensibly noted that,as a matter of fact, this instrument is spread von neolitischenLigurien bis zu den Azteken.42 Recent classifications, anyhow, in-clude South-East Asia, Indonesia, Madagascar and East Africa in thedistribution area of the transverse or skte-Jblown conch trumpet.43All three MPE specimens of Somali Triton shell trumpets are side-blown, so that this item fits in nicely into HornbosteFs Hovaclass. I will simply note, however, that the apex-blown, and not theside-blown type, is predominant in Indonesia as well as in India.44According to MPE catalogue, the Somali Triton conch {bun) is usedby the sagalle or sub-chiefs, and is blown to call pacific meetingsas well as to assemble warriors in wartime. Specimens of it werecollected in Merca, Brava and Gelib.41 Ant ropo log ia etc., III, 9 0 - 9 2 . I m a y a d d tha t M P E collections also includea five-str ing lyre f rom Somalia ( N o . 61826), obviously of Abyssinian model .42 Schilde, op. cit., 123-124. Endblown Triton shells are used to this day byshepherds in southern Italy to call their flocks.43 E. M. v. Hornboslel, The Ethnology of African Sound-Instruments, AfricaVI (1933), 29t. A. Schaeffner, Origine des instruments de musique, 260, Paris1936, merely follows Sachs' classification, and does not include East Africain the Triton's distribution area. G. Montandon, on the other hand, who assignsthe end-blown shell to the cycle du totem (?), puts the side-blown one down tothe cycles austronsode et soudanode (L'olognse culturelle 720-721, Paris1934).44 C. Sachs, Die Musikinstrumente Indiens und Indonesiens etc., Berlin 1915,which I quote by memory as it was temporarily unattainable to me; J. Kunst enC. J. A. Kunst-v. Wely, De Toonkunst van Bali, 233, Batavia 1925; A. Lavignac,Enc. de la musique, I, 356-357, Paris 1913.173Downloaded by [University of North Carolina] at 23:10 11 November 2014 Ethnos No. 4. 1947Fig. X. Triton-shell trumpets. 1, Somali (MPE 89097). 2, Sumatra (MPE 1J69).VII. Puccioni's clarinet trumpet (No. 19197, MNA) is an in-teresting item. It has been fully described by him, so I need notrepeat these details, but I reproduce a so-far unpublished drawing ofthe instrument (fig. XI, 4) along with that of a similar MPE item(fig. XI, 1). This instrument had already been noted in Jubalandby Ferrandi, who called it clarinet. He said it was used by theGubahin (Somali ex-slaves of Negro origin) who give it the Swahiliname parapanda.4* It probably corresponds to the instrument thatHornell noticed further south, and to which he refers as bellmouthed clarinet.46 Puccioni has not recorded the Somali name,but the two (fairly identical) MPE specimens of this item are givenas simbar or sombar trumpet.47 They were both collected in theBimal (Merca) region, and are said to be used by ex-slaves of thecoast, when they dance the fantasia called modundo or mudundu:a dance I witnessed myself many years ago on the Uebi Scebeli.45 U. Ferrandi, Lugh, 187. On the Gubahin, see also V. Bottego, L'esplorazionedel Giuba, p. 282, Roma 1900.46 J. Hornell, Indonesian Influence on East African Culture. Journal R. An-thropological Institute 1934, 319.47 One was collected in Merca, the other in Gelib (Juba river) (nos. 89239 and96531). The original catalogues written by local Govt. Residents also mention aninstrument called zumari in Brava, and described as a piva made of brass, hornand wood. Now zumari or zomari is the Swahili corruption of 2amr; but theZanzibar oboe shown in the rather poor photograph in Ingrams, op. cit., pl. facingp. 416 is the Malay and Chinese, not the Arab type. The Brava item, whichprobably constituted the missing link between the oriental and the Somali types,unfortunately never reached Rome.174Downloaded by [University of North Carolina] at 23:10 11 November 2014 Vinigi L. Grottanelli: Asiatic influences on Somali cultureThis fantasia takes place at night, and more often than not, be itsaid incidentally, it degenerates into a wild orgy before dawn.The above mentioned authors have been mistaken as to the in-strument's nature. It is neither a clarinet nor a trumpet, but anoboe, functioning on the double reed principle, and can be classifiedunder Hornbostel's class of the (Pre-Christian West-Asiatic) conicaloboe/8 There is little need to stress that it is not an originallyAfrican instrument. As a matter of fact, I was not aware that inEast Africa it occurred elsewhere than along the Swahili coast untilI came across Barblan's illustration of a similar instrument fromAbyssinia49: a rare specimen there, no doubt, and certainly an im-ported one. A somewhat similar, five-holed oboe, is also used inMadagascar, where Rouget states it to be instrument islamique.50In "West Africa, on the other hand, Hausa and Fulbe offshoots ofthe Arab ghaitah show little resemblance to our item.51Must an islamic origin also be assigned to the latter? Thoughthe oboe, in several slightly different varieties, is spread in all themoslem countries of western and central Asia, I cannot notice anydirect similarity between our item on one side and, on the other,the Persian zourna or sni, the Arab zamr and ghaitah, the Indiansni and ngarra.52 In spite of rougher making (wood employedinstead of brass for the bell) and of the reduced number of finger-holes, the East-African instrument bears a much more striking resem-48 v. Hornbostel, op. cit., 293.49 G. Barblan, Musiche e strumenti musicali dell'Africa Orientale Italiana,72-73 , Napoli 1941. It is said to have a thin, nasal sound, and to be playedat countryside parties and games. The number of finger-holes (from 10 to 7 inAsiatic oboes, five in Somali ones) varies here from 5 to 6.50 J. Fauble, L'ethnographie de Madagascar, 92, Paris 1946.51 For the Arab oboe, see C. Sachs, Geist u. Werden der Musikinstrumente,Taf. 16, fig. 118, Berlin 1929; Schaefjner, op. cit., 279 ff.; J. Kouanet, Arabi(Musica) in EIT, III, 1929.52 For the Persian oboe, see Sachs, Geist etc., fig. 198; and a better illustrationin L. Dubeux, La Perse, tab. 86 fig. 1, Paris 1841 (wrongly given as espce deflte); for the Indian oboe, see G. Knosp, India (Musica) in EIT XIX, 80;Lavignac, op. cit., I, 355.l 175Downloaded by [University of North Carolina] at 23:10 11 November 2014 Ethnos No. 4. 1947Fig. XI. Bell-mouthed oboe. 1, Somali (MPE .89239). 2, Sandakan, Borneo(MPE 8IO6) . 3, China (MPE 31234). 4, Somali (MNA 19197).blance to the corresponding Indonesian item (fig. XI, 2) which inturn appears to have been imported from China53 (fig. XI, 3).It is very difficult to form for oneself a correct idea of the culturalhistory of Asiatic musical instruments in general on the basis ofavailable literature, whose data are often far from complete andsatisfactory. Of this far-eastern oboe, for instance, there is no proper53 This oboe is wide-spread in North Borneo. The Dusuns, "Witti reports inhis diary, are a tribe musical in brass, and their bell metal pans are going onall day long. People further inland he adds have bamboo instrumentsinstead, which tends to confirm that the instrument used along the coast (seeabove, fig. XI, 2) is imported. See H. Ling Roth, The Natives of Sarawak etc.,II, 264, London 1896.176Downloaded by [University of North Carolina] at 23:10 11 November 2014 Vinigi L. Gro11anelli: Asiatic influences on Somali culturemention in such fundamental works as Amiot's, van Aalst's andLaloy's.54 When it is actually dealt with, it is called by differentauthors with a confusing variety of names,55 and hardly any attemptis ever made to trace this, or any other instrument, back to itscultural origin.In my opinion, it is more than probable that this cultural elementreached East Africa from China via Indonesia; it is not the firstmusical instrument which we know to have come to Africa from thelatter area.66 To ascertain its actual birth-place is however a muchmore difficult problem, and one I am not prepared to solve. E.Palazzo, when stating that this oboe is not of Chinese origin, possiblyhad good reasons for this inference. At any rate, it must be kept inmind that China has not only created a number of instruments,thence spread throughout the eastern hemisphere, but has also ab-sorbed and imitated a score of foreign instruments; and has donethis to such an extent, that das heutige chinesische Orchester bestehtfast zu 80 % aus auslndischen Instrumenten.57 It is quite possiblethat the Asiatic oboe had its cradle somewhat in west-central Asia,and that it merely followed the longer route of the Far East andIndonesia before it reached the African shores.54 P. J. Amiot, De la musique des Chinois, etc. (in Mm. . . . par les Mis-sionaires de Pe-kin, Paris 1780, vol. VI); J. A. van Aalst, Chinese Music, Shanghai1884; L. Laloy, La musique chinoise, Paris s. d. (1912).55 Kin kheou kyo according to Lavignac (No. 94); hlang teih according toNataletti, Catalogo descrittivo degli strumenti, etc., Roma 1936 (same name inAlbum . . . du Muse du Cons. Roy. de Mus., Bruxelles s.d., XII, 119); so-naaccording to E. Palazzo (Musica e strumenti di terre lontane, Roma 1933, 41).The latter name recalls the oboe's name in Persia and India (sanai, zourna).56 J. Kunst, Ein musikologischer Beweis fr Kulturzusammenhnge zwischenIndonesien . . . und Zentralafrika, Anthropos XXXI (1936), 131-140; G. Lind-blom, op. cit. Our oboe is probably only one of the many Chinese cultural ele-ments which Indonesians carried with them in the course of their westward jour-neys. The Malays Burckhardt met in Mecca were wearing Chinese dresses andkerchiefs, and using Chinese teapots and copper implements (Viaggi in. Arabia,280, Prato 1844). Also see Burton, A pilgrimage etc., I, 172.57 Kwang-Chi Wang, Musikalische Beziehungen zwischen China und dem We-sten etc., (Studien zur Gesch. u. Kult, des nahen u. fernen Ostens, P. Kahlezum 60. Geb., Leiden 1935, 221).177Downloaded by [University of North Carolina] at 23:10 11 November 2014 Ethnos No. 4. 1947VIII. We have no indication as to the period in which theseAsiatic elements reached present Somalia. All we know is that Asiaticinfluence in the country is far older 5S than the arrival there of theSomali themselves, the latter event having gradually taken placeduring the last eight or nine centuries. At a time when the Somaliwere still settled in the Gulf of Aden region, and had only reachedthe Ocean coast in its northernmost corner, and when in central andsouthern Somalia the Bantu were still fighting to stave off the south-ward Galla invasion,59 Arab settlers and merchants from the Persian. Gulf had already established themselves along the coast, bringingtrade, wealth and culture with them.60 The hinterland was still in-habited by Negroes, at least so far as southern Somalia is concerned.In the XIII century Ibn Batttah already describes Mogadiscio as aSomali, as well as an Arab, city-state, but more or less in the sameepoch al-Dimsqi relates that near the town is the country ofKalbah, inhabited by moslem Zeng.61 The first oriental settlersmust have found here an ethnical situation similar to that of theSwahili coast from Lamu and Kilwa right down to Sofala.If the Arabo-Persian and, to a lesser degree, the Indian 62 coloni-zation of the Somali coast are relatively weH-known to us in theirbroad outline, of the relations between East Africa and the Far Eastwe know fairly little. The Chinese element in the Mischkultur isusually not reckoned with, but this does not mean that it must bewaived aside. The Chinese are known to have imported zang(Negro) slaves as early as the beginning of the IX century,63 but in58 See B. Ankermann, Kulturkreise und Kulturschichten in Afrika, ZfEXXXVII 1905, 55.59 Apart from Arab and Galla traditions, Bantu reminiscences confirm thisancient struggle (Guillain, op. cit., III, 245).60 See especially E. Cerulli, Iscrizioni e documenti arabi per la storia deltaSomalia, RSO 1926; Le popolazioni della Somalia nella tradizione storica locale,RRAL 1926; Somalia (Storia) in EIT XXXII (1936); La citt di Merca e tre sueiscrizioni arabe, OM XXIII, 1943.61 C. Conti Rossini, Storia d'Etiopia, Bergamo 1928, 330.62 W. Seidel, Die Verbreitung der Inder in Ost- 11. Sdafrika, Mitt. Geogr. Ges.Rostock, 1937, was unattainable for me. But see Monneret, Note, 328, 345.65 G. Ferrand, L'lment persan dans les textes nautiques arabes, J. As. 1924,241, quoted by Monneret, Note etc., 346.178Downloaded by [University of North Carolina] at 23:10 11 November 2014 Vinigi L. G rottan eilt: Asiatic influences on Somali culturewhat points of the Indian Ocean coast the slave trade centres were,we shall probably never know. Coins found at Mogadiscio, Kilwaand Mafia were identified with the K'ai Yuan from A. D. 713742,and others from 845, and from the three following centuries.64 Itwas only at a later epoch, apparently, that the Chinese entered intrade relations with what was then the Somali country, i. e. presentBritish Somaliland: Chinese pottery fragments found near Zeila weredated from the Sung and Ming dynasties (XIIXVI century).63The records of the Ming dynasty, extracts of which were publishedby Bretschneider, contain an interesting and fairly accurate descrip-tion of the Somali coast, in the course of which the name of Moga-discio (Mu ku tu su) is mentioned.68 But even if we are to creditthe record of a whole Chinese fleet visiting this town in the XVcentury, it still remains probable that trade between China andAfrica was mainly carried on by seamen of countries geographicallysituated between the two areas.67Indonesia certainly was an important link in this imposing com-mercial chain, but what we do not know is whether Indonesiannavigators reached the East-African coast from the south, as a furtherstage after the colonization of Madagascar, or from the north ontheir way there: in other words, whether the Indonesian culturalelements which we find in East Africa are due to direct (Indonesian)or indirect (Malagasy) influence. That the Malagasy, and particu-larly the Sakalava, carried out extensive raids to the Comore islandsand to the Mosambique coast, is a well-known fact68; but how much64 J. Strandes, Die Portugiesenzeit von Deutsch- u. Englisch-Ostafrika, 88,Berlin 1899; Monneret, Note etc., 340 and 346; Ingrams, op. cit., 88.65 A. T. Curie, The Ruined Towns of Somaliland, Antiquity Sept. 1937,320-321.68 Ingrams, op cit., 9 0 - 9 1 .67 J. T. Reinaud, Relation des voyages faits par les Arabes et les Persans dansl'Inde et la Chine dans le IXe sicle, Paris 1845, Vol. I, XXVIII ff.; and onIndian shipping, p. XXXVIII.68 M. O. McCarthy, Iles africanes de la mer des Indes . Les Iles Arabes , 133 ff.,Par is 1885; G. Ferrand, Les musulmans Madagascar et aux les Comores , 125-126, Par is 1902; C. Keller, Die ostafrikanischen Inseln, 126 ff., 106, Berlin 1898;K. Weule, Neger leben in Ostafr ika , 315, Leipzig 1908.179Downloaded by [University of North Carolina] at 23:10 11 November 2014 Ethnos No. 4. 1947further north they arrived is still an unsettled problem. The Saka-lava colonization of Pemba only rests, as far as I know, on a curi-ously muddled tradition that may well be a legend.69 Monneret'sopinion, that people from al-Qomr (Madagascar) should have reachedthe island of Pte (Patta) is based on slight toponomastical evid-ence,70 and he himself does not give much credit to Ibn-al-Mugwir's(XIII century) text, according to which al-Qomr navigators shouldat one time have occupied Aden.71 Conti Rossini dismisses this lastsupposition as fantastic, but wonders whether it could not be inter-preted as the exaggeration of a nucleus of truth, i. e. the memory ofminor Malagasy landings on the Somali coast.72 Now the Indonesiancultural elements pointed out by Puccioni, and the ones I have dis-cussed above, bear positive evidence that some contacts existed, andleft their traces, at least in the southern regions of Somalia; and thatin order to influence the Somali proper, who did not reach this areaearlier than the XII century, they must have lasted at least as lateas this period. Some of these elements, as Puccioni had alreadypointed out, are not common to the coastal Mischkultur, so thatthe theory of an indirect penetration of them through Arab orPersian agency does not hold. The Indonesian (or Malagasy) navi-gators who brought these elements with them apparently did notcome in great numbers: Parenti, who recently studied Puccioni'swould-be Indonesoid Bajuni, found no more than a limited Indo-nesoid influence in their physical characters, their main anthropo-logical kinship being with the Arab element ; but he does not alto-gether discard the possibility of a Malagasy streak among them.In my opinion, however, this last explanation is not the moreprobable one, and we should by no means dismiss the possibility of69 W. H. Ingrams, op. cit., p . 124; also R . F. Burton, Zanz iba r , L o n d o n 1872,I , 380 and elsewhere, speaks of a Madagascar quar ter in Zanzibar .70 Note etc., p. 34;.71 Ibid., p. 316.72 C. Conti Rossini, Pubblicazioni etiopistiche dal 1936 al 1945, RSE IV,1944-45, P. 34.71 R. Parenti, op. cit., p. 187-188.180Downloaded by [University of North Carolina] at 23:10 11 November 2014 Vinigi L. Gr o 11an elli: Asiatic influences on Somali culturemore or less direct Indonesian influences from the northern route.Apart from relatively recent influences by the constant, if thin,stream of Indonesian moslem pilgrims, some of whom may well havehalted on the East-African horn,74 I am rather inclined to agree withHorn-ell in supposing that also the earlier Indonesian migrationsalong the African coast were performed in stages via South Indiaand the Arabian coast (Aden), and thence southward to Madagascar not viceversa. Traces of their passage are slight after so manycenturies, but then a thorough in loco investigation of this problemhas never been undertaken, and ethnological as well as archaeologicalresearch is, after all, only at its first steps.Drawings by O. La Bella and A. Grillo.ABBREVIATIONS:EIT = Enciclopedia Italiana Treccani.MC = Museo Coloniale, Roma.MME = Museo Missionario-Etnologico Laterno, Roma.MNA = Museo Nazionale di Antropologia, Firenze.MPE = Museo Nazionale Preistorico-Etnografico L. Pigorini, Roma.PRM = Pitt-Rivers Museum, Oxford.OM = Oriente Moderno, Roma.RRAL = Rendiconti della R. Accademia dei Lincei, Roma.RSE = Rassegna di Studi Etiopici, Roma.RSO = Rivista di Studi Orientali, Roma.SEM = Statens Etnografiska Museum, Stockholm.SNR = Sudan Notes and Records, Khartoum.74 See p. 177, footnote 56.181Downloaded by [University of North Carolina] at 23:10 11 November 2014