Saturday, 26 June 2010


One of the places I now wish I’d visited when I lived on Pemba island in 1994-96 is Kojani. It's one of the least accessible villages, located on an islet off the eastern coast of the main island. At the last count Kojani was home to more than nine thousand people. New arrivals are liable to be met by throngs of excited children wading into the water to ogle at them. The menfolk of the village are often absent for months on end fishing up and down the East African coast, while their wives and other family members are left behind to farm and eke out a bare living. And bare it is, because this is the region of Pemba most prone to periodic drought and famine and recurrent problems of public health and nutrition (Walsh 2009b).

Kojani is one of a handful of comparatively remote Pemban villages that has resisted the spread of Kiunguja, the dialect of Swahili that is spoken in Zanzibar town and throughout the plantation areas of both Unguja and Pemba (Whiteley 1958: 8). When the late John Middleton surveyed the systems of land tenure in Zanzibar in 1958 he thought that Kojani was probably the part of Pemba “least affected by clove growing and modern change” (1961: 56). Another mark of this is the survival of various aspects of traditional political organisation and community ritual. A few days ago I stumbled across Odile Racine-Issa’s (2001) fascinating description of the annual mwaka or New Year ceremony in Kojani, a three-day rite of passage and communal cleansing during which the islet is effectively closed to visitors.

One of the explicit purposes of this and other Swahili mwaka ceremonies is to refresh the relationship between the community and its guardian spirits. Interestingly, Kojani was one of the few villages on Pemba that resisted penetration by Popobawa, the malevolent entity that caused mass panic on the island in the run-up to the first multi-party national elections in 1995 (Walsh 2009a). I heard two different versions of what had happened. The first was related by Salim, my watchman-cum-gardener in Limbani, who’d been told this story at Chwale junction. A mainlander from the Tanga area had appeared there one day asking the way to Kojani. He had a scar right round his neck and spoke in a faint throaty voice. People asked him who he was going to see in Kojani and he said a local mganga or healer. But no one believed this and they accused him of being Popobawa, following which he was seized and brought to the police at Wete. Salim didn’t know what had subsequently happened to him, but thought it likely that he really was an innocent visitor in search of traditional treatment.

The second, much longer version, was recorded by my research assistant, Jamila. On this account a stranger had appeared one day on the shore by the ferry crossing to Kojani. For a long time he just stayed there watching people being ferried to and fro. Eventually an old man approached him and asked if he wanted to go across to Kojani himself. “Yes, I do want to go there” he replied, pointing towards Kojani. But he didn’t budge, and was still there clutching his bag when the sun went down. Meanwhile, in Kojani itself, a local man had fallen into a possession trance and his possessory spirit announced that Popobawa was trying to cross to the village. However, the benevolent spirits of Kojani had tied up his tongue and the rest of his body, and he was unable to move. “If you want to confirm that he really is Popobawa, then go and open his bag and look inside!”

Hearing this the men of Kojani jumped into their canoes and sped across to the main island. Without further ado they laid into the stranger on the shore, some with sticks and stones, and he was thoroughly beaten. Some of his assailants grabbed hold of his bag and found that it was full of the paraphernalia of local medicine and perhaps sorcery: the roots of plants, charms, and something that looked like a face-mask. Eventually sailors from the local detachment of the Anti-Smuggling Unit (KMKM, Kikosi Maalum cha Kuzuia Magendo) intervened to stop him from being beaten further.

The Kojani men wanted to finish him off, and told the sailors to examine the contents of his bag. They were shocked by the horrible smell of one of the charms: the same stinking odour that was said to follow Popobawa during his nocturnal assaults. When the sailors quizzed the roughed-up stranger he claimed to be on his way to Kojani to seek traditional treatment after a long period of hospitalisation and the prescription of modern medicines had failed to cure him of illness. But he was unable to name anyone in Kojani or explain satisfactorily why he hadn’t crossed to the islet. And he was unable to explain the contents of his bag, claiming that he hadn’t visited any mganga since falling ill.

Following this the local KMKM commander decided that he should be handed over to the police, and his staff telephoned the station in Wete. The police came in a vehicle at around 9 o’clock at night and interviewed everyone present before taking the stranger away. Afterwards it was said that the police only kept him in gaol overnight for his own safety. He was released early the next morning and – or so many people thought – allowed to continue his nefarious activities as a manifestation of Popobawa. This just helped to confirm the widespread belief that the authorities were themselves responsible for bringing Popobawa to the island, to punish Pembans for supporting the opposition and distract them from political campaigning.

Thus ends Jamila’s account. This wasn’t the only occasion in 1995 when a stranger was identified as Popobawa and attacked by a mob. In Zanzibar town and elsewhere on Unguja island this led to a number of deaths, the most notorious of which was widely reported (Jansen 1996). Whatever the truth behind the Pemban narratives of victimisation and/or political conspiracy that I’ve paraphrased here, it’s striking that they support the perception that Kojani is indeed a spiritual fortress, guarded by the very spirits that the annual mwaka rituals are intended to appease. And they carry a warning to would-be travellers to Kojani: don’t go there without good reason, especially if you look weird and have strange stuff in your bag.


Jansen, Henriette 1996. Popobawa is Dead! Tanzanian Affairs 53: 22-24.

Middleton, John 1961. Land Tenure in Zanzibar (Colonial Research Studies No. 33). London: Her Majesty’s Stationery Office.

Racine-Issa, Odile 2001. Le Mwaka de Kojani (Pemba). In Bridget Drinka and Derek Nurse (eds) African Language and Culture in Historical Perspective: Essays in Memory of Edgar C. Polome. Special Issue of General Linguistics 38: 199-229.

Walsh, Martin 2009a. The Politicisation of Popobawa: Changing Explanations of a Collective Panic in Zanzibar. Journal of Humanities 1 (1): 23-33.

Walsh, Martin 2009b. The Use of Wild and Cultivated Plants as Famine Foods on Pemba Island, Zanzibar. Etudes Océan Indien 42-43: 217-241.

Whiteley, W. H. 1958. The Dialects and Verse of Pemba: An Introduction (Studies in Swahili Dialect IV). Kampala: East African Swahili Committee.

Saturday, 19 June 2010


A few years ago I wrote a note for the journal Azania on the rare use of a nasalised dental click in Digo, the Mijikenda language that’s spoken on both sides of Kenya-Tanzania border. This click occurs in a couple of interjections-cum-ideophones, n|a ‘go away, get lost!’, and n|akule ‘miniscule, minute’, where n| represents a dental click with voiced velar nasal accompaniment (to approximate this sound try tutting with your tongue while humming through your nose). Linguists had missed these words, and I couldn’t find any record of the similar use of accompanied clicks in other Mijikenda languages. However, given that my field research for this article comprised little more than a chance encounter at a bus-stop in Tanga, a friendly conversation on the overnight train between Mombasa and Nairobi, and a chat followed by a quick tape-recording session in a taxi in Dar es Salaam, I fully expected other examples of click-bearing words to turn up.

I’m still waiting for evidence of these in Mijikenda. But I have come across a case in Swahili, recorded in Mombasa at the end of the 19th century by the Rev. W. E. Taylor. William Ernest Taylor (1856-1927) has been described as “England’s greatest Swahili scholar (Frankl 1999), while his Giryama Vocabulary and Collections (1890) remains the best lexicon of any of the Mijikenda languages, not least because he marked the phonological nuances that other missionaries of his time didn’t. It’s hardly surprising then that Taylor should have picked up on this unusual feature in the Mvita dialect of Swahili that was spoken by his informants in Mombasa. Here’s the relevant entry in his African Aphorisms (1891: 93):

Taylor was evidently at a loss as to how to write down this unusual word, and it’s difficult to know how to interpret his tentative transliteration. Zulu ‘c’ is a dental click and an educated guess would be that the word that Taylor heard (perhaps better transcribed mn|wa) includes the same nasalised dental click that occurs in Digo. The only other reference to this "difficult interjection" that I can find is in the article on 'Phonetics' that Taylor himself wrote for Mrs. F. Burt’s Swahili Grammar and Vocabulary (1910: 13, fn 1). There’s no sign of it in the fascinating collection of Swahili exclamations made by Carol Eastman and Yahya Ali Omar (1985), and Zanzibaris I’ve asked recall nothing like it in the Unguja dialect. But who knows what targeted research will produce? Is anyone listening?


Burt, F. 1910. Swahili Grammar and Vocabulary. London: Society for Promoting Christian Knowledge.

Eastman, Carol M. and Yahya Ali Omar 1985. Swahili Gestures: Comments (Vielezi) and Exclamations (Viingizi). Bulletin of the School of Oriental and African Studies 48 (2): 321-332.

Frankl, P. J. L. 1999. W. E. Taylor (1856-1927): England’s Greatest Swahili Scholar. Afrikanistische Arbeitspapiere 60: 161-174.

Taylor, W. E. 1890. Giryama Vocabulary and Collections. London: Society for Promoting Christian Knowledge.

Taylor, W. E. 1891. African Aphorisms; or, Saws from Swahili-land. London: Society for Promoting Christian Knowledge.

Walsh, Martin T. 2006. A Click in Digo and its Historical Interpretation. Azania 41: 158-166.

Saturday, 12 June 2010


About October the dark clouds begin to gather in the east, lingering about the crests of the ridges before breaking in spectacular thunderstorms. While it is still raining the hot sun comes out, the earth steams and rainbows appear over the lakes and dams. At this moment, the people say, “The hyaena is giving birth to twins.” Overnight scarlet lilies appear like balls of fire among the sprouting millet; shrubs along the cattle-paths break into leaf and in a few days are covered with pink scented flowers; bees hum everywhere; and a red velvet mite which crawls over the damp earth is welcomed as “the child of the rain” (mwambua) - a favourite personal name for anyone born at this time. Along the edge of the cattle-paths grass begins to appear. By late December, in a normal year, the rain may be so heavy that much of the country becomes impassable.” (Jellicoe 1978: 6-7)

Marguerite Jellicoe’s evocative description of the start of the annual rains in Singida includes a rare reference to the cultural significance of red velvet mites in Tanzania. I first came across these brightly coloured arachnids (family Trombidiidae) in 1981, at the start of my second wet season in the village of Utengule in Usangu (in what is now Mbarali District). I was away when the rains began on 2nd December, but when I returned to the village two days later these small crimson creatures were everywhere on the newly dampened earth. Sangu-speakers called them inkhadupa, their generic name for ticks and mites, and told me that they were thought to fall down from the sky along with the rain. In this respect they were similar to ground pangolins (Manis temminckii), another creature believed to fall from the heavens, from whence they were sent by the ancestors, amanguluvi.

At the time I had no idea what kind of tick or mite these might be. I later found a reference to them in adventurer Marius Fortie's book Black and Beautiful (1938). When he was in Dodoma in December 1933, Fortie was told by an American naturalist that the "velvety red bugs" crawling on wet sand after heavy rainfall were a kind of tick (1938: 251). In May 1995 I wrote to the National Museums of Kenya in Nairobi asking if they could help identify the creature that both Fortie and I had seen. In the absence of a specimen they couldn't. I therefore referred to it as a "small red tick" in the paper that I wrote about 'The Ritual Sacrifice of Pangolins among the Sangu of South-west Tanzania' (1995/96: 165).

The growth of the internet has since made searching for this kind of information much easier than it once was, and when I eventually revisited this question it didn't take me long to find out that these 'ticks' were in fact red velvet mites - a perfect description of their external appearance. According to Joanna Makol, an authority on the Trombidiidae and their relatives, the red velvet mites that appear in large numbers after heavy rain are most likely members of the genus Dinothrombium (Oudemans, 1910), only two species of which have been recorded in Tanzania, D. tinctorium (Linnaeus, 1767) and D. zarniki (Krausse, 1916). But definite identification of the heavenly mites that make such a dramatic seasonal appearance in Usangu and elsewhere in the Eastern Rift must await the collection and description of more specimens.

Whatever species are involved, it's clear that red velvet mites are known throughout the region. A post on the Tanzania Odyssey blog in February 2009 describes their sudden appearance in Ruaha National Park, when "in some areas the whole ground can take on an almost red sheen and to walk without standing on one is almost impossible." But their emergence at the start of the rains isn't everywhere ascribed to heavenly intervention. Writing in 2002, my Hehe-speaking research assistant in Iringa, John Justin Kitinye, described pink-coloured ngudupa or ngudupasa mwihala (lit. 'bush mites', as distinct from ngudupasa mnyumba or kaye, 'house ticks') quite accurately as creatures which could dig down and hide in sand for months on end during the long dry season. No mention of them falling from heaven here, but wondrous animals nonetheless.

I couldn't have written this note without the information provided by the late Ngwila Simuhongole and other villagers in Utengule, as well as by John Justin Kitinye in Iringa. I'm also grateful to Richard Bagine and David Moyer for answering my queries, and especially to Joanna Makol for her email correspondence in February 2008 and for sending me extracts from her paper in Annales Zoologici (2007).


Fortie, Marius 1938. Black and Beautiful: A Life in Safari Land. London: Robert Hale.

Jellicoe, Marguerite 1978. The Long Path: A Case Study of Social Change in Wahi, Singida District, Tanzania. Nairobi: East African Publishing House.

Makol, Joanna 2007. Generic Level Review and Phylogeny of Trombidiidae and Podothrombiidae (Acari: Actinotrichida: Trombidioidea) of the World. Annales Zoologici 57 (1): 1-194.

Walsh, Martin 1995/96. The Ritual Sacrifice of Pangolins among the Sangu of South-west Tanzania. Bulletin of the International Committee on Urgent Anthropological and Ethnological Research 37/38: 155-170.

Sunday, 6 June 2010


One of my favourite examples of linguistic innovation in Tanzania was the rapid transformation in 1996 of a politician’s name into a term of ridicule in Swahili. Ramadhani Kihiyo became CCM MP for Temeke constituency in Dar es Salaam in the first multi-party general elections in October 1995. But his election was challenged, and in the subsequent petition hearing it emerged that he had lied about his academic qualifications (there’s a nice summary of these proceedings in Tanzanian Affairs, No.55). Kihiyo resigned his seat at the end of May 1996, and his name became a byword for fake certificates and invented resumés.

I remember this well because it was the running joke among my travelling companions (government officials and a driver from Dar) when I was visiting villages in Pawaga, in Iringa District, in September 1996. ‘Fulani ana kihiyo’, and the plural form ‘Fulani ana vihiyo’, ‘So-and-so has bogus qualifications’, was the expression I heard in endless variations and hilarious applications as we drove along the dusty dry-season roads. As this became widespread usage in the country it was said that people who shared Kihiyo’s Sambaa name were dropping it to avoid its unfortunate associations and the embarrassment thus caused.

Fortunately for them, time has since begun to erase memories of the Kihiyo scandal, and many Tanzanians don’t know what a kihiyo is, though the practice it refers to is all-too familiar. A quick search of the internet produces a few examples, all of them using kihiyo as a term for the person with the invented qualifications (e.g. ‘huyo kihiyo!’, ‘that faker!’) rather than the made-up achievements (paper or otherwise) themselves.

The only other example of eponymy in Swahili that I can think of was coined during the Lewinsky scandal in 1998, when US President Bill Clinton’s extra-marital relationship with White House intern Monica Lewinsky was a major international news story. Among at least some urban Tanzanians kumk(i)linton mtu, ‘to “clinton” someone’, is now a jocular euphemism for fellation, kunyonya (lit. ‘sucking’) in everyday Swahili.

If this was an English expression you’d expect it to be a reference to whatever the president did to his young intern. But as Clinton told the courts, he didn’t really do anything, and the Swahili slang refers to the act that Monica is alleged to have performed on him. So while it would be correct to say that Monica “clintoned” Bill (Monica alimklinton Bill), you can’t use this eponymous verb to say that he had “clintoned” her – or indeed any other woman. It is possible, however, for one man to “clinton” another, linguistically-speaking that is.

It’s perhaps no coincidence that these two examples of contemporary eponymy derive from political scandals and the prominence given to them in the newly-liberalised Tanzanian media. But the first seems not to have stuck and I don’t know how widely used or understood the second is, beyond the circle of Zanzibari women and Bongo dwellers that I’ve heard it from. I can’t imagine seeing the likes of ‘kihiyo, n. vi-’ or ‘k(i)linton, v.’ in a printed Swahili dictionary anytime soon, but would love to be proved wrong.