PULA Journal of African Studies. vo.1 11 nO.2 (1997)
Book Review
Tudor Parfitt, Journey to the Vanished City: the Search for a Lost Tribe of
Israel London: Phoenix (Orion Books), 1997. 373pp. Paperback 0-7538-
0267-8. UK Price £7.99 (1st edn. 1992)
This book is the story of the author's travels in Africa and southern Arabia in his quest
for the history of the so-called Black Jews, which had previously brought him to study
the Falasha of Ethiopia and which now took him to the Lemba people of the Northern
Province of South Africa and southern Zimbabwe. The Lemba live in scattered small
communities among Venda and Shona! Kalanga peoples, distinguished not by language
but by common traditions and Semitic-seeming customs, including male circumcision,
strict dietary rules, and restrictions on marrying out.
    After living getting to know Lemba people in South Africa and Zimbabwe over the
course of many months living with them, the author succeeded in tracing back their
declared male ancestry via the Zambezi and the Zanj coast of Tanzania, to the port of
Sayhut on the southern coast of Yemen (in the former Aden Protectorate), and thence up
the Wadi Masilah river to the now tiny and almost forgotten ancient town ofSena in the
Hadramaut valley. An account of origins which may at first sound altogether too
fantastic to be true. This reviewer was extremely dismissive when he first heard it put
forward by the author on a radio programme.
    But this book is no ordinary travelogue and the author is no ordinary amateur. It is a
scholarly work, with a sceptical and questioning approach to evidence. The author,
Tudor Parfitt, is chairman of the Centre of Middle Eastern Studies at the School of
Oriental and African Studies in London, as well as Reader in Modem Jewish Studies at
SOAS.
    Parfitt is careful to disassociate himself from the once fashionable rubbish about the
non-black exotic origins of ancient civilisation in black Africa, put forward by the likes
of Cecil Rhodes and still being pushed by the white settlers whom Parfitt disparages as
"Rhodies". He has no truck with the racist myth of ancient Phoenician sailors from the
Mediterreanean, come down the Red Sea into the Indian Ocean, to build Great
Zimbabwe. With the Lemba, as he says, he is dealing with people whose ancestry and
culture is predominantly local African. They claim only their direct male lineage
running back to Jews in southern Arabia-a claim confirmed by comparative genetic
analysis on Lemba and Yemeni blood samples cited in an appendix to the 1997 edition
of the book. The Lemba were specialist craftworkers, traders and medical practitioners,
who may well have acted as masons laying the better stone walls in trench foundations.
 But Parfitt finds the Lemba claim to have wholly built Great Zimbabwe etc. scarcely
 credible, and puts this down to "feed-back" in oral tradition from Christian missionaries.
There are, he concedes, too many questions left begging about Lemba relations with the
 Venda in particular-as well the south-western Shona, i.e. Kalanga, in general-in
 unravelling the precolonial history of Zimbabwe.
     Parfitt runs the story "backwards" in time from his journeys in the present, and thus
 makes no attempt to contruct a new narrative running "forwards" from remote times.
 While he tackles in detail oral tradition and existing literature on the Lemba in South
 Africa and Zimbabwe, and includes fascinating snippets of evidence from further north,
 he has no space or inclination to fit the ,story of t?e .Lembe/ and their "VaMwenye"
 ancestors into the macro history of the IndIan Ocean III Its days of glory as a centre of the
 world economy before the Portuguese conquest c.lSOO.
                                             191


     That task must be left to others. What Parfitt shows or suggests is this. The port of
Sayhut on the south Arabian coast, which served the Hadramaut valley of the interior,
was uniquely situated to catch the winds and currents to East Africa, unlike the coast
nearer Aden from which ships sailed to India and Indonesia. The link with Africa still
exists today: an old man he met in the Hadramaut claimed to have sons who were
cabinet ministers in Tanzania. The sub-clan names used among the Lemba today are
also still family names in the Hadramaut. As for the ancient people of the south Arabian
coast, they seem to have followed the Old Testament religion of Baal and Astoroth until
cAOOAD, when they were converted to Judaism by male immigrants-which remained
their religion (unless some were also converted to the aberrant Jewish cult called
Christianity) at least until the Muslim conquest c.700. Thereafter only a minority of
Yemeni or South Arabians in the Hadramaut remained Jews, staying through the
centuries until they decamped to Israel in or after 1948. Meanwhile the ancient city of
Sena, based on irrigation, remained the jewel ofthe Hadramaut until the local dam broke
open (probably about c.1200) and the area became stifled by dessication.
     It is well known that there was extensive trade in Arabian ships, and possibly
Persian and Indian ships, along the Zanj coast of East Africain pre-Muslim times. The
famous Periplus of the Erythrean Sea, published c.IOO AD, indicates this; and the
archaeological site of Chibuene on the Sofala coast of Mozambique Gust north of the
Limpopo), dated before c.600, contains Arabian beads and sherds of Persian pottery. The
first Arabians to sail the Sofala coast may well therefore have been Jews by religion. It
may even have been the Islamic conquest of Arabia which made a few Arabian Jewish
males decide not to return home but to settle on the Sofala coast. Parfitt has found
evidence of a pre-Portuguese settlement near Sofala called Sena. A third town called
Sena, which still exists today, on the Zambezi, probably came later. Arabian traders
probably did not penetrate the Zambezi until after c.lOOOwhen trade in gold and ivory
began to boom.
     Lemba traditions record two ancestor figures of the Lemba coming from Arabia-one
 called Baramina from the city of Sena, and the other called Saidi. "Baramina" (possibly
 a group rather than a person) appears to represent the earlier Jewish element in Lemba
 ancestry. "Saidi" could possibly, according to Parfitt, be derived from fleeing Zaidi
 heretics of the Shi'ite Muslim leader Said ibn Ali, who was killed in Arabia in the year
 739. Whatever the case, "Saidi" may represent a later Muslim element among the
 Lemba, which has been not totally absorbed into the dominant Jewish element of their
 culture. Parfitt's Lemba informants in South Africa may be adamant that they are Jews
 and not Muslims (or "Arabs"), but Parfitt discovers that their innermost rituals use
 Muslim Arabic rather than Jewish Hebrew words. (There are also Lemba references to an
 ancestral figure called "Sulaiman". He may also have been a Muslim immigrant. There
 is no reason to equate him with the Biblical king Solomon.)
      The Lemba links with Judaism on one hand, and with Great Zimbabwe on the other,
 are nothing new to scholarship. As Parfitt shows, there is a considerable existing
 literature on this, but its implications for the study of the history of Africa have been
 skated over. It simply did not fit in with the "Merrie Africa" version of indigenous
 initiative and development that Africanists propagated in the heady days of African
 nationalism forty years ago. It looked too much like the (racist or mystic) paradigm of
 "primitive" earthlings living in ignorance until "civilized" aliens brought enlightenment.
      But the profession of African history is now much more self-confident, with the basic
 outlines of indigenous initiative and development well established. We should therefore
 positively welcome scholarship that seeks historical connections outside Africa,
 especially in Asia. Thus the Lemba might have become skilled craftworkers in metal
 and stone, but we know full well that mining in iron and copper long predates the
 coming of the "VaMwenye", as indigenous developments in the interior to the north;
 and even the earliest building of houses with stone walls on the Zimbabwe plateau
                                               192


seems to have come from the southern interior. Garlake (1973) showed long ago that the
dry-stone architecture of Great Zimbabwe can be shown to be an indigenous
development on site, and is quite distinct from the mortared stone "Arabic" architecture
of the Zanj coast. But that would not have precluded the VaMwenye from joining and
even taking over the masons' guild at Great Zimbabwe.
     Where the Mwenye/ Lemba factor may be most useful to historians and
archaeologists is in helping to unravel questions of the origins and selective spread of
circumcision in Eastern and Southern Africa. The archaeologist Tom Huffman (1996)
raises pertinent questions about the "great enclosure" at Great Zimbabwe, and related
ruins as far west as the stone circles of Kubu island in the Makgadikgadi salt pans,
asking whether their architecture cannot be best explained as the sites of circumcision
schools. The great conundrum being that Shona peoples on the central Zimbabwe
plateau, where the ruins of Great Zimbabwe are located, do not circumcise and seem to
have no traditions of having once abandoned circumcision. While on the other hand the
Venda and Sotho/ Tswana to the south of the Limpopo did, at least until recently,
practice male circumcision. Huffman looks to the links of the Venda rulers with the
rulers of Great Zimbabwe. But the Lemba, once closely connected with the Venda rulers,
may be the explanation for the much wider spread of male circumcision, of which they
were skil1edpractioners.
     The anthroplogist Gina Buijs (1998), at the University of Venda, has noted another
 curiosity relating to the religio-magical culture of the Venda and Sotho-Tswana. Their
 divining dice (thangu or ditaola) are apparently like no others in Africa, and yet appear
 to reflect mathematical concepts otherwise found in South Arabia.
     There is no denying that, as photographs in Partitt's book show, the conical towers
 inside the "great enclosure" at Great Zimbabwe do look very much like conical towers in
 the Hadramaut of South Yemen. Strangely enough this parallel was tumbled to a century
 ago by the antiquarian J. Theodore Bent, an expert on South Arabia who was sent by
 Cecil Rhodes to inspect Great Zimbabwe. But Bent was too obsessed with "ancient"
 times to make a "medieval" connection.
                                                                                         N.P.
                                         References
  Buijs, Gina (1998) 'Sources of power: the role of divining dice (thangu) in Venda cosmology'
      Presented at History/ Archaeology Seminar, University of Botswana, 16 April.
  Garlake, Peter (1973) Great Zimbabwe London: Thames & Hudson
  Huffman, T.H. (1996) Snakes and Crocodiles. Power and Symbolism in Ancient Zimbabwe
      Cape Town: David Philip & London: James Currey
                                                193