Creating the Cape Colony: The Political Economy of Settler Colonization

G een, E ik Book C ea ing he Cape Colony: The Poli ical Economy o Se le Coloniza ion P o ided in Coope a ion wi h: Bloomsbu y Academic Sugges ed Ci a ion: G een, E ik (2022) : C ea ing he Cape Colony: The Poli ical Economy o Se le Coloniza ion, ISBN 978-1-350-26321-5, Bloomsbu y Academic, London, h ps://doi.o g/10.5040/9781350263215 This Ve sion is a ailable a : h ps://hdl.handle.ne /10419/281245 S anda d-Nu zungsbedingungen: Die Dokumen e au EconS o dü en zu eigenen wissenscha lichen Zwecken und zum P i a geb auch gespeiche und kopie we den. Sie dü en die Dokumen e nich ü ö en liche ode komme zielle Zwecke e iel äl igen, ö en lich auss ellen, ö en lich zugänglich machen, e eiben ode ande wei ig nu zen. So e n die Ve asse die Dokumen e un e Open-Con en -Lizenzen (insbesonde e CC-Lizenzen) zu Ve ügung ges ell haben soll en, gel en abweichend on diesen Nu zungsbedingungen die in de do genann en Lizenz gewäh en Nu zungs ech e. Te ms o use: Documen s in EconS o may be sa ed and copied o you pe sonal and schola ly pu poses. You a e no o copy documen s o public o comme cial pu poses, o exhibi he documen s publicly, o make hem publicly a ailable on he in e ne , o o dis ibu e o o he wise use he documen s in public. I he documen s ha e been made a ailable unde an Open Con en Licence (especially C ea i e Commons Licences), you may exe cise u he usage igh s as speci ied in he indica ed licence. h ps://c ea i ecommons.o g/licenses/by-nc-nd/3.0/ C ea ing he Cape Colony ii C ea ing he Cape Colony The Poli ical Economy o Se le Coloniza ion E ik G een BLOOMSBURY ACADEMIC Bloomsbu y Publishing Plc 50 Bed o d Squa e, London, WC1B 3DP, UK 1385 B oadway, New Yo k, NY 10018, USA 29 Ea ls o Te ace, Dublin 2, I eland BLOOMSBURY, BLOOMSBURY ACADEMIC and he Diana logo a e adema ks o Bloomsbu y Publishing Plc Fi s published in G ea B i ain 2022 Copy igh © E ik G een, 2022 E ik G een has asse ed his igh unde he Copy igh , Designs and Pa en s Ac , 1988, o be iden i ied as Au ho o his wo k. Fo legal pu poses he Acknowledgemen s on pp. x–xi cons i u e an ex ension o his copy igh page. Co e image: Ox-Span o Co e ed Wagon and Plow, by Jan B andes, 1806. Du ch Colonial pain ing o a Du ch se le ’s a m in Lochne , Sou h A ica. Cou esy o Rijksmuseum. This wo k is published open access subjec o a C ea i e Commons A ibu ion- NonComme cial-NoDe i a i es 3.0 licence (CC BY-NC-ND 3.0, h ps://c ea i ecommons.o g/ licenses/by-nc-nd/3.0/). You may e-use, dis ibu e and ep oduce his wo k in any medium o non-comme cial pu poses, p o ided you gi e a ibu ion o he copy igh holde and he publishe and p o ide a link o he C ea i e Commons licence. Bloomsbu y Publishing Plc does no ha e any con ol o e , o esponsibili y o , any hi d-pa y websi es e e ed o o in his book. All in e ne add esses gi en in his book we e co ec a he ime o going o p ess. The au ho and publishe eg e any incon enience caused i add esses ha e changed o si es ha e ceased o exis , bu can accep no esponsibili y o any such changes. A ca alogue eco d o his book is a ailable om he B i ish Lib a y. A ca alog eco d o his book is a ailable om he Lib a y o Cong ess. ISBN: HB: 978-1-3502-5823-5 ePDF: 978-1-3502-5824-2 eBook: 978-1-3502-5825-9 Typese by In eg a So wa e Se ices P . L d. To ind ou mo e abou ou au ho s and books isi www.bloomsbu y.com and sign up o ou newsle e s. Lis o maps i Lis o igu es ii Lis o ables iii Acknowledgemen s x 1 Unde s anding he c ea ion and expansion o se le colonies 1 2 Indigenous agency, he cos o ade and ini ial s eps owa ds a se le colony 23 3 Fac o endowmen s, ins i u ions and he expansion o he on ie 41 4 Was he Cape Colony a sla e economy? 75 5 Unequal we s and 103 6 Eli es, coali ions and se le esis ance 129 Concluding ema ks 151 Re e ences 155 Index 167 Con en s Lis o maps 2.1 Ske ch map o he Cape, showing app oxima e posi ions o Khoesan g oupings in he se en een h and eigh een h cen u ies 26 3.1 Map o he Cape Colony, compa ing 1682, 1705, 1731 and 1795 bounda ies wi h mode n-day bounda ies 42 3.2 Se le expansion a he Cape h ough 1750 48 3.3 Bounda ies o eehold a ms a Cape Colony, 1717 (Cape Peninsula excluded) 55 3.1 GDP pe capi a (1990 in e na ional Gea y–Khamis dolla s) o he Cape Colony, 1701–93 45 3.2 Compa ison o GDP pe capi a es ima es, 1701–73 46 3.3 Eu opean popula ion in he Cape Colony, 1703–93 49 3.4 Eu opean and sla e popula ion in he Cape Colony, 1703–93 50 4.1 Es ima ed numbe o Khoesan, 1701–95 90 5.1 Weal h inequali y among se le - a me s in Cape Colony, 1663–1757 111 6.1 G aph o se le p i ilege and inco po a ion in o go e nance 132 Lis o igu es C ea ing he Cape Colony2 In 1652, he Du ch Eas Indian Company (he eina e VOC, Ve eenigde Oos indische Compagnie) es ablished a ading pos a he Cape o Good Hope. One hund ed and i y yea s la e , he VOC was in con ol o an a ea he size o I aly. This book is an a emp o iden i y and analyse he ac o s ha led o he c ea ion o he colony and d o e i s con inued geog aphical expansion. This expansion was nei he in ended no planned. How did i come abou ? This book will show ha he ounding and g ow h o he Cape Colony can only be unde s ood as a combined e ec o se le and he indigenous Khoesan agency, leading o an ins i u ional o de ha ein o ced he enla gemen o he colony’s bo de s. Focusing on he g adual and unce ain p ocesses ha cha ac e ized he c ea ion o he Cape Colony helps us o unde s and he e olu ion o a colonial ins i u ional o de . Wi hou a p ope unde s anding o hese p ocesses, mo eo e , we canno accu a ely analyse he long- e m legacies o colonialism. The book is based on he ich li e a u e examining he social, poli ical and economic his o y o he Cape Colony. Schola s like Robe Ross, Nigel Wo den, Leona d Guelke, He mann Giliomee, Wayne Dooling, Susan New on-King and Shula Ma ks, o name a ew, ha e g ea ly con ibu ed o ou unde s anding o he his o y o he Cape Colony and ha e p o ided in aluable sou ces o in o ma ion and inspi a ion in he w i ing o his book. In ecen yea s we ha e seen a e ised in e es in he economic his o y o he Cape, wi h impo an publica ions om a new gene a ion o schola s like Johan Fou ie, Die e on Fin el, Jeanne Cillie s, Ka e Ekma, Calume Links and Igo Ma ins. Combining his o y wi h s a is ical me hods and economic heo y, his g oup o schola s ha e p o ided new insigh s in o he Cape Colony’s long- un de elopmen ajec o y. This book has g ea ly bene i ed om hei wo k. The eade may eac o he ac ha he book does no ci e any o he high quali y schola ly wo k ha has been published in A ikaans. Tha is indeed a limi a ion o he book and s ems om he ac ha I do no ead A ikaans. I would no eel igh o men ion and/o ci e li e a u e ha I am only able o unde s and on a supe icial le el. Reade s will also no ice ha no p ima y sou ces a e ci ed in he book. Ha ing said ha , he book does o a signi ican ex en ely on my own use o p ima y sou ces in pape s ha ha e al eady been published. O special impo ance o his book has been my join wo k wi h Johan Fou ie, Die e on Fin el, Jeanne Cillie s, Calume Links, Auke Rijpma Igo Ma ins and Robe Ross in cons uc ing he Cape o Good Hope Panel da abase, he longes - unning and mos de ailed such da ase in exis ence o he Global Sou h. The panel da ase will con ain longi udinal, demog aphic and economic in o ma iona he indi idual le el o e a pe iod o 180 yea s (1660–1840). The cons uc ion o he da ase is s ill in p og ess bu has al eady C ea ion and Expansion o Se le Colonies 3 p o ided us wi h he abili y o analyse and e ise a gumen s made in he la ge li e a u e on se le economies a ound he globe. Below, I o e a b ie e iew o he li e a u e on se le colonialism wi h a ocus on schola ly wo k ha uses he on ie as a concep ual ool o unde s and he p ocesses by which Eu opean se le socie ies we e es ablished. This is ollowed by an ou line o he heo e ical poin o depa u e o his book. Ra he han p esen ing a ixed model, he sec ion iden i ies he key ac o s ha in luence he expansion o se le socie ies. Fi s , howe e , I p o ide a e y b ie desc ip ion o he se en een h- and eigh een h-cen u y Cape in o de o gi e he eade a ough sense o he con ex be o e mo ing on o he p e ious li e a u e on se le economies and he analy ical and concep ual guidelines ha a e used in his book. To unde s and how se le colonies expanded, one needs, as a poin o depa u e, an analy ical amewo k ha akes in o accoun he unce ain and agile p ocesses ha cha ac e ize he e olu ion o a new social o de , one ini ia ed by in ude s in egions whe e e y di e en social o de s we e al eady p esen . These unce ain ies and ac al de elopmen s a ec ed he ins i u ional ab ic in a way ha laid he g oundwo k o he ins i u ions ha we can iden i y in he nine een h-cen u y Cape and in wen ie h-cen u y Sou h A ica, including acial seg ega ion, unequal land dis ibu ion and ex eme weal h exis ing side by side wi h deep po e y. To cap u e hese g adual de elopmen s, I e u n o a concep ha a ew decades ago inspi ed he his o ians and geog aphe s who s udied se le colonies bu has in ecen yea s played a ma ginal ole in he ield: he concep o he on ie (Lama and Thompson 1981; Elphick and Giliomee 1989a; Penn 2005; Willebald 2016). I de ine he on ie in economic/sociological e ms and dis inguish be ween an open and a closed on ie . In simple e ms, an open on ie ep esen s a si ua ion in which no clea ins i u ional o de has ye been es ablished and no g oup has gained powe and judicial con ol o e a e i o y. A closed on ie ep esen s he opposi e, ha is, a clea and hie a chical poli ical and economic o de based on ins i u ions ha a e i mly embedded in he socie y, bo h in de ju e and de ac o e ms. F om his pe spec i e, he book unde s ands he i s 150 yea s o he Cape Colony’s his o y as a p ocess o on ie closu e in which ce ain ‘Eu opean’ ins i u ions did de elop na u ally as a consequence o economic and poli ical s uggles o e economic esou ces. Theapp oach aken he e di e s om he a ional- choice poli ical economy li e a u e in a meaning ul way. While ha schola ship ocuses on he ole o ins i u ions, echnology, human capi al, inequali y and g ow h in a closed on ie , his book cen es on he p ocess ha led o he closing o he on ie . C ea ing he Cape Colony4 A e y b ie his o y o he Cape Colony, 1652–1796 Eu opeans i s se led a he Cape o Good Hope in 1652 unde he auspices o  he Du ch Eas India Company wi h he goal o adminis e ing a e eshmen s a ion o passing ships. Due o hei ailu e o p ocu e su icien p o isions om he indigenous Khoesan popula ion, he VOC e en ually decided o allow Eu opeans o es ablish a ms nea he o . In 1657, he i s hand ul o company se an s eed om hei con ac s we e g an ed lands along he Liesbeek Ri e , behind Table Moun ain. This spa ked a geog aphical expansion o he Eu opean se le communi y in o he in e io , which i s mo ed no h and hen eas along he coas . Twen y- i e yea s a e he es ablishmen o he ading pos , he a ea o he VOC’s e i o ial con ol was s ill small, encompassing oughly 650 km2. By he end o he eigh een h cen u y, i had expanded o oughly 250,000 km2. Th oughou he pe iod, he Eu opean popula ion a he Cape was doubling app oxima ely e e y gene a ion (Ross 1993). By 1793, he e we e wen y- i e housand Eu opean se le s a he Cape, wen y housand sla es and abou wen y- i e housand indigenous Khoesan (Guelke 1985; Fou ie and G een 2015; Dye and La C oix 2020). Se le ag icul u e o med he basis o he colonial economy. Eu opean a me s in he sou h-wes e n Cape in es ed mainly in whea a ming and i icul u e. Fa he in o he in e io , pas o alism domina ed. The a ms we e smalle han plan a ions in he Ame icas, wi h an a e age size o 50 hec a es. The se le s ope a ed hei a ms wi h he help o impo ed sla es and indigenous Khoesan labou e s. Sla eholding was widesp ead. In 1773, 75 pe cen o all se le s owned sla es (Wo den 1985). In con as o he plan a ions o La in and No h Ame ica, he a e age a m a he Cape had a low numbe o sla es. By he mid-eigh een h cen u y, mo e han 50 pe cen o a me s held ewe han i e sla es, compa ed wi h he ‘median suga es a e’ in Jamaica, which held an a e age o 204 sla es (Fou ie 2013a; G een 2014). P e ious es ima es indica e high le els o economic inequali y a he Cape, bo h among se le s and o he en i e popula ion (Guelke and Shell 1983; Fou ie and on Fin el 2010a; Cillie s and G een 2018). A ew se le s became e y a luen , while o he s ha dly made ends mee . The es ablishmen and geog aphical expansion o he Cape Colony was me wi h esis ance om he Khoesan. Se e al wa s and con lic s b oke ou as se le s mo ed eas wa d (Ma ks 1972; Links, G een and Fou ie 2020). Be ween wa s, esis ance con inued by o he means. The Khoesan ecu en ly s ole ca le om he Eu opeans, which occasionally o ced se le s o abandon hei C ea ion and Expansion o Se le Colonies 5 a ms (Penn 2005). Meanwhile, se le s ea ed hei indigenous labou e s wi h b u ali y, which spu ed u he esis ance om he Khoesan (New on- King 1999). Se le economies and a ional-choice poli ical economy Views o se le colonialism ha e s ikingly shi ed back and o h since he ea ly wen ie h cen u y. Ea ly on, libe als and socialis s alike saw he di usion o Eu opean powe s ac oss he globe as necessa y o unleash he p oduc i e o ces o non-Eu opean egions (Lenin 1917; see also Mamdani 2015 o a c i ical e iew o he ea ly li e a u e on ‘Ame ican excep ionalism’). By he middle o he wen ie h cen u y, wi h he ise o independence mo emen s in A ica, his posi i e iew o se le colonialism shi ed d ama ically. I was by ha poin o en desc ibed as a ype o ‘dieha d colonialism’ cha ac e ized by agg essi eness and exploi a ion (Ve acini 2013: 313). In he 1980s, his iew pa ly changed as schola s became inc easingly in e es ed in unde s anding why ce ain pa s o o me Eu opean colonies had pe o med dis inc i ely be e in economic e ms han did o he s. The la e colonies we e ound in opical Asia, A ica and La in Ame ica and he o me in he empe a e zones in he sou he n and no he n hemisphe es. In his p ominen compa a i e wo k, Denoon (1983) explained his di e ence by dis inguishing be ween he numbe o Eu opeans who se led in each egion. These de eloped in o se le capi alis economies, d i en by a con inued expansion o he land on ie . The di e si y o economic de elopmen pa hs among o me colonies has mo e ecen ly been picked up by he li e a u e ha Aus in (2008) desc ibes as a ional-choice poli ical economy (RPE). The RPE li e a u e, in line wi h Denoon, a gues ha egions whe e Eu opeans se led in g ea e numbe s (so-called neo-Eu opes; see C osby 1986) ha e pe o med be e economically and socially han o me colonies whe e he Eu opeans we e ela i ely ew (Enge man and Sokolo 2000, 2002, 2005; Acemoglu, Johnson and Robinson 2002; Acemoglu and Robinson 2012; Eas e ly and Le ine 2012). The explana ion o his, acco ding o he RPE li e a u e, is o be ound in he ins i u ional o de es ablished by he colonize s. In egions whe e Eu opeans se led in g ea e numbe s, hey b ough wi h hem ins i u ions – such as p i a e p ope y igh s and he ule o law – ha acili a ed economic g ow h. In colonies wi h ew se le s, an ins i u ional o de was es ablished C ea ing he Cape Colony6 ha enabled hese se le s o ex ac na u al and human esou ces bu also c ea ed high le els o inequali y and hinde ed long- e m economic g ow h. The cu en income gap be ween La in and No h Ame ica is commonly used o epi omize he di e gen legacies o colonialism. La in Ame ica is poo e oday, i is a gued, because i a ac ed ewe se le s, leading o an ex ac i e ins i u ional o de ha p e en ed long- e m economic de elopmen . In con as , he la ge numbe o Eu opeans ha se led in No h Ame ica o med a c i ical mass ha ensu ed he es ablishmen o g ow h-enhancing ins i u ions. A numbe o schola s – while hey iden i y di e en ansmission channels, such as echnology and human capi al o ma ion – ha e len suppo o he idea o a posi i e ela ionship be ween he numbe o se le s and Eu opean popula ion densi ies in he colonies ac oss he globe (e.g. Angeles 2007; Bake , B unnschweile and Bul e 2008; B uhn and Gallego 2008; Pu e man and Weil 2010; Eas e ly and Le ine 2012). I seems ha we a e back in he ea ly wen ie h cen u y, wi h a ange o iewpoin s seeing he di usion o Eu opean ins i u ions and knowledge as ha ing acili a ed long- e m economic de elopmen . The RPE li e a u e has been exposed o c i ique. The common g ound o he c i ique can bes be summa ized by he wo ds o Gonzales and Mon e o, who claim ha he RPE li e a u e is ‘a basically an i-empi ical way o easoning’ (2010: 255). I does no o e an empi ical in es iga ion o how ins i u ions, inequali y o economic g ow h e ol ed o e ime in di e en colonies. Ins ead, i assumes ha Eu opean se le s es ablished speci ic ins i u ional o de s ha pe sis ed o e ime. A numbe o schola s a gue ha hese assump ions lack empi ical suppo (Coa swo h 2008; F ankema 2009; Williamson 2015). La in Ame ica was nei he poo e no mo e unequal han he Uni ed S a es du ing he colonial pe iod. I was only a he end o he nine een h cen u y, abou h ee hund ed yea s a e being colonized, ha he wo egions di e ged. Coa swo h (2008: 553) concludes ha ‘wha li le quan i a i e e idence he e is does no sugges ha owne ship o land, o o he asse s o ha ma e , was mo e concen a ed in La in Ame ica han in he Uni ed S a es’. The RPE li e a u e no only lacks an empi ical unde s anding o he di e gen ins i u ional o de s once hey had been es ablished, bu also sees i s es ablishmen as an exogenous e en plainly d i en by Eu opean in e es s (Lloyd and Me ze 2013; F ankema, G een and Hillbom 2016). W i ing o he es ablishmen o he ins i u ional o de in La in Ame ica, Acemoglu, Johnson and Robinson (2002: 1375) s a e: ‘Soon a e he conques , he Spanish c own g an ed igh s o land and labou ( he encomienda) and se up a complex C ea ion and Expansion o Se le Colonies 7 me can ilis sys em o monopolies and ade egula ions o ex ac esou ces om he colonies.’ Simila ly, Enge man and Sokolo conclude ha La in Ame ica ‘ea ly on’ de eloped an ins i u ional o de based on high le els o inequali y and he ex ac ion o economic en s, as compa ed o No h Ame ica, which began ‘wi h ela i e equali y and homogenei y’, leading o he de elopmen o g ow h- acili a ing ins i u ions (2013: 67). In nei he o he wo cases a e we p o ided wi h any de ailed accoun s o how hese ins i u ional s uc u es eme ged. His o ians ha e con incingly shown ha he p ocess o es ablishing o e seas se le socie ies was complex and can ha dly be unde s ood as exogenous (Lama and Thompson 1981; Elphick and Giliomee 1989a; Wea e 2003). Wha we imagine oday o be in ended ou comes we e g adual p ocesses ha nei he happened o e nigh no we e ine i able. Se le economies and hei ins i u ional ab ics e ol ed g adually and cons i u ed endogenous p ocesses a he han exogenous e en s (F ankema, G een and Hillbom 2016). C i ics ha e also acknowledged ha wen ie h-cen u y colonialism in A ica di e ed om he b oad pa e ns desc ibed in RPE schola ship. Fi s , he ela ionship be ween Eu opean popula ion densi ies and he deg ee o ex ac i e ac i i ies seems o be posi i e, a he han nega i e (A ighi 1970; Aus in 2008; Fibaek and G een 2019). Ins ead o lobbying o g ow h- enhancing ins i u ions, se le s pushed o a coe ci e and seg ega ionis ins i u ional o de . Sou h A ica and colonial Zimbabwe (Sou he n Rhodesia) a e o en ci ed as an example o how se le s success ully managed o pe suade he colonial au ho i ies o es ablish an ex ac i e ins i u ional o de . In colonies wi h small se le popula ions, in con as , coe ci e policies we e less common and le els o inequali y lowe . Second, he ela ionship be ween Eu opean popula ion densi ies and ela i e economic p ospe i y wi hin A ica is a om clea -cu (Hillbom and G een 2019). Acco ding o he A ican De elopmen Bank, Sou h A ica and Kenya, bo h o me se le colonies, we e in 2019 on he lis o he en iches coun ies on he con inen in e ms o income pe capi a (Sou hA ica was numbe wo and Kenya numbe nine). Meanwhile, he o me non-se le colonies Nige ia and E hiopia made i on o he lis a numbe one and numbe eigh , espec i ely. The ques ion ha needs o be asked in ligh o he inconclusi e e idence o La in and No h Ame ica is whe he A ica may be he con inen ha p o es he RPE li e a u e w ong. Wha is needed is a mo e sys ema ic s udy o how speci ic colonial ins i u ional o de s g adually e ol e. The so-called on ie li e a u e p o ides a use ul s a ing poin o such an exe cise. C ea ing he Cape Colony8 The ea ly on ie li e a u e and he people wi hou his o y Few publica ions had an impac on he schola ship on Eu opean se le colonialism in he i s hal o he wen ie h cen u y as Tu ne ’s ‘The signi icance o he on ie in Ame ican his o y’ (o iginally gi en as a pape a he Ame ican His o ical Associa ion mee ing in Chicago in 1893 and ep in ed in 1921). The a icle does no deal wi h he es ablishmen o a se le socie y pe se, bu wi h he con inued enla gemen o a se le socie y’s bo de s. The expansion o he on ie was o Tu ne a p ocess in which he ‘disin eg a ing o ces o ci iliza ion en e ed he wilde ness’ (1921: 21). I was a g adual p ocess, o en beginning wi h indi idual ade s mo ing a he wes , la e o be ollowed by g oups o se le s looking o economic oppo uni ies. In Tu ne ’s wo ds, ‘Each expedi ion was an epi ome o he p e ious ac o s o wes e n ad ance’ (24). Tu ne ’s wo k ocused solely on he Eu opean se le s who managed o make p oduc i e use o wha Tu ne ega ded as emp y i gin land on he wes e n on ie . In some cases, he expansion was hal ed by esis ance om na i e Ame icans, bu ha was empo a y and success ully s uck down by he supe io weapons and mo e sophis ica ed o ganiza ion o he se le s (23). This expansion c ea ed i s own unique ins i u ions and cul u es. F ee om he eudal chains s ill el in Eu ope, he expansion and es ablishmen o amily a ms p omo ed indi idualism, mobili y, equal oppo uni ies and poli ical eedom. Acco ding o Tu ne , ‘The exis ence o an a ea o ee land, i s con inuous ecession and he ad ance o Ame ican se lemen wes wa d, explain Ame ican de elopmen ’ (1). Al hough no always explici ly men ioned, Tu ne ’s on ie hesis inspi ed much o he schola ship in he i s hal o he wen ie h cen u y on Eu opean se lemen expansion o e seas (e.g. Bowman 1928; P ice 1950). I is also no able how Tu ne ’s idea o he sp ead o indi idual ‘pionee ing’ amily a me s o ms he ounda ion o he unde s anding o he his o y o No h Ame ica in he mo e ecen RPE wo k on se le economies. The di e gence be ween No h and La in Ame ica is p ima ily explained by an ins i u ional o de ha in he o me case o med a ound independen se le amily a ms ha could ope a e a a m’s leng h o he go e nmen . Simila a gumen s de eloped in he ea ly wen ie h cen u y ega ding he es ablishmen o he Cape Colony. As ea ly as 1909, Leo Fouché ga e a lec u e on ‘Die E olu ie an die T ekboe ’ in which he desc ibed he eme gence o a class o wande ing pas o al a me s – he T ekboe s – in he i s i y yea s o he Cape Colony. Acco ding o Fouché, hese became he ‘spi i ual a he s’ o he se le s who, a cen u y la e , would mig a e no h-eas in C ea ion and Expansion o Se le Colonies 9 wha was known as he G ea T ek, he his o ical expe ience hough o ha e laid he ounda ion o Sou h A ica as a na ion. The old on ie li e a u e can be c i icized on se e al accoun s. Wi h oday’s eyes, i is easy o see how i p esen s a oman ic iew o he on ie smen and he whole p ocess o on ie ‘opening’. Con lic s and b u al iolence a e ei he neglec ed o men ioned only in passing. One ea ly excep ion o his is he libe al his o ian E ic Walke who, al eady in 1930, a gued ha he oo s o acism in Sou h A ica lay in a speci ic on ie cul u e. In gene al, howe e , he Eu opeans a e desc ibed as he agen s o ci iliza ion, while he indigenous peoples a e ei he neglec ed o ega ded as passi e agen s. One can also ques ion, as Fu niss (2006) does, whe he Tu ne ’s ag a ian populis accoun – which sees he main economic and poli ical con lic s as hose be ween go e nmen and big business on he one hand and he o dina y people on he o he – accu a ely cap u e he dynamics o a on ie socie y. Tu ne igno ed o he ac o s ha p esumably a ec ed he sociopoli ical s uc u e o he on ie socie ies, including class con lic , u baniza ion and sla e y. The idea ha i was a ela i ely isola ed on ie ha shaped he sociopoli ical ins i u ions in he colony has also been ques ioned. The ea ly li e a u e ul ima ely ea ed he on ie in a a he s a ic way. The opening and closing o on ie s: A dynamic accoun A e being neglec ed o decades, he on ie concep e-eme ged in he 1960s and 1970s. The eme ging on ie li e a u e c i icized he ea lie wo ks o neglec ing he ole o indigenous people and social and economic con lic s. Tu ne ’s hesis was especially a acked, accused o being no only simplis ic, bu o some ex en e en acis (Be kho e 1981). In his wo k on US his o y, Fo bes (1968: 207) c i iqued Tu ne ’s on ie hypo hesis as one-sided in i s ocus on Eu opeans. Ins ead, he a gued he on ie mus be unde s ood as ‘an in e - g oup si ua ion’ (wi hou discussing in any g ea de ail he in e play be ween he g oups). Slo kin (1992) wen u he when he desc ibed he Ame ican on ie as a eas ma ked by con lic s, domina ion, ep ession and ma ginaliza ion. The ea lie li e a u e was also c i icized o ea ing he on ie as isola ed om he poli ical and economic cen e o he economy. Looking a he Cape Colony, Legassick (1980) claimed ha he on ie should no be seen in isola ion om he cen e o he colonial economy, bu as an ex ension o i . Acco ding o Legassick he acism ha libe al his o ians desc ibed as a on ie phenomenon was ins ead b ough om he cen e o he on ie . C ea ing he Cape Colony10 The ocal poin in he la e li e a u e is on on ie closu e. Lama and Thompson (1981) de ine he on ie as a e i o y whe e wo o mo e dis inc socie ies mee . The social and economic o de is luid and unce ain. Acco ding o hese au ho s, he on ie emains open un il a single poli ical au ho i y has es ablished hegemony o e he a ea. This au ho i y can ei he ep esen one o he g oups o cons i u e a hyb id o hem. Taking ou poin o depa u e as on ie closu e leads us o conclude ha i is he in e ac ion and con lic s be ween a ious g oups and socie ies in he opening on ie ha de e mine he ins i u ional o de ha will domina e once he on ie is closed. The dis inc ion be ween an open and a closed on ie has been applied o he case o Sou h A ica as well, mos explici ly by Giliomee (1989), who a gued ha he on ie has wo dimensions: social and geog aphical. The la e emains open as long as he e is s ill a possibili y o he in ude s o mo e spa ially. The social dimension is abou poli ical and social con ol. In an open on ie , he social o de is luid. Va ious g oups a e igh ing o gain poli ical powe . No poli ical au ho i y is ecognized as legi ima e by all pa ies, which makes he use o coe cion common. A closed on ie is cha ac e ized by he es ablishmen o a poli ical au ho i y and a social o de ha can be mo e o less egali a ian. In his wo k, Giliomee ocused mainly on he expansion o he eas e n on ie a he Cape. W i ing me ely wo decades la e , Penn (2005) used a simila amewo k o analyse he expansion o Eu opean se le s in o he no he n on ie o he Cape. The ou come o on ie closu e is unce ain. I may be poli ical and economic ic o y o he Eu opean se le popula ion, bu i may also be a de ea . The unce ain y shapes how people beha e and hence he e olu ion o ins i u ions. This is pa ly neglec ed in he RPE li e a u e as i deals wi h on ie s ha ha e al eady been closed – ha is, a eas whe e he Eu opean se le s managed o es ablish poli ical au ho i y and whe e he indigenous popula ions ha e become ma ginalized o – e en wo se – elimina ed. Neglec ing he p ocess o on ie closu e means ha he RPE li e a u e misses ou c ucial ac o s ha de e mined how he socio-economic ins i u ions e ol ed o e ime and why ce ain ins i u ions de eloped, pe sis ed o anished. As Denoon (1983: 27) obse es, ‘i is he quali ies o he indigenous socie y which p o oundly in luenced he kind o se le socie y which could be supe imposed upon i , o which migh en i ely eplace i ’. Al hough he on ie li e a u e p o ides use ul insigh s in o unde s anding he es ablishmen and expansion o a se le colony, i is no wi hou p oblems. A majo limi a ion, Lama and Thompson (1981) poin ed ou , is he lack o C ea ion and Expansion o Se le Colonies 11 a cohe en and analy ically s ingen way o de ining he on ie o he key a iables shaping he on ie p ocess. Does he expansion imply a di usion o capi alism, ma ke s and comme cializa ion o p oduc ion as well as ela ions o p oduc ion? O , as he ea lie li e a u e a gued, does he expansion o a on ie en ail he c ea ion o a unique se o ins i u ions based on amily businesses ha a e only pa ially in eg a ed in o he capi alis economy? How should we unde s and con lic s a he on ie and he g oups in ol ed in hese con lic s? Should one simply classi y he la e in o wo g oups, he a i ing se le s and he al eady p esen indigenous popula ions? O is he p ocess mo e complex, wi h a ious in e es g oups and/o social classes ha do no necessa ily align wi h hei e hnici ies? And how exac ly should one de ine a on ie ? When is a on ie ‘opened’ and when is i ‘closed’? Le us s a wi h he ques ion o de ining he opening and closing o a on ie . As I men ioned, Lama and Thompson (1981) sugges ha he opening o a on ie occu s when he i s ‘ ep esen a i e’ o an in usi e socie y a i es in a speci ic egion. I closes when he in ude s ha e gained poli ical au ho i y o e ha egion. This de ini ion neglec s he possibili y o he on ie being closed i he in ude s a e chased away o when he poli ical au ho i y con ains a hyb id be ween new and old in e es g oups among he in ude s and he al eady p esen popula ions. Reducing he opening o a on ie o a clash and/ o in e ac ion be ween wo dis inc socie ies bo h simpli ies he p ocess and con uses i wi h he p ocess o closing he on ie (Be kho e 1981). To a oid his pi all, I de ine he opening and closing o a on ie om an economic pe spec i e. A on ie  is opened when a new g oup o people a i e in a speci ic loca ion and use he esou ces a ailable he e in a way ha changes he ex ac ion o esou ces and he economic ela ions be ween g oups. I is no necessa ily o only closed when a ce ain g oup has gained poli ical con ol o e he e i o y, bu a he when a new way o ex ac ing esou ces has been es ablished and has become necessa y o ep oducing he popula ion o he egion. This de ini ion allows us o see he opening o a on ie no only as a con lic be ween wo o mo e g oups o e poli ical con ol, bu as con lic s (and coope a ion) among a ious economic in e es s ha g adually (o , in excep ional cases, d ama ically) lead o a ans o ma ion o he economy in ha a ea. This way o using he concep o opening a on ie inco po a es he ac ha he e a e al eady people p esen a he on ie who a e engaged in a ious economic ac i i ies. These ac i i ies may no immedia ely be h ea ened by he a i al o a new g oup, as long as he e is an abundance o esou ces ha can be pu in o p oduc i e use. This is he iew aken by Giliomee (1989) and o C ea ing he Cape Colony18 he coali ion, which makes he na u al s a e inhe en ly agile and ulne able o exogenous shocks. These shocks can easily lead o iolence o o majo shi s in he composi ion o he eli es ha make up he go e ning coali ion. In his book, I ega d on ie closu e as a p ocess leading owa ds he es ablishmen o a na u al s a e based on a limi ed-access social o de . My poin o depa u e is ha he expansion o he colonial on ie is no d i en by he au ho i ies. I is la gely a p ocess ini ia ed by he se le s. Once he se le s ha e es ablished hemsel es a he on ie , he au ho i ies y o gain con ol o e he e i o y. This p ocess can be cha ac e ized as a na u al s a e ha ex ends i s e i o ial con ol. The ex en o which his will be manageable depends on he balance o powe o he exis ing coali ion and how i is a ec ed by he es ablishmen o new se le a ms and he po en ial ise o a new eli e. In ha ega d, he closu e o he on ie in ol es he o ma ion o a new ‘ on ie ’ eli e, which is some hing ha needs o be deal wi h by he exis ing coali ion o eli es. This implies ha he economic and poli ical de elopmen s a he on ie canno be unde s ood in isola ion o he de elopmen s a he economic cen e. Gi en ha we a e dealing wi h a se le colony go e ned by a cha e ed company ( he VOC), he connec ion be ween he local eli es (a speci ic g oup o se le s) and he local au ho i ies may be blu y because he au ho i ies, a leas ini ially, ep esen he in e es s o hei sha eholde s. Does he coali ion consis o eli es in he me opole and he au ho i ies in he colony, o is i made up o local eli es? In a s udy o he (B i ish) Eas Indian Company, Boga (2015) analyses i s ac ions h ough he lens o changes in he eli e coali ions in G ea B i ain, ins ead o i s in e ac ion wi h he eli es in India. This book ocuses on he ela ionship be ween he local eli es and he colonial au ho i ies. The chie eason is ha he Cape Colony was ne e a p io i ized a ea o in e es o he VOC. The Company did no ha e any majo in es men s in he colony ha i needed o p o ec . I was up o he local au ho i ies o inance hei ecu en expendi u es and he VOC would no in e ene as long as he local au ho i ies con inued o ensu e a s eady supply o p o isions o passing ships. A p oblem wi h No h, Wallis and Weingas (2009) is ha hey do no o e concep ual ools o help iden i y he glue ha held he eli es oge he . To iden i y how eli e coali ions a e o med and kep in ac unde such complex ci cums ances, one could use he concep o social con ac s, mo e speci ically Nugen ’s (2010) classi ica ion o p oduc i e, pe missi e and coe ci e social con ac s. In a coe ci e con ac , he igh o go e n is linked o he capaci y o he ule s o impose hei will h ough iolence o he h ea o iolence. Ul ima ely, subjec s gi e up hei poli ical in luence in exchange o being spa ed C ea ion and Expansion o Se le Colonies 19 om ha assmen . The poli ical au ho i y gains ull de ac o powe , whe e de ju e elemen s a e negligible. The pe missi e con ac implies some so o nego ia ion be ween ule s and subjec s as o how he la e can ha e i s in eg i y p ese ed h ough he ule o he o me . I he de ju e powe can be ag eed upon, he de ac o powe can jus be a unc ion o i . In a p oduc i e con ac , de ju e and de ac o powe coexis in a speci ic balance. While he ule claims de ju e powe , i does no ully exe cise i de ac o, c ea ing a so o g ey a ea whe e social no ms de e mine wha is pe mi ed and wha is no . The powe o he ule , howe e , es s on he ac ha his balance is a delibe a e choice; in o he wo ds, he ule can exe cise de ac o powe any ime i s powe is h ea ened. In his book, I ake Nugen ’s h ee ypes o social con ac as a poin o depa u e bu ex end and modi y hem. As an ex ension, I employ he con ac s o unde s and he ela ionship no only be ween au ho i ies and he popula ion a la ge bu also be ween di e en classes o he popula ion. I modi y hem by no only allowing o he exis ence o mul iple con ac s in a socie y a any gi en poin in ime, bu also by a guing ha he exis ence o one con ac can be p edic ed based on he p esence o ano he . Fo example, he ela ionship be ween he au ho i ies and ce ain, p esumably weal hy, g oups in he socie y may be p oduc i e, bu only so long as hose g oups a e allowed o use coe ci e measu es agains o he g oups in he socie y. The di e en mix o con ac s ul ima ely depends on nego ia ions among he eli e g oups and he poli ical and economic s eng h o hose g oups. A he same ime, changes in one con ac may a ec all con ac s and hence – in he longe un – change he balance o powe among he eli e coali ions in a way ha also a ec s he ins i u ional o de as he on ie is closed. The s uc u e o he book Taking inspi a ion om he on ie li e a u e and he li e a u e on ins i u ions, ac o endowmen s and social con ac s his book now begins i s jou ney, which s a s in he se en een h cen u y and ends in he la e eigh een h cen u y, when he VOC loses con ol o e he e i o y. The book is s uc u ed as ollows. Chap e 2 p o ides backg ound con ex on he VOC’s p esence a he Cape. I will e eal a dialec ic p ocess. The VOC ne e in ended o es ablish a se le colony. Ins ead, i ounded a ading pos as a way o lowe ing he moni o ing and en o cemen cos s o ading wi h he indigenous Khoesan. The e ec was la gely he opposi e, leading he VOC o C ea ing he Cape Colony20 allow a ew se le s o es ablish hei own a ms o ensu e supply o p o isions o VOC ships. This unin en ionally spa ked a p ocess o expanding se lemen o e which he VOC had li le con ol. Chap e 3 ocuses on he expansion o he colonial bo de s. While shi ing ac o endowmen s was pa o he s o y, he ins i u ions ha go e ned landed p ope y igh s we e equally impo an . Bo de expansion con inued e en when he land–labou a io was low, mainly because weak p ope y igh s led o a dispe sion o he Eu opean se le popula ion. The weak p ope y igh s sys em was la gely an e ec o Khoesan esis ance, which caused se le s o wan o minimize ixed in es men s in landed p ope y. The sad i ony is ha Khoesan esis ance ga e se le s incen i es o expand, which hen weakened he Khoesan u he . In Chap e 4, I analyse he ex en o which he Cape Colony was a sla e economy, ha is, he deg ee o which sla e y as an economic ins i u ion was necessa y o he g ow h o he se le a ming economy. I concludes ha he Cape was indeed a sla e economy, bu no o he easons gi en by p e ious schola s. While sla e labou played an impo an ole o se le - a me s, especially a able a me s, I a gue ha p io esea ch has exagge a ed bo h he ole and e iciency o sla e labou . Di ec ly ela ed o his is he endency o neglec he ole o Khoesan labou . By including he Khoesan, I show ha he e u ns o sla e labou we e well below hose o he sla e economies o La in Ame ica and he Uni ed S a es. The eason sla e y was impo an o he Cape economy, I a gue, is ha i ga e sla eholde s access o bo h labou and capi al. Sla es we e no only an ag icul u al inpu . They could also be used as colla e al and leased ou o o he a me s. This gene a ed much-needed capi al o he se le - a me s. Chap e 5 deals wi h inequali y, especially inequali y le els and ends among he se le a ming popula ion. I ela es o Chap e 4 by again highligh ing he ole Khoesan labou played in he e olu ion o inequali y a he Cape. In his chap e , I show ha he inc eased inequali y seen om he 1740s onwa ds was la gely d i en by a concen a ion o weal h and income in he uppe ail o he dis ibu ion. While p e ious esea ch has implici ly o explici ly discussed inequali y a leng h, less ime has been de o ed o iden i ying he d i e s o inequali y. Human capi al, access o land and sla es, and immig a ion policies ha e been men ioned as possible causes. I a gue ha in o de o unde s and he e olu ion o inequali y a he Cape we need o look a he capaci y o ini ially accumula e weal h and income. In he ea ly yea s, his could be done by aking con ol o domes ic ade channels, in es ing in u ban se ices and/o in es ing C ea ion and Expansion o Se le Colonies 21 in a ming. A small g oup, wi h poli ical connec ions, gene a ed subs an ial amoun s o weal h h ough ade and in es men s in u ban se ices. Howe e , a la ge sha e o se le s, who would become he Cape gen y, ini ially ook on an accumula ion pa h based on a ming. Key o his s a egy was no possession o sla es, which a he ime we e oo expensi e, bu a he he abili y o a ac a su icien numbe o Khoesan labou e s o hei a ms. The i s o es ablish a ms in he sou h-wes e n Cape ook ad an age o a lack o clea land policies, by gaining con ol o as much landas hey possibly could. They hen used his land o a ac weakened Khoesan o se le on hei a ms, allowing hem o es ock hei ca le in exchange o labou . The ew who had he capaci y o a ac a su icien numbe o Khoesan could make p oduc i e in es men s in land ha gene a ed he weal h needed o ini ia e a mo e sla e-based ag icul u al sys em. Finally, he chap e discusses he pe sis ence o weal h. I concludes ha he e was a signi ican deg ee o eli e pe sis ence. This is somewha su p ising gi en he sys em o pa ible inhe i ance, bu i can be explained by ma iage s a egies ha kep weal h wi hin amilies, e en hough owne ship o landed p ope y ecu en ly changed hands. Focusing on he ela ionship be ween he VOC and he se le s, I a gue ha he VOC au ho i ies we e weak and ha hei e y exis ence depended on con inuous coope a ion wi h he se le eli es. This was achie ed by gi ing a small g oup o se le s con ol o e he highly luc a i e ade in wine, g ain and li es ock. While his p o ided a ai ly s able eli e colla ion, i occasionally came unde h ea , mainly because o opposi ion om ising new eli es. On one occasion, he se le eli es e en managed o ge he go e no o he VOC eplaced. A po en ial h ea o he eli e came om he majo i y o he se le s who we e no allowed o engage in he ade. To ensu e ha his g oup accep ed he s a us quo, he VOC allowed hem a la ge measu e o eedom in hei ea men o and obliga ions owa ds he Khoesan labou e s. On a dis ic le el, his eedom was ensu ed h ough he Cape gen y’s in luence o e dis ic au ho i ies. In sho , he chap e e u es he no ion ha he se le s we e a homogenous g oup whose in e es s aligned wi h he VOC. On he con a y, hei in e es s di e ged, which ende ed he VOC p esence a he Cape inhe en ly agile. The las chap e p o ides a b ie summa y o he lessons ha can be d awn om he local economic his o y o he Du ch Cape Colony o  he global s udy o se le colonialism. I a gue ha he es ablishmen o a se le colony is bes unde s ood as a g adual and complex p ocess, wi h impo an oles played by he ac ions o a ange o g oups wi h di e gen economic in e es s – including he indigenous popula ions. 22 In oduc ion The es ablishmen o he Cape Colony was no pa o a g and plan. I was a g adual and unin ended consequence o he VOC’s de e mina ion o inc ease i s con ol o he luc a i e ade wi h he Eas Indies by p o iding a e eshmen s a ion a he Cape. The chie p oblem was how o secu e supplies o mea o ships passing on hei way o and om Asia. To unde s and why his e en ually led o he es ablishmen o a se le colony, we need o pu he agency o he indigenous Khoesan a he o e on o he analysis. I was he inc eased di icul ies he VOC expe ienced coo dina ing he ade wi h he Khoesan ha e en ually led i o es ablish a ading pos . Bu his did no sol e he VOC’s p oblems. I any hing, he cos o ade wi h he Khoesan inc eased, e en ually leading he go e no o allow Eu opeans o se le as a me s in he sou h-wes e n Cape. This ma ked he beginning o he expansion. This expansion con inued h oughou he e a o Du ch ule, and he VOC had li le con ol o e i , as subsequen chap e s will show. The indigenous people a he Cape When VOC employees i s se led on he sho es o Table Bay in he mid- se en een h cen u y, hey encoun e ed people whom hey called ‘Bushmen’ ( he San) and ‘Ho en o s’ ( he Khoe). The e ms ‘San’ and ‘Khoe’ a e s ill in use, bu an h opological and his o ical esea ch has shown ha hese a e no disc e e e hnic ca ego ies. I used o be commonly belie ed ha he San we e smalle in s a u e han he Khoe, bu schola s like Shula Ma ks (1972) say he e idence is inconclusi e. His o ians he e o e commonly e e o he Khoe and 2 Indigenous agency, he cos o ade and ini ial s eps owa ds a se le colony 24 C ea ing he Cape Colony he San as one g oup, he ‘Khoesan’. An al e na i e is o ollow Elphick (1977) and Penn (2005) and dis inguish be ween he Khoe and hun e -ga he e s, ca ego izing he indigenous people on he basis o hei economic ac i i ies a he han along e hnic lines. Penn a gues ha his dis inc ion is necessa y as he hun e -ga he e s and he pas o al Khoe eac ed di e en ly o he expansion o he colonial on ie . The Du ch hemsel es did no dis inguish be ween Khoe and San by physical appea ance bu classi ied hem acco ding o hei socio- economic sys ems: Khoe owned ca le, San did no . Khoe could be eclassi ied i hey los hei li es ock and became hun e -ga he e s, and hun e -ga he e s could be eclassi ied i hey ob ained li es ock (Ma ks 1972). To a oid con usion, I ollow cu en his o iog aphical con en ion by using he e m ‘Khoesan’ in his book, dis inguishing whe e necessa y be ween pas o al Khoesan and hun e - ga he e Khoesan. Hun e -ga he e s had been p esen a he sou he n ip o A ica o abou wel e housand yea s. They li ed in small, amily-based o aging bands o usually en o i y people, some imes coming oge he in hun ing g oups ha could con ain se e al hund ed people. These la ge bands we e seldom ixed bu could o m and dissol e seasonally. The bands o en mo ed wi hin a de ined a ea and li ed in makeshi shel e s o ca es (Adhika i 2010). The Khoesan we e a nomadic pas o al people descended om mig a o y g oups who a i ed in he Wes e n Cape a ound 1000 ce, o igina ing om no he n Bo swana. They we e o iginally hun e -ga he e s, bu some ime a ound he i s millennium bce hey unde wen a pas o al ‘ e olu ion’. The need o keep mo ing o new pas u es o suppo hei g owing he ds e en ually led hem o he Cape. E idence o how hey mig a ed is sca ce. Acco ding o Elphick and Malhe be (1989), hey i s mo ed in o he g assland o he wes e n pa o wha is now Zimbabwe and hen sou h in o wha is now Gau eng, in Sou h A ica. They hen spli in o wo g oups, one going sou h-wes o e en ually each he Cape o Good Hope and he o he , known as he Nama, going wes wa d along he O ange Ri e un il hey eached he A lan ic. The g oup ha wen o he Cape we e pas o alis s who he ded ca le and sheep. They g ew no c ops excep dagga (cannabis). The hun e -ga he e s, al eady p esen in wha was o become he Cape Colony, came in o con ac wi h he Khoesan in a ious ways: h ough a med con lic , ade and economic coope a ion, o by being inco po a ed in o a Khoesan g oup as subo dina es o h ough in e ma iage (Elphick 1977). Con lic s mainly a ose o e access o esou ces. The pas o al Khoesan needed g azing o hei li es ock; he hun e -ga he e s wan ed he land p ese ed o game. T ade consis ed o goods like honey, mea , skins and bows and a ows. The adecould Indigenous Agency, Cos o T ade, Se le Colony 25 lead o he hun e -ga he e s becoming in eg a ed as pa o Khoesan socie y, al hough p obably li ing unde hei own chie in an encampmen close o he Khoesan. By he ime he Eu opeans a i ed, he hun e -ga he e s and he Khoesan had in e mixed o cen u ies, making i impossible o dis inguish be ween hem among e hnic lines; hence he e m ‘Khoesan’. The Khoesan socie ies a he Cape we e cons an ly on he mo e, dispe sing in o as a eas in sea ch o esh pas u es. As long as popula ion densi ies we e low, he e was no need o egula e access o land and he socie ies we e o ganized acco ding o kinship and possession o ca le, a he han land. Kinship g oups we e ela ed o each o he h ough he male line. I was qui e common ha hese pa ilineal clans we e loosely g ouped as ‘ ibes’, each wi h a chie a i s head. In gene al, howe e , he chie s had li le powe o e he clans, who could simply mo e on wi h hei ca le, which happened o en (Elphick and Malhe be 1989). The pas o al economies we e agile. Weal h ook he o m o li es ock (ca le and sheep), which we e ulne able o he by o he Khoesan, a ack by wild animals and loss om disease and d ough . As he li es ock we e owned by amilies a he han clans, a weal hy amily could be impo e ished o e nigh . The amily was hen aced wi h a choice o ei he s ealing li es ock om neighbou s o hi ing hemsel es ou as he de s o weal hy Khoesan who would pay hem p ima ily in ca le. A hi d op ion was o all back on hun ing and ga he ing. The hun e -ga he e s he Du ch called ‘Bushmen’ may hus ha e been Khoesan who had los hei li es ock (Elphick 1979). A he ime he VOC ading pos was se up, many Khoesan g oups we e li ing in he Cape Peninsula o close by (see Map 2.1). In he peninsula i sel , he Go inghaiqua, he Go achouqua, he Go inghaicona and o he sca e ed g oups we e all li ing unde a single chie . No h o he bay he e we e he Cochoqua and, a he no h, he Gu iqua. Wes o he Bay he e we e he Chainouqua and a he wes he Hessequa. Hun e -ga he e g oups we e sca e ed a ound he egion. The e a e no eliable es ima es o he size o he indigenous popula ions in he sou h-wes e n Cape a he ime o Eu opean a i al. Guelke (1974) es ima es ha he e we e app oxima ely wo hund ed housand in he mid-se en een h cen u y, which is also he numbe la e used by Fou ie and G een (2015). Elphick (1977) sugges s a mo e modes es ima e o one hund ed housand, while Elphick and Malhe be (1989) and Guelke and Shell (1992) sugges ha in 1652 he e we e abou i y housand. The i s commande o he Cape, Jan an Riebeeck, es ima ed in 1660 ha he e we e abou hi y- ou housand indigenous people li ing in he peninsula (i.e. mos likely only including he Chainouqua and he Cochoqua). In his ecen e iew o p e ious es ima es, La C oix (2018) inds Map 2.1 Ske ch map o he Cape, showing app oxima e posi ions o Khoesan g oupings in he se en een h and eigh een h cen u ies. Indigenous Agency, Cos o T ade, Se le Colony 27 ha i y housand is he mos accu a e es ima e o he size o he Khoesan popula ion, al hough he admi s ha his is p obably a sligh unde es ima ion. The la ge di e ences in he es ima es a e due o di e en me hods being used (such as backwa d p ojec ions o assump ions abou he ca ying capaci y o he soil) and a ia ions in he size o he egion ha is included. An es ima ed i y housand people in he sou h-wes e n Cape, an a ea o app oxima ely 130,000 km2, gi es us a popula ion densi y o 0.38 pe km2, while an es ima e o wo hund ed housand gi es us 0.65 pe km2. In ei he case, he Cape was, by 1650, a ypical open on ie , no emp y bu spa sely popula ed by sca e ed g oups. Coo dina ion cos s and he es ablishmen o he e eshmen s a ion By he end o he six een h cen u y, he Po uguese comme cial empi e was acing in ensi ied compe i ion om England and he Uni ed P o inces o he Ne he lands. Bo h he English and he Du ch s opped a he Cape o p o isions on hei ou wa d and homewa d oyages. The Po uguese had a ound i y ading pos s o o s on A ica’s eas and wes coas bu , by using he Cape, he English and he Du ch could a oid he eas coas o A ica and cu di ec ly ac oss he Indian Ocean. The Du ch Eas Indian Company (VOC) had been es ablished in 1602 by uni ing a numbe o comme cial unde akings, each go e ned by sha eholding di ec o s (bewindhebbe ). Se en een o hese di ec o s, he ‘He en XVII’ o in English he ‘Lo ds XVII’, o med he execu i e council. Ab oad, he Company’s a ai s we e managed by he go e no -gene al and Council o India (loca ed in Ba a ia) (Shu e 1989). Ini ially, he ela ionship be ween he VOC and he Khoesan was ela i ely peace ul. Bo h sides wan ed o a oid con lic – he Khoesan because hey wan ed Eu opean goods and he Eu opeans because hey needed mea o supply hei ships. Bu ha monious ela ionships did no las long. A e a decade o exchanging ca le o i on and coppe , in 1610 he Khoesan began demanding paymen in coppe only (Elphick 1977). To make ma e s wo se, hey especially wan ed coppe bea en in o shee s, no in lumps, so ha i could be u ned in o jewelle y. Once he Eu opeans had adjus ed o he new demand, he Khoesan s a ed asking o b ass. These changes made i di icul o he VOC o coo dina e hei ading ac i i ies. Ships a i ed om Holland loaded wi h i on, only o ind ha 34 C ea ing he Cape Colony he eina e e e ed o as ‘se le - a me s’ – we e g an ed land in exchange o aising ca le on behal o he VOC. Ini ially, he idea was o combine his wi h he p omo ion o in ensi e a ming in a limi ed geog aphical a ea. Van Riebeeck was e y op imis ic abou he po en ial. Sho ly a e his a i al a he Cape, he inspec ed he eas e n side o Table Moun ain. He concluded ha e en i he e we e housands o Chinese o o he ille s hey could no ake up o cul i a e mo e han a en h pa o his land. I is mo eo e so e ile and ich ha nei he Fo mosa, which I ha e seen, no New Ne he lands, which I ha e hea d o , can be compa ed wi h i . (Thom 1954: 35–6) The plan was no o es ablish a a ming se le socie y ha could ou compe e Asian ag icul u e; a he , VOC o icials belie ed ha he abundance o wa e and he e ile soil and a ou able clima e would enable a small numbe o Eu opeans o supply he VOC wi h su icien mea and g ain h ough in ensi e a ming p ac ices. Each a me was g an ed a po ion o land o only 28 ac es. The Company would emain in con ol o he de elopmen s, as he g an s depended on hese a me s handing o e 10 pe cen o hei annual p oduce. The es could be sold, bu only o he VOC a a ixed p ice (Guelke 1979). The in ensi e a ming plan ailed. The sou h-eas e ly winds we e o en s ong and long-las ing du ing he ha es season, causing ecu en , se e e losses o whea c ops. Mo e impo an ly, he labou needed o in ensi e a ming was in sho supply. Low popula ion densi y made i di icul o a ac su icien numbe s o Khoesan o wo k on he a ms. Sys ema ic use o coe cion was, a leas on pape , p ohibi ed by he VOC. The al e na i e was o employ Eu opean company se an s, impo sla es, o bo h. As I a gue in Chap e 4, nei he op ion was iable as long as he p o i abili y o Eu opean a ming emained low. I was simply oo expensi e o employ Eu opeans o buy sla es. Ins ead, he a me s came o ely on in ensi e a ming, in which ca le played an impo an pa . In 1666, Wagenae , he second Cape commande , s a ed ha he policy o es ablishing Eu opean a ms on he Cape Peninsula had been a ailu e and ha some Eu opean a me s had ‘abandoned all hopes, all inclina ion o a ming’ (quo ed in Ka zen 1969: 195). By he la e 1670s, only abou a hi d o se le s we e engaged in c op a ming ( an de Me we [1938] 1995). No only did he in ensi e a ming plan p o e o be a ailu e, bu he subsequen decision o allow a small g oup o VOC employees o es ablish a ms beyond Table Moun ain made i e en mo e di icul o he Company o sa is ac o ily con ol ade wi h he Khoesan. Two ac o s can explain his. The i s is ha he Indigenous Agency, Cos o T ade, Se le Colony 35 es ablishmen o a ew Eu opean se le a ms nea he o inc eased ensions be ween he VOC and he Khoesan, especially because he Du ch a me s we e beginning o mo e a he in o he in e io in sea ch o g azing land o hei li es ock. This h ea ened Khoesan access o pas u es and wa e , and hos ili ies almos immedia ely b oke ou . I was especially he Eu opeans – a his poin ela i ely ee o VOC con ol – who misbeha ed. Thie ing and ha assmen we e common (Guelke 1979). This led o he a o emen ioned i s wa be ween he Khoesan and he Eu opeans, in 1659. The VOC e en ually managed o nego ia e a peace deal, bu i did no e eal a clea winne . The Khoesan had o accep he p esence o he se le s and allow hem o g aze hei li es ock. Meanwhile, and agains he se le s’ wishes, he Khoesan we e allowed o s ay in he a ea and use g azing land unoccupied by he whi e a me s (Elphick 1977). Mis us inc eased, making VOC ade wi h he Khoesan inc easingly di icul . The second ac o is ha , in o de o su i e, he Eu opean se le - a me s also engaged in ade wi h he Khoesan. In 1652, he Company banned ade be ween Khoesan and Du ch se le - a me s. I ea ed ha allowing his ade would push up p ices and e en ually s ip he Khoesan o hei li es ock al oge he and des oy he sou ce o supply. They also ea ed ha he ade would encou age he a me s o go on hei own ading expedi ions in o he in e io , beyond he con ol o he VOC, and po en ially h ea en he p esence o he Company by ac ing in a hos ile manne owa ds he Khoesan. The VOC au ho i ies a he o soon hea d epo s o a me s no only ading li es ock di ec ly wi h he Khoesan, bu some imes coe cing hem o sell (Penn 2005). The ension be ween he se le - a me s and he local au ho i ies was exace ba ed by he Company’s p icing policy. The a me s complained abou bo h low pu chasing p ices and delayed in o ma ion abou he p ices. In 1658, a me s u ged he VOC, in he Jou nal o Cape Go e no s, o ‘Se us a p ice soon, because we will a m no land as long as we do no know he p ice, o we do no wish o be Company sla es’ (quo ed in an de Me we 1995: 2–3). An al eady agile si ua ion became wo se, and he VOC had o ace he eali y and accep ha hey we e losing con ol. The i s s eps owa ds on ie expansion Unde s anding ha he plan o es ablish land-in ensi e whea a ming was a ailu e, and ha ing managed o weaken he Khoesan li ing on he Cape Peninsula, he VOC ocused once again on building up a mo e sus ained ading 36 C ea ing he Cape Colony ela ionship wi h he Khoesan li ing nea he peninsula, especially wi h he Cochoqua, who li ed jus no hwes (see Map 2.1). The p esumably much- weal hie Cochoqua had emained neu al du ing he i s Khoesan–Du ch wa , which, acco ding o bo h Ma ks (1972) and Elphick (1977), is one eason why he Du ch managed o su i e he wa and nego ia e peace. No longe being dependen on he Khoesan in he peninsula o he same ex en as be o e he wa , he VOC decided o in ensi y expedi ions in o he in e io o es ablish o imp o e con ac s wi h o he Khoesan g oups. The Lo ds XVII app o ed he decision and ga e a boun y o explo e s. Du ing he se e al expedi ions ha we e made, he explo e s we e especially s uck by he a ou able condi ions hey ound no h-eas o he peninsula, in a place ha la e came o be known as Ho en o s Holland. As ea ly as 1657, Go e no an Riebeeck desc ibed he a ea in his jou nal: Th ee eemen had wi hou ou knowledge p oceeded inland abou i een hou s’walk, mainly sou hwa ds … In he icini y o a e y beau i ul i e , on bo h sides o which bi e almond ees g ew in p o usion and in such ine, e ile soil ha , acco ding o hem, he Cape alleys bea no compa ison he ewi h, hey ha e ound wo na i e encampmen s, numbe ing abou i e o six hund ed people … who … called his place hei Fa he land o Holland, o gi e ou men an idea o he abundance o ood and excellen pas u age o be ound he e. (quo ed in Heap 1993: 18) Imp essed by he su oundings, which hey deemed a mo e sui able o in ensi e whea a ming, he VOC made a second a emp o es ablish a ew se le a ms. In 1672, a numbe o company se an s we e sen in o he egion o sow g ain he e. O he s ollowed a ew yea s la e and se up a ms in he Tijge be g egion ( an de Me we [1938] 1995). Jus as wi h he p e ious a emp o es ablish in ensi e ag icul u e, howe e , he de elopmen s in he Ho en o s Holland a ea we e no as good as ini ially hoped. Th ee yea s a e he ini ia i e, he a ea whe e he VOC had en isaged a second se lemen had me ely become a ca le ading pos . A majo eason o he ailu e was ha he a emp o es ablish a se lemen in he a ea p o oked a second, longe wa (1673–7). Ini ially he con lic was no ela ed o he expansion o se le ag icul u e bu o a g owing mis us be ween he VOC and Gonnema, he in luen ial chie o a sub-g oup o he Cochoqua. Unlike o he Cochoqua chie s, he gene ally a oided con ac s wi h he colonis s. In he ea ly 1670s, he colonial au ho i ies became con inced ha Gonnema was planning a se ies o a acks on he Eu opeans. I his was ue, i is di icul o ell. The Company had e idence o only one case o Gonnema being Indigenous Agency, Cos o T ade, Se le Colony 37 in ol ed in a acks on Eu opean a me s. Thei suspicions we e mos ly based on accusa ions made by Khoesan enemies o Gonnema. Ne e heless, o ou yea s he VOC made ou puni i e expedi ions agains Gonnema, o en wi h suppo om o he Khoesan g oups. Only once did Gonnema make a coun e a ack. Ins ead, he usually used a mo e e icien weapon: he old his people o dispe se hei li es ock o e as a eas, making i nea ly impossible o he VOC o de ea him. In us a ion, he colonis s ins ead s a ed o a ack o he Khoesan g oups who had only loose connec ions wi h he ecalci an chie . E en ually a peace ag eemen was eached, in 1677. The wa , which had included se e al Khoesan g oups eas and no h o he Cape Peninsula, c ea ed ensions be ween hem and he Eu opean a me s and empo a ily pu an end o he idea o u he expansion. In 1679, he e we e only abou se en y Eu opean a me s in ha a ea o he peninsula (Guelke 1979). Agains his backg ound, he VOC again changed i s policy, in 1679. A e wen y yea s o con olled expansion, he Company decided o allow se le s o mo e eely beyond he Cape Peninsula. No legal limi s we e placed on he size o an indi idual’s land claim, bu he land had o be cul i a ed wi hin h ee yea s o be o ei ed (Guelke and Shell 1983). Fa me s mo ed eas o wha we e o become he S ellenbosch and D akens ein dis ic s and a ew yea s la e hey also mo ed up no h o Tijge be g. Soon he VOC would lose con ol o he expansion. Unin ended consequences and indigenous agency The plan o es ablish a se le colony a he Cape was no pa o a g and long- e m well- hough -ou s a egy. I a ose ou o a se ies o e en s ha s a ed a cumula i e p ocess leading o an ini ial expansion o he colony’s e ec i e bo de s. As I a gue in he ollowing chap e , once his p ocess was on he mo e, i c ea ed i s own logic o expansion. The decision o es ablish a ading pos and i s ea ly de elopmen canno be unde s ood wi hou acknowledging ha Khoesan ac ions and eac ions had a undamen al impac on how i de eloped. I was he way he VOC handled he unwillingness o inabili y o he Khoesan o ade hei ca le ha d o e he p ocess. I onically, e e y s ep he VOC ook – om es ablishing a ading pos o allowing a small numbe o se le s o es ablish a ms – inc eased he cos o ade and obliged he VOC o ake addi ional s eps o encou age he es ablishmen o se le ag icul u e. I would hus be a mis ake o ea he es ablishmen o he Cape Colony in i s ea ly yea s 38 C ea ing he Cape Colony as an exogenous e en , solely imposed om he ou side. I was a he a g adual p ocess, d i en by indigenous agency and unin ended consequences. I am ha dly he i s o acknowledge he impo ance o he indigenous popula ions. This is well in line wi h ecen esea ch in A ican economic his o y in gene al and among Cape his o ians in pa icula (see especially Aus in 2008; F ankema, G een and Hillbom 2016). Ye , his undamen al ac has been su p isingly neglec ed in ecen esea ch on he economics o se le colonialism. The p ocess ha led o he es ablishmen o a se le colony a he sou he n ip o A ica was complex and en ailed he encoun e o a a ie y o g oups wi h di e en in e es s and demands, gi ing ise o new o ms o con lic and coope a ion. The ansi ion om a ading pos o a ew a ms being es ablished beyond Table Moun ain was la gely he consequence o a clash be ween wo economic sys ems ha could no de i e mu ual bene i h ough ade. On one side, we ha e a Company whose comme cial in e es s we e ounded on he luc a i e ade wi h Asia; on he o he we e nomadic pas o alis s whose agile li elihood depended on main aining hei he ds in ac . Gi en he ulne abili y o he pas o al economy, he Khoesan we e seldom willing o supply he VOC wi h as much mea as i equi ed. F om he Company’s pe spec i e, his p oblem could no be sol ed by exe cising inc eased con ol o e one o se e al Khoesan g oups. In he sho e m, his could ha e helped o inc ease he p o ision o li es ock, bu i would e en ually ha e led o he collapse o he Khoesan pas o al economy and pu an end o he supply. Wha he VOC could and did do was o engage in ade wi h a la ge numbe o Khoesan socie ies o e ime. The p oblem hen was ha he cos o ade inc eased and made he VOC e en mo e dependen on he a ious Khoesan g oups wi h whom hey some imes li ed in peace and o he imes in con lic . To make ma e s wo se, he weal hie Khoesan socie ies ended o be loca ed a he away om he Cape Peninsula, whe e he clima e was mo e sui able o pas o alism. To access hese sou ces o ca le, he VOC needed o collabo a e wi h he Khoesan in he peninsula, whom hey needed as guides and in e p e e s. All his mean ha he ade wi h he Khoesan demanded he es ablishmen o a us ing ela ionship, no only be ween he VOC and he peninsula Khoesan bu also be ween he peninsula Khoesan and hose in he in e io . Es ablishing a ading pos was he VOC’s a emp o c ea e mo e sus ained ading ela ions wi h he Khoesan. Ins ead, he e ec was he e e se: i wo sened he ela ionship. When he Khoesan ealized ha he Eu opeans had come o s ay, hey began o igh back – no all g oups, and no a he same ime, Indigenous Agency, Cos o T ade, Se le Colony 39 bu a pa e n o ecu en esis ance made he si ua ion unce ain o he VOC and he se le s. In he ea ly yea s, i was a om clea who would come ou as he winne in his s uggle. In he end, he Khoesan los because hey could no p e en he VOC om encou aging Eu opean se lemen . Once his decision had been aken o allow a small numbe o Eu opean se le s o es ablish a ms, he logic o an expanding se le colony changed cha ac e . I came o be d i en less by indigenous agency and mo e by he way Eu opean se lemen a ec ed ac o endowmen s and ins i u ions, as I a gue in he nex chap e . 40 In oduc ion The his o y o he Du ch Cape Colony is, o a la ge ex en , a his o y o geog aphical expansion. Twen y- i e yea s a e he es ablishmen o he ading pos he a ea o he VOC’s e i o ial con ol was s ill small, encompassing oughly 500–650 km2. Expansion ook o om 1680. By ea ly 1700, he a ea con olled by he VOC o alled abou 1,440 km2 (see Map 3.1). By he end o he eigh een h cen u y i was oughly 250,000 km2, o abou he same size as mode n I aly – a uly ema kable expansion (Guelke 1985). This chap e deals p ima ily wi h his expansion and he ac o s ha made i bo h necessa y and possible. The li e a u e commonly iden i ies ou cha ac e is ics o a se le economy: ini ial low popula ion densi ies, a con inuous in low o mig an s, ex ensi e economic g ow h h ough an expansion o bo de s, he la e being suppo ed by secu e landed p ope y igh s (as discussed in Chap e 1). Only wo o hese we e ue o he Cape Colony: ini ial low popula ion densi ies and ex ensi e economic g ow h h ough he expansion o he on ie . Immig a ion o Eu opeans was low and hei p ope y igh s we e weak. Cape his o ians ha e a gued ha he expansion o he on ie was caused by he na u al g ow h o he se le popula ion. In his chap e , I modi y his demog aphic accoun o he expansion. I a gue ha inc eased popula ion p essu e alone is no su icien o explain he expansion o he colony’s bo de s. Ra he , i was popula ion g ow h in combina ion wi h a se o ins i u ions go e ning landed p ope y igh s ha spu ed mig a ion in o he in e io . Con a y o wha one migh expec , he se le s’ landed p ope y igh s we e in gene al e ms weak and luid. P i a e p ope y igh s we e only implemen ed in he sou h-wes e n Cape, a pa o he colony ha was closed by he end o he 3 Fac o endowmen s, ins i u ions and he expansion o he on ie Map 3.1 Map o he Cape Colony, compa ing 1682, 1705, 1731 and 1795 bounda ies wi h mode n-day bounda ies. Fac o Endowmen s, Ins i u ions, F on ie Expansion 43 se en een h cen u y, bu in combina ion wi h mo e luid igh s o g azing land. Beyond he bounda ies o he sou h-wes e n Cape, se le s e en ually go hei land egis e ed, bu only as leaseholds (locally known as loan a ms), and he land was nei he su eyed no p ope ly dema ca ed. On pape , he VOC could wi hd aw he leasehold con ac s on an annual basis, pu ing he se le s in he hands o he Company’s good will. I would a gue ha insecu e and luid landed p ope y igh s, oge he wi h he p ac ice o he pa ible inhe i ance o landed p ope y, encou aged con inual expansion o he on ie and ex ensi e economic g ow h. The ques ion is: why did se le s accep insecu e p ope y igh s? The eason is ound in he luid and unce ain social o de s ha cha ac e ized he ime o open on ie s. The isk o con lic s wi h he indigenous people who could o ce se le s o abandon hei a ms made hem unwilling o make any majo ixed in es men s in land, such as buying i o paying he cos o su eying and p ope dema ca ion. This chap e con ibu es in h ee ways o he la ge li e a u e on se le economies, ac o endowmen s and p ope y igh s. Fi s , he p e ious li e a u e on se le colonies assumes a one-di ec ional ela ionship be ween he ac o endowmen s and he ins i u ional o de o a speci ic socie y. Fac o endowmen s a e aken as gi en and p esumed o ha e had a signi ican impac on he shaping o economic ins i u ions. I a gue, in con as , ha one mus ecognize ha he ela ionship is mul i-causal. Changes in he ela i e supply o ac o s o p oduc ion can igge ins i u ional change, bu ins i u ions also a ec supply and access o economic esou ces. In his chap e , I a gue ha he la e mechanism is mo e impo an han he o me o he unde s anding o he expansion o he colony’s bo de s. Second, I claim, in con as o much o he mo e ecen li e a u e on se le economies, ha insecu e p ope y acili a ed a he han impeded ex ensi e economic g ow h a he Cape. Se le - a me s a he on ie had an in e es in keeping he cos o ixed in es men s as low as possible, gi en he isk o losing one’s p ope y. While he sys em o insu icien ly dema ca ed loan a ms kep he cos o ixed in es men s low, i also pushed se le - a me s deepe in o he in e io and he eby inc eased he likelihood o con lic s wi h he indigenous people. This leads o he hi d con ibu ion: my a gumen ha in o de o unde s and how he ins i u ions e ol ed o e ime, indigenous agency, la gely neglec ed in he economic li e a u e on se le socie ies, mus be aken in o accoun . Be o e I analyse he ole o ac o endowmen s, ins i u ions and agency I will b ie ly summa ize he economic de elopmen o he se en een h- and eigh een h-cen u y Cape in o de o pu he expansion in o a p ope mac oeconomic con ex . 50 C ea ing he Cape Colony he eigh een h cen u y ha a signi ican and s able inc ease began. Figu e 3.3 shows ha he numbe o Eu opeans inc eased om oughly h ee housand in 1703 o o e wen y housand in 1793. This gi es us a popula ion g ow h a e o abou 2.6 pe cen pe annum. This could be compa ed wi h A ica’s popula ion g ow h o 2.52 pe cen in 2019. Despi e he g ow h o he Eu opean popula ion, he Cape Colony emained spa sely popula ed. Guelke’s (1985) es ima e o oughly 250,000 km2 gi es us a se le popula ion densi y o 0.08 people pe km2 by he end o he eigh een h cen u y. This unde es ima es he ac ual popula ion densi y as i does no include he sla e popula ion o he indigenous people. Rega ding he o me , we ha e ai ly eliable sou ces. Figu e 3.4 shows he Eu opean and he sla e popula ions. Including he sla es would gi e us a popula ion densi y o oughly 0.18 people pe km2. The challenge is o cap u e he size o he indigenous popula ion. The i s p ope census aken o he indigenous people, om 1806, es ima ed ha hi y housand indigenous people we e li ing wi hin he bounda ies o he colony (Fou ie and G een 2015). Assuming ha his es ima e oughly ep esen s he size o he indigenous popula ion in he la e eigh een h cen u y, he popula ion densi y eaches 0.3 people pe km2, which is s ill s ikingly low. As explained in Chap e 2, he bigges challenge is o accoun o he size o he indigenous people Figu e 3.4 Eu opean and sla e popula ion in he Cape Colony, 1703–93. Sou ce: Dye and La C oix (2020: 38). No es on he sou ce: o he Eu opean popula ion. Es ima es o he sla e popula ion a e om Shell (1994). Fac o Endowmen s, Ins i u ions, F on ie Expansion 51 o ea lie pe iods. Se e al schola s ha e a gued ha he size o he Khoesan popula ions declined h oughou he eigh een h cen u y because o he Eu opean se le expansion and he ou b eaks o smallpox in 1713 and 1755. Howe e , he e is no consensus on he pace a which he decline occu ed – i depends on which ‘es ima e’ you use as a benchma k. Es ima es o he size o he indigenous popula ion in he sou h-wes e n Cape upon he a i al o he Eu opeans di e wi h a ac o o 9, om ele en housand o one hund ed housand (La C oix 2018). I we ake he uppe bound es ima es o he sou h-wes e n Cape he popula ion densi y would ha e been oughly 26 people pe km2 – signi ican ly highe , bu no ex eme. Howe e , his is p obably an o e es ima e, e en i we accep ha he e could ha e been a hund ed housand Khoesan in he egion. The eason is he Khoesan p ac ised anshumance, a a ie y o pas o alism in which li es ock a e mo ed seasonally be ween summe and win e pas u es, which mean ha he Khoesan mo ed in an a ea signi ican ly la ge han he sou h-wes e n Cape. Despi e he lack o eliable es ima es i seems sa e o conclude ha a e age Khoesan popula ion densi ies a he Cape emained low h oughou he eigh een h cen u y. Wha caused he g ow h o he se le popula ion? Se e al Cape his o ians ha e acknowledged ha , unlike wha happened in o he se le colonies ac oss he wo ld, immig a ion o he Cape did no play a decisi e ole in he expansion o he on ie ( an de Me we [1938] 1995; Guelke 1974; Ross 1993). In 1707, he VOC a he Cape decided no o u he encou age immig a ion, a policy ha was ein o ced in he 1717 decision ha he u u e o he Cape economy would ely on sla e labou and no Eu opean immig an labou . The chie cause o he popula ion g ow h was he high e ili y le els among he se le popula ion. Cillie s and Ma io i (2019) p o ide he i s sys ema ic and longi udinal analysis o e ili y le els in he Cape Colony. They ind high and s able e ili y a es in he colony h oughou he eigh een h cen u y ha con inue in he i s hal o he nine een h cen u y. The a e age se le woman had se en child en. Cillie s and Ma io i ound no signi ican di e ences be ween egions. Fe ili y le els we e as high in u ban Cape Town as in he u al on ie egions. Thei es ima es a e on a pa wi h indings om o he se le socie ies, such as Aus alia and he Uni ed S a es (Binion 2001 and Moyle 2017). The high e ili y a es ensu ed a s eady g ow h o he se le popula ion a he Cape. One o he ew schola s o ques ion his in e p e a ion is Shell (2005). He a gues ha i he expansion was due o na u al popula ion g ow h, hen we should expec o ind an equal sex a io among he second o hi d gene a ion o he se le popula ion. Using he annual ax censuses, he inds ha om 52 C ea ing he Cape Colony he mid-se en een h cen u y o he end o he eigh een h cen u y he se le popula ion ne e had an equal sex a io; males always p edomina ed. This leads him o a gue ha immig a ion mus ha e played an impo an ole in he expansion o he bo de s. Un o una ely, he e a e no eliable es ima es o immig a ion o he Cape. Shell he e o e uses he numbe o applica ions o he s a us o ye bu ghe ( ee ci izen) and o loan a ms (leaseholds) as p oxies. He assumes ha on i s a i ing a he Cape he immig an s o e ed hei se ices o he Company. Wages we e low, bu hey enabled an immig an o buy ime and become amilia wi h he egion be o e applying o s a us as a ee bu ghe and es ablishing his own a m. The p oblem is ha while Shell inds an inc ease in he numbe o bo h applica ions o become a ee bu ghe and applica ions o a loan a m, he numbe s emain e y low o he yea s co e ed (1738–49). The highes numbe is in 1749 when wen y- i e applica ions o ee ci izen s a us we e submi ed. Fu he , Shell igno es he s eady decline in he sex a io. By he mid-se en een h cen u y he a io was six hund ed males o one hund ed emales, and a he end o he eigh een h cen u y i had d opped o one-hund ed- i y males o one hund ed emales. In line wi h Ross (1993) we he e o e conclude ha while Eu opeans con inued o mig a e o he Cape h oughou he eigh een h cen u y, he numbe s mus ha e been ai ly small and no su icien o explain he g ow h o he se le popula ion. The g ow h o he se le popula ion – independen ly o i s cause – is used o explain why se le s mo ed a he in o he in e io . Guelke, o example (1974: 25), concludes ha ‘ he on ie mo emen was basically demog aphic in cha ac e ’, while Ross (1993:1) says ha ‘a he simples le el, he eason o he expansion o whi e se lemen is qui e e iden . I lies in he immense popula ion inc ease o he A ikane s [ he whi e se le s]’ (quo ed in Shell 2005: 2). Schola s disag ee on whe he he mo e in o he in e io was caused by push o pull ac o s. Guelke (1974) a gues ha mig a ion was a las eso o se le s who could no longe access a su icien amoun o land in he closing o closed on ie s. O he s, like Neuma k (1957), a gue ins ead o pull ac o s: se le s ealized he economic po en ial in mo ing owa ds a new on ie and es ablishing p o i able ca le a ms. The wo iews, as New on-King (1999) ecognizes, a e no mu ually exclusi e. Some may ha e mo ed as a su i al s a egy, while o he s may ha e been d awn by he oppo uni ies o se ing up p o i able ag icul u al businesses. Es ima es by Cillie s and G een (2018), o example, e eal signi ican weal h di e ences in he on ie dis ic o G aa -Reine : a small g oup o a luen a me s and a majo i y who seem o ha e li ed om hand o mou h. Fac o Endowmen s, Ins i u ions, F on ie Expansion 53 Independen ly o he mo i es, bo h iews poin o land sca ci y o explain se le mo emen s in o he in e io . The s and o he li e a u e ha emphasizes pull ac o s assumes ha he e was insu icien land le o es ablish new a ms in he sou h-wes e n Cape, lea ing he se le s wi h no choice bu o y hei luck a he inland. Those who emphasize push ac o s likewise assume ha land had become ai ly sca ce – lea ing enough o a ms bu no enough o p o i able ag icul u al businesses. I do no ques ion ha sca ci y o land ma e ed o he expansion o he on ie s. In con adic ion o p e ious li e a u e, howe e , I a gue ha he expansion canno be explained by ac o endowmen s alone bu needs o be combined wi h an ins i u ional pe spec i e. People mo ed in o he in e io be o e land became sca ce in he physical sense o he e m, because he way ha land igh s we e egula ed mean ha he a ms became widely dispe sed and his ein o ced he expansion o he on ie . Fac o endowmen s, ins i u ions and he closing o he sou h-wes e n on ie A he u n o he se en een h cen u y, when se le s we e beginning o mo e beyond he bo de s o he sou h-wes e n Cape o es ablish a ms on hei own ini ia i e, all he a able land alloca ed in he sou h-wes e n Cape was in he o m o eehold a ms, i.e. p i a e p ope y. The size o he eehold a ms was smalle han he plan a ions in he ‘New Wo ld’, a e aging om 35 o 80 hec a es (Dye and La C oix 2020). Guelke (1979) es ima es ha he o al size o all eehold a ms es ablished in he sou h-wes e n Cape om 1657 o 1717 equalled 3 pe cen o he o al land a ea alloca ed o a me s (see Table 3.1). Table 3.1 F eehold land g an s, 1657–1717. Dis ic Numbe o g an s A ea o g an s (km²) To al a ea se led (km²) Cape 107 45.6 1,300 S ellenbosch 108 53.3 1,300 D akens ein 189 94.0 3,900 To al 404 192.9 6,500 Sou ce: Guelke (1979: 49). 54 C ea ing he Cape Colony Fo una ely, as he a ms in he sou h-wes e n Cape we e clea ly dema ca ed he e a e a chi al sou ces ha help us es ablish hei loca ion. Using in o ma ion om he Deeds eco ds in he Cape Town a chi es, Guelke, Bonne and Maisoneu e (1987) p oduced a map which shows he loca ion o he eehold a ms and yea o es ablishmen . In a pilo s udy, Fou ie and G een (2018) digi ized he map and compa ed he loca ion o a ms wi h soil sui abili y, access o wa e and amoun and annual a iance o ain all. Unsu p isingly, hey ound ha hose who we e i s o es ablish a ms chose a eas mo e sui able o a ming han he la ecome s. By 1717, a me s had al eady gained con ol o e many o he wa e esou ces, as can be seen in Map 3.3, which shows a m bounda ies (bu omi s he Cape Peninsula). La ge pa s o he i e s and pe ennial s eams in he S ellenbosch, D akens ein and F anschhoek a eas we e al eady enclosed by a mland. Al hough i was becoming mo e di icul o a p ospec i e a me o ind land wi h a pe ennial s eam, i would s ill ha e been possible o es ablish a eehold a m, wi h access o less eliable wa e esou ces bu in an a ea wi h mo e eliable ain all and a clima e sui able o a able a ming. Fa he no h, in he Land an Wa e en ( oday Tulbagh) and he Swa land, i was s ill possible o loca e a ms wi h secu e access o wa e . Fa me s ne e heless wen e en a he in o he eas and he no h. To unde s and why hey did his we need o ake a close look a p ope y igh s and land con ol in he sou h-wes e n Cape. I we conside only he land unde eehold enu e, we unde es ima e he o al amoun o land unde se le con ol as his neglec s he impo ance o pas o al a ming in ha pa o he colony. Including he amoun o land alloca ed o g azing indica es ha land was indeed becoming a sca ce esou ce, as does he way g azing land was egula ed di e en ly om a able land, which caused ins i u ionally induced land sca ci y. I is common o dis inguish be ween he sou h-wes e n Cape and wha would become he no he n and eas e n on ie s o he colony in ega d o ag icul u al ac i i ies. I is claimed ha he eehold a me s in he sou h-wes e n Cape engaged mainly in whea and wine a ming, while he on ie se le s engaged in pas o al a ming. While his may ha e been ue in he la e-eigh een h cen u y Cape, i was no an accu a e desc ip ion o he economic ac i i ies o he egion du ing he i s eigh y yea s o se lemen . Du ing his pe iod pas o al a ming was a key indus y in he sou h-wes e n Cape. Table 3.2 shows he es ima ed a e age head o ca le and sheep a he Cape. I we ocus only on he pe iod 1663 o 1702, when we can be ce ain ha all he a ms in he sou h-wes e n Cape we e eehold, we see a signi ican inc ease in li es ock holdings. In 1662, an Map 3.3 Bounda ies o eehold a ms a Cape Colony, 1717 (Cape Peninsula excluded). 56 C ea ing he Cape Colony a e age se le - a me had abou 12 head o ca le and 45 sheep. Fo y yea s la e , his had inc eased o 37 head o ca le and 201 sheep. This can be compa ed o he a e age ou pu o c op p oduc ion pe a m (Table 3.3). Ou pu luc ua ed signi ican ly, p obably as an e ec o changing wea he condi ions, bu shows no clea upwa d end. Tha s ock a ming played an impo an ole in he ea ly yea s o se le ag icul u e is no su p ising. I equi ed less in es men in labou , which was sca ce (see Chap e 4), and he closeness o Cape Town made Table 3.2 Mean se le li es ock holdings a he Cape Colony, 1663–1773 (selec ed yea s). Yea A e age no. o ca le pe Eu opean a m A e age no. o sheep pe Eu opean a m 1663 11.70 44.74 1670 8.57 134.09 1678 21.40 122.98 1682 15.83 124.73 1685 26.07 170.60 1688 20.34 239.26 1692 22.77 213.00 1695 23.71 145.03 1700 35.08 192.73 1702 37.01 201.20 1712 42.76 281.54 1719 39.09 162.28 1731 42.32 218.54 1738 49.35 221.65 1741 41.35 193.04 1752 29.24 158.84 1757 27.81 150.47 1762 22.87 135.94 1773 23.71 172.29 Sou ce: Fou ie and on Fin el (2010b: 21). Fac o Endowmen s, Ins i u ions, F on ie Expansion 57 Table 3.3 Mean se le p oduc ion o wine, whea , ye and ba ley in he Cape Colony, 1663–1773 (selec ed yea s). Yea Wine (league ) Whea (muids) Rye (muids) Ba ley (muids) 1663 0 0 0 0 1670 0 0 0 0 1678 0 11.05 1.86 3.02 1682 0 21.28 3.20 12.49 1685 0 18.98 3.19 3.73 1688 0 9.57 2.85 1.69 1692 0 25.89 5.30 7.70 1695 0.56 12.06 1.78 2.16 1700 4.19 14.25 9.15 8.44 1702 3.11 11.54 5.18 4.34 1712 2.65 30.86 1.75 16.87 1719 2.76 29.27 0.61 15.43 1731 2.87 29.28 3.10 18.57 1738 1.44 23.06 1.91 19.58 1741 1.36 38.18 2.70 20.55 1752 2.90 21.66 1.33 16.26 1757 3.32 13.17 0.40 11.97 1762 2.50 18.85 0.48 13.62 1773 4.62 20.74 0.05 9.61 Sou ce: Fou ie and on Fin el (2010b: 20). i easonably easy and cheap o anspo li es ock o he ma ke he e ( an de Me we [1938] 1995). I was in he middle o he eigh een h cen u y ha se le - a me s in he sou h-wes e n Cape began inc easingly u ning o a able a ming. Be ween 1701 and 1740, he o al numbe o ca le in S ellenbosch and D akens ein, he wo main a able a ming dis ic s, inc eased om 5,500 o 24,090 head, an inc ease o almos 400 pe cen ( an Duin and Ross 1987: appendix 9). As la e as 1748, 58 C ea ing he Cape Colony 40 pe cen o all a ms in S ellenbosch, o example, we e pu ely pas o al (Ross 1993). Bu as eehold igh s did no p o ide su icien g azing land, he a me s g azed hei ca le on he common land be ween he a ms and occasionally also beyond hei bo de s. In o de o main ain con ol o he si ua ion, Go e no Simon an de S el used bo h s icks and ca o s. Al eady by 1687 he had designa ed common g azing land in S ellenbosch. A he same ime, he o bade se le s o mo e li es ock ac oss dis ic bounda ies. The se le s we e no e en allowed o keep each o he ’s li es ock among hei own he ds o paymen ( an de Me we [1938] 1995). Fa me s, howe e , con inued o mo e beyond he on ie . The VOC u ged he se le s o keep hei li es ock no a he away han one day’s jou ney om hei eehold p ope ies. Bu his exho a ion had li le e ec , leading he adminis a ion in 1703 o e e se i s policy o keeping ee g azing land and equi ing a me s o apply o g azing pe mi s. These would be issued o ee and would gi e a me s access o g azing land wi hin a speci ic geog aphical a ea. Each g azing pos had o be a leas one hou ’s walk om he nex g azing pos , which in p ac ice mean ha a pe mi holde had an exclusi e igh o a leas 2,420 hec a es o g azing land. O icially, a a me had he igh o only one licensed pos , bu in p ac ice mos a me s mo ed be ween se e al ( an de Me we [1938] 1995). We lack in o ma ion on he numbe o g azing igh s alloca ed pe a m. Ra he conse a i ely, we migh assume ha each a m had one g azing pos . On his basis, we can e ise he amoun o land alloca ed o a ming in he sou h-wes e n Cape. Table 3.4 shows he e ised es ima es. When g azing land is included, and i is assumed ha each a me had one g azing pos , he o al a ea alloca ed o a ming inc eases, expanding Table 3.4 Re ised es ima es o o al land alloca ed o a ming, 1657–1717. Dis ic Numbe o g an s A ea o g an s (km²) Es ima ed o al amoun o land alloca ed o a ming¹ To al a ea se led (km²) Cape 107 45.6 450 1,300 S ellenbosch 108 53.3 491.6 1,300 D akens ein 189 94.0 714.3 3,900 To al 404 192.9 1,655.9 6,500 Sou ce: Guelke (1979: 49). Assuming on a e age one ading pos pe a m.¹ Fac o Endowmen s, Ins i u ions, F on ie Expansion 59 well beyond he bo de s o he dis ic s. I he e o e comes as no su p ise ha Go e no an de S el s opped u he se lemen in S ellenbosch as ea ly as 1687. The alloca ed g azing pos s had by his ime pushed a me s beyond he dis ic bounda ies. The expansion o pas o al a ming in he sou h-wes e n Cape quickly made land a sca ce esou ce. Al hough we lack su icien in o ma ion, i seems likely ha he sys em o g azing pos s a ou ed ea ly a i als. The way g azing land was egula ed made u he se lemen di icul . G azing land was nei he dema ca ed no su eyed. The loca ion o he g azing pos was desc ibed only in ague e ms. Van de Me we men ions a licence ha ga e he a me he igh o g aze his li es ock in he ‘dunes behind he g aauwe heu el [bu ial mounds]’ ([1938] 1995: 55). The only egula ion imposed, as shown abo e, was ha a a me could no loca e a g azing pos nea e han a leas one hou ’s walking dis ance om he nex nea es ading pos . In p ac ice, his sys em a ou ed he se le s who a i ed ea ly. Two a me s who had al eady se led could posi ion hei ading pos s so as o p e en a hi d a me om accessing pas u eland in he a ea. By loca ing hei g azing pos s jus less han wo hou s’ walk away om each o he , hey made su e ha he e could be no ading pos es ablished be ween he wo a ms. This ga e he wo se led a me s access o mo e pas u eland han hey we e necessa ily en i led o, and a he same ime p e en ed compe i ion om o he a me s who migh wan o se le in he a ea. This sys em, oge he wi h he abili y o con ol wa e esou ces, enabled he ea ly a i als o p e en u he se lemen e en i a able land had no ye become sca ce (Guelke 1984). P ope y igh s and he expansion o he on ie As land became sca ce in physical and ins i u ional e ms in he sou h-wes e n Cape, se le s mo ed a he no h and eas wa ds. Two hings a e s iking abou his mobili y: i con inued h oughou he eigh een h cen u y, and i was ma ked by a change owa ds mo e loosely de ined and p esumably mo e insecu e p ope y igh s. Some, like an de Me we ([1938] 1995), ha e a gued ha he con inual mo emen was pa o he cul u e o he on ie a me s, who we e always on he mo e in sea ch o be e land and ended o wan o keep he go e nmen a a m’s leng h. Howe e , he limi ed da a a ailable do no suppo his claim. Mos o he se le - a me s did no mo e once hey had es ablished a a m. Guelke (1985), o example, no es ha an e alua ion o he loan a ms in 1731 ound ha he a e age on ie a m had been occupied o wen y yea s 66 C ea ing he Cape Colony inspec ion and ound ha he dis ance be ween he homes eads was a enough, bu ha ‘ he poo condi ions o he pas u e on [Rossouw’s land] consis ing only o b ambles and bushes, and wi hou any sui able capaci y o g azing nea by he a m in ques ion, mus be conside ed’ ( an de Me we [1938] 1995: 66). Unde hese condi ions, whe e he ules o he game we e unce ain and he sys em a ou ed hose al eady se led, new a i als o en decided o loca e hei a ms a he away han he egula ed one hou ’s walk. This was a s a egy o a oiding con lic s and he isk o losing he igh o he land. In 1806, he a el w i e Ba ow epo ed ha ‘ he s akes a e so placed ha , on an a e age h oughou he colony, he a ms a e wice he dis ance, and consequen ly con ain ou imes he quan i y o land allowed by he Go e nmen ’ (quo ed in an de Me we [1938] 1995: 97). In ha way, as al eady no ed, he loan- a m sys em, wi h i s unclea bounda ies and he e e -p esen isk o con lic be ween se le s o e he loca ion o landed p ope y, encou aged he dispe sal o he a ms and consequen ly he expansion o he bo de s. This does no explain why he se le s accep ed weak p ope y igh s, howe e . Unlike Dye and La C oix (2020), I a gue ha he con inual h ea o clashes wi h he indigenous people made i a ional o he se le s o op o insecu e p ope y igh s because hey educed he cos o such con lic s. Buying land was a signi ican ixed in es men ha a me s wan ed o a oid as long as he e was a isk ha hey may ha e o abandon hei a m. Dye and La C oix a e co ec in a guing ha he powe s o he indigenous people declined o e ime. In his, hey ollow a long adi ion in Cape his o iog aphy. Ma ais, o example, a gued ha ‘ om he yea 1689 onwa ds he e was no esis ance on he pa o he Ho en o s o he Eu opean ad ance’ and ha ‘unlike he Ban u, hey we e su p isingly eady o ba e away hei ca le in exchange o coppe , beads, obacco and … b andy and a ack’ (1957: 6–7). Van de Me we ([1938] 1995) a i ed a a simila conclusion, main aining ha i was only in 1770, when he Eu opean se le s i s clashed wi h Xhosa on he eas e n on ie , ha hey began o encoun e obs acles o expansion in he o m o esis ance by indigenous people. These in e p e a ions, which la gely ea he Khoesan as passi e ic ims, ha e been ques ioned by schola s who ha e con incingly shown ha Khoesan esis ance was a mo e common han ini ially belie ed. As Ma ks a gued, ‘The pa e n o iolence and coun e - iolence, aid and coun e - aid, punc ua ed e e y decade o he eigh een h cen u y’ (1972: 70). To ame i di e en ly, social and economic con lic s did no disappea wi h he closing o he on ie ; hey ook di e en o ms. The VOC lacked he capaci y o paci y he Khoesan. Fac o Endowmen s, Ins i u ions, F on ie Expansion 67 Asla e as 1774, Ma inus Ad iaan Be g, he land os o he S ellenbosch dis ic , a o me on ie a ea ha had been closed o nea ly six y yea s, ins uc ed Commandan Oppe man o educe he indigenous people o ‘a pe manen s a e o peace and quie , o o he wise, in case o necessi y, en i ely des oy hem’ (quo ed in New on-King 1999: 105). The con lic s we e no a he le el o o ganized esis ance. Ra he he clashes occu ed be ween indi idual Khoesan and se le s abou access o g azing land, wo king condi ions and so on. Khoesan a m labou e s commonly played a key ole in eigh een h-cen u y up isings agains he colonial in ude s and could swi ch du ing hei li e ime om labou e o ebel (Malhe be 1978; New on-King 1981). A he on ie s, o ganized and wa -like con lic s we e mo e common and some imes o ced he Eu opeans o abandon hei a ms empo a ily o pe manen ly. In 1715, o example, a g oup o Khoesan li ing in he moun ains a ound Land an Wa e en a acked Eu opean a me s in an a emp o eplenish hei he ds. They managed o lea e wi h wo hund ed ca le and se e al housand sheep (New on-King 1999). In he 1730s, a e a decade and a hal o se le expansion in o he Oli an s Ri e alley in he no hwes o he colony, majo con lic s a ose be ween he Khoesan and he colonial in ude s. In he ea ly yea s o se lemen in he a ea, he social ela ionship be ween he Khoesan and he se le s had been luid and ela i ely peace ul. The Khoesan could con inue as pas o alis s alongside he Eu opeans. Bu as he yea s wen by and mo e se le s a i ed, he ela ionship became inc easingly hie a chical and con lic s o e land e up ed. This led o o ganized esis ance by he Khoesan. A leas en Eu opean a ms we e a acked and ano he o y-eigh we e abandoned in he la e 1730s. Ano he inciden occu ed in he 1770s, when se le s had eached he Sneeube g, in wha en yea s la e would o m he no he n pa o G aa - Reine dis ic . They we e a acked by he Khoesan, and many had o abandon hei a ms and mo e a he down sou h (Cillie s, G een and Ross 2021). Meanwhile, less o ganized a acks occu ed in pa allel on he open on ie . In 1719, o example, se en hund ed ca le we e s olen om he ca le s a ion a Ri ie sonde end and in he 1750s se e al a acks we e eco ded as he se le s wen beyond he bo de s o Swellendam (Ma ks 1972; Penn 2005). Rumou s o planned Khoesan a acks commonly sp ead among he se le s. In 1712, i was epo ed ha no ewe han i e housand o he G ea Namaqua we e mo ing owa ds he Oli an s Ri e a he no hwes on ie . The VOC ound no signs o his and he local Khoesan claimed ne e o ha e hea d o any planned a ack (Penn 2005). Such ecu en umou s show ha he se le - a me s we e awa e and a aid o he isk o being a acked. 68 C ea ing he Cape Colony The se le s we e a om innocen . No only did hey gi e hemsel es he igh o a m on land ha was al eady being used by indigenous pas o al a me s, hey also egula ly s ole ca le om he Khoesan and ea ed he Khoesan labou e s wi h a signi ican deg ee o b u ali y (New on-King 1999). The eason I ocus on Khoesan esis ance and a acks is ha i helps us unde s and why Eu opeans a he on ie p e e ed o access land unde a loan- a m sys em a he han unde a sys em o p i a e eehold p ope y igh s. The VOC’s ailed a emp o e o m he loan- a m sys em cons i u es e idence o his. In 1732, he au ho i ies in oduced a qui en sys em. The plan was o u he p omo e se le a ming by allowing se le s o ex end he size o hei loan a ms by leasing land annexed o hei own a m on a i een-yea con ac . The en hey paid would di e acco ding o he quali y o he land. This oppo uni y was no me wi h any en husiasm by he se le - a me s. Only hi y- i e g an s we e issued be o e he sys em was abandoned (Fishe 1984). The pa e n o a acks and coun e -a acks made i necessa y o he se le s o keep he cos o ixed capi al in es men as low as possible. Whe e he e is a cons an isk o losing landed p ope y, a pe son na u ally wishes o educe he ixed cos o acqui ing ha p ope y. I hus a gue, con a y o he claims made by Dye and La C oix (2020), ha i was no he absence o esis ance om he Khoesan ha made he loan- a m sys em a ac i e o he se le s; a he , i was he h ea posed by he indigenous people ha made he sys em so a ac i e ha i las ed o mo e han a hund ed yea s. The eason he VOC ailed o make he one-yea enewable loan- a m sys em mo e secu e was ha as long as he e was a high isk o con lic wi h he indigenous people, he a me s emained eluc an o sign any long- e m leases o o buy land ou igh , as hey would in a sys em o eehold p ope y. Expansion and inhe i ance As I explained in Chap e 1, he sys em o inhe i ance ma e ed o bo h he concen a ion o landed weal h and he dispe sal o a ms. In Chap e 5, I ocus on he o me . In his sec ion, I show how he sys em o pa ible inhe i ance u he acili a ed he expansion o he bo de . Acco ding o he Roman-Du ch law ha was used a he Cape, when a husband o wi e died, he su i ing spouse ecei ed one-hal o wo- hi ds o he p ope y and he es was equally di ided among he child en. The law ga e each hei he igh o a ‘legi ima e po ion’ o he es a e, which p e en ed p ope y owne s om disinhe i ing Fac o Endowmen s, Ins i u ions, F on ie Expansion 69 hei s (Dooling 2005). The implica ion o such a sys em, as explained by Ross (1993: 141), was ha p ope y was di ided and edi ided a egula in e als … This en ailed ei he ha he a m as a unning conce n would be sold, and he p oceeds di ided among he hei s, o ha a alua ion would be made o he es a e wi h he hei s ecei ing hei po ions in cash. In he la e case, he su i o would in all p obabili y be o ced o ake ou a loan o pay o his o he child en and s epchild en. Subdi ision mean ha he a ms became smalle o e ime. To a oid his, one o he su i o s could buy he en i e a m om he o he hei s. This equi ed access o c edi , which was di icul o ob ain wi hou esou ces o use as colla e al, such as a m equipmen o sla es. This leads Dooling (2005) o conclude ha he p ac ice o se le s buying each o he ou was uncommon, wi h he excep ion o he mos a luen , who owned mo e han one a m. In mos cases, he p ope y o he deceased was ins ead auc ioned. E en hei s o ai ly a luen owne s o eehold a ms would auc ion he whole a m ins ead o keeping i in he amily. Wolwedans, one o he bigges whea a ms in he S ellenbosch dis ic , changed hands nine imes be ween 1751 and 1835. A neighbou ing a m aced simila u no e , wi h i e ans e s be ween 1757 and 1815, and he Hoo nbosch a ms changed hands six een imes be ween 1751 o 1835, oughly e e y i h yea (Dooling 2005). A he on ie s, whe e he loan- a m sys em domina ed, subdi ision o land was in p ac ice impossible. The loan a m was based on wo c i e ia: a ixed annual en s ipula ed in a lease con ac be ween he VOC and an indi idual se le , and, a leas in heo y, a ixed size o he a m (a minimum o one hou ’s walk be ween he homes eads o wo a ms). These c i e ia in e ec p e en ed he subdi ision o land. Fo people a he on ie , he e we e wo al e na i es: ei he o ake o e a con ac and buy o he s ou , o o auc ion o he a m equipmen , including sla es. The la e was he bes op ion o mos people i hey lacked su icien capi al o buy one ano he ou . Independen ly o whe he he hei s we e bough ou o he whole a m was auc ioned, he capi al ecei ed pe child would in any case be insu icien o buy a new a m o make signi ican capi al in es men s, such as pu chasing sla es (see Chap e 4 in his book, as well as du Plessis, Jansen and on Fin el2013). Fo mos people, bo h hose who g ew up on eehold a ms and hose who g ew up on loan a ms, mo ing and es ablishing hei own a ms was he main op ion le . As explained abo e, eehold a ms we e much mo e aluable han 70 C ea ing he Cape Colony he loan a ms, so i seems easonable ha only a e y ew o he weal hies could e-es ablish hemsel es by buying a eehold a m. Buying access o a loan a m was easie , bu s ill equi ed access o c edi , which was sca ce a he ime. Ins ead, he younge gene a ion mig a ed a he in o he in e io o es ablish hei own a ms. P ope y igh s and ag icul u al in es men s In ‘new’ ins i u ional economics and economic his o y, a well-de ined and secu e p ope y igh s sys em (i.e. p i a e p ope y igh s) is o en iden i ied as a key ins i u ion o enhancing long- e m economic de elopmen (No h and Thomas 1973; Fede and Feeny 1991; de So o 2000; Acemoglu, Au o and Lyle 2004). The a gumen is ha secu e p ope y igh s c ea e incen i es o p oduc i e in es men and acili a e capi al accumula ion by educing he isk o ge ing he land con isca ed. Wi h his in mind, one would expec ag icul u al in es men s o be signi ican ly lowe on he loan a ms han on he eehold a ms. The ex en o which eehold a ms a he Cape p o ided a mo e secu e o m o p ope y igh s, and he eby p omo ed ag icul u al in es men s mo e han loan a ms did, is ul ima ely an empi ical ques ion. Schola s like Guelke (1974), New on-King (1999) and Penn (2005) ha e a gued ha in p ac ice he e was li le di e ence be ween he eehold and he loan- a m sys em. Bo h p o ided secu e p ope y igh s. To quo e Penn, ‘in p ac ice loan a ms became he p i a e and i ually inalienable p ope y o hei owne s’ (2005: 43). Al hough he lessee did no own he land, he o she had he igh s o all imp o emen s made on i , including buildings and ag icul u al equipmen . Con isca ion was ex emely a e. E en in cases in which a me s did no ollow he s ipula ions and enew hei licences on an annual basis, he VOC did no ake ac ion o o ce he a me o lea e he land ( an de Me we [1938] 1995). Smi h (1974: 9) no es ha ‘while he Company did no legally su ende i s igh o ake back a loan a m, his was so seldom done ha he Boe s came o accep ha he a ms we e hei own, un il hey decided o lea e; e en he ailu e o pay ecogni ion ees did no esul in he e oca ion o a pe mi ’. Despi e no being exposed o any eal h ea o con isca ion on he pa o he VOC, loan a me s a he on ie in es ed less in hei a ms han a me s in he sou h-wes e n Cape did. Guelke (1989: 86) ound ha he alue o houses and o he ixed imp o emen s a e aged less han 300 o 500 ixdolla s on loan a ms compa ed o 1,700 o 3,300 ixdolla s on eehold a ms. How can we explain Fac o Endowmen s, Ins i u ions, F on ie Expansion 71 his gap? Swanepoel (2017) in es iga ed whe he his di e ence was because he loan- a m p ope y igh s sys em p e en ed hese in es men s o because loan a me s we e on a e age less weal hy and hence spen less on hem. Using p oba e in en o ies, she iden i ied ypes o a ms, hei alue and he a me s’ le els o deb . The epo s o alue we e inconsis en and he e o e di icul o use, so she used le el o deb as a p oxy o weal h, assuming ha he e was a posi i e co ela ion be ween weal h and deb . The e a e good easons o his assump ion. Se e al schola s ha e ound ha weal hie people in p e-indus ial socie ies we e also mo e indeb ed (e.g. Muld ew 2012; Ogil ie, Küpke and Maeg ai h 2012). Swanepoel ound ha eehold a me s we e on a e age mo e indeb ed, which would suppo he iew ha i was he de ju e igh s ha ma e ed. The a gumen is ha a eehold a me could – unlike a holde o a loan a m – use land as colla e al o access c edi ha could be used o a m in es men s. Howe e , as Swanepoel igh ly ecognized, he e e se may also be ue: eehold a me s may ha e had mo e deb s because hey needed c edi o buy he a m. I could also be, as a gued by Guelke and Malhe be (1989), ha i he eehold a me s we e weal hie , i was no because o he p ope y igh s sys em, bu because o hei close p oximi y o Cape Town. The da a did no allow Swanepoel o in es iga e whe he dis ance o Cape Town was he main ac o ha de e mined di e ences in weal h. Howe e , by using he deb le el o he i s bo n, she was able o con ol he ex en o which i was he p ope y igh s sys em ha de e mined le els o weal h. The i s bo n we e on a e age weal hie and when hey we e included in he eg ession he co ela ion be ween eehold a ms and le els o deb s became insigni ican . Al hough we canno obse e his di ec ly, Swanepoel’s s udy does seem o sugges ha he di e ence in p ope y igh s sys ems did no ha e any majo e ec s on deb s. This is some hing o a puzzle because o he sca e ed e idence ha loan a me s in es ed less in ixed imp o emen s on hei a ms. The puzzle can be sol ed i we shi ou ocus om he p ope y igh s sys ems pe se and look ins ead a he di e ences be ween an open and a closed on ie . I a gued ea lie ha a me s p e e ed o lease land on an annual basis because hey ea ed losing hei p ope y in he e en o con lic s wi h he indigenous popula ions. Fo he same eason, in es men s in ixed equipmen like buildings we e limi ed as long as he on ie was open and he economic and poli ical si ua ion unce ain. C edi was ins ead used o o he ypes o ag icul u e- ela ed in es men s and o smoo hing ou consump ion pa e ns among he pas o al a me s a he on ie . Long dis ances o he ma ke mean ha weal hie a me s in es ed 72 C ea ing he Cape Colony mo e in mobile asse s like ca s, wagons and ho ses, as Cillie s and G een (2018) a gue was he case in G aa -Reine . To conclude, he e is li le e idence ha loan a ms – he less secu e p ope y igh s sys em – impeded ag icul u al in es men , p obably because he isk o land being con isca ed by he VOC was e y low. The di e ence in he pa e n o in es men s p obably e lec ed he ac ha mos o he loan a ms we e loca ed on he open on ie , whe e a me s we e a isk o losing p ope y in imes o con lic wi h he Khoesan, a he han di e ences in he p ope y igh s sys ems. Conclusion As wi h se le socie ies in o he pa s o he wo ld, he his o y o he Cape Colony was o a la ge ex en he his o y o expanding on ie s. Schola s ha e a gued ha he expansion o he bo de s o se le colonies was an e ec o he con inuous immig a ion o Eu opeans o he colonies. The g ow h o he popula ion made land sca ce, o cing a i ing se le s o mo e pas he exis ing on ie in sea ch o esh land. The in low o Eu opeans and he abili y o expand he land on ie enabled se le economies o pu sue a pa h o ex ensi e economic g ow h. Cape his o ians pain a simila pic u e o he i s 150 yea s o he colony, wi h one modi ica ion: ha he expansion o he on ie was caused no by immig a ion bu by an ex ensi e g ow h in he se le popula ion al eady p esen in he colony. While I do no ques ion he impo ance o popula ion g ow h o unde s anding he expansion o he on ie , I a gue in his chap e ha he explana ion needs o be modi ied. Fac o endowmen s and geog aphy played an impo an ole in he dispe sion o se le - a me s, bu o ully g asp he dynamics we also ha e o ake he e olu ion o p ope y igh s sys ems in o accoun . I is emp ing o see his expansion as a planned p ocess whe eby he VOC sough o enla ge he e i o y i con olled in o de o enhance i s p o i s. Such a claim lacks his o ical e idence. As I a gue in Chap e 6, he VOC ne e managed o make he Cape Colony a p o i able en e p ise. On he con a y, he Company ha dly managed o make ends mee . So how would we hen explain he expansion o he colonial on ie ? In ou case, i was la gely a p ocess induced by he se le s, bu in which he ac ions o he Khoesan played a pa . A majo eason o he con inued expansion was he luid landed p ope y igh s sys em. Wi h he excep ion o he sou h-wes e n Cape, he p ope y igh s we e on pape insecu e and in p ac ice luid. Fa me s leased land on an annual basis wi h a Fac o Endowmen s, Ins i u ions, F on ie Expansion 73 isk – in heo y – ha he VOC would wi hd aw he con ac . In p ac ice, his ne e happened, and one should no exagge a e he insecu i y o he leasehold sys em. Con ac s we e ha dly e e cancelled. Wha ma e ed was he luidi y. Land was no su eyed o clea ly dema ca ed. This made he bo de s be ween a ms uzzy. In o de o a oid land dispu es, se le - a me s sp ead hemsel es ou o e as a eas. The VOC’s weak adminis a i e and inancial capaci y can explain why he sys em o p ope y igh s e ol ed as i did in he ini ial yea s, bu no why i pe sis ed h oughou he 150 yea s ha he VOC go e ned he colony. Al hough i emained poo , he VOC’s adminis a i e capaci y did imp o e o e ime. The pe sis ence o he insecu e and luid sys em was ins ead a consequence o he po en ial isk he se le - a me s aced in being chased away by he indigenous people. As long as he on ie was no comple ely closed, his was a eal isk. The many on ie wa s bea wi ness o his. Unde hese ci cums ances, se le - a me s wan ed o keep he ixed in es men s in land as low as possible. Buying land o ge ing you leasehold con e ed o clea ly dema ca ed eehold p ope y was cos ly. Se le - a me s we e no eady o make such in es men s as long as hey aced a isk ha hey could lose hei land. The consequence was a con inued dispe sion o se le a ms and a con inuous expansion o he colonial on ie . 74 In oduc ion Sho ly a e he VOC es ablished i s ading pos , i began o impo sla es. Be ween 1652 and 1808, abou six y- h ee housand sla es we e impo ed om Asia and elsewhe e in A ica. Nume ous schola s ha e a gued ha he use o sla es cons i u ed a majo con ibu ion o he g ow h o Eu opean se lemen a he Cape (Legassick 1980; Wo den 1985; A ms ong and Wo den 1989; Feins ein 2005; Wo den and G oenewald 2005). In o de o explain he widesp ead use o sla e y, au ho s poin a he p e ailing ac o endowmen s: he abundance o land and he sca ci y o labou (Wo den 1985; Feins ein 2005; G een 2014). Though schola s do no always e e o i explici ly, his easoning is inspi ed by he so-called Nieboe -Doma hypo hesis (Nieboe 1900; Doma 1970). This hypo hesis holds ha in cases o land abundance and an absence o ba ie s o people o access land, people will ake up a ming ins ead o selling hei labou , o cing landlo ds in need o addi ional labou o pay wages so high ha i ea s up any po en ial p o i s. Landlo ds will hus ins ead eso o he use o coe cion (sla e y) o access labou . As he ensla emen o he Khoesan was no allowed, se le - a me s ins ead became dependen upon he use o impo ed sla e labou . In his chap e , I eassess he economic ole o sla e y a he Cape and a gue ha p e ious li e a u e pa ly exagge a es he ole o sla e labou and simul aneously unde es ima es he impo ance o indigenous Khoesan labou . The p o i abili y o using sla es depended upon how i could be combined wi h o he o ms o labou , including Khoesan wage labou . Fu he , he economic ole o sla e y canno be na owly unde s ood in e ms o labou inpu . While land–labou a ios can explain he eme gence o sla e y in pa , i s pe sis ence canno ully be g asped h ough a na ow ocus on sla e y as labou . Inspi ed by ecen esea ch by Ma ins (2020b) and Ma ins and G een (2021), I make he 4 Was he Cape Colony a sla e economy? 82 C ea ing he Cape Colony da a we do ha e la gely con i ms his iew. Table 4.1 shows he dis ibu ion o sla eholding on a able and pas o al a ms in he Cape and in he S ellenbosch, D akens ein and Swellendam dis ic s om 1705 o 1800. In all dis ic s, he mean holding o adul male sla es is indeed lowe on he pas o al a ms han on he a able a ms. Table 4.1 also e eals ha he close he a ms we e, geog aphically, o Cape Town (a he ime, he only ma ke o ag icul u al p oduce in he colony), he la ge he mean size o sla eholding. Indeed, he mean holding o sla es on pas o al a ms in he Cape dis ic was la ge han o he a able a ms in D akens ein. Fa he away, a he on ie , he mean holding o sla es we e e en smalle . In G aa -Reine – a dis ic es ablished in 1786 a he eas e n on ie – he mean numbe o sla es emained a ound one pe a m o he pe iod om 1798 o 1828 (Cillie s and G een 2018). This is also in line wi h expec a ions as a ms in he comme cial sou h-wes e n Cape on a e age we e mo e p o i able han hose a he in o he in e io . I seems in ui i e o assume ha weal hie a me s we e likely o ha emo e sla es. Ye i is di icul o es ablish his empi ically, as mos s udies ha analyse weal h dis ibu ion ac oss he se le popula ion include sla es as a weal h a iable (Guelke and Shell 1983; Ross 1999; Dooling 2005; Fou ie and on Fin el 2010a, 2011; Cillie s and G een 2018). Wha can be de e mined o ce ain, as I discuss in Chap e 5, is ha he se le a ming communi y was highly s a i ied. This is also no ed in he une en dis ibu ion o sla eholdings, as shown in Table4.2. The as majo i y owned nine sla es o less, while only a iny numbe o se le s we e in possession o mo e han wen y sla es each. O e ime, he la ges sla eholde s and hose who held be ween en and nine een sla es inc eased in p opo ion o he smalles sla eowne s, which indica es ha some managed o accumula e mo e sla es. So a , i seems ha he case o Cape Colony la gely con i ms he Nieboe - Doma hypo hesis. Due o low land–labou a ios, sla e y g adually de eloped in o a iable solu ion o he p oblem o labou sho ages. Ini ially, only he weal hies could a o d o in es in sla es, bu o e ime mo e and mo e a me s became sla eholde s, al hough a a modes scale. The c ucial ques ion is: did hese in es men s make economic sense om he landholde s’ poin o iew? Gi en he ac o s men ioned by Wo den (1985) – dominance o pas o al a ming, ai ly small a ms and limi ed demand on he wo ld ma ke o Cape p oduce – i is a om ob ious ha in es men s in sla e labou we e p o i able. Was he Cape Colony a Sla e Economy? 83 Table 4.2 Pe cen age dis ibu ion o holdings o adul male sla es by dis ic , 1705–99/1800 (selec ed yea s). Cape dis ic S ellenbosch dis ic D akens ein dis ic Yea 1 o 9 10 o 19 > 20 1 o 9 10 o 19 > 20 1 o 9 10 o 19 > 20 1705 85.8 10.7 3.5 84.2 15.8 0 97.5 2.5 0 1723 61.5 25.0 13.5 77.1 12.5 10.4 94.2 4.8 1.0 1731 53.3 27.6 19.1 70.8 16.9 12.3 88.3 10.4 1.3 1741 54.3 31.1 23.6 73.2 15.5 11.3 87.6 10.0 2.4 1752 55.0 24.8 20.2 83.7 11.9 4.4 92.1 6.4 1.5 1762 56.0 31.4 12.6 76.3 15.0 8.7 90.5 9.1 0.4 1773 53.0 31.8 15.2 83.3 12.1 4.6 87.3 12.1 0.6 1797/8 70.6 20.7 8.7 76 19.5 4.5 1799/1800 31.8 41.6 26.6 Sou ce: Wo den (1985: 32). The economics o sla e labou Schola s w i ing on sla e y a he Cape ha e o en a gued ha i was he p e e ed sys em because he cos o sla es was signi ican ly lowe han he cos o employing Eu opean labou . This iew is also echoed in VOC epo s. In 1717, Fiscal Co nelis an Beaumon w o e: Each labou e ’s wage was eckoned a he a e o a soldie ’s pay – 9 pe mon h – o he would ha dly accep less e en o wo k done by sla es, add o his he cos o his boa d and u he emolumen s, and a g ea di e ence will be ound, as a sla e cos annually (e e y hing included) abou 40. Se agains his he ac ha he emolumen s o a soldie a y om 8 o 10 gulden, le alone he wages which he Company would ha e o pay. Fu he , one will no ind Eu opeans o nea ly as much use as sla es, especially in he daily menial asks; besides i is mo e i ing ha sla es a he han Eu opeans should be used. (quo ed in Schoeman 2007: 196) The la e poin – he unwillingness o Eu opeans o pe o m he ha d wo k o sla es – was epea edly b ough up. In 1699, Cap ain-Lieu enan K. J. Slo sboo 84 C ea ing he Cape Colony ema ked ha ‘no ma e how poo a [Eu opean] is, he will no accus om himsel o pe o m he wo k o he sla es, as he hinks in his way o dis inguish himsel om a sla e. Mo eo e , he ac ha hey ha e le hei coun y makes hem hink ha hey should lead an easie li e han a home’ (quo ed in Schoeman 2007: 197). A hund ed yea s la e , Fiscal W. W. an Rijne eld concluded ha sla e y had become a ‘necessa y e il, ha canno be emo ed wi hou sac i icing he Colony, and pe haps he poo sla es hemsel es ha a e in i ’ (quo ed in Schoeman 2012: 324). No e e yone ag eed ha sla e labou was cheap. Cap ain Dominique Pasques de Cha onnes, he nephew o he Go e no , concluded in he la e se en een h cen u y ha Eu opean labou would be cheape . He a gued ha an o dina y sla e would cos 80 o 150 ixdolla s and o an ensla ed he dsman he p ice would be 150 o 300 ixdolla s. To his he e we e addi ional cos s o clo hing and eeding as well as ‘ he ne e -ending sickness, acciden s, maimings, dea hs and bu ials and wha e e is s olen by his class o pe son, and he ac ha h ee sla es a e equi ed o do he wo k o wo Eu opeans’ (quo ed in Schoeman 2007: 198). A ew obse e s claimed ha sla e y made Eu opeans lazy and hinde ed he Cape’s economic de elopmen . In 1743, Ba on on Imho , a mino nobleman who had spen six een yea s in se ice o he Company, claimed ha ‘a ca pen e each ea ns om eigh o nine schellingen a day and in addi ion ecei es ood and d ink, and wi hal does no do as much as a hal - ained a isanin Eu ope’. This was, acco ding o him, a ‘bu den ha he Colony could no bea ’. I was de as a ing, especially o he de elopmen o ag icul u e in he colony. I would ha e been much be e , acco ding o Imho , i he colony in i s ea ly yea s had ied o a ac Eu opean labou e s, ins ead o impo ing sla es (Schoeman 2012: 326–7). These wo examples ep esen excep ions, howe e . Mos con empo a y wi nesses ag eed on he bene i s o using impo ed sla es o e Eu opean immig an s as a sou ce o a m labou . Tha sla es we e cheape han Eu opean labou does no mean ha i was inexpensi e o a me s o acqui e hem. Da a on sla e p ices a e impe ec . The bes sou ce a ailable a e he auc ion olls, which gi e he p ice o sla es ha we e sold upon he dea h o his o he owne . Un o una ely, he auc ion olls do no p o ide in o ma ion on he heal h and s a us o he sla e sold. They do, howe e , p o ide he p ices o all sla es sold a he a m, which enable us o iden i y he p ice ange o each yea . Table 4.3 shows he pe cen age dis ibu ion and median p ice o sla es om he auc ion olls. The p ice o sla es a ied signi ican ly ac oss ime and space. The p ice changes we e, acco ding o Wo den, closely Table 4.3 Pe cen age dis ibu ion and median p ice o adul male sla es sold a u al auc ions in ixdolla s, 1682–1795. P ice Yea To al numbe 1–99 100–99 200–99 300–99 400–99 500–99 600–99 700–99 800–99 900–99 > 1,000 1692–1714 103 29.1 62.1 8.8 0 0 0 0 0 0 0 103 1715–21 67 16.4 50.7 28.3 4.6 0 0 0 0 0 0 130 1722–9 131 38.2 48.8 12.2 0.8 0 0 0 0 0 0 105 1730–7 197 21.3 58.8 13.2 5.0 1.0 0 0.7 0 0 0 134 1738–47 151 32.4 44.4 16.5 5.9 0.8 0 0 0 0 0 120 1748–56 123 17.8 40.6 30.9 6.5 1.8 0 0 0 0 0 192 1757–64 123 17.0 52.0 17.9 4.1 6.5 1.6 0.9 0 0 0 156 1765–72 219 18.7 47.5 20.1 9.1 4.1 0.5 0 0 0 0 163 1773–9 175 13.7 35.4 28.0 14.9 4.6 3.4 0 0 0 0 195 1780–4 189 3.3 16.6 23.2 19.8 14.4 7.2 6.1 2.2 0 5.5 310 1785–90 199 6.5 19.1 19.1 23.1 14.6 4.5 7.6 1.5 1.5 2.5 305 1791–5 201 5.4 21.9 21.9 26.4 16.9 6 2.5 1.5 0.5 1.0 345 Sou ce: Wo den (1985: 74). 86 C ea ing he Cape Colony ela ed o ac o s o demand and supply. The median p ice ose g adually un il 1740 and mo e sha ply in he 1740s and 1750s due o a sho age o supply caused by inc eased sla e mo ali y as a esul o an epidemic o gas oen e i is and possibly also measles. The p ice inc ease in he 1780s was caused by in la ion oge he wi h a cu back in sla e impo s (Wo den 1985). To pu he p ices o sla es in con ex , we can compa e hem wi h he a e age incomes o he se le - a me s. We lack eliable da a on se le a m incomes, bu ough es ima es by Fou ie and on Fin el (2010a) show an a e age annual income o 192 ixdolla s in 1723 and 107 ixdolla s in 1757. This would allow he pu chase o oughly one sla e o ‘lowe quali y’. Income s a i ica ion was signi ican (see he mo e de ailed discussion o his in Chap e 5) and g oups o a me s ea ned signi ican ly abo e he a e age income. Howe e , o he b oad masses o se le - a me s, buying a sla e ep esen ed a conside able in es men . The long- e m end o sla e p ices e eals a signi ican inc ease, which has o be unde s ood in he con ex o he VOC’s me can ilis ade policies. As I discuss in Chap e 6, he VOC bough some o he wine, whea and li es ock a adminis a i ely se p ices while he es was pu chased by indi idual se le s wi h which i con ac ed. The p ice o e ed by he VOC was s able o wine while i modes ly declined o whea du ing he eigh een h cen u y. P ices paid by he con ac o s luc ua ed mo e. The p ice o whea was lowe han he VOC p ice h oughou he pe iod, while he p ice o wine was highe a e 1740. While he p ices paid by he VOC s agna ed, he p ices paid by he con ac ed ade s did no inc ease a he same a e as he p ice o sla es (see Table 4.4), sugges ing ha he e u ns o sla e labou dec eased o e ime. This begs he ques ion: why con inue o in es in sla es ha we e expensi e and whose e u ns dec eased o e ime? I seems plausible ha a me s would no ha e con inued o p ac ice ensla emen i sla es we e no a p o i able in es men . Bu i is di icul o es ablish any p ecise es ima e o he p o i abili y o sla es, o wo easons. Fi s , we do no know o su e wha ype o wo k he sla es we e doing. Some quali a i e e idence gi es us examples o he a m asks sla es we e engaged in, bu i is di icul o gene alize om hese obse a ions. Second, ou knowledge o he ole o indigenous Khoesan a m wo ke s is a he limi ed. Due o lack o quan i iable da a, schola s ha e un il ecen ly excluded hem om any assessmen o he p o i abili y o sla es. Wo den (1985) is o my knowledge he only schola who has ied o sys ema ically assess he p oduc i i y o sla e labou a he Cape on a ew Was he Cape Colony a Sla e Economy? 87 Table 4.4 A e age p ice o ag a ian p oduce a public auc ions and O icial Company p ice (in ixdolla s), 1716–1800 (selec ed yea s). P ice Whea (pe muid) Whi e wine (pe league ) Yea Auc ion VOC Auc ion VOC 1716 2.0 2.6 21 27 1720 2.2 2.6 22 27 1730 2.4 2.6 26 27 1740 2.4 2.6 28 27 1750 1.6 2.5 30 27 1760 1.5 2.5 38 27 1770 1.0 2.5 22 27 1780 2.4 2.5 38 27 1790 2.6 2.5 35 27 1800 2.8 2.5 30 27 Sou ce: Wo den (1985: 69). selec ed a ms o which con empo a y sou ces do exis . These es ima es e eal a es o e u n o be ween 5 and 11 pe cen , which equal hose ound in he Ca ibbean and in he Ame ican Sou h. Howe e , as Wo den himsel admi s, hees ima es mus be ea ed wi h a g ea deal o cau ion, o wo easons. Fi s , he ew selec ed a ms ep esen he la ge and mo e p o i able ones, p oducing on la ge scale o he ma ke , and possibly achie ing economies o scale. Second, no all cos s o p oduc ion a e always included. Fo some a ms, o example, he cos o hi ed Eu opean labou , (knech s) pe sonal expenses and main aining he household a e excluded. To deal wi h his, Wo den c ea es h ee di e en ypes o model a ms ( o a de ailed o e iew o his me hods, see Wo den 1985: chap. 6). Like he con empo a y sou ces, he model a ms e eal a a e o e u n o be ween 5and 12 pe cen . Al hough no a linea ela ionship, la ge a ms on a e age show g ea e e u ns pe sla e. The exis ence o economies o scale has been suppo ed by some schola s (Guelke and Shell 1983; Fou ie and on Fin el 2010a) and ques ioned by o he s (Williams 2013). Meanwhile, a undamen al and pa ly o e looked p oblem wi h Wo den’s calcula ions is ha while hey do include he employmen o Eu opean a m 88 C ea ing he Cape Colony labou e s (knech s), hey neglec he ole o Khoesan labou e s. In ac , Khoesan labou e s a e no included a all in he equa ion employed o he model a ms. I seems plausible o assume ha he Eu opean wo ke s we e no assigned he same wo k asks as he sla es, while he Khoesan would ha e wo ked side by side wi h he sla es. This phenomenon dis inguished he Cape Colony om he sla esocie ies o he New Wo ld (Wo den and G oenewald 2005). In a census ha includes indigenous a m labou e s in S ellenbosch and D akens ein dis ic s in 1806 i is eco ded ha indigenous labou accoun ed o abou 30 pe cen o he o al labou o ce. These we e he wo dis ic s whe e sla e-based ag icul u e was mos widesp ead. In he pas o al on ie dis ic s, he p opo ions we e g ea e . In G aa -Reine in 1798, o example, he a e age a me employed eigh Khoesan and ewe han one sla e (Cillie s and G een 2018). Khoesan labou Eu opean se le s may ha e ega ded sla e y as p e e able o Eu opean wage labou . I is less clea whe he hey ega ded sla e labou as supe io o he employmen o Khoesan labou . I seems plausible o assume ha i was ini ially di icul o ely on Khoesan labou as long as land was in abundance, bu ha his changed o e ime as he se le communi y expanded and he Khoesan los access o land and li es ock, o cing hem o seek employmen on Eu opean a ms. Gi en hei expe ience as pas o alis s and hei knowledge o he egion, hey we e commonly employed as he de s (Penn 2005). Cou eco ds also show ha he Khoesan we e o en pe manen ly a ached o he a able a ms in he sou h-wes e n pa o he colony, especially in S ellenbosch and D akens ein, wo dis ic s whe e he use o sla es is known o ha e been widesp ead. The Khoesan mus ha e li ed and wo ked alongside he sla es (Wo den 1985). This sugges s ha he e is li le e idence o wa an he exclusion o he Khoesan om ea lie quan i a i e es ima es. The VOC had been eluc an o employ Khoesan o wo k on hei pas o al and a able a ms. In he o me case, he company ea ed he ; in he la e , i was belie ed Khoesan lacked he necessa y skills o ill he land. Se le s, howe e , seem o ha e been less eluc an o employ Khoesan. Eu opean se le s who mo ed in o S ellenbosch and D akens ein dis ic s in he e y ea ly yea s o on ie expansion in gene al lacked he means o buy su icien numbe s o sla es. In 1695, he e we e 0.65 sla es o e e y Eu opean se le in S ellenbosch and only 0.27 in D akens ein. To ill he gap, he a me s ec ui ed Khoesan Was he Cape Colony a Sla e Economy? 89 labou e s. By 1688, Khoesan labou e s on a ms had eached such numbe s ha Go e no Simon an de S el compa ed hem o he in lux o mig an wo ke s om Wes phalia in o he Ne he lands (Guelke and Shell 1983). Acco ding o Elphick, mos Khoesan men in he sou h-wes e n Cape we e by he end o he se en een h cen u y ei he in se ice in Cape Town o on he Eu opean a ms (1979). The wo k o he indigenous people was o en desc ibed in app ecia i e e ms. In 1695, Johannes Willem de G e enbock e e ed o he Khoesan labou a he en husias ically: They ain oxen o use in ploughing […] and a e ound exceedingly quick a inspanning o ou spanning o guiding eams. Some o hem a e e y accomplished ide s, and ha e lea ned o b eak ho ses and mas e hem […]. They make us y bea e s, po e s, ca ie s, pos boys and cou ie s. They chop wood, mind he i e, wo k in he ki chen, p une ines, ga he g apes, o wo k he wine p ess indus iously […]. Wi hou elaxa ion hey plough, sow, and ha ow. (Elphick and Malhe be 1989: 17) In 1713, he Cape expe ienced a de as a ing smallpox epidemic. The e is no consensus as o how badly he Khoesan we e a ec ed. Wo den a gues ha almos 90 pe cen o he Khoesan popula ion died (1985), while Ross (1977) p oduces a much lowe es ima e o 30 pe cen o less. The u h may be somewhe e in he middle o hese wo es ima es. Rega dless, he epidemic led o a signi ican educ ion in he Khoesan popula ion. This in u n led o a signi ican inc ease in sla e impo s as se le s needed o eplace he Khoesan labou e s who had died in he epidemic (A ms ong 1979). This e eals ha Khoesan ea ly on played an impo an ole as a m labou e s and ha he use o sla e labou was a om always he p e e ed choice among he se le - a me s. One could specula e abou wha would happen i he Cape had no been hi by he smallpox epidemic in 1713. I is no un easonable o assume ha he impo a ion o sla e labou would ha e been signi ican ly lowe . Despi e he inc eased labou sho ages caused by he epidemic, he e a e good easons o belie e ha Khoesan labou con inued o play an impo an ole h oughou he eigh een h cen u y. Wo den (1985) assumes ha mos o he Khoesan mig a ed as a consequence o p imi i e accumula ion. Schola s like Elphick (1977), Penn (1989) and Fou ie and G een (2015), howe e , claim he con a y, ha he p ocesses o p imi i e accumula ion caused an inc ease in he supply o indigenous labou o e ime. We lack eliable es ima es o he o al size o he Khoesan popula ion li ing wi hin he bounda ies o he colony. The i s es ima es o he numbe o Khoesan on Eu opean a ms da e o 1806. Johan Fou ie and Jan Lui en an 90 C ea ing he Cape Colony Zanden (2013) ha e cons uc ed a GDP se ies o he Cape Colony based on he consump ion pa e ns o ou popula ion g oups: se le s, VOC employees, sla es and Khoesan. To accoun o he la e , hey use a no el echnique: hey assume ha he numbe o indigenous people was p opo ional o he numbe o Khoesan who appea ed be o e he Cou o Jus ice in Cape Town. Using he Cou o Jus ice eco ds, hey a e able o es ima e he size o he Khoesan popula ion. Figu e 4.1 shows he es ima es. Khoesan numbe s a e much lowe han hose o he o he h ee popula ion g oups, hough hei g ow h a e is much he same. By he mid-1720s, he a e o Khoesan popula ion g ow h inc eases conside ably, ollowed by a empo a y and modes decline om 1740 o 1750; i hen con inues o inc ease o each mo e han en housand in 1795. The e is a undamen al p oblem wi h his accoun . The numbe o Khoesan who appea ed in cou may simply ha e been a unc ion o how de eloped he cou sys em was and how willing people we e o ake ma e s o cou o sol e hem by hemsel es. Ra he han being a use ul p oxy o he size o he Khoesan popula ion li ing in he colony, Fou ie and an Zanden’s es ima es should be ead as an a emp o measu e he numbe o Khoesan ha we e inco po a ed in he colonial economy. Thei es ima es con i m wha we would expec . As he on ie closed, he numbe o Khoesan ope a ing unde he ada o he colonial au ho i ies dec eased. Figu e 4.1 Es ima ed numbe o Khoesan, 1701–95 (based on hei p esence in cou eco ds). Sou ce: Fou ie and an Zanden (2013: 472). Was he Cape Colony a Sla e Economy? 91 In an a emp o imp o e hese es ima es, Fou ie and G een (2015) use ano he me hod, no o calcula e he size o he indigenous popula ion in o al, bu a he he numbe s o Khoesan wo king on Eu opean a ms. Assuming ha only Khoesan we e used o pas o al a ming, and es ima ing hei p oduc i i y by elying on a ew a elle s’ accoun s ha gi e he numbe o Khoesan p esen on indi idual a ms ha hey iden i y in he annual ax censuses, hey a e able o es ima e he size o he Khoesan labou o ce as a unc ion o numbe o ca le on he Eu opean a ms. The es ima es a e ound in Table 4.5. Table 4.5 Es ima ed numbe o Khoesan and sla es by egion, 1663–1773 (selec edyea s). All egions Cape, S ellenbosch and D akens ein dis ic s Swellendam* Yea Khoesan Sla es Khoesan Sla es Khoesan Sla es 1663 0 19 0 19 1682 0 192 0 192 1685 122 222 122 222 1692 406 303 406 303 1695 380 533 380 533 1700 526 807 526 807 1709 1338 1582 1398 1582 1712 1298 1766 1298 1766 1719 1076 2317 1076 2317 1723 1434 2508 1434 2508 1731 2016 3590 2016 3590 1738 2724 4769 2724 4769 1741 2617 4495 2617 4495 1757 5359 5007 2123 4711 3236 296 1773 6843 7101 3296 6369 3547 732 To al 26136 35211 19354 34183 6783 1028 *Swellendam was only p oclaimed as a sepa a e dis ic in 1754. Sou ce: Fou ie and G een (2015: 205). 98 C ea ing he Cape Colony Sla es as capi al a he Cape Labou was no he only challenge se le - a me s aced. Capi al was also a cons ain . C edi has played an impo an ole in mos ag icul u al economies, and he Cape Colony was no excep ion. Ross (1986) a gues ha he mos impo an sou ce o inance s emmed om he o ma ion o capi al on he a ms. Ye , as he also acknowledges, i is di icul o see how especially he la ge landholdings would su i e wi hou an ex e nal sou ce o capi al. The loca ion o he Cape in he global ade ne wo k o he Du ch Empi e made he ole o c edi e en mo e impo an , since he e was a ch onic sho age o Du ch ixdolla s. A numbe o schola s ha e no ed ha a signi ican sha e o he se le s in he Cape Colony we e highly indeb ed and used his obse a ion o make he claim ha he a e age a me was poo and could ha dly make ends mee (e.g. Guelke 1989; Schoeman 2012). Con empo a y wi nesses sha ed his iew. In 1795, o example, igh be o e he B i ish annexed he Cape Colony, he conce ned a me Johannes F ede ik Ki s en w o e o he B i ish go e nmen , complaining ha ‘by a he g ea e pa o he Fa me s and Inhabi an s o he Town a e Bank up s [sic], Table 4.11 Numbe o sla e men, women and child en in Cape Colony, 1687–1793 (selec ed yea s). Yea Men Women Child en Sex a io % o W o M 1687 230 44 36 5,23 18 1691 285 57 44 5,00 17 1701 702 109 80 6,44 13 1711 1,232 290 249 4,25 19 1723 2,224 408 290 5,45 16 1733 3,384 711 614 4,76 17 1743 3,804 815 742 4,67 18 1753 4,137 1,031 877 4,01 20 1763 5,072 1,214 929 4,18 19 1773 6,102 1,707 1,093 3,57 22 1783 7,808 2,533 1,609 3,08 24 1793 9,046 3,590 2,111 2,52 28 Sou ce: Wo den (1985: 53). Was he Cape Colony a Sla e Economy? 99 he es ha e hei p ope y unde Seques e , and e e y indi idual looks o wa d o impending uin’ (quo ed in Swanepoel 2017:21). The e is ample e idence ha a me s s uggled om ime o ime. Bu Swanepoel inds ha weal hie a me s a he Cape ended o be mo e indeb ed han poo e ones. Using auc ion olls om 1663 o 1784, she inds ha c edi played an impo an ole o se le - a me s (65 pe cen o he sample we e in ol ed in c edi ansac ions) and ha a me s who had mo e sla es, which she uses as a p oxy o weal h, also ended o be mo e indeb ed. This is, mo eo e , in line wi h indings om p e-indus ial Eu ope (Thoen and Soens 2009). I is wo h no ing ha he majo i y o he c edi was used o in es men s in p oduc ion a he han consump ion. O all he c edi in he sample, 56 pe cen was used o p oduc ion, 21 pe cen was alloca ed o consump ion and he es was used o und dona ions, axes, accoun s andinhe i ance. This unde sco es he impo ance o c edi o se le a ming a he Cape. The o mal c edi ma ke was limi ed and domina ed by majo playe s like he Company, he O phan Chambe , chu ches and a ew Company employees (Guelke 1989; Ross 1989; Schoeman 2012). Toge he , hese ins i u ions and indi iduals made up a small sha e o he c edi ansac ions a he Cape. In o al, 90 pe cen o he lende s a he eigh een h-cen u y Cape we e indi idual se le s (Swanepoel 2017). I is o hese ansac ions ha sla es played an impo an ole. The majo i y o se le s, who leased land unde he loan- a m sys em, could use nei he land no houses o buildings as colla e al. Theo e ically, hey could make use o a m equipmen as colla e al, bu i s alue was in gene al oo small o enable se le s o access any signi ican le els o c edi . Fa me s did, howe e , possess one asse hey could use as colla e al: sla es. Sla es we e he mos liquid and mobile asse s in he colony, se ing o ‘c edi o s [ o] gauge a deb o ’s iskiness’ (Swanepoel 2017: 42). In 1713, o example, sla eholde Johannes C aa mo gaged his sla es o ecei e c edi om he Du ch Re o med Chu ch. E idence sugges s ha om ha poin on, his p ac ice became widesp ead since sho - e m c edi was essen ial o he ope a ion o sla eholding es a es (Shell 1994; Dooling 2006). In he s udy, Swanepoel (2017; see also Fou ie and Swanepoel 2018) inds ha sla e owne ship was he only signi ican cha ac e is ic used o de e mine deb . Especially o olde sla eholde s, he sla es ‘we e hei bank’ (Shell 1994: 109). The use o sla es as colla e al was in no way unique o he Cape. In cases like p ecolonial Nige ia (Fenske 2012) and B azil (Schulz 2008), o example, sla es we e used as secu i y in c edi ansac ions. Jus as i was a he Cape, de ju e landed p ope y igh s we e weak in hese h ee cases, making i di icul o a me s o use hei land as colla e al. E en when de ju e landed p ope y igh s we e s ong, as hey we e in he Uni ed S a es, sla es he e we e commonly used as colla e al simply because hey we e mo eable asse s (Kilbou ne 2014). 100 C ea ing he Cape Colony Sla es we e no only used as colla e al. The e is ample e idence o sla eholde s leasing ou hei sla es o o he s. This made a lo o sense gi en he di e en ag icul u al seasons p e ailing in whea a ming and i icul u e. The lease ma ke o sla es mean ha la ge sla eholde s could lease ou pa o hei sla e labou o ce du ing hei slack season o a me s who had ewe sla es and needed addi ional labou du ing hei peak season. The impo ance o leasing o he sla eholde was acknowledged by he VOC. In 1786, he Council o Policy w o e: The acquisi ion o sla es ha ing inc eased by means o he g ea p o i s, pa o hem being used o ex end he ineya ds, while ano he pa belongs o such people as make a li ing by hi ing hem ou , so ha hey a e bound o lose hei sou ce o income on he dea h o hese sla es o a e bu dened wi h hei suppo when hey can no longe be hi ed ou . (quo ed in Schoeman 2012: 323–34) The VOC published a ecommended p ice o hi ing sla es, indica ing ha he p ac ice had become ins i u ionalized in Cape socie y. In 1775, he a e was 4 ixdolla s pe mon h plus ood and obacco. By 1783, he p ice had inc eased o 6 ixdolla s plus ood and obacco as an e ec o he inc easing p ice o pu chasing sla es. Hi ing ou sla es was a p o i able en e p ise. Fo example, Jan Nel, a igne on om S ellenbosch, hi ed ou wo sla es o he pe iod om Ma ch 1730 o No embe 1737. In e u n, he was paid 225 ixdolla s, which is almos equi alen o wha he would ha e ea ned by selling hem a auc ion. The VOC i sel hi ed sla es om sla eholde s and spen a subs an ial amoun in doing so. In 1787–8, o example, he Company alloca ed no less han 22,000 ixdolla s o his expense (A ms ong 1979; Shell 1994). The possibili y o using sla es o acqui e capi al is mani es ed in he e ec s o he sla e emancipa ion in 1834. The emancipa ion happened o y yea s a e he pe iod co e ed by his book, so I will no discuss i in any de ail. Bu i needs o be men ioned as i bea s wi ness o he impo ance sla es played o sla eholde s’ access o capi al. I has been a gued ha he emancipa ion led o an ag a ian c isis o he se le a ming communi y as labou wi hd awal caused a decline in ou pu (Dooling 2006; Wo den 2017). Meanwhile, one needs o acknowledge he limi ed choices acing he eed sla es. The as majo i y emained wi hin he dis ic s and we e le wi h no al e na i e o he han o p o ide hei labou se ices o he se le - a me s. This has led schola s like Shell (1994) o conclude ha he pos -emancipa ion c isis was sho li ed. S ill, Was he Cape Colony a Sla e Economy? 101 how can one explain he c isis, sho -li ed o no , i he sla es we e no in a posi ion o lea e hei o me mas e s? In a ecen s udy, Ma ins (2020) o e s he explana ion ha he decline in ou pu was o a g ea ex en an e ec o he capi al losses he se le - a me s aced due o he emancipa ion o sla es. No longe being hei p ope y, he se le s could no longe make use o sla es o access capi al. Ma ins es ima es ha 70 pe cen o he a ia ion in g ain ou pu and 20 pe cen o he a ia ion in wine ou pu a e he emancipa ion can be explained by he capi al losses ha a me s aced. Conclusion The cen al ques ion asked in his chap e was whe he he Cape Colony was a ‘sla e economy’ in Nieboe ’s sense o he e m, ha is, whe he sla e y was a key ins i u ion o he g ow h and su plus p oduc ion o he se le a ming sec o a he Cape. My conclusion is ha i was indeed a sla e economy, bu no necessa ily in he way desc ibed in p io li e a u e. In his chap e , I ha e made he case ha p e ious schola ship has exagge a ed he ole o sla es as labou and a he same ime has neglec ed he impo ance o Khoesan labou e s. Ra he han being a supe io o m o labou , sla es paid a key ole in he Cape economy as hey ga e se le - a me s access o bo h labou and capi al. Despi e he many ac o s ha heo e ically would p e en he p o i able use o sla es, he gene al conclusion is ha sla e labou was a key elemen in he g ow h o se le ag icul u e in he colony. My conclusion in his chap e – based on es ima es ha admi edly should be ea ed wi h a g ea deal o cau ion – is ha p io s udies ha e o e es ima ed he p oduc i i y o sla e labou . Recognizing he impo ance o Khoesan labou and including i in he equa ion causes es ima es o sla e labou p oduc i i y o all signi ican ly. Wi h he Khoesan included, he p oduc i i y o sla e labou a he Cape is shown o be a lowe han ha o sla es in he New Wo ld. To make ma e s wo se, he cos o sla e labou inc eased o e ime because o he way he Cape ma ke o ag icul u al p oduce was o ganized. Why hen was sla e owne ship so widesp ead a he Cape? Was i simply ha owning a sla e was associa ed wi h social s a us? Tha p obably was one ac o , bu equally impo an was he ac ha i ga e a me s access o much- needed capi al. La ge sla eholde s could accumula e capi al by specula ion on he sla e ma ke , bu also by leasing ou hei sla es o o he se le s who could 102 C ea ing he Cape Colony make use o hem. Fu he , in he absence o secu e, de ju e p ope y igh s in land, sla es played a key ole as colla e al enabling he capi al-cons ained se le s o access esou ces needed o keep hei businesses a loa . The Nieboe - Doma hypo hesis igh ly poin s o he ole o ac o endowmen s bu ocuses oo na owly on land–labou a ios. To unde s and sla e y as an economic ins i u ion, we need o include capi al in o he equa ion. Sla e y was no jus a o m o coe ced labou , bu a sys em ha ga e he sla eholde comple e igh s o e a mobile p ope y, a sys em ha allowed he owne no only o ex ac labou , bu also access capi al. In oduc ion A cen al conce n in he a ional-choice poli ical economy (RPE) li e a u e on se le colonialism is he di e gen le els o inequali y ac oss se le colonies and i s impac on long- e m economic de elopmen . The li e a u e a gues ha inequali y le els emained ela i ely low in egions whe e he Eu opeans se led in g ea e numbe s (Acemoglu, Johnson and Robinson 2002; Enge man and Sokolo 2005; Angeles 2007). Because o hei la ge numbe s, he Eu opeans cons i u ed a c i ical mass ha p essu ed he colonial au ho i ies o implemen so-called ‘de elopmen al’ o ‘inclusi e’ ins i u ions ha gua an eed b oad access o economic and poli ical ma ke s, secu ed p ope y igh s ( o he se le s) and used ax e enues o he p o ision o public goods. In egions whe e he numbe o Eu opean se le s was lowe , se le s ins ead pe suaded he colonial au ho i ies o implemen ex ac i e ins i u ions ha enabled hem o ex ac economic en s h ough he exploi a ion o human and na u al esou ces and a he same ime p e en se le s who a i ed la e o do he same. RPE schola s u he a gue ha ini ial le els o inequali y ha e pe sis ed o e ime and can explain cu en income gaps be ween o me se le colonies. While he ecen li e a u e on se le economies has had a signi ican impac on ou unde s anding o di e ences in inequali y among hose colonies, hey do no p o ide a sys ema ic empi ical analysis o he e olu ion o inequali y. In his chap e , I use he Cape as a case s udy o con ibu e o he deba e on se le colonialism, i s ins i u ional o de and i s e ec s on inequali y. The Cape Colony does no easily i wi h ei he side o he dicho omy p oposed by RPE. On he one hand, he numbe o Eu opeans was, in ela i e e ms, la ge and he a e age size o he se le a ms was compa a i ely small. In ha espec , he Cape esembles No h Ame ica, in which Eu opean amily a ms domina ed, leading o low le els o inequali y and an inclusi e ins i u ional o de . On he 5 Unequal we s and 104 C ea ing he Cape Colony o he hand, as I explain below, inequali y le els we e high a he Cape and he ins i u ional o de – which en ailed sla e y, weak de ju e p ope y igh s and egula ed ma ke s – was ex ac i e i we ollow he classi ica ion p oposed in he RPE li e a u e. This chap e will make use o ich his o ical sou ces, enabling mic o- le el longi udinal analysis o he e olu ion o inequali y a he Cape o ask a undamen al ques ion: wha ac o s enabled some se le s o accumula e weal h mo e quickly han o he s? P e ious wo k on he Cape has highligh ed a ange o ac o s o explain inequali y du ing his pe iod, including une en access o e ile land, sla e owne ship and di e ences in human capi al. While all o hese ac o s ma e ed, I will a gue ha di e ences in accessing and con olling labou we e he mos impo an . The e we e wo ways o accumula ing capi al in he ini ial yea s. Fi s , we ha e a g oup o se le s who ini ially gained weal h h ough non- a ming ac i i ies. This weal h was in es ed in land and e en ually in pu chases o sla es. This g oup made up a small p opo ion o he iches a me s. The la ge sha e o ela i ely weal hy a me s ini ially accumula ed hei weal h h ough a ming, whose p o i abili y du ing he i s i y yea s depended on he abili y o access and con ol indigenous Khoesan labou . Ope a ing in an open on ie , he Eu opean in ude s ini ially used a ange o me hods o ensu e a su icien numbe o indigenous labou e s, including b u e o ce and o he kinds o coe cion. Howe e , coe cion could no be employed on a sys ema ic basis. Ins ead, he i s wa e o se le s ook con ol o mo e land han hey had he means o pu in o p oduc i e use. This was done in o de o a ac pas o al Khoesan o s ay on he land as enan s. The weal h hey hen accumula ed h ough he use o Khoesan labou could be in es ed in sla es, and o e he yea s ha ing la ge sla eholdings became a key cha ac e is ic o he se le a ming eli e. This chap e looks a inequali y in he se le popula ion, a he han among he Khoesan, o wo easons. Fi s , we lack eliable in o ma ion abou income and weal h among he Khoesan. I is no con o e sial o assume ha he Khoesan – as a g oup – was signi ican ly poo e han he se le s. Howe e , we do no ha e enough in o ma ion on he weal h/income o he Khoesan o sys ema ically s udy changes in he di e ences in income/weal h be ween a ious Khoesan g oups and he se le s. Second, he ecen esea ch on inequali y and economic de elopmen in se le colonies has p ima ily conce ned i sel wi h s a i ica ion among he se le s. So, while ou na ow ocus on he se le s can igh ly be blamed o being Eu ocen ic, i can s ill p o ide impo an insigh s abou he cu en deba es on he legacy o se le colonialism. Unequal We S and 105 Explaining inequali y in p e-indus ial colonial socie ies Economic his o ians ha e long been in e es ed in explaining di e ences in inequali y o e ime and space. Economic g ow h in gene al and pa hs o s uc u al change in pa icula ha e been highligh ed as key ac o s de e mining inequali y ends. A p ominen con ibu ion s ems om he wo k o Simon Kuzne s (1955) and his amous in e ed U-shaped inequali y cu e. Kuzne s ela ed changes in inequali y o he s uc u al ans o ma ion o he economy, ha is, a shi away om an economy domina ed by low-p oduc i e sec o s (in mos cases ag icul u e) o one in which high-p oduc i e sec o s (commonly manu ac u ing) inc eased hei sha e o GDP. Du ing he ini ial phase o his ans o ma ion, he wage gap be ween he low- and high-p oduc i e sec o s would lead o inc eased inequali y. O e ime, howe e , inequali y will dec ease as mo e people mo e in o he high-p oduc i e sec o s, while he wage le els con e ge be ween he sec o s as labou sho ages in he o me low-p oduc i e sec o s s imula e labou -sa ing echnological changes. Kuzne s himsel admi ed ha one should e ain om gene alizing as each case, wi h i s own ins i u ional mix, would a ec he pa e n. This is also con i med in la e esea ch ha shows mixed esul s (Linde and Williamson 1985; Linde 1986; Feins ein 1988; Bou guignon and Mo ison 2002; Milano ic, Linde and Williamson 2010). Acemoglu and Robinson (2002) o e one explana ion o he mixed esul s, a guing ha he exis ence o a U-shape inequali y cu e depends on poli ical ac o s. In line wi h Kuzne s, hey a gue ha indus ializa ion led o inc eased inequali y bu a he same ime acili a ed an inc eased geog aphical concen a ion o he wo k o ce. This educed he cos o ha wo k o ce o mobilizing p o es s agains he economic and poli ical eli e. The h ea posed by an o ganized wo king class would, unde ce ain ci cums ances, make he eli es willing o implemen edis ibu i e policies leading o a decline in inequali y. This is wha happened in Wes e n Eu ope bu , o example, no in La in Ame ica, acco ding o he au ho s. In he la e case, ini ial inequali y was high and ci il socie y ela i ely weak, so p essu e o change he sys em was likewise weak. This begs he ques ion: why was ini ial inequali y much highe in ce ain egions han in o he s? The Kuzne s’ in e ed U-shape cu e implici ly assumes ha p e-indus ial socie ies we e cha ac e ized by ela i e low le els o inequali y. This is a odds wi h mo e ecen esea ch showing high le els o inequali y in a ious p e- indus ial socie ies, o which mos o he cases come om Eu ope (see Al ani 2021 o an o e iew). I is also widely accep ed ha inequali y le els we e qui e 106 C ea ing he Cape Colony high in p e-indus ial se le colonies. This is commonly explained by poli ical ac o s. The se le s we e a p i ileged g oup who, unde he p o ec ion o he colonial au ho i ies, could ensu e he implemen a ion o economic and social policies ha a ou ed hem a he expense o indigenous popula ions and/o la e- a i ing immig an s (Elkins and Pede sen 2005). Al e na i ely, al hough his is seldom discussed in he li e a u e, one could explain he high le els o inequali y in se le colonies as a consequence o popula ion g ow h and an ex ension o he land on ie . The g ow h o popula ion would lead o he opening o ma ginal land o cul i a ion, leading o so-called Rica dian en s. Acco ding o he classic economis Rica do ([1821] 2004), he en is he di e ence in he yield o a plo o supe io land as compa ed o he yields p oduced on ma ginal lands. This would lead o an ‘ea ly-a i al p emium’ in which la e a i als would ea n lowe incomes (Cillie s, G een and Ross 2021). Inequali y would inc ease as long as popula ion g ow h con inued and he land on ie expanded. A key ocus in he RPE li e a u e on colonialism is how inequali y di e s among colonies. To accoun o he di e ences, RPE schola s o e a modi ied poli ical explana ion. Keeping he assump ion ha le els o inequali y a e d i en by he in e es s o he se le s, RPE schola s a gue ha he size o he se le popula ion de e mined he ins i u ional o de ha was es ablished and he e o e also he deg ee o income and weal h inequali y. In a eas wi h ew se le s, hey would ely on ex ac ing economic en s and p e en u he se le immig a ion. This c ea ed high le els o inequali y ha ha e pe sis ed o e ime. In cases whe e Eu opeans se led in g ea e numbe s, hey became a poli ical o ce o equal oppo uni ies ( o he se le s) and inclusi e ins i u ions ( o he se le s), which c ea es socie ies o ela i ely low le els o inequali y (Enge mann and Sokolo 2000; Acemoglu, Johnson and Robinson 2002). La in Ame ica and No h Ame ica epi omize a di e en pa h. In he o me , se le s emained ew and le els o inequali y high, while in he la e he numbe o se le s was g ea e and le els o inequali y smalle . This in e p e a ion o inequali y has – di ec ly o indi ec ly – in luenced a g ea deal o wo k on colonialism and inequali y among economic his o ians and economis s o e he pas wo decades (Bake , B unnschweile and Bul e 2008; B uhn and Gallego 2008; Pu e man and Weil 2010; Eas e ly and Le ine 2012). A basic p oblem wi h his line o inqui y, howe e , is ha , in he wo ds o Gonzalez and Mon e o, i is ‘a basically an i-empi ical way o easoning’ (2010: 255). Ins ead o in es iga ing le els o economic inequali y he au ho s assume ha i was highe in ce ain pa s o he colonial wo ld, an assump ion hey subs an ia e by iden i ying he exis ence o ce ain ins i u ions, such as he Unequal We S and 107 hacienda plan a ions in La in Ame ica. No ac ual igu es on land dis ibu ion be ween egions a e o e ed (F ankema 2009). The ew schola s who ha e ied o empi ically in es iga e economic inequali y in he Ame icas ha e ound li le e idence ha La in Ame ica and No h Ame ica di e ed o any signi ican deg ee du ing he colonial pe iod (Coa swo h 2008; Williamson 2015). I was du ing he commodi y boom in he la e nine een h and ea ly wen ie h cen u ies ha he wo egions i s began o di e ge in e ms o inequali y. The nega i e co ela ion be ween numbe o se le s and inequali y can also be ques ioned when one looks a he A ican expe ience. Schola s on colonialism in A ica con incingly make he case ha inequali y was highe and ex ac ion mo e pe asi e in se le colonies han in non-se le colonies (Acemoglu and Robinson 2012; F ankema, G een and Hillbom 2016). Excluding Aus alia, Canada, New Zealand and he Uni ed S a es, whe e he indigenous popula ions we e me ely elimina ed, he same pa e n seems o exis o mos o me colonies, as Angeles (2007) has shown. To unde s and he e olu ion o inequali y o e ime, we ha e o e u n o he amewo k used in his book, which sugges s ha he ins i u ions go e ning an open on ie a e no he same as would be ound in a closed one. A he Cape, he e we e wo majo a enues o gene a ing weal h: ade o a ming. The o me c ea ed signi ican weal h, bu only o a e y small eli e, as I show in he inal chap e . The majo i y in es ed in a ming. One’s abili y o gene a e weal h om a ming was a combined e ec o he quali y o he land and he a me ’s capaci y o access and con ol a su icien numbe o labou e s. The on ie being open, one would expec he i s wa e o se le s o es ablish hei a ms in he clima ologically mos a ou able loca ions, i his was no p e en ed by esis ance om he indigenous popula ions. Howe e , clima ological di e ences would no be ansla ed in o di e en capaci ies o gene a e weal h as long as he se le s did no ha e a su icien numbe o labou e s o wo k he land. This has been ecognized by a g ea numbe o Cape his o ians and is one o he easons why many a gue ha sla e y played such an impo an ole o he economy (see Chap e 4). In Chap e 4, I a gued ha much o he exis ing esea ch has exagge a ed he ole o impo ed sla e labou . Sla es played a c ucial ole and o e ime a s ong co ela ion be ween sla e owne ship and weal h de eloped a he Cape. Howe e , o be able o pu chase sla es one al eady had o be able o accumula e su icien capi al as sla es we e expensi e. This was done by ensu ing a su icien supply o Khoesan labou e s. Those who es ablished a ms i s had an ad an age as hey we e allowed o ake con ol o as much land as hey wished. They commonly 114 C ea ing he Cape Colony Wha can we make o hese es ima es? I we ocus on he se le a ming communi y, he es ima es e eal high le els o bo h income and weal h inequali y. The long- un end seems o be owa ds an inc eased concen a ion o weal h and income o e ime among he weal hies se le s. This is a ascina ing inding. The ac ha mos a ms we e smalle han hose ound in he New Wo ld, he p ac ice o pa ible inhe i ance, which in heo y mean ha a ms we e subdi ided o each gene a ion, along wi h he ac ha ac o endowmen s did no allow o la ge-scale p oduc ion o opical p oduc s as hey did in he New Wo ld, would all seem o mili a e in a ou o ela i ely low le els o inequali y. Why do we ind he opposi e? Explaining inequali y ends among he se le a ming popula ion P e ious esea ch has iden i ied di e en d i e s o inequali y a he Cape. Inspi ed by Enge man and Sokolo (2005), Fou ie and on Fin el (2010a) a gue ha changes in immig a ion policies can explain how inequali y e ol ed o e ime. In he ea ly yea s, weal h inequali y among he se le - a me s was low, as he majo i y o he newly a i ed immig an s we e poo wi h no possessions. As hey began o in es in land, hey managed o accumula e mo e weal h while a he same ime he e was a con inued in low o poo mig an s. Unde hese condi ions, weal h inequali y g ew. A e 1700, immig a ion o he Cape declined, leading o a diminishing sha e o people a he bo om ail o he weal h dis ibu ion. Weal h inequali y consequen ly dec eased. The inc ease in inequali y a e 1740 was mainly d i en by changes in he uppe ail, whe e a g oup o a me s managed o accumula e weal h a a as e pace han he a e age. Acco ding o Fou ie and on Fin el, his was because he i s gene a ion o Table 5.5 Weal h and inequali y o se le s and Khoesan popula ion in Swellendam, 1825. G oup A e age weal h Gini Fa ming se le s wi h land and sla es 894.38 0.47 Fa ming se le s wi h land and no sla es 214.63 0.38 Fa ming Khoesan 42.56 0.57 Sou ce: Links, G een and on Fin el (2018). Unequal We S and 115 immig an s had been able o es ablish iable a ms ha we e success ully aken o e by he second gene a ion. Those who we e ela i ely be e o could eap he bene i s o economies o scale by employing mo e sla es (see also Gulke and Shell 1983 and Wo den 1985). The e is a ai ly la ge li e a u e om he 1980s ha implici ly analyses weal h dis ibu ion by ocusing on he ise o a speci ic g oup o Eu opean se le s, he Cape gen y. The in e es s ems om wo k done by he Ge man immig an O.F. Men zel, who li ed a he Cape in he 1730s. Men zel di ided he Eu opean se le s a he Cape in o ou g oups. Fi s , he e we e he poo es , who we e pas o al a me s li ing in he in e io . Abo e hem he iden i ied c op a me s who owned ew sla es and wo ked ha d o make a li ing. They we e, in Men zel’s e ms, bo h ‘mas e and knech ’. The second weal hies g oup we e se le s who ope a ed p o i able a ms, while he iches g oup consis ed o Eu opeans li ing in Cape Town and owning se e al p o i able a ms (Guelke and Shell 1983: 277). The la e Men zel desc ibes as ‘capi alis s … who li e in own bu own es a es in he coun y, which a e wo ked by sla es and managed by an o e see o s ewa d who has o ende a p ope accoun pe iodically o his employe ’ (1925–1944, ol. 2: 85). I is his second weal hies g oup o se le s ha Men zel de ined as a gen y. Men zel e e ed o hem as se le s who li e in he sou h-wes e n Cape and ‘who possess excellen a ms, paid o and luc a i e, who li e on hese a ms and p oduce mo e han hey consume’ ( ol. 3: 98–101). Al hough Men zel’s classi ica ion was ague, i has ne e heless inspi ed a numbe o schola s o c i ically analyse he ac o s ha explain hei ise. Ross (1983) e e s o he gen y as a g oup o a me s who p oduced o he ma ke s on an ex ensi e scale and con ends ha hey cons i u ed he b oad mass o a me s a he han a small eli e. Guelke (1979), on he o he hand, de ines he gen y as a small g oup o a me s ha had ela i ely la ge a ms ha hey ope a ed by using sla es. The p oblem wi h his is ha sla e owne ship was widesp ead among a me s (see Chap e 4) and Guelke ailed o p o ide a de ini ion o a la ge a m in he con ex o Cape se le ag icul u e. Dooling (2005) simila ly de ines he ‘gen y’ in e ms o ha ing access o la ge landholdings and sla es. To his he adds ha he gen y had a ce ain deg ee o in luence o e local poli ics (I discuss his in mo e dep h in Chap e 6). Williams (2013) igh ly poin s ou ha ha ing access o la ge ac s o land does no necessa ily pu a me s in o a ce ain class in he economic sense o he e m. Fa me s need su icien labou e s o pu he land in o p oduc i e use and i is hei di e en ial access o labou e s ha should de ine hem in o di e en classes. Gi en he impo ance o sla es, bo h as labou and capi al, i would make 116 C ea ing he Cape Colony sense o dis inguish he gen y om o he g oups o se le - a me s in e ms o numbe o sla es owned. This is in line wi h Giliomee (2003), who de ines he ‘gen y’ as a me s who owned six een sla es o mo e. Following his de ini ion and using Wo den’s (1985) es ima e o he dis ibu ion o sla e owne ship (see Chap e 4) we can see ha in he sou h-wes e n Cape, he size o he gen y class inc eased o e ime. The ques ion is: wha enabled he g ow h o he gen y? To answe ha , one mus di e en ia e be ween he ini ial condi ions enabling a g oup o a me s o accumula e weal h and he ac o s ha enabled hem o sus ain his accumula ion o e ime. S a ing wi h he o me , i seems likely ha he i s mo e ad an age played a ole. In Chap e 3, we saw ha he i s wa e o se le s o g an eehold land igh s loca ed hei a ms in mo e sui able a eas han g oups coming la e . Explo a i e wo k by Cillie s, G een and Ross (2021) sugges s ha he i s mo e ad an age con inued o play a ole as he colony expanded geog aphically. Bu , again, ha ing access o he bes land means li le i you lack he capaci y o u ilize i p oduc i ely. Fou ie and on Fin el (2012) a gue ha Hugueno s om wine-p oducing a eas o F ance who a i ed as e ugees a he Cape in 1688 had a compa a i e ad an age in i icul u e. Con olling o labou , soil quali y and o he impo an ac o s, hey a gue ha hese a me s on a e age p oduced mo e wine han o he se le s. They do no associa e he ise o i icul u e in he Cape wi h gen y a me s explici ly, bu a gue ha by he end o he eigh een h cen u y hese wine p oduce s we e among he weal hies igne ons in he Cape. I ha e no eason o ques ion his in e p e a ion, bu wo ac o s lead me o a gue o modi ying he ole o human capi al in i . Fi s , wine played a ma ginal ole in weal h accumula ion un il he second hal o he eigh een h cen u y. Expo s o wine we e small and o low quali y. Wine a elled badly and Cape wine had a poo epu a ion ab oad. Passing ships wan ed o pu chase b andy, no wine (Theal [1922] 1964; Ross 1993). Expo s o Cape wine did no s a o g ow un il he 1740s, when i began o be expo ed annually o Ba a ia (Guelke 1989; Bosho and Fou ie 2010). In he 1770s, wine su passed whea as he majo expo c op ( an Duin and Ross 1987). Be o e ha , he money one could make ou o i icul u e was limi ed and i is no su p ising ha Bosho and Fou ie (2010) ind no co ela ion be ween passing ships and he p oduc ion o wine. Ins ead, as I showed in Chap e 3, he mos impo an sou ce o weal h in he i s se en y yea s o se le a ming was in es ing in ca le and sla e labou . As discussed abo e, access o sla es co ela ed s ongly wi h a me s’ weal h. This was ue in he eigh een h cen u y when he use o sla e labou i s ook o . Be o e ha , he numbe o sla es emained low. In Chap e 4, I showed Unequal We S and 117 ha  sla es we e expensi e and equi ed a me s o ha e access o su icien capi al. The numbe o se le s wi h la ge sla eholdings was e y small un il he second decade o he eigh een h cen u y. In 1682, only 2 pe cen o se le s had mo e han i een sla es and none had mo e han wen y. By 1702, 3 pe cen had mo e han i een sla es, while 2 pe cen had be ween i een and wen y sla es and 1 pe cen had mo e han wen y- i e. I was no un il 1710 ha he sha e o se le s wi h mo e han i een sla es began o g ow and by 1731, 12 pe cen had be ween i een and wen y sla es while 8 pe cen had be ween wen y o wen y- i e and 6 pe cen had mo e han wen y- i e (Fou ie and on Fin el 2010a). In o he wo ds, he size o he gen y class – i de ined in e ms o numbe o sla es – was negligible un il 1710. To unde s and he d i e s o inequali y, we hus need o ake one s ep back and ask how he se le s managed o accumula e enough capi al o in es in sla es in he i s place. Inequali y and ini ial sou ces o capi al I is gene ally ag eed ha mos o he Eu opean immig an s o he Cape we e poo , al hough excep ions exis ed. High dea h a es and he low wages o e ed by he VOC made i di icul o a ac weal hy Eu opeans. Mos o he immig an s we e a me s, skilled a isans o labou e s (Guelke 1989). Once he VOC decided o p omo e Eu opean ag icul u e, he immig an s we e ini ially o e ed a ou able condi ions. They we e allowed o cul i a e as much land as hey could, and hey we e p o ided ag icul u al implemen s, seed and ca le on c edi (Theal [1922] 1964). To u n ag icul u e in o a p o i able business, howe e , hey needed labou , an inpu ha was no subsidized by he VOC. One possibili y was o in es in sla es, bu his equi ed capi al, which had o be gene a ed om non- a m ac i i ies. As Wo den (1985) and Williams (2013) ha e shown, many a me s did no ini ially accumula e weal h om a ming, bu by engaging in ade, anspo and o he non- a m ac i i ies. In es ing in he e ia y sec o was an impo an sou ce o income o many Eu opeans. Passing ships and hei c ew b ough money o Cape Town ha was spen on lodging, ood, d ink, p os i u ion and mino ade. Acco ding o an Duin and Ross (1987) his ‘may indeed ha e con ibu ed, h ough he mul iplie e ec , o he p ospe i y o [ he] Colony in ways we ha e been unable o measu e’ (quo ed in Bosho and Fou ie 2010: 473). A la ge po ion o he ac i i ies in he u ban e ia y sec o indeed wen un eco ded, bu a su ey o occupa ions in Cape Town in 1732 e eals ha abou 118 C ea ing he Cape Colony 20 pe cen o he se le s we e employed in he se ice sec o , while ano he 40 pe cen we e eco ded as ‘unce ain’ (Bosho and Fou ie 2010). Ha ing said ha , by a he mos p o i able business was o gain con ol o domes ic ade. In o de o ade, a licence (pach ) om he VOC was needed. Only a ew se le - a me s we e g an ed such igh s (see Chap e 6). This monopsony powe o e local ade gene a ed subs an ial p o i s (G oenewald 2007). The e a e se e al examples o weal hy a me s who began hei ca ee s ou side o ag icul u e. H. O. Eks een, o example, ini ially accumula ed his weal h by ge ing a licence o ade in alcohol. The money he made h ough ading he in es ed in a ming. By 1731, he had acqui ed a o al o se en eehold a ms, six loan a ms and one hund ed sla es (G oenewald 2007). In es ing in pas o al a ming and i icul u e was especially p o i able o his g oup as i could be in eg a ed wi h hei ading businesses and owne ship o slaugh e houses. In ha ega d, as Williams (2013) has obse ed, hei ad an ages did no lie in p oduc i e a ming pe se, bu in hei con ol o e he ma ke s. Some o he businessmen would become absen landlo ds and no a landed gen y as de ined in he li e a u e, while o he s would mo e ou o Cape Town o become ull- ime a me s. Mos se le s, howe e , lacked he inancial capi al and, equally impo an , he poli ical connec ions o he VOC o ge a ading licence. Ins ead, hey g adually accumula ed weal h by in es ing in ag icul u e. As shown abo e and in Chap e 3, in es men in i icul u e was ini ially ela i ely modes and no e y p o i able. The ma ke o whea was be e , bu g ow h in whea p oduc ion was slow un il he 1720s. Wha ini ially became a chie sou ce o accumula ion was pas o al a ming. The demand o mea was ini ially compa a i ely high and emained s able in he se en een h and ea ly eigh een h cen u ies. Acco ding o Robe son (1984: 114), he combina ion o eehold enu e and he associa ed igh o access pas u eland ga e indi idual se le - a me s an oppo uni y o accumula e weal h a a apid pace, and his led o inc eased inequali y among he se le s. He concludes ha : i was in he s ock a ming and/o ba e ing o o he wise acqui ing s ock om he Ho en o s, a ening i , and supplying he mo e buoyan ma ke s o animal p oduc s, ha excep ional weal h would seem o ha e accumula ed mos apidly in he hands o a limi ed numbe o a me -bu che s, whose weal h was no necessa ily e lec ed – o was only pa ially (and seconda ily) e lec ed – in ex ensi e eehold p ope ies. In sum, he weal hies li es ock a me s we e hose who had he capaci y o also ake con ol o he local mea ade and/o indus y. O e ime, hey inc eased Unequal We S and 119 hei in es men s in a able a ming, mos no ably i icul u e (Williams 2013). The gen y, on he o he hand, we e hose who in es ed mainly in li es ock ha hey sold in Cape Town, o he VOC o some imes di ec ly o passing ships. Labou and accumula ion in he se en een h-cen u y Cape Colony Un il 1717, ea ly a i als gene ally managed o access he mos sui able lands as aconsequence o he VOC policy o i s -come, i s -se ed. They also bene i ed om he ague o mula ion o p ope y igh s wi h ega d o pas u eland, which allowed hem o gain con ol o a he as land a eas (see Chap e 3). This led o a concen a ion o land owne ship ha bene i ed ea ly a i als o e la ecome s (Guelke and Shell 1983). As al eady men ioned, land concen a ion pe se would ha e limi ed e ec on weal h inequali y i a me s lacked access o he su icien numbe o labou e s needed o make p oduc i e use o he land unde hei con ol. Sla e labou was in pa a solu ion, bu in es ing in sla es was expensi e and only an op ion o he ew weal hy se le s who could in es in non- a ming ac i i ies. The majo i y had o g adually build up hei weal h by in es ing in ag icul u e, which, again, would gene a e p o i s i he se le s could manage o access and con ol su icien numbe s o Khoesan labou e s. Acco ding o Elphick (1977), se le s we e ini ially eluc an o employ Khoesan labou e s. They mis us ed he Khoesan as he de s as hey ea ed ha hey would s eal he ca le. They also conside ed he Khoesan o be ine icien as a m wo ke s as hey lacked skills in a able a ming. As he se le s expanded beyond he Cape Peninsula, howe e , he need o labou became acu e. Buying sla es o hi ing knech s was ini ially no an op ion o mos a me s because hey lacked su icien capi al. To ope a e hei a ms, hey needed access o Khoesan labou (Penn 2005). Low popula ion densi ies, he nomadic li es yle o he Khoesan and mos likely a gene al dislike o he Eu opeans among hem made i challenging o he se le s o ind su icien supply o labou e s. In his acis and p ejudiced e ms, Theal ([1922] 1964: 351) summa ized he si ua ion well as i was pe cei ed by he VOC and some se le s: The colonis s would gladly ha e employed some hund eds o Ho en o s [i.e. Khoe], i hey could ha e been induced o ake se ice, bu he men lo ed hei wild, ee, idle li e oo well o exchange i o one o oil. They had no objec ion, howe e , o do ligh wo k occasionally o ea n obacco and spi i s, and in 120 C ea ing he Cape Colony ha es ing especially hey we e use ul. They we e willing also o hi e ou hei emale child en, and by his means a ew household se an s we e ob ained and knowledge o he Du ch language was sp ead. None o hem had ye p og essed so a in ci iliza ion as o make ga dens o hemsel es, o in any way o cul i a e he g ound. Bu i was no ha he Khoesan people we e lazy ha made i challenging o he se le s o employ hem as a m wo ke s. Ins ead, i made li le sense o hem o wo k o he se le - a me s as long as hey could con inue as independen pas o al a me s selling ca le di ec ly o he se le s o he VOC. The si ua ion was pa ly eased by he expansion o he Eu opeans and he hos ili y be ween he a ious Khoesan socie ies ha his caused. This ga e some se le s he oppo uni y o employ Khoesan who ound hemsel es be e o ca le as a esul o in igh ing. I was commonly epo ed in he se en een h cen u y ha g oups o Khoesan mo ed in gangs om a m o a m (Elphick 1977). They used empo a y employmen as a s a egy o gain he means o es ock hei he ds. F om he landlo ds’ pe spec i e, howe e , his sou ce o labou was inhe en ly un eliable. Se le s could ne e be su e ha su icien numbe s o Khoesan labou e s would be a ound in ime o he peak seasons o inish hei asks be o e hey le . Al hough o en ea ed ha shly, he Khoesan gang labou e s he e o e ope a ed wi h some deg ee o eedom. To make ma e s wo se, he se le - a me s compe ed wi h he VOC o e Khoesan labou , as shown in Chap e 4. I seems plausible o assume ha he Khoesan would p e e o be employed by he VOC as he Company could o e be e p o ec ion and possibly also highe le els o emune a ion. The se le - a me s consequen ly needed o ake ce ain measu es o ob ain he se ice o su icien numbe s o Khoesan. The use o di ec coe ci e measu es on a sys ema ic le el was no possible. This esembles he si ua ion o Eu opean se le s in wen ie h-cen u y colonial A ica. Eu opeans in he la e e a depended on he colonial au ho i ies o ake measu es o ensu e a p ocess o p imi i e accumula ion. The au ho i ies we e able o inc ease he oppo uni y cos o A ican a ming o he ex en ha A icans we e willing o sell hei labou o he Eu opean landlo ds (A ighi 1970; F ankema, G een and Hillbom 2016; Fibaek and G een 2019). Wi h he excep ion o allowing se le s o es ablish a ms on land al eady occupied by Khoesan, no explici measu es we e aken by he VOC o acili a e an inc ease in he supply o Khoesan labou e s. They we e no axed, hey we e no emo ed o special designa ed a eas, and con ol o hei mobili y was no in oduced un il he nine een h cen u y. E en hough land was al eady sca ce in he sou h-wes e n Cape by he end o he eigh een h cen u y Unequal We S and 121 (see Chap e 3), Khoesan who we e s ill in possession o ca le heo e ically had he op ion o mo e a he in o he in e io . By he end o he se en een h cen u y, he expansion o he on ie had no ye gone a enough o comple ely disin eg a e he Khoesan in he sou h-wes e n Cape and i s en i ons. Nei he he economic no he poli ical o ces we e s ong enough o in oke a ull-scale p ocess o p imi i e accumula ion and u n he pas o al Khoesan in o a p ole a ia . This did no mean ha he Khoesan we e un ouched by he de elopmen ha did occu . Thei nomadic pas o al economy los access o aluable land as Eu opeans b ough he sou h-wes e n Cape unde hei con ol. This, oge he wi h he ca le ade and he wa s agains he Eu opean in ude s (see Chap e 2), pu he wo socie ies unde p essu e as i s ained hei capaci y o build up hei he ds. The s ain he Khoesan expe ienced is bes exempli ied by he expe ience o wo g oups: he Chainouqua, who dwelled on he ich pas u es o he sou he n plain wes o he colonial on ie , and he Cochoqua, p esen a he sou h- wes e n Cape. Toge he wi h he Hessequa, a he wes , he Chainouqua we e likely among he weal hies Khoesan socie ies in he mid-se en een h cen u y. This weal h empo a ily ga e he Chainouqua he uppe hand in con lic s wi h o he Khoesan socie ies like he Cochoqua. Fi y yea s la e , i s p ospe i y and powe had diminished conside ably. The Chainouqua we e in ol ed in a con inual eud wi h he Cochoqua. A he heigh o he second on ie wa (1671–7), hei alliance wi h he VOC enabled hem o se e ely weaken he la e g oup. This alliance came a a cos , howe e , as he Chainouqua, in exchange, had o ensu e a s eady supply o li es ock o he VOC. In 1693, Go e no Simon an de S el, unde p essu e om enemies o he Chainouqua, abandoned his iendship wi h hem. The Chainouqua we e al eady weakened due o diminishing he ds caused by he in ensi e ade wi h he VOC (Ross 1989). The once-powe ul socie y was on he e ge o disin eg a ing. No longe p o ec ed by he alliance wi h he VOC, many o i s membe s sough employmen as labou e s o he se le - a me s. While empo a y losses o ca le could be deal wi h on a socie al le el, indi idual Khoesan amilies ound i inc easingly di icul o manage majo losses. Those who owned jus a small numbe o li es ock we e ulne able o exogenous shocks like he o disease as he supply o ca le wi hin hei communi ies was educed h ough in ensi e ading and wa a e. The Eu opean se le - a me s could ake ad an age o his ulne abili y by o e ing Khoesan he hope o egaining hei s ock he way ha had o en been used adi ionally: by he ding o someone else. We lack eliable eco ds o emune a ion, bu 122 C ea ing he Cape Colony acco ding o Elphick (1977), he wages ecei ed by he Khoesan who wo ked as casual labou e s o se le - a me s we e likely oo low o enable hem o es ock and e u n o hei weakened k aals ( illages). Ins ead, he Khoesan would se le on he a ms and wo k as he de s as well as a m labou e s o an ex ended pe iod o ime. Fo his, hey we e o en paid in li es ock and we e allowed o he d i on he se le s’ land. This was he possibili y a ailable o he Khoesan o ebuild hei he ds in he sho es ime possible (Viljoen 2001; Penn 2005). By he beginning o he 1680s, se le - a me s in he sou h-wes e n Cape began o encou age he Khoesan o come and li e on hei a ms. Al hough e idence is only sca e ed and p e en s any quan i ica ion o he Khoesan who did his, he e a e nume ous e e ences o Khoesan amilies li ing on Eu opean a ms in he la e pa o he se en een h cen u y (Elpick 1977; an de Me we [1938] 1995). This pa on–clien ela ionship p o ided he se le s wi h a pa ial solu ion o he high en o cemen cos s associa ed wi h employing Khoesan on sho - e m con ac s. I was qui e easy o casual wo ke s o dese hei employmen when hey needed o. Thei k aals we e no e y a away, and he VOC lacked he capaci y o police hem (Ross 1993). Tying he Khoesan o he a ms on long- e m con ac s (in which hey we e paid in li es ock and also allowed o he d hei own ca le and sheep) lowe ed he incen i es o hem o un away. In 1688, a VOC cap ain epo ed ha he Khoesan we e li ing side by side wi h he Eu opeans in he sou h-wes e n Cape. In a he en husias ic e ms, he desc ibed ‘ hei lo e and a ec ion owa d us g owing mo e and mo e, so ha a he p esen du ing he p essu e o ha es and plowing ime, hey slip down among us jus like he Wes phalians in he Ne he lands’ (quo ed in an de Me we [1938] 1995: 24). The i s quan i a i e obse a ion we ha e o li es ock owne ship among Khoesan wo ke s is om 1798, almos 150 yea s a e he es ablishmen o he colony, when he expansion o he colony’s bo de s had weakened he Khoesan o he poin ha e y ew could s ill li e independen ly. A ha poin , he a e age Khoesan esiding on a Eu opean a m had i e ca le and wen y- h ee sheep (Ross 1993). I is likely ha he numbe s we e la ge a he end o he se en een h cen u y, when Khoesan socie ies had no ye collapsed. The need o ie Khoesan labou o he a m c ea ed incen i es o he se le s in he sou h-wes e n Cape o con ol mo e pas u eland han hey ac ually needed o hei own a ming. This no only acili a ed dispe sion a he in o he in e io , as we saw in Chap e 3, bu also c ea ed he inc eased concen a ion o land and o he asse s ha Guelke and Shell (1983) obse ed. As he se le s Unequal We S and 123 lacked he economic and poli ical capaci y o push he Khoesan ou on he labou ma ke , hose who had he means gained con ol o e la ge amoun s o land in o de o pull Khoesan o hei a ms as he de s. These a me s would e en ually cons i u e an impo an pa o he Cape gen y. Gende , inhe i ance and weal h accumula ion To wha ex en did he weal hy manage o keep hei o unes wi hin hei amilies ac oss gene a ions? The ac o s ha explain he ini ial accumula ion o capi al a e no he same as hose ha explain hei pe sis ence. To unde s and he con inui y o weal h, we need o d aw ou a en ion o inhe i ance p ac ices a he Cape. The Cape ollowed he Du ch model o pa ible inhe i ance. The implica ion o his sys em was desc ibed in Chap e 3. Ross (1993: 141) summa izes i in his way: p ope y was di ided and edi ided a egula in e als … This en ailed ei he ha he a m as a unning conce n would be sold, and he p oceeds di ided among he hei s, o ha a alua ion would be made o he es a e wi h he hei s ecei ing hei po ions in cash. In he la e case, he su i o would in all p obabili y be o ced o ake ou a loan o pay o his o he child en and s epchild en. The sys em o pa ible inhe i ance made i mo e di icul o pass on weal h ac oss gene a ions han was he case in p imogeni u e sys ems, in which weal h was bequea hed o he oldes child (mos commonly he son). Schola s like Fou ie and on Fin el (2010a) also emphasize he nega i e e ec s he sys em o inhe i ance could ha e on weal h accumula ion. No ably, i led o smalle a ms. Con lic s o e p oduc i e esou ces among amily membe s we e common, o en ending wi h one o mo e membe s mo ing a he in o he in e io in sea ch o esh land (see Chap e 3). The a me s in he sou h-wes e n Cape owned hei a ms, while se le s a he in o he in e io leased hei land om he VOC on an annual basis (see Chap e 3). Thei land could no be subdi ided o sold. Loan a me s did own pe manen buildings, a m equipmen and sla es, howe e , and hese could be auc ioned a e he dea h o a a me . In p ac ice, buying he pe manen buildings ga e he pu chase he igh o ake o e he lease con ac . Con empo a y wi nesses a gued ha i he inhe i ance sys em allowed a ms o be sold o o subdi ided, i would s ymie long- e m accumula ion. In he 130 C ea ing he Cape Colony g ew oo as , o i s abili y o ex ac en s diminished, he sys em would all apa , ei he g adually o in a mo e d ama ic way. This implies ha a na u al s a e is inhe en ly agile. As long as he on ie is open and expanding, new compe ing eli es may a ise as he economy e ol es. These new eli es ei he ha e o be included in he coali ion o ep essed. I we accep his a gumen , hen we canno po ay he ela ionship be ween colonial se le eli es and he au ho i ies as s a ic. To ease ou he changing ela ionship be ween he VOC, he se le a ming communi y and he Khoesan, I begin wi h Nugen ’s classi ica ion o social con ac s. In his a icle ‘S a es and social con ac s in A ica’ (2010), Nugen c i ically e iews s udies on s a es in A ica and concludes ha hey igno e he way he ela ionships be ween he s a es and he popula ions a y. To cap u e he a ia ion, Nugen p oposes h ee di e en con ac s: coe ci e, p oduc i e and pe missi e. A coe ci e con ac is one in which he igh o go e n is p edica ed on he ule s’ capaci y o make li e mise able o he subjec s. In ex eme cases, he uled gi e away hei poli ical oice in exchange o no being exposed o p eda o y ac s om he au ho i ies. A p oduc i e con ac ep esen s he opposi e o a coe ci e one. In his case, he ule s and he uled ha e, h ough nego ia ions, ag eed ha he ule s will ha e he igh o go e n as long as hey imp o e he li e o he uled. Las ly, he pe missi e con ac ep esen s a si ua ion in which he go e nmen has he capaci y o be coe ci e bu exe cises es ain in o de o ensu e some kind o de ac o compliance. Al hough Nugen ’s classi ica ion scheme allows o a mo e complexpic u e o he A ican colonial and pos colonial s a e, i has some limi a ions. Fi s , when iden i ying he social con ac s, Nugen la gely ocuses on he ela ionship be ween he go e nmen and ce ain sec ions o he popula ion. In he case o Ghana, o example, he ocus is on he ela ionship be ween he au ho i ies and he cocoa g owe s, he la e being c ucial o he colonial economy, bu ne e heless cons i u ing a mino i y o he popula ion. Second, Nugen does no allow o se e al social con ac s o exis in pa allel. I could be ha he exis ence o one con ac be ween he s a e and pa s o he popula ion is p edica ed on he p esence o ano he con ac be ween he s a e and o he pa s o he popula ion. To ake an ex eme example, one may conclude ha he ela ionship be ween he go e nmen and he whi e mino i y in apa heid Sou h A ica was based on a p oduc i e con ac ha elied hea ily on a coe ci e con ac be ween he go e nmen and he majo i y o he black A ican popula ion. Las ly, Nugen does no ake in o accoun ha con ac s may change o e ime as a consequence o shi ing powe balances Eli es, Coali ions and Se le Resis ance 131 be ween he au ho i ies and he a ious segmen s o he popula ion as well as among hose di e en segmen s. In his chap e , I build on Nugen ’s no ion o social con ac s by allowing o all h ee ypes o exis in pa allel and o depend upon one ano he . I also allow o con ac s o change o e ime. Using he exis ence o mul iple social con ac s and aking he limi a ions o he VOC’s powe s, he di e gen in e es s o he se le s and he agency o he Khoesan in o accoun , his chap e p o ides an analysis o he poli ical economy o he Cape Colony. Typologies o se le colonialism A numbe o schola s ha e poin ed ou ha se le colonies became ela i ely independen o me opoli an ule. This au onomy is o en used o dis inguish be ween se le colonies and o he o ms o colonialism (see Ve acini 2013). Se le s used hei ela i e independence o asse con ol o e he colonial au ho i ies and ensu e ha an ins i u ional o de was es ablished ha ac ed in hei in e es s (A ighi 1970; F ankema, G een and Hillbom 2016). Mosley (1983) and Os e hammel (1997) a gue ha wha dis inguished se le om non- se le colonies in wen ie h-cen u y A ica was no he numbe o se le s, bu whe he se le s’ in e es s had ep esen a i es in he local colonial go e nmen . In sho , a p econdi ion o a se le colony o eme ge and sus ain i sel was ha he se le s’ in e es s we e p o ec ed by he local colonial au ho i ies. Elkins and Pede sen (2005) ske ch wha hey e e o as a ‘beginning o a ypology’ o se le colonialism in an a emp o mo e beyond he assump ion o a gene ically s ong ela ionship be ween se le s and local au ho i ies. The ypology concen a es on ou a iables: high and low le els o ins i u ionaliza ion o se le p i ilege and high and low le els o he inco po a ion o se le s in o go e nance. Figu e 6.1 summa izes his ypology and shows whe e Elkins and Pede sen place di e en se le colonies. Wha he ypology sugges s is ha he e a e cases whe e se le s ha e been p i ileged ye had no in luence o e local adminis a ion (e.g. Nazi Ge many’s occupa ion o Eas e n Eu ope). The e a e also cases in which se le s ha e been pa o he local go e nmen bu ha e no expe ienced a ou able ea men (e.g. No he n I eland). The e a e e en cases whe e he se le s ha e been nei he ep esen ed no p i ileged (e.g. Japan’s colonies in Asia). Unlike Elkins and Pede sen’s cases om he wen ie h cen u y, he Cape was go e ned by a Company and was no a na ion s a e. A cha e ed company’s 132 C ea ing he Cape Colony chie objec i e is o c ea e p o i s o i s sha eholde s by engaging in global ade. The inancial in e es s o he Company we e always he o e all p io i y. Fo a cha e ed company, con ol o e e i o y is ne e an objec i e in i sel , bu can be deemed necessa y o make ade p o i able. Fu he mo e, he Cape Colony was no a sou ce o he luc a i e spices and o he ade goods which Asian se lemen s o e ed. I was me ely a ansi poin o Asian luxu y goods (Baa man 2015). To a oid expenses, he Lo ds XVII o he VOC pushed he local au ho i ies o keep he colony’s ecu en unning cos s as low as possible. In a way, he VOC ep esen ed an ex eme o m o a ‘ga e-keeping s a e’, a concep in oduced by he A ican his o ian F ede ick Coope in 2002 o desc ibe a colonial and pos colonial A ican s a e ha is adminis a i ely and inancially weak and whose exis ence depends on making e enues by axing ade, o ‘gua ding he ga e’. To make he Cape Colony a p o i able en e p ise, he VOC had o ensu e ha Eu opean se le s p oduced a su plus and ha i would, o a la ge ex en , con ol he p o i s. Using Elkins and Pede sen’s ypology, one would place he Cape Colony in he same posi ion as Kenya and Sou he n Rhodesia. The expansion o he Cape Colony and i s bo de s was no pa o a long- e m, planned p ocess. Fu he , he inancial capaci y o he local au ho i ies was se e ely limi ed because he managing Lo ds XVII emained unwilling o make any majo in es men s in he colony and, h oughou he eigh een h cen u y, he income om expo s was no su icien o make ends mee (as discussed u he below). While his made he colony inancially weak, i also mean ha he se le s could ac ei he ela i ely Figu e 6.1 G aph o se le p i ilege and inco po a ion in o go e nance. Sou ce: Elkins and Pede sen (2005: 5). Eli es, Coali ions and Se le Resis ance 133 independen ly o he local colonial au ho i ies and he VOC headqua e s in Ams e dam o could in luence he local au ho i ies o ensu e ha hey wo ked in he se le s’ in e es s, al hough he se le s lacked o mal ep esen a ion beyond he dis ic le els. Gi en ha we a e dealing wi h a on ie economy, we need o ecognize he impo ance o indigenous agency. The ac ha bo h he VOC and he se le s ini ially depended on he Khoesan o ade and labou made he si ua ion delica e. To a oid losing a majo ading pa ne , he VOC needed o ensu e ha he Khoesan socie ies did no comple ely collapse, hough a he same ime i was weakening hei pas o al economies by deple ing hei li es ock and employing hem as labou e s. To analyse he poli ical economy a he Cape we hus need o ake in o accoun a web o di e en in e es s. To unde s and hese we i s need o look a he iscal capaci y o he VOC. The Company’s es ic ed inancial and adminis a i e capaci y is an essen ial elemen o unde s anding he poli ical economy o he Cape. The iscal capaci y o he VOC As he VOC was a sha eholde company, sho - e m p o i abili y was impo an o i . This p o i abili y would be ensu ed h ough con ol o ade a he han long- e m in es men in p oduc i e ac i i ies. Con ol o e i o ies was hus ne e an aim pe se, al hough ha was some imes deemed necessa y as a means o s imula ing ade. To enable i o compe e economically wi h he Po uguese and Spanish in he Eas Indies, he VOC was g an ed he igh o monopolize all in e egional ade in he Eas Indies and o con ol ade in all o he a eas ha we e unde i s con ol, including he Cape Colony (Fou ie, Jansen and Sieb i s 2013). As men ioned in Chap e 2, he Lo ds XVII ne e planned o es ablish a pe manen Eu opean se lemen a he Cape. They in ended only o ound a small pe manen s a ion a he Cape o Good Hope o u he acili a e ade wi h he Khoesan and passing ships. They ne e conside ed he Cape s a ion o be comme cially p o i able o he VOC and consequen ly he Cape was no one o he Company’s p io i ies (Ross 1989). Once he expansion o he colonial bo de s began, i was le o he VOC a he Cape o und i s own ecu en cos s. I s main sou ce o e enue came om ade, bu e enues om expo s emained low. Whea was p ima ily expo ed o ac o ies in Asia whe e i was Table 6.1 Re enues, expendi u es and de ici s (in guilde s) as pe cen ages o GDP. Yea Re enue Expendi u e De ici GDP Re enue/GDP a io Expendi u e/ GDP a io De ici /GDP a io 1777 200,000 454,000 254,000 2,391,000 8% 19% 11% 1787 282,000 1,586,000 1,304,000 4,014,000 7% 40% 32% 1789 341,000 1,744,000 1,403,000 3,472,000 10% 50% 40% 1791 … 1,300,000 … 3,499,000 … 37% … 1792 416,000 758,000 342,000 3,253,000 13% 23% 11% 1794 588,000 818,000 230,000 … … … … Sou ce: de Kock (1924: 78). Eli es, Coali ions and Se le Resis ance 135 made in o b ead o Eu opean ade s, while wine p oduced a he Cape was long ega ded as in e io and was no in demand in Eu ope o Asia. Expo s o li es ock p oduc s we e limi ed. Sal ed mea could be sold as p o isions o passing ships, bu he e was no expo ma ke as he e was no echnology o keeping he mea esh o long pe iods o ime (Ross 1989). The limi ed e enue om ade, in combina ion wi h an ine icien sys em o collec ing axes, mean ha h oughou i s pe iod o ule, he VOC was unning on budge de ici s (Ross 1989). O , o quo e G oenewald (2012: 1), he VOC a he Cape ‘ an a an eno mous loss’. Es ima es by de Kock (1924) o he la e eigh een h cen u y, shown in Table 6.1, con i m his iew. To manage he si ua ion, he VOC had o keep hei ecu en expendi u es as low as possible. In 1795, o al expendi u e was 790,000 guilde s, o which 44 pe cen wen o law and o de , 43 pe cen o he mili a y and he emainde o main enance o he ading pos s (de Kock 1924). The dis ic s we e expec ed o be sel - inancing. Dis ic au ho i ies we e esponsible o in as uc u e (main enance o oads, b idges and chu ches) and emune a ion (sala ies o hose employed by he dis ic ). The dis ic au ho i ies commonly an de ici s. Table 6.2 shows es ima es o accumula ed deb o selec ed yea s in he S ellenbosch and Swellendam dis ic s. Wo h no ing is he subs an ial inc ease in deb in S ellenbosch, he home o many o he weal hie a me s a he Cape. The inancially weak VOC esembles he ‘ga e-keeping’ s a es o wen ie h- cen u y colonial A ica (Coope 2002). I s exis ence depended on he possibili y Table 6.2 Accumula ed deb o S ellenbosch and Swellendam dis ic s in selec ed yea s, 1702–93. S ellenbosch Swellendam Yea Guilde s Yea Rixdolla s 1702 4,025 1760 4,333 1720 2,000 1770 4,333 1729 4,000 1780 4,333 1740 4,000 1753 9,000 1766 18,500 1794 13,000 Sou ce: P. J. Ven e (1960: 157), aken om Fou ie, Jansen and Sieb i s (2013: 64). 136 C ea ing he Cape Colony o making e enue by con olling ade. Being a ‘na u al s a e’, his in u n equi ed he c ea ion o a coali ion be ween he au ho i ies and he se le eli es. T ade, p o i s and he se le eli es The Cape Colony di e ed om o he possessions unde VOC con ol in a c ucial way: i s economy o e ime became dependen on es ablishing p o i able Eu opean ag icul u e. Wi h he excep ion o he Banda a chipelago in he Du ch Eas Indies, he e we e no o he a eas con olled by he VOC whe e a la ge numbe o Eu opean se le s o ganized ag icul u al p oduc ion. Al hough he ole o he Cape in he global economy was p ima ily o p o ision passing ships, i was a key conce n o he VOC o ea n as much e enue as possible om his ade. The ini ial idea o ensu ing p o isions h ough ade wi h he Khoesan was g adually abandoned as Eu opeans became he main supplie s o mea , g ain and wine (Ross and Sch ikke 2012). In An Inqui y In o he Na u e and Causes o he Weal h o Na ions, Adam Smi h c i icized he Du ch p ac ice o go e ning hei colonies h ough p i a e monopolies, saying ha : ‘O all he expedien s ha can well be con i ed o s un he na u al g ow h o a new Colony, ha o an exclusi e company is undoub edly he mos e ec ual’ ([1776] 1982: 44). This c i icism was g ounded in his i m belie ha ee ade was undamen al o long- e m economic g ow h as i would spu he di ision o labou . Fo him, he VOC’s p ac ices ep esen ed he e y opposi e, as he assumed ha hey had a monopoly o e ade in he e i o ies unde hei con ol. A numbe o schola s wo king on he Cape ha e con i med his o e all nega i e imp ession. The VOC used i s monopsony powe o e ade a he Cape o buy Eu opean se le p oduce a p ices ha i se . I hen sold he p oduce o he passing ships a much in la ed p ices (Wo den 1985; Fou ie, Jansen and Sieb i s 2013). This would imply he exis ence o a coe ci e con ac ha enabled he VOC o ex ac en s om he se le - a me s. As expec ed, and as will be discussed below, his c ea ed ensions be ween he VOC and pa s o he se le communi y. The VOC’s con ol o e ade should no , howe e , be exagge a ed (Ross 1983, 1989). As an Duin and Ross (1987) poin ed ou , i he VOC ex ac ed p o i s om a me s by se ing low p ices and i he le el o ex ac ion inc eased o e ime, we would likely see a decline in ag icul u al p oduc ion. This was no he case, as I showed in Chap e 3. Ins ead, p oduc ion inc eased h oughou he eigh een h cen u y. Van Duin and Ross (1987) conclude ha his o ians in Eli es, Coali ions and Se le Resis ance 137 gene al ha e ended o unc i ically accep he iews o he VOC’s opponen s, mos no ably he Cape Pa io s, who complained abou VOC con ol o e he ma ke and i s p icing policies (see he u he discussion below). The Pa io s did no necessa ily ep esen he in e es s o he se le - a me s a la ge. In eali y he VOC’s abili y o con ol ade was es ic ed. I had no di ec con ol o e impo s, which we e in he hands o p i a e ade s (Ross and Sch ikke 2012). I s con ol o e expo s was only pa ial. In he 1770s, he VOC di ec ly con olled only one- i h o he wine ma ke and hal o he mea ma ke . The es was in he hands o licensed p i a e ade s (pach e s) who sold he wine and mea ei he on he local ma ke o o passing o eign ships. The VOC did, howe e , domina e he whea ma ke ( an Duin and Ross 1987). I s con ol o he domes ic ma ke was pa ial and indi ec , bu ne e heless luc a i e. Ins ead o di ec ly engaging in ade, he VOC con olled he e enue s eam by selling ading licences (pach s) o he se le s, which b ough in subs an ial e enue, as Table 6.3 shows. Being able o g an one o hese licences allowed indi idual se le s o accumula e signi ican amoun s o weal h (as seen in Chap e 5). I was a sys em in which he au ho i ies managed o es ablish a coali ion be ween he VOC and he se le eli es, egula ed by a p oduc i e social con ac om which bo h pa ies bene i ed. In gene al, he p ices paid by pach e s on he auc ion loo s we e lowe han hose paid by he VOC. This is pa ly due o he na u e o he ade. The se le s bough p oduce a he auc ions ha ook place h oughou he colony, while he VOC only bough p oduce a he Cape Town ma ke . When selling o he VOC, he se le - a me s co e ed hei own anspo cos s, while hese we e deduc ed by he p i a e ade s buying ag icul u al ou pu a auc ion. As we do no know he loca ions o he a ious auc ions, i is impossible o calcula e he anspo cos s. Ra he han compa ing le els, we can look in o p ice ends. Wo den’s (1985) es ima es o p ices a he auc ions show ha auc ion p ices o Table 6.3 VOC income om pach as p opo ion o o al e enues (selec ed yea s). Yea P opo ion o pach income (%) 1777 41.1 1787 68.9 1789 65.2 1792 30.6 Sou ce: an Duin and Ross (1987: 148–9). 138 C ea ing he Cape Colony wine emained ai ly s able o e ime, while whea p ices declined signi ican ly in he pe iod om 1741 h ough he 1770s. The same did no happen wi h he whea p ices paid by he VOC, which emained s able h oughou he pe iod (see Table 6.4). The declining p ices a auc ion equi e an explana ion. Was he ma ke sa u a ed, as Guelke and Shell (1983) ha e claimed, o was he decline a consequence o local ade s’ inc eased abili y o ex ac a la ge sha e o p o i s om he whea a me s, using non-ma ke measu es? The pe iod in which p ices o whea declined on he auc ions coincided wi h he pe iod when ax e asion peaked. One possible eason o he inc eased ax e asion was he whea a me s’ de e io a ing inancial si ua ion. In he 1740s, hey began o complain o he VOC abou he Company’s i egula demand o whea . The a me s o en had o s ockpile p oduce be o e i could be sold ( an Duin and Ross 1987). I ha e no e idence ha he numbe o auc ions declined du ing he pe iod o ha he e was a declining demand o whea on he expo ma ke . On he con a y, an Duin and Ross’s es ima es show ha he demand o whea inc eased by a ac o o 5–6 be ween 1704 and 1793. Al hough di icul o subs an ia e empi ically, one possible eason o he declining p ices is ha he p i a e ade s could use he i egula demand o whea om he VOC o hei ad an age. They we e willing and able o ade egula ly wi h he whea a me s, bu only a declining p ices. Table 6.4 A e age p ice o whea (pe muid) and wine (pe league ) in ixdolla s paid by VOC and a he auc ion loo s. Yea Whea (pe muid) Whi e wine (pe league ) 1716 2.0 21 1720 2.2 22 1730 2.4 26 1740 2.4 28 1750 1.6 30 1760 1.5 38 1770 1.0 22 1780 2.4 38 1790 2.6 35 1800 2.8 30 O icial Company p ice 2.6 ( o 1741) 2.5 (a e 1741) 27 Sou ce: Wo den (1985: 69). Eli es, Coali ions and Se le Resis ance 139 The sys em o ade egula ions ha ga e he VOC only indi ec con ol c ea ed a s ong coali ion be ween he uling and economic eli es a he Cape. As I a gued in Chap e 5, i was he weal hies se le s who had he means and esou ces o con ol domes ic ade by buying ading licences. This ga e hem he oppo uni y o ex ac en s om he majo i y o he se le - a me s. The VOC gene a ed subs an ial e enues om his sys em, and he coali ion o eli es las ed only as long as he VOC bene i ed om i inancially. Meanwhile, he sys em gene a ed e enues only as long as a b oad g oup o se le s could p oduce a su plus o be sold on he domes ic ma ke s, which in u n depended o a la ge ex en on hei abili y o access and con ol enough labou e s. Landlo ds and labou : P oduc i e o pe missi e con ac s? As I ha e shown in he p eceding chap e s, a undamen al d i e o se le ag icul u e was Eu opean a me s’ access o labou . In Chap e s 4 and 5, I a gued ha he e a e good easons o belie e ha he Khoesan played a mo e impo an ole – o bo h he es ablishmen and subsequen expansion o se le ag icul u e a he Cape – han p e iously acknowledged. To unde s and how he indigenous people we e inco po a ed in o he Cape se le economy in gene al and how he se le s managed o make p oduc i e use o hem, I analyse he implici con ac s egula ing he ela ionship be ween he Khoesan and he se le s. Du ing he eigh een h cen u y, he VOC ins i u ed legisla ion ha limi ed he eedom o he Khoesan. G adually, in o mal ways o con olling Khoesan labou we e o malized h ough a ious legisla i e measu es (Elphick and Giliomee 1989b). Tha is, o e ime, he con ac be ween he VOC and he la ge g oup o se le s was ans o med om pe missi e o p oduc i e. This de elopmen may seem puzzling. Why was he VOC no p epa ed o p o ide legisla ion o suppo he coe cion o Khoesan labou in he ini ial yea s, when Khoesan communi ies we e s ill ai ly independen ? And why did i mo e owa ds mo e ep essi e legisla ion in he la e hal o he eigh een h cen u y when mos o he Khoesan communi ies had al eady los hei independence and had li le choice o he han o be inco po a ed in o he colonial economy? I a gue ha se e al ac o s explain he VOC’s beha iou , bu in mo e gene al e ms i was only when he on ie was closing ha he VOC achie ed he capabili y o s ep in and legally back he labou ela ions al eady in place on he se le a ms.