ThirdWorld Quarterly, Vol 18, No 2, pp 287± 302, 1997

Comeback ? Questioning a collapsedstate

VIRGINIALULING

Somaliahas cometo beseen as aprimeexample of a`collapsedstate’ . Together withLiberia, it isabywordfor anarchy; ` becominganother Somalia’ is thefate tobeavoidedby everyAfrican state. And yet this is whatwas previouslyhailed as themost ` uni®ed’ state in Africa, free fromthe ethnic divisions that plague nearlyall the others, with a singleculture, language and religion. Somalia remindsus todistinguish between the underlying causes ofstate collapse, and thefault lines along which the collapse occurs. These do notneed to beregional, linguistic,religious or ` ethnic’. InSomalia, rivalries between descent groups (clans)served the same purpose. Inthis paper I shallconsider the process ofdisintegration, and ask whatthe prospectsfor reconstruction are. 1 Thoughthe same generalconsiderations apply toall the Somali territory, I dealonly with southern Somalia (the once Italian colony)and have not attempted to look at ` ’, theself-declared republicin the Northwest.

Somaliain 1996 Wellbefore the departure of the last UN troops,Somalia had almost vanished fromthe international news media. It had its brief period of notoriety, with the famineyear of 1991± 92, when it was ratedsecond (after Mozambique) in the worldscale ofhuman misery, and with the US/ UNinterventionthat followed. Sincethen, the Somali people have been left, with greatly reduced support and interferencefrom abroad, to struggle with the problem of what comes next. Thisis notthe place to analyse the international intervention in Somalia, whichbegan in such optimism as the® rst fruitsof the New World Order, and becamea chronicleof muddle,waste of resources,and pointless bloodshed, with theforces thathad been welcomed as rescuers comingto seem tomany of the Somalisas invaders.Politically, the intervention failed entirely to impose any order.At most it ` frozethe war in place’ for a time.It could point to its humanitarianachievements, but even there the extent has beenquestioned. 2 Its maineffects wereto reinforce the power of the ` warlords’it set outto destroy, establishSomalia’ s dollareconomy, and raise theexpectations of variousinterest groupsto a levelwhich could never again be met. Sincethen the country remains fragmented. Besides thebreakaway Republic ofSomalilandin thenorthwest, there are several de facto independentareas. The

VirginiaLuling is atSurvival, 11± 15 EmeraldStreet, London WC1N 3QL, UK.

0143-6597/97/020287-16$7.00 Ó 1997Third World Quarterly 287 VIRGINIA LULING formercapital and port , with its hinterland, remains split between the factionsheaded by Cali Mahdi Maxamed and (up to July 1996) by General MaxamedFaarax Caydiid, then by his son Xuseen. 3 Eachof them attempts to unitethe whole country through its own alliance of clan and faction groupings: Caydiidwith the Somali National Alliance ( SNA)andCali Mahdi with the SomaliSalvation Alliance ( SSA).Bothclaim to head central governments and havedistributed ministerial posts to their followers. The Northeast and the port ofBoosaaso is controlledby the Somali Salvation Democratic Front ( SSDF). In thefar south,the valley of the Jubba river with its plantations remains divided betweenthe forces ofGeneral Maxamed Sayid Xirsi ` Morgan’in the port of Kismaayo,and those of Caydiid’ s ally,Colonel Axmed Cumar Jees inthe hinterland.In theplain between the Jubba and Shabeelle rivers (the centre of the 1992famine), the Digil± Miri¯ e orRaxanweyn group of clans set upa supreme councilin the town of Baydhaba (Baidoa) to administer the region, but were dislodgedwhen Caydiid took over the town in September 1995. A further splintergroup emerged in 1995 when Caydiid split with his former right-hand man,the millionaire businessman Cusmaan Xuseen Cali ` Caato’. Thoughthe situation in 1996 is peacefulcompared to the out and out warfare of1991± 92, spasmodic violence continues between the militias of the SNA and SSA,andbetween the supporters of Caydiid and those of his former supporter Caato.Both, given the narrowness of theirown clan support bases, are becoming increasinglyreliant on mercenary militia from other clans. A focusof this con¯ict is the` bananawar’ for the lucrative European export trade from the plantationson the Shabeelle river. The Somalfruit Company, which was estab- lishedin the 1980s under Siyad Barre, confronts Sombana, a subsidiaryof the USDolecompany, which began operating in 1993 under cover of the US intervention.Caato has interestsin Somalfruit,while Caydiid (and now Xuseen) protectSombana. Apart from this, there is alotof simplebanditry, with in most areas nolegal force able to control it. Andyet economic life goes on and even thrives. Somalia is aplace(as the title of an NGO reportputs it) of ` paradoxesof private prosperity, poverty pockets,relative vulnerability and public pauperisation’ . Thecentre of Mogadishumay be bombedout, but the markets on theoutskirts are busy;there maybe no banks but there are plentyof money-changers; with a wellfunction- ing,privately run satellite service it is far easier totelephone Somalia than it was beforethe war. Hotels charge exorbitant prices, even while the roads are lined withbeggars, and the aid agencies report high levels of malnutrition. Theevents of the last ® veyears havecaused massive displacementof people, bothwithin the former borders of Somaliaand into Kenya, and . Thetowns of Kismaayo in thesouth and Boosaaso in thenorthwest, in particular, havetheir populations swollen with refugees from many different clan groupings.

Stateand society Canthe Somali state be puttogether again? Before even asking the question, one needsto see whathas disintegratedand why. Thestate which has disappearedwas, as inmost of Africa, an essentially 288 SOMALIA: QUESTIONING ACOLLAPSED STATE arti®cial one, ` suspendedabove’ a societywhich would never have produced it anddid not demand it. This ` mis-match’between state and society is theessence ofthe problem. 4 Somaliintellectuals are acutelyaware of this, most often castigatingnot so muchthe inappropriate nature of the state as thedivisiveness oftheir inheritance from a societyconsisting largely of nomadic livestock herdersÐthe ` nomadicmentality’ . Thusfor instance Abdalle Omar Mansur: `Themost serious problem in Somalia today is thatour cultural traditions are notcompatible with the construct of a modernstate. We are prisoners ofa culturethat we had created in the past and one which we refuse tore- examine¼we mustreinvent ourselves and in the process launchthe construction ofa new,viable state.’ 5 As longas thestate in question was acolonialone (or rather two colonial onesÐthe British and the Italian) the mis-match was patentand, in a sense, unproblematic.However, it was continuedby thesuccessor, post-colonialnation state,which retained the same centralisedstructure and even strengthened it. Economicallythere was reallynot the base tosustaina stateof the kind to which theSomali elite aspired, and the lifestyle to which they became accustomed. Henceit subsisted off foreign aid. Thesociety over which this state hovered is largelythough not exclusively pastoraland nomadic in background and origin, and is effectivelystructured throughclan membershipÐ asystem ofgroupslinked genealogically by descent throughthe male line, both dividing and uniting the Somali nation. I say `effectively’, bothin the sense thatclanship is ahighlyeffective system; andthat thoseÐnon-pastoralÐ sections of Somali society where clanship is less import- ant,are alsoless powerful,so thatother principles of organisation are pushed intothe background. Forclan divisions are notthe only ones inherited by Somali society. In spite ofits reputation for homogeneity it has itscultural and ` ethnic’splits. The dominanttradition is thatof pastoral nomadism, and most of the political elite havebeen from this background or separatedfrom it byonlya generationor so; however,beyond the pastoral nomadic clans, there are thefarming clans of the south,different from them in history, culture and dialect. Visibly different are thepopulations of the southern coastal cities who are ofArab± Persian descent, andthe so-called ` Bantu’groups, largely the descendants of slaves imported fromfurther down the East African coast. Yet it is precisely not along these divisionsthat the battle lines have been drawn. These ` minority’groups have untilrecently ® guredin the civil war only as victims.The ® ghtinghas been entirelybetween people within thepastoral± nomadic tradition (including those whoare urbanised,but ultimately spring from it). Hence it is their clan system thathas structuredthe con¯ ict. Toquoteone Somali commentator: ` Because ofthe strong belief of almostthe entirepopulation in clanism as anecessary andinevitable political regime, no politicalorganisation with an open-door non-clan-based membership has yet appearedin the political landscape. Any member of theSomali society who aims foran active political participation is requiredto present himself as aclan memberwho plans to fosterhis clan interest. The notion of politicsoutside clan af® liationshas noroots in our minds.’ 6 Itis truethat clan ties are manipulated 289 VIRGINIA LULING asatoolby politicians, but they would not be able to do this if those ties were nota realityin peoples minds and lives. Arti®cial state and clan-based society, unable to work together healthily, insteadset upapoisonousinteraction which produced corruption in the state and distortionof theclan-tie. Clans had always competed for resources suchas land, grazingand water, but now that control of all these resources andmuch more was vestedin the state, competition between clans, which before had been only oneaspect of their existence, became its permanent condition. The state was boththe arena within which they fought and the prize for which they contended. Sothe state, which had been intended to unify the Somali people, actually becamethe mechanism of their disintegration. Andyet a majortheme of Somali history in the last 50 years has beenthe struggleto transcend the divisive aspect of the clan inheritance and attain a unitedSomali identity. Independence was wonin 1960 on a greatwave of enthusiasmfor national unity (though one may question how far thiswave really carriedbeyond the intellectual elite). There was aparadoxhere, since the genealogicalmodel, the myth of commondescent, which had been the traditional ideologicalbasis forthe unity of the Somali, had to be cancelled, and a new foundationfound in Western-derived nationalism. Yet clan-identity remained a basicreality in peoples lives, and determined the politics of theSomali Republic from1960 to 1969. In that year the military took over and the tired parliamen- tarydemocracy found no defenders. The new regime, with General Maxamed SiyadBarre atits head proclaiming ` scienti®c socialism’, stageda `burialof clanism’with a great¯ ourishof publicity. But what was buriedwas infactÐas CabdullaahiDool points out 7Ðonlythe most trivial and harmless aspects of clanship.All public use ofclan names was forbidden;yet all the while Siyad Barre was practising` clanclientelism’ , distributingarms andmoney to his friends,` encouragingthem to attack the common clan enemies who, of course, wereaccused of divisive ª tribalismºby the master tribalist¼By destroying his country’s economy,Siyad also directly promoted those conditions of general lackof resources andinsecurity on which clan loyalty thrives, since clan solidarityoffers theonly hope of survival.’ 8 Atthe same timethe state underminedthe benignant side of theclan systemÐ the customary law ( xeer) and mechanismsfor con¯ ict resolution at the local level. The result was, in Ahmed Samatar’s words,` theperversion of kinship through a decouplingof blood ties (particularlymale lineage) from the great civilising or universalising factors of Somaliculture: customary law and Islamic precepts. Without these companion moraldirectives, Somali male behaviour becomes unrestrained and is likelyto turninto egotistic criminality.’ 9 Yet,in thesame period,new strata had emerged in Somali society. Built into thefoundation of the state had been ` theassumption that a minusculeWestern- educatedclass wouldrule Somalia, since only they could administer a modern state’ .10 Nowmerchants, bureaucrats and employees of state agencies, the intelligentsiaand landowners found their material and social welfare depended onthe state, whether as employeror as dispenserof contracts. Thetwo main sources ofthe new wealth were foreign aid and land. During thecold-war period came whatAnna Simons has calledthe ` aidavalanche’ , ®rst 290 SOMALIA: QUESTIONING ACOLLAPSED STATE fromthe West, then from the USSR, and then from the West (principally the USA)again.This wealth was inthegift of thosewho controlled the state. While thosein power made their fortunes, 11 forpeople in the lower government echelons,whose of® cial salaries wouldhardly feed them for two days, some formof ` corruption’was unavoidable.Those, meanwhile, who did not have access tothis fountain of wealth, wondered when their turn would come. With 1992came another` aidavalanche’ , andthis time the formerly excluded were determinedto get their share andmore. Thesecond source of wealth was the` landgrab’ . Largeareas ofagricultural land,originally mainly in the hands of smallsubsistence farmers, hadbeen taken overfor plantation agriculture by the Italian colonists. After independence, the existingplantations increasingly came intothe hands of Somalientrepreneurs. At thesame timethe expropriation of landfrom small farmers continued,especially afterthe passing of the 1975 land law, which, by putting all land in state ownership,actually facilitated takeover by the ruling elite, who were able to manipulatethe tenancy rules. Therewas certainlyan embryonic class structurehere; butcompetition betweenmembers ofthese new strata turned out to be stronger than any emergingsense ofcommon interest. And when they split, they split along clan lines.This was thelegacy that 30 years ofthe post-colonial stateÐ 10 under civiliangovernment and 20 under military dictatorshipÐ bequeathed to the Somalipeople. The state had already in effect collapsed when the disintegration afterthe ¯ ightof Siyad Barre fromMogadishu in January 1991 made the conditionplain. Inspite of this, for many people in the urban population before the war, outsidethe central circle of power,clan identity had become less important.The idealismof the early days of independence, the propaganda of the Siyad Barre years,and the general melting pot effect of urban life, were not without effect. Peopleoften knew each other without even wondering about one another’ s clan origin.One may question whether this meant the real spread of a pan-Somali identity,or was ratherthe sign of a class unityamong the elite. (It is perhaps signi®cant, too, that nearly all the people who have mentioned this to me were women.)In any case, thecivil war and its atrocities have torn this emerging unityapart. As onewoman said, ` Weused not to know our clans. Now we have to.’

Clanship anddescent Ihavewritten so far as thoughthe Somali ` clan’was some kindof simple unity.This is far fromtrue, and I mustnow analyse the system, the ` macro- genealogicalgrid’ 12 inmore detail, since it is highlyrelevant to the present politicaland military situation. This is notto imply that clanship in itself somehowcauses thecon¯ ict. It is simplya formof organisation, and in order tofollowevents in Somaliaone needs to understandit, just as oneneeds to grasp thefeudal system tounderstandmediaeval history, or theoperation of the nation stateto make sense ofevents in modern Europe. 13 TheSomali clan system consistsof groupsbased on descentthrough the male 291 VIRGINIA LULING line.That much is easilygrasped by most outsiders. What they commonly ® nd harderto deal with is theway groups recruited in this manner do not remain constantbut split up into sub-groups, each with its own life. The UNOSOM commandmust have felt like the sorcerer’ sapprentice.This is easier to comprehendwhen one views the situation not as largegroups splitting into fragments,but as smallgroups forming larger, but intrinsically unstable, combi- nations.The former UN Envoyto Somalia, Mohamed Sahnoun, was among thosenon-Somali who made the most genuine efforts to understand this situation;before his recall he compared Somali society to a vase thathas been smashed intosmall fragments. This is avividexpression of the task he felt he was facing,and is accurateup to apoint;but it is alsomisleading. A vase is not madeup ofseparablecomponents, and when smashed itcan break at any point. Butthe Somali descent system is madeto take to pieces. It is morelike a scaffoldingwhich has collapsed. Throughouttraditional Somali society, the basic politico-legal unit is the lineageÐagroup numbering generally a fewhundred though sometimes up to 1000±2000 individuals, tracing their descent to a commonancestor anything fromsix to 14generations back. The small size ofthelineage unit, while making itmanageable, also makes itvulnerable; hence it never stands on its own, but existsonly in association with other similar units. Lineages are combinedinto largerunits (generally on thebasis ofsharinga moreremote ancestor), and those intolarger ones again, in a pyramidalstructure. That is tosay, smaller groups mayat need combine or ally into larger groups, which in turn unite into yet largerunits, and so on.However, these larger groupings will fall apart into their constituentswhen the situation which led them to unite alters, a process of `constantdecomposition and recomposition’ . 14 Orderand security are thusonly achievedthrough the balance of opposing groups, and the effort to achieve this balanceleads to the shifting political alliances that are acommonfeature of Somalipolitics. It is whenthe balance breaks down that con¯ ict follows. Studentsof Somali society identify various levels of grouping and use a specialisedvocabulary to distinguish them: ` primary` ,`secondary’and even `tertiary’lineages; ` clan’and ` clan-family’Ð theselast being the major group- ingsof a millionor more that make up the Somali people: Daarood, Haawiye, ,, Digil and Raxanweyn (or Miri¯ e). 15 Ultimately,according to the genealogicalmyth, all these are descendedfrom the same ancestor,so that all Somalisare onefamily. Somalis themselves use terms like reer and quabiil in a¯exibleway, without tying them to any particular levelÐ which is not confusingwhen one lives within the system, since the context makes itclear whatone is talkingabout. Thesystem grewup in a pastoraltraditional situation, and it continues to operatein `modern’contexts, because it is notonly emotionally compelling, but ef® cientÐfrom the point of view of the individual, an effective method of organisingin a situationof competition. It is thisvery effectiveness, the ease withwhich it can be used to form a ma® a,which makes thedescent system so potentiallydestructive of wider unity, and hence an anathema to nationalists, whoyet cannot free themselvesfrom it. Contrary to some journalisticassertions, thereis nothing` irrational’about the clan system. Indeed many of those who 292 SOMALIA: QUESTIONING ACOLLAPSED STATE makeuse ofit are beinghighly ` rational’, inthe sense ofengaging in the calculatedpursuit of ® nancialgain. It is notablethat when Somalis accuse someoneof ` tribalism’, whatthey are generallyimputing to him is notblind atavisticloyalty, but self-interest. This is notto deny that belief in common descentand kinship is alsoa strongemotional force. Indeed it is theinteraction ofthe emotional and the instrumental aspects whichgives this form of organis- ationits power. It is noteasy foran outsiderto appreciate what it is liketo live ina system wherethe emotional identity is strong,and yet can shift according tocontext; but I believethat Anna Simons is nearestthe mark: `Genealogy works¼mapping trustworthiness Ð bycharting who has trustedwhom in the pastand where this has ledin terms ofthicker or thinner, and sustained and brokenrelationships...Breaks between lineages indicate breaks, or at the least gaps,in trust.’ 16

Clans, alliances andfactions Inthe traditional pastoral situation, competition was (andstill is) between relativelysmall groups in a localsetting. The clan-family, that large unwieldy categoryof tens of thousands of people, scattered across hundredsof miles of territory,never acted as onebodyÐ never, for example, went to warall together. Suchmobilisation occurred at lower levels of segmentation, between clans or lineageswho clashed over matters such as thecontrol of wells or trade routes. Inthe modern situation, the arena is muchlargerÐ hence the need for wider alliances.These have been formed in two modes. The® rst appliedduring the existence of theSomali state, and in the® rst phase ofthe civil war. Groups competed at the higher level of descent organisationÐ thelarge clans and even the ` clan-family’Ð agroupcounting its membership in millionsÐbecame the political units. Thus during the time of civilian govern- mentthe Daarood and Haawiye clan-families competed for government posts. DuringSiyad Barre’ sreignit was theDaarood who held power through himÐbut not all of them. The three clans whom he used as hischief support- ersÐ the ` MOD’allianceÐwere his own clan the Mareexaan, and the Ogaadeen andDhulbahante (connected to him through marriage). All these are Daarood, butso are theMajeerteen who became one of the main opposition forces, as eventuallydid a sectionof the Ogaadeen. At the same timethere was ®erce competition within theselarge groupings, between their constituent lineages, for jobsand the spoils of of® ce. Theanti-government movements which overthrew Siyad, only to become the contestantsin the civil war, were based on clans or clan-families, but adopted ideologicaltitles in English, shortened to sets ofinitials. (Several of these organisationshad started with a genuineattempt to start a non-clanbased movement,but without success.) Thusthe ( USC) representsthe Haawiye clan-family, the Somali Salvation Democratic Front (SSDF)theMajeerteen clan, the Somali Patriotic Movement ( SPM)theOgaadeen clan.Eventually about 10 such movements emerged. Somalia-watchers carry checklists, since there is nological way of relating the initials to the group theyrepresent. The exception to this is thealliance of the Digil and Miri¯ e 293 VIRGINIA LULING clan-families,who helpfully named their movement the Somali Democratic Movement (SDM),thusincorporating their real initials. (They had done the same thing40 years earlierwhen they ® eldeda politicalparty.) Inthe course of the civil war these larger units divided into two alliancesÐ GeneralCaydiid’ s SomaliNational Alliance ( SNA),andCali Mahdi’ s Somali SalvationAlliance ( SSA).These,however, are nowformed in the second mode. Inevery case wherean organisation moves into one or other camp, some ofits componentclans or lineages, with reasons oftheirown for dissatisfaction, have splitoff and joined the opposite one. Hence a doublelist of initials,for instance USC±SNA versus USC±SSA. The SDM,forinstance, originated in a meetingof intellectuals from the Digil±Miri¯ e groupof clans, held in Saudi Arabia in 1987. Unlike most of the otheranti-Siyad movements at that time, it was non-violent,and had no militia. Itwas notuntil 1991 that the SDM beganto raise amilitia.During the following years,the Digil± Miri¯ e territory,round the town of Baydhaba,became the battle groundbetween Caydiid’ s forces andthose still supporting ex-President Siyad, andwas plunderedand ravaged by both. The SDM continuedits resistance as aguerrillaforce in the bush. In 1992 one ofits leaders, Colonel Maxamed Nuur Caliow, asked for support from General Caydiidand his SNA alliance.This, however, did not ® ndfavour with his allies, anda splitdeveloped between those in the SDM whowere in alliance with Caydiid’ s SNA andthose who continued to favour alliance with Cali Mahdi and the SSA.Theformer consisted mainly of one of the most important Digil clans, theEelay, together with lineages split from other clans. Therewas anattempt to reunify the SDM ata conferenceheld at the village ofBonka near Baydhaba in 1993. The only result of this, however, was to produceyet another faction; one which held aloof from both alliances and attemptedto assert theDigil± Miri¯ e as athirdforce. It was thisorganisation that tookall the seats inthe UNOSOM-organiseddistrict councils in the Baydhaba area. Hencethere were three SDM factions: SDM±`Manifesto’, SDM±SNA and SDM± Bonka. Thusthe clan families and larger clans have progressively split and re-split intotheir component parts. Each alliance has its` ®fthcolumn’ working hard to encouragethis process byseducing the sub-clans and lineages of the other, so thatwhat we have now is alliancescutting across clansand clan-families, with splitsat every level. Thegenealogical grid, far frombeing a ®xedframework which imposes itself onpeople’ s behaviour,is manipulatedaccording to political exigencies. An exampleis thearea roundthe southern port of Kismaayo, where politics are dominatedby the power struggle between the two major Daarood groups (both ofthem immigrants into the area fromthe northeast region): the Ogaadeen and the` ’. Thelatter, in the northeast where it originates, is aspeci®c genealogicalgrouping of four clans only; but in Kismaayo it has become acatch-allterm to include all Daarood other than the Ogaadeen, including thegenealogically quite distant Mareexaan. The genealogy, in fact, is being manipulatedto form an alliance to counterbalance the powerful Ogaadeen. Thechange from the old clan system to` neo-clanism’(in Gerard Prunier’ s 294 SOMALIA: QUESTIONING ACOLLAPSED STATE phrase)has beenone of substance as wellas form.The clan and the minor groupswhich make it up used to be self-governing direct democracies; the clan elders,and the local sultans and ` kings’were simply its leading members. These ®guresare stillthere, but today it isthepoliticians and the military leaders who callthe tune to which the elders and the sultans are forcedto dance. What is noteworthyis thatthe same formof organisation which worked in the old rural conditionshas provedso (disastrously)well adapted to the new ones. Under the centralisedstate, people had used the clan network to compete by political manoeuvres.With the collapse of government, the means becamemilitary. All sorts ofresources are inplay, but one of the most signi® cant is thecontrol of agriculturalland, especially the plantations along the two rivers, since this has ledto amajorshift of population.Under Siyad, plantation land was heldmainly byhisfellow-clansmen and protege Âs; withhis fall, ` liberators’from the Haawiye clanstook them over. On the Jubba they were driven out again when the Daaroodagain established their ascendancy; while on the Shabeelle river, the HaawiyeÐmainly Habar GidirÐ took over the plantations and kept them. In eithercase, peopleof other groups, the earlier inhabitants of the land, have lost out.17

TheIslamic movements Asigni®cant, though not dominant, element in theSomali political scene are the Islamicmovements. This is animportant and complex subject, which has notyet beenstudied with the thoroughness it demands. It can only be touchedon here. Islam is thereligion of all the Somali people, whatever clan they belong to, anintrinsicpart of theiridentity. In being thus assimilated and totally part of the nationalway of life, the faith has inevitablymade its compromises with local custom,and certainly many Somalis are less carefulin their practice than they are staunchin their allegiance. Nevertheless the Somali people have in the courseof theirhistory known many movements intended to revive and purify the faith,of whichthe best known is thatin theearly 20th century headed by Sayid MaxamedCabdille Hassan (theso-called ` madMullah’ ). Inhis case, as inmany others,the determination to reform the people’ s religiouslife was associated withresistance to foreign control. The various Islamic movements in Somalia todaytherefore have a historyto draw on. Yet the organisations of today are fundedfrom abroad, and come out of other Islamic traditions than those of Somalia.They appeal to thosewho see submissionto Islamiclaw in itsfullness, astheyunderstand it, as thesolution to the lawlessness ofthe country. In 1996 Islamicgroups were in controland Sharia courts were operating in thenortheast, innorth Mogadishu, in Beled Weyn and Jowhar on the Middle Shabeelle, in Luuqon the Upper Jubba, and in Raas Kaambooniin the far south. Thegrowth of these movements goes back to the time of Siyad Barre, when theybegan to ¯ourishin place of the banned political parties, and ever since they haveremained linked to certain clan groups. However much a religiousmove- mentmay reject clanism and try to transcendit, it isalmostimpossible to avoid runningin thatgroove. The movements are notunited (if they were their power wouldno doubtbe greater); there are atleastfour separate organisations: Ittihad 295 VIRGINIA LULING al-Islami (IslamicUnion); al-Ikhwanal-Muslimun (MuslimBrothers); Wuhdat al-Shababal-Islami (Unionof Islamic Youth); Majmaal-Ulama (Unionof the Clerics).18 Allof them are receivingfunds from overseas, but not all from the same sources.They differ in their programmes and objectives, and in their degreeof acceptanceby thepeople. Al-Ittihad, funded from Saudia Arabia, is the groupmost active in the northeast; they are alsothe predominant group in the Kismaayozone. The organisation in charge in Luuq on the Upper Jubba on the otherhand has Iranianbacking. 19 Probablythe most signi® cant of allis theSharia court set upby Al Ittihad in NorthMogadishu, the zone under the authority of Cali Mahdi’ s USC. In calling inthis group to control the lawlessness ofthe streets hehas founda powerful ally,but one which shows signs of becominghis master. The court is supported morallyand ® nanciallyby Saudia Arabia and Sudan. It has succeededÐwhere UNOSOM failedÐin disarming the local population. It controls the airport, the barriersbetween North and South Mogadishu, and the police. The of® cial line aboutthe North Mogadishu court is thatit isabranchof the SSA administration, alongwith other ministries. It has beencalled in on a temporarybasisÐ by this accountÐto deal with the lawless situation, and once a regulargovernment is established,` thepeople’ will decide if the court should keep its power or have itcurtailed. But one may doubt whether this would in fact be so easy. Thereis nodispute that, where the Islamic courts rule, there is ameasure of peaceand security, and hence they enjoy much public support. Punishments such as amputationand stoning to death, which outrage opinion in Europe and the USA,are notunpopular with the Somali public, since they are seen asbringing safetyto the streets. Thereare criticalvoices however. ` Itis notthe Sharia itself that is wrongbut theway these groups handle it. They cut off the hand of a poorboy who steals abananaÐbut people who steal millions are stilldoing business.’ There is distrustof their political agenda. ` Theyare exploitingthe funds from foreign countries.They lead to ® ghtingof brother against brother. They only succeed dueto poverty.’ Many are affrontedat the idea of presuming to ` islamise’an alreadyMuslim nation. Uncommitted opinion is perhapsbest summed up bythe manwho said that if hehadto choose between the court and anarchy, he would choosethe court. In 1996 it seemed thatthe Islamic organisations remained ` a factor,but not a majorfactor in the situationÐ in default of other attractive options.’ 20

Whatprospects forpeace? Amongmany Somalis there is amoodof despair:` Peoplehere have no faith in thefuture’ . Yetthe corollary of this is awar-wearinesswhich may itself be a forcefor peace. ` Theydestroy, destroy’ , saida womanwho had lost eight of her relatives;` we’ve hadit uptohere’. Bothinside Somalia and among the Somalis ofthe diaspora, there is arecognitionthat they will not achieve their aims throughmilitary means: `Theyused to run towards the sound of ® ghtingÐbut nowthey run away. People abroad used to raise fundsfor their groups, but now thatfeeling is dying.Now they are talkingabout ª peacewith our neighboursº .’ 296 SOMALIA: QUESTIONING ACOLLAPSED STATE

Whatform could such a peacetake? Would reconciliation between the SNA and the SSA beenough? Must ` Somaliland’in the northwest be involved, or shouldit go its own way? Would it mean agreement on a centralauthority, or agreementto do without one? Thereare effortstowards reconciliation going on, a wholerange of initiatives bygroupsof politiciansand intellectuals, both in Somalia(eg between north and southMogadishu) and overseas. ` Weare tryingto bring together twelve to twenty-fourª cleanºpeople, such as formerstatesmen, who could be the link betweenthe international community and the Somalis, who can advise the internationalcommunity and the regions.’ 21 `Aconferenceof selected leaders fromeach clan; genuine representatives of thepeopleÐ 100 people maximum.’ 22 Perhapsmore importantly, at thelocal level there are realpeace initiatives by genuine` elders’, forinstance those of the Xawaadle and in Beled Weyn,who in April1996 made a pactto seek forpeace, rejecting the politicians whothey claim do not represent them. At another level deals are doneby the morepowerful merchants, such as thosewho formed a `Peace Committee’in Mogadishu,needing stability to run their business. Womenand women’ s groupsare oftencited as aforcefor peace. This is not necessarilyso. Women have been no less partisanthan men. Here as inother wars,women have egged their menfolk on (their traditional role in Somali warfare); helpedthem, as whenGeneral Morgan used women to bring guns and munitionsinto Kismaayo under their robes; andon occasion fought and killed. Itis truethat women have an equally well established peace-making role. Because Somalis(especially among the pastoralists) generally marry outside theirclan, it followsthat women have ties to boththeir own and their husband’ s clan,and so are bothmotivated to help in reconciliation, and useful as go-betweens.In traditional society, when peace was establishedbetween two groups,women from one group were often married into the other as asignof reconciliation(which in fact means menusing women for diplomatic purposes). Theexperience of the sufferings of war has turnedwomen’ s mindsto this peace-makingfunction, while at the same timethey have acquired new roles in organisedgroups. The support given by international agencies to such groups has enabledthem to take part in peaceconferences like that at AddisAbaba in 1993 (wheresome ofthem went on a nine-dayhunger strike), and to travel as far as Beijing.However, such groups are mostlyclan-tied just as thereligious organi- sationsare. The activity of women’s NGOsinthehumanitarian and medical ® elds is notable;but it may be that in the long run they will have more signi® cance forthe history of gender relations in Somali society than for the peace process. Finally,forces outsideSomalia itself may be crucialin anypeace process. The UnitedNations, putting the ® asco ofarmed intervention behind it, continues to playa littlepublicised role, encouraging peace initiatives. So do other inter- nationalbodies, such as theEuropean Union and the Arab League (to which Somaliaused to belong). Somali intellectuals generally believe that these organisations,and others such as theOrganisation of AfricanUnity ( OAU) as well astheirneighbouring countries, should play a moreactive role (for instance by imposingan effective arms embargo).Some even argue that the UN should assume atrusteeshiprole. 297 VIRGINIA LULING

Whilethe real effectiveness of these large international bodies may be doubted,operating as theydo mainly at the level of politicians and roving intellectuals,smaller aid organisations, both non-governmental and offshoots of the UN like UNICEF,playan important part on the ground. In the absence of functioningstate institutions, they not only provide essential services, especially medicalones, but in a sense canbe said to assume some statefunctionsÐ since thepriorities for aid and development are set,not by the Somalis but by the organisations.They can take on the task of building up local institutions with administrativefunctions, as doesthe Swedish Life and Peace Institute,which has beenhelping to organise local councils at grassroots level. Atthe same time,these organisations too become associated with particular clangroups, which bene® t bothfrom their services andfrom the employment theyoffer. Willy-nilly the NGOs®ndthemselves identi® ed with ` their’clans. Hencesome clanshave waged inter-clan warfare by driving out one another’ s NGOs. (GeneralMorgan has reportedlybeen very successful inthis tactic in Kismaayo.)Caydiid’ s governmentmeanwhile imposed a registrationsystem on NGOs. Soaid organisations become players in the game, rather then arbitrators. Thereis, then, not only a generalcraving for peace but a numberof initiatives workingto attain it. All these efforts face formidableobstacles, however. Wedo not need to suppose that the cause ofthe violence is simplythe nationalcharacterÐ though it is aviewexpressed not only by stereotyping foreigners,but by embittered Somalis: ` Thegreatobstacle is theSomalisÐ they becomewild, they have to ® ghtevery day.’ There are enoughother dif® culties. Thereis theabsence of legitimate recognised political representatives among the Somalis.Who has theright to represent whom? Who has theauthority to enforce anaccord? (This is whatwrecked the UNOSOM-sponsoredconferences.) The publicdo not necessarily obey either elders or politicians; nor do the militias obeytheir commanders unless it suits them. As ColonelJees reportedlysaid to aUNrepresentative:` If onlyI couldmake my militia do what they don’ t want todo¼’ Partlythis has todowiththe non-hierarchical nature of Somalisociety, partlyalso with the poor quality of leadership and the lack of trust people feel intheirpoliticians. The gap in communicationis increasednow that many of the politicianslive outside the country. Thereis theseemingly limitless supply of arms availableÐboth the heavy weaponryleft over from the Siyad Barre period,and the light arms and ammunitionthat continue to ¯ oodin unchecked. There are thepowerful vested interestsin the absence of lawÐ `warlords’ and militia leaders, the young men who® ndliving by the gun more pro® table and more fun than herding camels, orwhohave no otheremployment; ma® a-likegangs who run protection rackets. Anotherset ofpeople who, while they need peace, have reason to fear whata ruleof law might bring, is thosewho have taken over propertyÐ whether real estatein Mogadishu, pasture lands or plantations. Those who have pro® ted do notwant the process reversed. Thebuilding up of local coalitions, such as thoseof the Daarood in the northeastand along the Jubba, is liableto beseen asÐindeed to beÐ not a `®rst step’to national unity but an alliance threatening others, a preparationfor war, notpeace. There are limitations,too, on local grassroots initiatives and reliance 298 SOMALIA: QUESTIONING ACOLLAPSED STATE onlocal elders. These men do not necessarily have the authority they once enjoyed,in theface ofmilitialeaders and armed gangs of youths.Moreover local eldersare notalways as localor as traditionalas theyseemÐ they may be part ofan armed land takeover. Thereis aproblemwith the continuing involvement of the United Nations and theEUÐ apartfrom the fact that many Somalis regret the ` gravydays’ of 1992±93, and would like a higherlevel of commitment from them. This is that theyareÐ correctlyÐ seen as sidingwith the SSA against the SNA. There are historicalreasons forthis since it was AliMahdi’ s governmentwhich originally calledin the UN, andit is understandablegiven Caydiid’ s behaviour,but it underminestheir standing as animpartial force. Finally,there is thedream of national unity, or command. In the absence of acommonenemy or anationwideconsensus, unity can only come by force,that is tosay byconquest. This was Caydiid’s method,perhaps one he saw as compelledby the supreme need to unite the country. In practice, however, it means thatuni® cation is notthe guarantor but the chief enemy of peace. Meanwhilethe two alliances continue their confrontational rhetoric, even thoughboth are rivenwith divisions. The SNA has,one might say, too much leadership,the SSA toolittle. The SSA triesdesperately to forge a realcommon policyfrom a mixtureof groupswith different agendas, a pointmade contemp- tuouslyby the other side: `Theyhave no proper principles, no leader no platform.All they can do is criticise.We don’ t recognisethem as apolitical association.Like people in LondonÐ Hyde Park. We ignore themÐ that is our policy.’ 23 The SNA,onthe other hand, was overwhelminglyidenti® ed with its leader,Caydiid, and it remains to be seen ifitcan hold together in his absence. Formany, Caydiid himself was themain obstacle to peace. ` Caydiidis behavinglike a banditÐwe have tried and tried to cooperate, but people are losingpatience. In the last resort we willhave to use thesame methodshe does todefeat him.’ 24 Tosome heseemed anunstoppable force, capturing one city afteranother until he appeared about to take over the whole of southernSomalia. Othershowever pointed out that he only partially controlled the territory he claimedÐ`he relies on splinter groups’ Ð that,though he had plenty of weaponry,his limited manpower was overstretched,that he owed his success to thelack of unityof theother side, and that his military domination had not been translatedinto a civilianadministration. ` Itis importantnot to confusethe power toattackand destroy with the power to govern.’ 25 Hewas describednot unjustly as yesterday’s man,still living in the 1980s in the world where to be president was thekey to limitless foreign aid. As anEU representativesaid, ` Hedoesn’t realisenobody is interestedany more’ . Hisdeath was followedby con® dent propheciesthat the way would be open for everyone else toreach a settlement. Suchhopes are unlikelyto be ful® lled.

Conclusion: goodfences makegood neighbours Itis clearthat the old days of the centralised state are gone:; `Wemustnever goback to over-centralisation. People want a largesay intheir local affairs, withoutimposing a willfrom the top.’ 26 Behindthis is thepremise that ` a 299 VIRGINIA LULING centralisedgovernment necessarily favoursone clan.’ 27 `Federalism’and ` decen- tralisation’are thewords on everyone’ s lips.Even the SNA thinkalong these lines,in spite of Caydiid’ s denunciationof the ` dismembering’of Somalia, and theirrefusal to join in any discussions on the topic: ` Ourpolicy is decentralis- ingÐfederalisation with regional autonomy.’ 28 Theproblem is whatthese words will mean in practice.Exiled politicians and intellectualsendlessly debate the possibilities. Should there be a federationof ®vestates, or four, or three? Should its components be based on clanship, or territory,since every piece of territory can be claimed by two or more clans, whileclans and clan-families are geographicallyscattered? Now that the clan maphas beenredrawn, would the federal entities follow the old map or thenew? Howcan their boundaries be agreed upon? Whilethe intellectuals and politicians argue, the people on the ground make theirown arrangements. As oneyoung man put it: ` Because ofthe past ® ve years,Somalis have come to rely on themselves, and this is working.No one is helpingthem, there is nogovernmentÐ and business is boomingand everyone is workingfor him or herself, there is nowelfare system. So who is interested ingovernment? No one is interested.We only need a governmentto represent thename of SomaliaÐ we don’ t needone which interferes in the affairs ofthe people.’ Themost convincing, and generally agreed, proposition is thatreconstruction muststart from the bottom level. Many would concur with the leader of a women’s organisation:` Idon’t believethe country can be united in the near future.It should be built up from the bottom from the smallest possible neighbourhoodsor villages.’ There are evencalls for what is ineffect a return totraditional xeer law:` Eachclan should sort itself out and neighbouring clans shouldhave arrangements for compensation.’ Unityby consentmay come at some futuretime, the genuine underlying sense ofSomaliidentity reassert itself,but the reality in themeantime is thepatchwork of`clanmini-states’ , andthe old logic of thegenealogical grid: that people unite toconfront a commonenemy, then split up again. 29 Theabsence of a state structurewith which to conduct relations poses aproblemfor foreign govern- mentsand the UN, butit is onethey are alreadybeginning to address. The immediateissue beforethe Somali people (as perhapsbefore those of other collapsedstates) is whetherthis non-unity can ® ndits own stability, with good governanceat locallevel. The setting up oflocalcouncils is designedto leadto suchan outcome. The question is whethersuch units can keep within and betweenthem, if not total peace, then enough of it to make life livable. If they cannot,then the prospects for eventual reuni® cation will also be poor.

Notes 1 Theresearch forthis paper was carried outin January± May 1996, thanks to a MacArthur NGO Visiting Fellowshipat theDepartment ofWar Studies,King’ s College,London. Interviews were heldin Nairobi, Mogadishu,Kismaayo andMarka in February 1996. 2 Accordingto one donor ` Thenumber of Somali lives saved by the intervention of UNITAF/UNOSOM is 300 SOMALIA: QUESTIONING ACOLLAPSED STATE

probablyonly a small fractionof what is claimed¼ Muchof the work had been done already in the latter halfof 1992,by UNICEF and various NGOswithairlifts andpublic health inputs.’ J Prendergast TheGunTalks Louderthan the Voice ,Washington,DC: Centre ofConcern, 1994, p 5.Anotherview is thataid as awhole caused thefamine rather thancuring it; this is argued,for instance, by Sam Rileyin The Times, 25 April 1996:` In1992 a shortfamine inSomalia caused bycivil war¼ was deliberatelymanipulated by the warlords,because the ICRC andother agencies, underthe glare ofTV lights,would pay any price tofeed Somalis,in a year inwhich domestic food production was insurplus¼ spending $100 000 a week on ªprotectionºin Mogadishu, the ICRC gavethe warlords a ®nancialincentive to maintain starvation at the highestpossible level. Far fewer mighthave perished if the famine hadbeen ignored entirely.’ 3 TheSomali language has itsown standard spelling system, andI haveused it here forpersonal and place names. Itmay strikethe English reader as oddat ®rst,but not more sothan, eg Portuguese or Danish,and ifnames inthose languages are normallyspelt in the native manner Idonot see whythe same shouldnot applyto Somali. I havemade anexception of ` Mogadishu’(rather than` Muqdisho’), onthe grounds that thisform, like ` Warsaw’ or` Munich’, has become anEnglish convention. The English reader shouldbear inmind that: (i) ` x’stands for the ` voiceless pharyngealfricative’ ,anh-like sound made at theback of the throat;hence Xasan 5 Hassan; (ii)` c’ isthesound ’ ayn,found in , butinaudible to most Europeans; hence Cali 5 Ali,Caydiid 5 Aydid.A problemwith personal names is thatSomalis on the international scene (particularlythose who grew upbeforethe Somali writing system was introducedin 1972)habitually spelltheir names inthe English or Italian style. Here Ihaveused the Somali spelling when referring to a personor quoting an interview, but retained the anglicised form when citing a writtenwork. 4 MDoornbos& JMarkakis,` Societyand state incrisis: whatwent wrong in Somalia?’ ,MAMohamedSalih &LWohlgemuth(eds), Crisis Managementand the Politics of Reconciliation in Somalia , Uppsala, NordiskaAfrikainstitutet, 1994, p 16. 5 AbdalleOmar Mansur` Contraryto a nation:the cancer oftheSomali state’ ,inAli Jimale Ahmed(ed) The Inventionof Somalia ,Lawrenceville, RedSea Press, 1995,p 115. 6 AbdiA Mohamed(` Baffo’), `Somere¯ ections on current Somali politics and its future directions’ , presentationat theNairobi Peace Forum,1995, p 8. 7 CabdullaahiDool, Soomaaliya:Maxay Kala Tirsaneysaa Maryooley? (Somalia:what does it mean tothe ordinaryperson?) London: Horn Heritage, 1995,p 32. 8 IMLewis, `Segmentarynationalism and the collapse of theSomali state; in Bloodand Bone ,Lawrenceville, RedSea Press, 1994,p 231. 9 AhmedI Samatar (ed), TheSomali Challenge: From Catastrophe to Renewal ,Boulder,CO: Lynne Rienner, 1994, p 9. 10 Doornbos& Markakis,` Societyand state incrisis’ ,p14. 11 Aleaked1991 World Bank document on publicexpenditure management makes itclear thatthe budget of virtuallyevery ministry and project was plunderedby the families thatdominated the government. Of the expenditureaggregate about a quarterappears as `unidenti®ed’ ie embezzled. Scams involvingstate farms are detailedin K Menkhaus& KCraven,` Landalienation and the imposition of state farms inthe lower Jubbavalley’ , inCBesteman &LVCassanelli, TheStruggle for Land in Southern Somalia: the War Behind the War,Boulder,CO/ London,Westview Press/Haan, 1996. 12 BHelander,` Vulnerableminorities in Somalia and Somaliland’ , IndigenousAffairs (Copenhagen),No 2, 1995, p 21. 13 Sucha system is notunique to the SomaliÐ the same sortof thingexists for instance among Bedouin Arabs. Butthe political system inthe Arab nationsis notgenerally dominated by people of Bedouinbackground, as is thecase inSomalia. 14 MohamedAbdi Mohamed ` Crise d’identiteen Somalie’ ,papergiven to the European Association of Somali StudiesConference, London, 1993, p 14. 15 Theproblem with referring to a groupinglike the Haawiye orIsaaq as a`clan’ ,as theEuropean and US media havebecome accustomed todo, is thatto an ordinary Westerner thissuggests a strongdegree of cohesionand common organisation, perhaps even a singleleader. Ithas tobeemphasised thatin traditional Somalisociety this is notthe case; these high-levelgroups have never been anything but unstable alliances. 16 A Simons, Networks ofDissolution; Somalia Undone ,Boulder,CO: Westview Press, 1995,p 139. 17 Onthis whole subject see Besteman &Cassanelli, TheStruggle for Land ;andR Omaar &AdeWaal, `Land tenure,the creation of famine, andprospects for peace inSomalia’ (DiscussionPaper No 1), London: African Rights,1993 (reprinted in Mohamed Salih & Wohlgemuth, Crisis Management ). 18 MohamedAbdi Mohamed, ` Crise d’identite ÂenSomalie’ ,p11. 19 ACairobroadcast declared there were camps forIslamic extremists near Luuq,containing people of many nationalities( SWB 10July 1996). 20 Hassan Cali Mirre,interview. 21 Maxamed Abshir` Waldo’, interview. 22 Dr Hassan ShIbraahim, interview. 23 MQanyere Afrah,interview. 24 Xuseen Xaaji Maxamed,` Bood’, interview. 301 VIRGINIA LULING

25 KMenkhaus,unpublished ® eldtrip report: ` Somalia’ ,September1994, p 12. 26 Ambassador CabdilleIsmaaciil, interview. 27 PStahl,` Somaliproposals for federal structuresof governancein Somalia’ ,Mogadishu,Policy and Planning Group, UNOSOM II,1994, p 2. 28 Maxamed Qanyere Afrah,interview. 29 See IMLewis, `Theuncentralised Somali legacy’ , in AStudyof Decentralised Political Structures for Somalia:A menuof Options ,reportprepared by consultants form the London School of Economics and PoliticalScience, 1995.

302