The Evolution of 'Portuguese' Identity: Luso-Africans on the Upper Guinea Coast from the Sixteenth...

The Evolution of 'Portuguese' Identity: Luso-Africans on the Upper Guinea Coast from theSixteenth to the Early Nineteenth CenturyAuthor(s): Peter MarkReviewed work(s):Source: The Journal of African History, Vol. 40, No. 2 (1999), pp. 173-191Published by: Cambridge University PressStable URL: http://www.jstor.org/stable/183545 .

Accessed: 12/02/2013 06:21

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp

.JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range ofcontent in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact [email protected].

.

Cambridge University Press is collaborating with JSTOR to digitize, preserve and extend access to TheJournal of African History.

http://www.jstor.org

This content downloaded on Tue, 12 Feb 2013 06:21:28 AMAll use subject to JSTOR Terms and Conditions

http://www.jstor.org/action/showPublisher?publisherCode=cup

http://www.jstor.org/stable/183545?origin=JSTOR-pdf

http://www.jstor.org/page/info/about/policies/terms.jsp


Journal of African History, 40 (I999), pp. 173-191 Printed in the United Kingdom I73 ? I999 Cambridge University Press

THE EVOLUTION OF 'PORTUGUESE' IDENTITY: LUSO-AFRICANS ON THE UPPER GUINEA COAST

FROM THE SIXTEENTH TO THE EARLY NINETEENTH CENTURY*

BY PETER MARK

Wesleyan University

DURING the late fifteenth and early sixteenth century, Portugal established a trading presence along the Upper Guinea Coast from Senegal to Sierra Leone. Emigrants from Portugal known as lanfados' - some of them Jews seeking to escape religious persecution2 - settled along the coast, where many of them married women from local communities. By the early sixteenth century, Luso-Africans, or 'Portuguese' as they called themselves, were established at trading centers from the Petite Cote in Senegal, south to Sierra Leone. Descendants of Portuguese immigrants, of Cape Verde islanders, and of West Africans, the Luso-Africans developed a culture that was itself a synthesis of African and European elements. Rich historical documentation allows a case study of the changing ways Luso-Africans identified themselves over the course of three centuries.3

The earliest lanf_ados established themselves along the coast as commercial middlemen between African and European traders4 and as coastal traders between Sierra Leone and Senegambia.5 Their position was formally

* I wish to thank Thomas Spear and David Robinson for their extensive editorial and substantive comments and Amy Retlew for editorial advice. A Wesleyan University travel grant enabled me to carry out research in France and Africa.

1 See Maria Emilia Madeira Santos, 'Origem e desenvolvimento da colonizacao: Os primeiros 'lancados' na costa da Guine, aventureiros e comerciantes', in Luis de Albuquerque (ed.), Portugal no Mundo (Lisboa, I988), ii, I25-36.

2 The Inquisition was reestablished in Portugal in I536. Several Cape Verdean merchants, including Andre Alvares d'Almada (fl. I590) and

Andre Donelha (fl. I570-I625) wrote accounts of the coastal trade. Their descriptions present the Luso-Africans from the perspective of the Cape Verdean elite.

4 On the economic history of Luso-Africans see George Brooks, Africa, I000-I630

(Boulder, I993). See also idem, 'Perspectives on Luso-African commerce and settlement in the Gambia and Guinea-Bissau region' (Boston University, African Studies Center, Working Paper no. 24) (Boston, I980); Jean Boulegue, Les Luso-Africains de Se'ne'gambie (Lisbon, I989). For the seventeenth century Petite COte, see Nize Isabel de Moraes, A la decouverte de la Petite C65te au XVIIe siecle (2 vols.) (Dakar, I993). For Serra Leoa, the southeastern part of 'Guinea of Cabo Verde', the history of the Luso-African community in the sixteenth and seventeenth century is covered by the richly annotated translations of Portuguese, Dutch and English documents published by P. E. H. Hair in Africana Research Bulletin between I 970 and I98I.

5 In I623, Dierick Ruiters, who had travelled to Guinea in the first decade of the century, described Luso-African trade at Cacheu: 'The trade of the Portuguese in Cacheu is of two kinds, first, trade from Portugal, second coastal trade ... mostly undertaken in small ships, pinnacles and launches, by Portuguese who live on Santiago Island'; Dierick Ruiters, Toortse der Zeevaert (P. E. H. Hair, trans.), in 'Sources for early Sierra Leone (2): Andrade (1582), Ruiters (I623), Carvalho (I632)', Africana Research Bulletin, I(I970), 5I.



I74 PETER MARK

discouraged by the Portuguese Crown until the second decade of the sixteenth century, but they nevertheless played an important role in trade with Portugal and the Cape Verde islands. Lanf ado communities were permanently settled on the Petite Cote, while in Sierra Leone and Rio Nunez much early commerce was in the hands of lanf_ados who sailed there regularly from S. Domingos, north of present day Bissau.6 The offspring of these lanfados and African women were called filhos de terra and were generally considered to be 'Portuguese'.7

Throughout the sixteenth century, the descendants of the lanf ados maintained close commercial ties with the Cape Verde islands. Cape Verdeans were themselves the offspring of mixed Portuguese and West African marriages. Sharing elements of a common culture and united by marriage and economic ties, mainland Luso-Africans and Cape Verdeans represented a socially complex and geographically dispersed community. Cape Verdeans, like mainland Luso-Africans, resolutely maintained that they were 'Portuguese', and both sub-groups employed the same essentially cultural criteria of group identification.

THE CHARACTERISTICS OF LUSO-AFRICAN IDENTITY

By the second half of the sixteenth century, membership in the Luso-African community was not associated with physical features. Rather, 'Portuguese' were defined, broadly speaking, by cultural and socio-economic characteristics.8 To be 'Portuguese' was to be a trader, much as to be 'Juula' in Senegambia implied being a long distance merchant.9 'Portuguese' identity

6 For references to this presence, see Francisco de Andrade, 'Relacao de Francisco de Andrade sobre as ilhas de Cabo Verde (I582)', in A. Brasio, Monumenta Missionaria Africana (Lisboa, I964), second series, iII, I02. Donelha exists in both English and French translations. The English translation as well as the annotations in both editions are by P. E. H. Hair. I refer primarily to the French edition: A. Teixeira da Mota (ed.) and Leon Bourdon (trans.), Description de la Serra Leoa et des Rios de Guine' du Cabo Verde (I625) (Lisbon, I977). Donelha, I63, mentions a tangomao named Luis Lopes Rabelo who had spent years at Rio Nunez. He also writes that south of the Rio Grande and as far as Sierra Leone 'There is no large river to which ships might travel to trade, but the barques of the tangomaos do come there to purchase a few Blacks, wax, and ivory, by traveling from one port to another', I79.

7 See Madeira Santos, 'Os primeiros " lancados" 8 Elsewhere, I have analyzed the material culture upon which 'Portuguese' identity

was based; Peter Mark, 'Constructing identity: sixteenth and seventeenth century architecture in the Gambia-Geba region and the articulation of Luso-African ethnicity', History in Africa, 22 (I995), 307-27. See also idem, ''Portuguese' architecture and Luso- African identity in Senegambia and Guinea, I730-i890', History in Africa, 23 (I996). Having focused on Luso-African architecture in these earlier essays, I shall de-emphasize the discussion of architecture in the following pages, but focus instead on the dynamics that underlay the process of 'Portuguese' identity formation.

9 In contemporary Mande society, one's profession is presumed to reflect one's ancestry in the case of those belonging to nyamakalaw groups. Individuals or groups may, however, change profession, a change sometimes recognized by ascription of a new social identity to the individuals in question. See Robert Launay, 'The Dieli of Korhogo', in David Conrad and Barbara Frank (eds.), Status and Identity in West Africa (Bloomington, I995), I53-69.



THE EVOLUTION OF 'PORTUGUESE' IDENTITY 175

Cape Verde Islands Petite Cote

[3 ~~~~Joal SENEGAMBIA

F-- 350 miles f

The main Cape Verde ] --tuCacheu e_ _ l

Map I. enegamba andetegiap Verou Isad hoigtain etr Islanandsce

into~ ~ ~ S Droeo romino. so Cigulnconon vocarim y eiedfo

Pothgee main Cap Veramatca stucur Geieb etAlni

theassmiltiv ntur ofthecutur wih wichitwasassciaed

0~~~

Bissagos I.

The3~iI~ d l n o w n0 miles 50 0 Miles 50

Map i. Senegambia and the Cape Verde Islands, showing trading centers and 'escales'.

was also defined by language, initially Portuguese, but gradually developing into Creole or Crioulo. As Crioulo conjoins vocabulary derived from Portuguese with a grammatical structure derived from West Atlantic languages, it is the result of a process of cultural assimilation. Luso-African language thus emphasized the hybrid aspect of the community, and defined the assimilative nature of the culture with which it was associated.

The development of Crioulo was a long process. While historical sources do not permit us to date this process precisely and regional variants may have formed at different times, it is likely that a form of Crioulo evolved in the Petite CO'te-Gambia region by the first half of the seventeenth century. In 15~82, Francisco de Andrade, Sargento-mor of Santiago, wrote that African traders on the Petite C6^te spoke French and Spanish, but he does not mention any trading language being spoken at the time.'0 In 154 Andre' Alvares de Almada observed that Bainunk of the Cacheu region, 'because of the close contact they have always had with our people, speak Portuguese

10 ' RelaVao de Francisco de Andrade', in Brasio, Monumenta Missionaria Africana, Ini, I03.



I76 PETER MARK

very well'." And in i6oo, Soares de Albergaria reported that many Africans along the Rio de S. Domingos near Cacheu also spoke Portuguese.12 Likewise in i6o6, the Jesuit, Balthasar Barreira, wrote of a Sierra Leonean king who, 'having been brought up with the Portuguese, understands and speaks our language quite reasonably'.13 For extended conversations, the king could meet with Portuguese who were fluent in his language.14 That conversations were in Portuguese or local African languages strongly suggests the absence of a common trading language, like Creole, in Sierra Leone at that time.15 In I602, however, Pieter de Marees noted that local people at Portudal on the Petite Cote 'have their own language, a mixture of many different languages'." This reference strongly suggests the existence of a hybrid or trading language along the Petite Cote by the early seventeenth century. Given the preponderance of Luso-Africans and Portuguese lanf ados along this part of the coast, such a language would surely have incorporated Portuguese.

Further south in the Gambia in i66i, English traders relied on an African marabout who spoke Portuguese as their interpreter.17 Gambian Portuguese may already have been transformed into a non-standard version of the language, however, as suggested by a i 646 report of an Andalusian Capuchin mission. At Juffure in the North Bank kingdom of Niumi, the missionaries reported 'many Christians ... who have never taken Confession and who speak a little Portuguese, by means of which they express themselves as best they can'."8 This reference to Luso-African traders, together with de Marees' earlier report, implies that a form of Crioulo or non-standard Portuguese was spoken on the Petite Cote and along the lower Gambia by the early to mid- seventeenth century.

Another characteristic of 'Portuguese' identity was their Catholic religion. Along the Upper Guinea Coast, it was common for each group to be characterized by its own religious practices. Thus, Floup (Joola) and Bagnun (Bainunk) both had their own shrines and associated rituals, although a powerful and efficacious shrine might be acquired by one group from another.'9 Luso-African religion was actually an amalgam of Christian,

" A. Teixeira da Mota (ed.) and P. E. H. Hair (trans., intro. and notes on chs. I3-19),

'An interim and makeshift edition of Andre Alvares de Almada's Brief Treatise on the Rivers of Guinea... ', Department of History, University of Liverpool, 1984, 87. I wish to thank Professor Hair for making this translation available to me.

12 Relavao de Lopo Soares de Albergaria', in Brisio, Monumenta Missionaria Africana, IV, 3.

13 See P. E. H. Hair, 'Sources on early Sierra Leone (5): Barreira', Africana Research Bulletin, 5 (I974), 4, 90. 14 Ibid. 4, 97.

15 Ibid. 4, 115. Hair observes that many African ruling groups in Sierra Leone, including both men and women, 'had been in extensive contact with the Portuguese in childhood and had acquired some European customs and a knowledge of the Portuguese language'. 16 de Marees, cited in I, 55.

17 Anonymous, cited in Nize de Moraes, A la decouverte de la Petite Co6te, II, 4I6. 18 Nize de Moraes, A la decouverte de la Petite Co6te, II, 36I (italics mine). Juffure was

adjacent to a Luso-African village called San Domingos (not to be confused with S. Domingos).

19 For an excellent study of the inter-ethnic and even international following attracted to a Manjak shrine, see Eve Crowley, 'Contracts with the spirits: religion, asylum and ethnic identity in the Cacheu region of Guinea-Bissau' (Ph.D. dissertation, Yale University, I990).



THE EVOLUTION OF 'PORTUGUESE' IDENTITY I77

Jewish and African practices. Among lanfados settled on the Petite Cote in the sixteenth and early seventeenth century were Jews and 'New Christians' (recently converted Jews).20 Few priests visited the coast, so Christians lacked regular access to the sacraments. As a result of such isolation, some 'Portuguese' participated in African religious rituals, but unorthodox Christian practices are poorly documented. This is hardly surprising, as 'Portuguese' who followed syncretic rituals would not have broadcast the fact to mission priests.

New Christians were granted the right to settle in Portugal's overseas possessions in i6o i. Additional New Christians soon arrived on the Petite COte.22 At Portudale in i6o6, Father Barreira noted 'a village of ioo Portuguese who follow the laws of Moses '.23 Some of these Jews had previously lived in Amsterdam and established ties with Dutch merchants.24 In Senegambia, many New Christians took advantage of their distance from Portuguese secular and religious authority to return to their ancestral religion. They maintained discretion, however, as Portuguese reprisals against their commerce with the Dutch were a constant threat.25 Hence, 'Portuguese' Jews may well have been content to be perceived as Christians by outsiders.

The Jewish presence on the Petite Cote gradually diminished after the first quarter of the seventeenth century, but 'Portuguese' Christians whose religion was mixed with African rituals increased over time. References describe both Africanized 'Portuguese' and Christianized Africans. In I 607, Father Barreira decried the presence in Sierra Leone of

Christian Blacks who ... by contact with the heathen had so forgotten the obligations of our holy faith that either they possessed chinas26 themselves or they allowed their slaves to do so and they had dealings with these chinas and made them

27 offerings.

Elsewhere, he observed Portuguese 'turned wild, whose very way of life is

20 Almada describes a Jew named Ferreira who travelled from the Gambia to the Grand Fulo, one of whose daughters he married. 'Almada's Brief Treatise on the Rivers of Guinea', ch. I, 23.

21 See Nize de Moraes, A la d6couverte de la Petite Co6te, I, 32. The right to move to the territories was revoked by Philip III in i6io.

22 An important source for seventeenth century Jewish presence in Guinea is the Jesuit Mission to Cape Verde (effectively I604-17). See Brisio, Monumenta Missionaria Africana, IV, 97-113, 159-74. See also P. E. H. Hair, 'Sources on early Sierra Leone (7): Barreira', Africana Research Bulletin, 6 (1975), 2; idem, 'Sources on early Sierra Leone ( 7)', Africana Research Bulletin, i i ( I98I), 1-2, 92-140.

23 Letter of Balthasar Barreira to Father J. Alvares, in Nize de Moraes, A la decouverte de la Petite Co6te, i, i I 6.

24 In i6o8, P. van den Broecke visited in Portodale (Portudale) with one Simon Rodrigos, who had lived in Amsterdam. That Rodrigos is characterized as 'an excellent trader' indicates that this was not simply a social visit. Nize de Moraes, A la de'couverte de la Petite Co6te, I, 138.

25 On the efforts of one Governor of Cape Verde to extirpate Jewish merchants from the Petite Cote, see Nize de Moraes, A la d&couverte de la Petite Co6te, II, 305; for the original document see Brisio, Monumenta Missionaria Africana, iv, 698, D. Francesco de Moura, I622. 26 Chinas were local shrines.

27 Hair, 'Sources on early Sierra Leone (7)', 2, 63.



I78 PETER MARK

more heathen than Christian, men who go many years without sacraments or mass .28

Initially, Luso-Africans' Christian religious affiliation called attention to family ties and cultural connections to Europe. Yet if Christian identity served to distinguish and differentiate its practitioners, this exclusivist aspect was counterbalanced by the religion 's claims to universality, a stance intimately tied to the missionary presence that was an important factor in Luso-African/African relations. Ironically, missionary activity served to efface the very distinction Christianity created between members of the 'Portuguese' community and their African neighbors. When the few Catholic priests sent to Senegambia converted members of local African communities, the cultural boundaries between Luso-Africans and their neighbors became less clearly defined. A case in point may be the Floup (Joola) village of Bolole, adjacent to S. Domingos, in present day Guinea- Bissau. By the late seventeenth century, many villagers had Christian names and some may have converted to Christianity.29 On an individual level, conversion to Christianity was an important factor in cultural assimilation and opened the possibility of further cultural transformations. In such cases, conversion was a first step in the process of becoming 'Portuguese'.30

The final characteristic that defined the 'Portuguese' communities of Senegambia and Guinea was their material culture. Luso-Africans were identified by the distinctive architecture of their houses.31 These houses had a vestibule at the entrance or, alternatively, were surrounded by a verandah or porch, known as an alpainter.32 The dwellings were rectangular and their exterior walls were whitened with either a wash of clay or with lime. Along the coast from Senegal to Bissau, this style came to be called architecture 'a la portugaise. Indeed, the sobriquet was employed in the eighteenth century for buildings as far away as the island of Reunion.33

Portuguese style dwellings were ideally suited both to the climate and to Luso-Africans' role as commercial middlemen, as the owner could receive traveling merchants in the vestibule. A clear description of the style, together with an illustration of its function is provided by de la Courbe in I685. Traveling south from the Gambia, he was welcomed by a signare:

She received us most civilly in a Portuguese-style house, which is to say having walls of whitened earth and a small vestibule in front of the entry where we were seated upon mats, in the fresh air.34

28 Hair, 'Sources on early Sierra Leone (5)', 4, 88. 29 Sieur de la Courbe reports that the people of Bolole were idolatres, but the names

suggest religious syncretism or at least nominal acceptance of Christianity: P. Cultru (ed.), Premier voyage du Sieur Jajolet de la Courbe faite 'a la coste de l'Afrique en I685 (Nendeln, I973), 260.

30 See below for the case of Ventura, son of a Sierra Leonean king. 31 On the nexus between Luso-African architecture and 'Portuguese' identity, see

Mark, 'Constructing identity'. 32 Alpendre is the Portuguese term for 'porch'. 33 See Yves Goasguen Leven, 'Architecture coloniale a l'ile de la Reunion' (these de

doctorat, Universite Lyon II, I997), 23, citing Pere Houbert, c. 1720.

34 De la Courbe in Cultru, Premier voyage du SieurJajolet de la Courbe, 196: 'Elle nous receut [sic] fort civilement, dans une case a la portugaise, c'est a dire ayant des murailles de terre blanchie et un petit vestibule devant la porte oui l'on nous fit asseoir a l'air sur des nattes'.



THE EVOLUTION OF 'PORTUGUESE IDENTITY I79

Architectural style could, of course, be readily appropriated by neighboring populations, and the spread of architectural elements - from 'Portuguese' to Africans and from Africans to 'Portuguese' -was characteristic of public and vernacular architecture in the Gambia-Bissau region from the sixteenth to the eighteenth century.

Each of the other defining characteristics of the 'Portuguese' community could be similarly shared by members of adjacent populations. The more isolated a Luso-African community became from the Cape Verde islands and from the few centers of Luso-African population on the coast, the greater the likelihood that its members would lose their culturally-defined distinctiveness. Consequently, identity transformations in both directions were undoubtedly rather common, especially as the isolation of smaller Luso- African communities increased after I700. It is precisely under such conditions of isolation and impending loss of cultural specificity, however, that one might expect 'Portuguese' to assert the exclusivity of those few cultural markers they could control most emphatically.

THE SIXTEENTH CENTURY: EARLY PORTUGUESE-AFRICAN

INTERACTION

Luso-African communities along the Upper Guinea Coast were closely linked to the Cape Verde islands by sea-borne commerce and close personal ties. From the early sixteenth century, islanders carried out extensive trade with the mainland, from the Petite Cote to Sierra Leone. In I582, Andrade wrote that along the entire coast from Sierra Leone to S. Domingos: 'there are many Portuguese who carry out trade with the Negroes of the land and send out the ships and armaments that go to the island (of Santiago).35 At trading centers along the Rio Grande, as many as twenty or thirty ships might be seen trading for ivory, gold and slaves from the interior.36

Andrade's report is confirmed by the more detailed observations of the Cape Verdean trader Andre Donelha, whose account, written about i625, was based upon first hand knowledge of the coast obtained primarily in the decade before I585.37 Donelha described lanfados- both Christians and Jews - living among Wolof of the Petite Cote.38 In the Gambia, he met many Portuguese tangomaos while, further south, Cacheu had replaced S. Domingos as an important trading center in the I 580s.39

Both Donelha and Almada present a nuanced picture of the complex relations that prevailed between the Cape Verde islands and the mainland. This situation was characterized by two-way migration of individuals and by an active process of cultural interaction and borrowing between the various Portuguese groups and their African trading partners as well as a continuing process of intermarriage between Cape Verdeans, Africans, and the descendants of lanfados.

35 Andrade, 'RelavAo', in Brisio, Monumenta Missionaria Africana, iii, 102. 36 Ibid. I05. 37 Donelha, Description de la Serra Leoa, intro., I5ff. 38 Ibid. I29. 39 Ibid. I67, I7I-3. S. Domingos' importance had declined due in part to warfare

among the Cassangas (linguistically related to Bagnun). The fighting had closed the Casamance River to long distance commerce.



i8o PETER MARK

Two Sapes40 rulers from Sierra Leone exemplify this process of cultural assimilation and, perhaps, ethnic transformation. Donelha recounts that these rulers had fled the Mane invasion about I550. The daughter of one king subsequently became a Christian and moved to Santiago, returning to Guinea during the I583 famine. A son of the other ruler also converted, taking the name Ventura, and during the I 583 famine, he also returned to the mainland near Cacheu, where he and his fellow Sapes refugees established their own settlement. According to Almada:

They live together in a separate section of the town, with a king ... a Christian called Ventura de Siqueira. He can read and write as he was brought up in Santiago Island. All the other Blacks of this settlement are Christians and he has all the babies who are born there baptised.4'

The movement of individuals back and forth between the physical spaces of Sierra Leone, S. Domingos and Santiago and, more significantly, between the cultural spaces of African, lanfjado and Cape Verdean societies suggests a crucially important characteristic of mainland Luso-African society: it was not firmly bounded, nor was it exclusionary of those of African origin. Rather, Luso-African culture was open to individual assimilation at the margins.

The Portuguese maintained an active commercial role in the flourishing Gambia River trade in the I 620S, in spite of growing competition from other European nations.42 The most important Portuguese port in the Gambia was located sixty leagues upstream on the north bank at Casao. There, Donelha reported: 'I found many well-known tangomaos'. 4 The tangomaos traded with Manding Muslim merchants, 'the best traders in Guinea'.44 These Muslims, or bixiiris,45 sometimes traveled as far as S. Domingos and the Rio Grande themselves.46

Donelha found nine ships (likely from Santiago) at Casao.47 In addition, the free movement of individuals to Sierra Leone, Rio S. Domingos and Casao linked the Cape Verde islands with the coast. In Casao, Donelha met Gaspar Vaz, a tailor whom he had known in Santiago as the slave of a neighbor.48 The two men quickly established a close commercial relationship. Vaz helped Donelha to purchase merchandise at the price prevailing for exchanges between Manding traders, rather than at the significantly higher prices that the traders charged tangomaos.49 It is not clear from Donelha's

40 The identity of 'Sapes' is problematic. Almada, who is inconsistent in his use of the term, uses 'Sapes' to refer to several different groups. He writes: 'In this kingdom of the Sapes are the following nations of people: Bagas, Tagunchos (?), Sapes, ... Temenes, Limbas ... and all these understand each other', Almada ch. 14, 17 (Hair translation). Elsewhere, however, (ch. 13, 6) Almada appears to distinguish between Bagas and Sapes when he writes 'The Baga blacks extend as far as Cape Verga where the Sapes begin.'

41 'Almada's Brief Treatise on the Rivers of Guinea', 8i. 42 See Donelha, Description de la Serra Leoa, 141 ff. 4 Ibid. I47. 4 Ibid. i6i. 45 Hair suggests that the term derives from serigne and ultimately from the Arabic

mubecherin, 'one who spreads the faith'; Hair, in Donelha, Description de la Serra Leoa, 303. 46 Ibid. i6i

47 Ibid. The author does not specify the origins of these vessels. Elsewhere, however, he is specific when vessels he mentions came from European ports. 48 Ibid. 147 ff.

49 Ibid. 147 ff. Clearly, European traders were not the only ones who established artificially high prices when trading on the African coast.



THE EVOLUTION OF 'PORTUGUESE' IDENTITY i8i

account precisely how Vaz had come to be a slave in Santiago, nor how he re-established his earlier social status in the Gambia, where in addition to being a tailor and a merchant,50 Vaz claimed to be the nephew of the Sandeguil (satigi) of Casao, a term he translates as 'Duke .51

It is clear that Gaspar Vaz, like Ventura, the son of the exiled Sierra Leonean king, was able to move back and forth between Cape Verdean 'Portuguese' and African society. Having returned to the Gambia, he brought with him professional skills that he had refined in Santiago. He spoke Portuguese, which enabled him to serve as interpreter for Donelha and, presumably, other visiting merchants. Vaz, like Ventura, was able to navigate in the social and economic spaces of two cultures. In the process, these men undoubtedly served as conduits for the transmission of cultural traits between 'Portuguese' and African society.52

Material culture in Casao reflected extensive intercultural contact. The remarkable architecture there attracted Donelha's attention:

The village of Casao is small and circular. The houses are round, made of sun- dried bricks that are whitened with a white mud that resembles lime. Some of them like that of the Duke or Sandeguil have a second floor and most of them have banquettes in the interior, made of unfired bricks ... All have doors. Their locks and wooden keys were the first I had ever seen.53

In a subsequent passage Donelha describes the courthouse or tribunal which, in addition to having a two-storey elevation and whitewashed brick walls, was rectangular.

It is noteworthy that in late-sixteenth-century Casao, the tribunal and houses associated with the Sandigil - the preeminent expressions of public and private architectural space in this Manding community - incorporated elements associated by Europeans with Luso-African vernacular architecture. Some of these elements probably developed independently among West African and European builders as logical responses to the challenge of constructing houses that would provide a cool interior in a hot climate, but it is significant that the architectural form widely known as 'Portuguese' derived many of its defining characteristics from West African forms.

Gaspar Vaz traded with Luso-African merchants, had lived in the Cape Verde islands and he spoke Portuguese. In addition, as his name indicates, he was a Christian (at least when not in Casao). As a Christian in Mande

50 Vaz served as Donelha's host and facilitated his commercial ventures. Trade in Gambian Manding society was based upon such individual host-guest relationships; the guest would be expected to share some of his profits with the host.

51 The term satigi or saltigi, in Manding, may be translated as maftre de la route (chief trader). It designates rulers (or chief traders) in both Manding and Peul communities; see Jean-Loup Amselle, Logiques m6tisses: Anthropologie de l'identite en Afrique et ailleurs (Paris, 1990), 73; Hair, in Donelha, Description de la Serra Leoa, 301.

52 In early seventeenth century Sierra Leone, too, European traders relied on individuals whose economic and cultural position served as a bridge between European, 'Portuguese' and African society. Ruiters mentions two such traders, one named Mathew Fernandos and the other Francisco Mendes. See P. E. H. Hair, 'Sources for early Sierra Leone (4): Ruiters (I623)', Africana Research Bulletin, 5 (i974), 3, 61-62 and 7on. Even contemporary descriptions do not always clearly indicate whether such men were Luso- African or African. Indeed, the distinction was probably blurred.

53 Donelha, Description de la Serra Leoa, 151.



I82 PETER MARK

society, his position was highly unusual.54 He, and others of comparable stature who lived in both African and Portuguese society, were conduits through whom cultural elements were exchanged between the two societies." In the Gambia and Bissau in the late sixteenth and seventeenth century, architectural style, religion and language all reflected extensive contacts among the populations of the region.

AN INDIGENOUS MODEL OF IDENTITY FORMATION

During the sixteenth and seventeenth centuries, Luso-African identity was consistent with the historically fluid and contextually defined manner in which ethnic identity was established throughout the Gambia-Bissau region before the colonial period.56 The permeable nature of 'Portuguese' identity reflected the fact that many of the lanfados and their descendants had married local African women. Luso-Africans maintained an understanding of group membership based essentially on cultural characteristics and this facilitated the assimilation into 'Portuguese' society both of filhos da terra and, as suggested by the lives of people like Vaz and Ventura, of individual creolized African men and women.

I find remarkably few indications that sixteenth century 'Portuguese' had recourse to a sense of boundaries in establishing and maintaining their community or that they discouraged assimilation into their society. Implicit contrasts certainly existed, as between free men and slaves or Christians and Muslims, but even here the boundaries were fluid rather than fixed indicators of the 'otherness' of opposed populations. Vaz, for example, was evidently both slave and noble, Christian and Muslim. In this respect, Luso- Africans represent a model of identity formation quite rare in the modern world, in marked contrast to Fredrik Barth's bipolar model of identity formation.57 While Barth, focusing on 'the boundaries that define groups', rightly observes that boundaries persist in spite of the movement of individuals across them, he overstates - or overschematizes - the distinction between group members and others when he writes of 'a dichotomization of others as strangers'.58 Not all boundaries are alike, and some are more

5 That Vaz's position as a Christian in a predominantly Muslim society was politically delicate is suggested by Donelha's description of him. At the time they met in Casao, Vaz was wearing Muslim robes. He assured the Cape Verdean that once he became Sandegil, however, he would reassert his Christian faith, whereupon the two became trading partners. Vaz may have understood his Cape Verdean counterpart better than Donelha understood him. Donelha, Description de la Serra Leoa, I49.

5 Later, during the eighteenth century, the idea that members of the Manding elite living along the lower Gambia had assimilated elements of Luso-African architectural style was expressed in popularized European travel accounts. In 1745, in his compendium of travel narratives, Thomas Astley argued from his reading of Labat that cazas, or Portuguese houses, had spread to the kingdom of Barra, where 'the king and his lords have of them': Astley, A New General Collection of Voyages and Travels (London, 1745),

246. This of course does not prove the primacy of the Luso-African model; influences may just as well have moved in the other direction.

56 On changing identity in precolonial Casamance, see Peter Mark, The Wild Bull and the Sacred Forest (Cambridge, I992); for the lower Gambia, Donald Wright, The World and a Very Small Place in Africa (New York, 1997).

5 Fredrik Barth, 'Ethnic groups and boundaries', in his Process and Form in Social Life, Selected essays of Fredrik Barth (London, I969), i, I98-227. 58 Ibid. 204.



'HE EVOLUTION OF 'PORTUGUESE IDENTITY I83

permeable than others. One is led, if not to deny that coastal Luso-Africans conceptualized 'otherness' in the construction of their own sense of being 'Portuguese,' at least to suggest that this sense of the 'other' played a relatively circumscribed role in creating their image of who they were.

This model of identity formation - flexible, malleable, and based on cultural and socio-economic factors - was characteristic of societies along the Upper Guinea Coast and derived from a local identity paradigm. Positing a simple bipolar model overlooks the fact that 'Portuguese' were closely related culturally and physically to their African neighbors, at times virtually indistinguishable from them, and that individuals could move back and forth between Luso-African and African society. A bipolar model presumes a later, essentially Western approach to identity, one that is not appropriate for the sixteenth and early seventeenth century Upper Guinea Coast.

THE LATE SEVENTEENTH AND EIGHTEENTH CENTURY: THE

'PORTUGUESE REDEFINED

During the late seventeenth and eighteenth centuries, three significant historical developments had a profound impact on Luso-African communities, affecting not only their economic livelihood but also the manner in which they defined themselves. The first was the rapid commercial expansion by Dutch, French and English trading companies. By the time that the French captured Goree Island from the Dutch in I677, Portuguese commercial ascendancy along the Petite Cote and the lower Gambia was but a memory.59 The second development was that these Europeans brought with them a conception of identity that contrasted with the fluid, culturally-based sense of identity characteristic of the 'Portuguese' and their West African neighbors. Finally, outside of Bissau, Cacheu and the Gambia, Luso- Africans, deprived of their economic base and increasingly isolated from the Cape Verde islands, began to lose the cultural traits that had been the foundation of their identity.

These developments had several consequences for 'Portuguese' identity. First, many smaller, isolated 'Portuguese' trading settlements south of Bissau and along the Petite Cote disappeared as their inhabitants either left or were assimilated into neighboring African cultures. Second, the more remote Luso-African communities that survived altered the way they defined themselves in order to stress their distinctiveness from their African neighbors. Along the Petite Cote, for example, religion took on an increasingly important role as a marker of cultural difference. Finally, 'Portuguese' in the Gambia and Bissau entered into the new European discourse on identity. While these 'Portuguese' responses to the new historical developments are not directly recorded by Luso-African chroniclers (the last important Cape Verdean source was Lemos Coelho, writing c. i68460), one can partially reconstruct them by means of careful interpolation and interpretation of the European sources.

" English commercial ascendancy in the Gambia may be roughly dated to the capture of James Island in i66i.

60 On the life of Francisco de Lemos Coelho and the dating of his 1669 manuscript, expanded about I684, see P. E. H. Hair, 'Introduction', in Lemos Coelho, Description of the Coast of Guinea (I684), privately issued, I985.



I 84 PETER MARK

A distinction between European and Luso-African parameters of identity was already evident by the early seventeenth century. In i 62o, the English traveler in the Gambia, Richard Jobson, observed:

another people ... as they call themselves, Portingales, and some few of them seem the same; others of them are Molatoes ... but the most part as black as the natural inhabitants .... They do generally employ themselves in buying ... commodities ... still reserving carefully the use of the Portingall tongue and with a kind of affectionate zeal, the name of Christian, taking it a great disdain, be they never [sic] so black, to be called a Negro.6"

For Jobson, the salient characteristic of the Luso-Africans was their skin color, which conflicted with his own conception of what (European) Portuguese should look like. At the same time, his indulgent tone does not betray a strongly negative reaction to their self-identification. As for 'Portuguese' themselves, their refusal to accept the label Negro or Black would remain central to their sense of identity for over two hundred years.

Revealing a more judgmental, even disdainful attitude than Jobson, the Frenchman, de la Courbe, in I685 expressed the conviction that a people who were not white could not, a priori, be Portuguese. He described the Luso-Africans in the Gambia as 'certain Negroes and mulattoes who call themselves Portuguese because they are descended from some Portuguese who formerly lived there'.62 Still more pejorative was Le Maire's I695 commentary on Luso-Africans of the Petite Cote:

a species of Portuguese, people who refer to themselves this way because they used to serve, and are descended from, those who first lived along this coast ... from the Negresses whom they married, were born these mulattoes, from whom in turn came even darker ones.63

Dark skin is here associated with lower status and, in Le Maire's view, a heritage of servitude. The association of skin color with social inferiority undoubtedly reflects the growing influence of the Atlantic slave trade, which in turn influenced European attitudes about race.64

Eighteenth century European descriptions of Luso-Africans progressively betray even more negative judgments, frequently tinged with irony or sarcasm. Francis Moore, the English factor in the Gambia whose extensive contact with Luso-Africans during his I732 stay seems not to have engendered much respect for them, described 'Portuguese' as speaking 'a sort of bastard Portuguese language'.65 Moore was presumably referring to

61 Richard Jobson, The Golden Trade, or a Discovery of the River Gambra (London, I623), 28-30.

62 De la Courbe in Cultru, Premier voyage du Sieur Jajolet de la Courbe, 193 (author's trans).

63 Le Maire, Les Voyages du Sieur Le Maire aux Isles Canaries, Cap- Verd, Senegal et Gambie (Paris, I695), 38 (author's trans).

64 On European attitudes towards Africans before the development of the Atlantic slave trade, see Peter Mark, Blacks in European Eyes: The Representation of Black Africans in fourteenth and fifteenth century Europe (Syracuse, 1974); for a contrasting view, which focuses on fifteenth century Portuguese conceptions of Africans, see Jos6 da Silva Horta, 'A represantaiao do Africano no literatura de viagens do Senegal a Serra Leoa (1453-1508)', Mare Liberum, 2(1991), 209-338.

65 Francis Moore, Travels into the Inland Parts of Africa (London, 1738), 29.




a form of Creole, but he viewed it as the cultural equivalent of the physical mixture of Portuguese and Manding: 'they are now very near as Black as [the Africans]' in spite of the fact 'they reckon themselves still as well as if they were actually white'.66 For Moore, light skin color conferred elevated standing, and complexion served, significantly, as the primary marker of identity.

Moore's words were later cited almost verbatim by Thomas Astley in his I745 compilation of travel narratives. Also citing Jobson and Labat (who plagiarized de la Courbe), Astley's comments reflected the spread in England of a popular image of Luso-Africans that clearly contested their identity as well as the criteria on which they based their own sense of self: 'nothing angers them more than to be called Negroes. This proceeds from their not understanding the true meaning of the word, which they use only for slaves '67

A generation later John Mathews, traveling among the isolated descendants of Luso-African traders in the Rio Pongo, voiced an even more sarcastic and pejorative image of 'Portuguese':

the principal people call themselves Portuguese ... though they do not retain the smallest trace of European extraction; but having had a white man once in the family is sufficient to give them the appellation. They also profess the Roman Catholic religion ... yet the most enlightened among them are merely nominal Christians.68

Not only was Mathews' conception of identity based on physical characteristics, but these traits determined one's status in a hierarchical order in which to be European was to be superior. The connection between physical appearance and level of cultural development implied by Moore earlier was here made explicit. Furthermore, Luso-Africans' deviance from European Christian norms became a measure of 'Portuguese' decadence, com- mensurate with the darkening of their complexion over the generations.

The origins of Luso-Africans' own use of the terms 'white' and 'black' can be traced to the sixteenth century.69 Almada used the term 'black' to refer to Africans and 'white' to Portuguese and Cape Verdeans, probably reflecting attitudes about color prevalent in late-sixteenth century Portugal.70 At the same time, however, Almada's terminology did not follow a simple oppositional model, and he transformed pre-existing Iberian terminology. Cape Verdean society was already widely intermarried in the sixteenth century and Almada himself was apparently a mulatto.7' He nevertheless

66 Ibid. 67 Astley, New General Collection of Voyages, 245. 68 John Mathews, A Voyage to the River Sierra Leone (London, 1788), 13-14. 69 On the use of brancos and negros in the writings of Almada and Donelha, see Jose da

Silva Horta, 'Portuguese accounts of the Guine do Cabo Verde and Luso-African identity (sixteenth to seventeenth century)', paper presented at the Mande Studies Association biennial meeting, Banjul, The Gambia, June I 998. I wish to express my gratitude to Jose da Silva Horta for sharing his profound knowledge of early Portuguese sources with me. My understanding of Luso-African self-identification and of the relationship of these concepts to earlier Portuguese models has been deeply influenced by our discussions.

7 Almada had travelled to Iberia and he may have written the Tratado breve while in Lisbon. See 'Almada's Brief Treatise on the Rivers of Guinea', notes by Hair, 2-3.

71 Ibid.



i86 PETER MARK

placed himself in the category 'white'. That his family also belonged to the island elite and owned slaves suggests that 'white' and 'black' referred not to skin color, but to social class.

A sense of the Luso-Africans' early use of color terminology can also be inferred from Jobson's observation that they objected, 'be they never so black' to being called 'Negroes'.72 To 'Portuguese', many of whom were professional slave traders, 'Negro' implied 'slave'. As social status remained relatively fluid in southern Senegambia in the seventeenth century, the distinction between free and slave did not imply the adoption of a rigid system of identity categories by the 'Portuguese'. Moore's I732 observation confirms the association of social status with the terms 'white' and 'black'. And in I 8 I 8, at the Luso-African trading community of Geba in present-day Guinea Bissau, Gaspard Mollien observed: 'Blacks and mulattoes who are nevertheless called white, because all who are free claim this title'. 73 From the late sixteenth to the nineteenth century, one perceives a consistent system of ascribed meaning, whereby 'white' implies both social status (slave trader/free man) and cultural and blood ties to Portugal.74

This system of meanings is similar to the language of identity of Maures of southern Mauritania today. There, as James Webb has demonstrated, the term 'white' (bidan) implies a cultural identity rather than skin color.75 Like 'Portuguese', Maures have a diverse heritage, one that incorporates Wolof and other Black Africans. And like Luso-Africans, many Maures were traders. Significantly, Bidan identity first developed in the late seventeenth century, not long after 'Portuguese' identity first emerged. Webb's charac- terization of 'white' and 'black' as 'regional constructs which refer to the cultural identities ... rather than to skin color'76 thus probably applies to Luso-Africans as well as to Bidan.

CHANGES IN 'PORTUGUESE RELIGION

John Mathews' reference to Luso-Africans as 'merely nominal Christians' raises the question of the nature of Luso-African religion in remote areas at the end of the eighteenth century. In I795, the Swedish naturalist Adam Afzelius described a characteristic instance of religious syncretism in Freetown, where Afzelius witnessed the poison ordeal of a man accused of witchcraft. The principal accuser was the elder Domingos, one of the few remaining Luso-Africans along this part of the coast. Domingos had a vested interest in the outcome of the trial, for those individuals found guilty of witchcraft were sold into slavery and he, not coincidentally, was the main slave trader in the region. The old man was fluent in Portuguese, prayed

72 Jobson, Golden Trade, 30. 7 Gaspard Mollien, Voyage a l'Interieur de l'Afrique aux Sources du Senegal et de la

Gambie (Paris, i820), II, 2i8 ff.: 'noirs et mulatres qu'on appelle cependant blanc, parce que tout ce qui est libre pretend a ce titre'.See also Mark, 'Constructing Identity'. In St Louis in the mid-nineteenth century, free Africans were also called 'whites'; see Anne Raffenel, Nouveau Voyage dans le pays des Negres (I856), 57.

7" In this latter sense, Mathews' sarcastic comments about European ancestry may contain a grain of truth.

7 James Webb, Desert Frontier: Ecological and Economic Change along the Western Sahel, i6oo-i850 (Wisconsin, 1995). 76 Ibid. xxv.



THE EVOLUTION OF 'PORTUGUESE IDENTITY I87

every day using a Portuguese missal and always wore a string of prayer beads given to him by a Portuguese priest.77 In addition,

He (had) expressed great concern that for some years past he had seen no priest ... he has left orders that as soon as he dies, 2 slaves shall be sent to Santiago to a priest there, who may intercede for him.78

During the poison ordeal, some local Temne placed their grisgris on the ground as a public invocation to ensure a guilty verdict. Then Domingos, 4 not being satisfied with the Timany Grisgris, or thinking them not powerful enough, let his own Portuguese Grisgris be brought forward and put near to the former' ... 79 Notwithstanding his intercession and prayers, however, the accused survived the ordeal and was found innocent.

This incident illustrates several points about Domingos' Christianity. His use of the prayer beads demonstrates a local West African ritual emphasis on instrumentality: grisgris served the specific function of achieving desired ends. The fact that Domingos was actively involved in witchcraft trials also strongly suggests that he had assimilated local metaphysical concepts concerning the interrelationship of the physical and spiritual realms. The diminished ritual and metaphysical distance separating Domingos from his Temne neighbors did not, however, deter him from using his Christianity as an important identity marker. His religion evoked religious, economic and cultural ties to the Cape Verde islands, where a priest would pray for his soul after his demise.

PORTUGUESE' RESPONSE: IDENTITY DEFENDED ON THE PETITE

COTE

Eighteenth century Europeans articulated a view of 'Portuguese' identity that was increasingly based on a priori characteristics, primarily skin color, rather than upon cultural traits. Luso-Africans were forced to respond to this ascribed identity at a moment when their increasing isolation was beginning to erode the cultural markers upon which their own sense of identity had been based. In response to the intrusion of an a priori identity model, they began to articulate a more assertive self-identification and an important change ensued in their self-conceptualization. Precisely because the domi- nant European model denied their distinctiveness from other African populations, 'Portuguese' came increasingly to define themselves by reference to who or what they were not.

An illustration of this process may be seen among Luso-Africans of the Petite Cote. There, two hundred years of Dutch and French commercial dominance had largely isolated the remaining 'Portuguese'. By the nineteenth century, the Jewish community had disappeared and the remnants of the 'Portuguese' at Joal now based their identity upon an oppositional model, distinguishing themselves from their Serer (Sereer) neighbors on the basis of religion and 'color'. These 'Portuguese' no longer spoke either

7 As a merchant, Domingos was well known to several English visitors, including Clarkson and Macaulay. Their observations are cited by Alexander Kup in his extensive annotations of Afzelius' manuscript: Adam Afzelius, Sierra Leone Journal 1795-1796 (Studia Ethnographica Upsaliensis, xxvii, i967) 89n.

78 Kup, in Afzelius, Sierra Leone Journal, 89n. 79 Ibid. 89.



i88 PETER MARK

Portuguese or Creole,80 yet they continued to give their children Portuguese names, and they resolutely continued to call themselves Christians.81 And as the Abbe Boilat explained, they steadfastly maintained that they were white:

although they are as black as the purest black Africans, they make the modest claim to be pure whites, and it is a great insult to consider them Negroes or Serers. They want to be called the whites of Joal, the Christians of Joal, because they are the direct descendants of the Portuguese... Succinctly stated, to be Christian is to be white; to be white is to be free... 82

For these Luso-Africans, calling themselves 'white' specifically differen- tiated them from other local populations. To be white was to be non-slave, recalling that their ancestors had been slave traders. Although their distinctiveness no longer resided in a unique Luso-African culture, the loss of cultural and physical markers of separation did not diminish their sense that they constituted a discrete group. Their dual identity, 'etre chretien, c'est etre blanc', implicitly evoked the existence of its opposite number, non- Christian and black, in a nineteenth-century bipolar model of identity.

The formation in Joal of an exclusivist 'Portuguese' identity differed, however, from the later process whereby, under the impetus of colonial administrators and ethnologists, 'ethnic' boundaries were established and then subsequently accepted by Africans themselves.83 The redefinition of Luso-African identity took place essentially before the colonial period. Furthermore, 'Portuguese' bore the brunt of the pre-colonial European discourse on identity, which largely spared other Senegambian peoples. The reason for them being singled out is clear: only the 'Portuguese', whose very existence was testimony to cultural and physical assimilation between Africans and Europeans, posed an ontological challenge to European identities which, by the time of the Enlightenment, were based on the premise of a non-white, non-European 'other' 84 By denying Luso-Africans' self-definition, eighteenth and early nineteenth century observers were beginning to articulate what subsequently evolved into a racially-based definition of peoples and cultures. And 'Portuguese', in order to maintain their sense that they too constituted a distinctive group, were being forced to adopt a model that stressed their differences from their African neighbors.

With this historical development, religion became the crucial parameter by which Luso-African communities defined themselves. Whereas in the sixteenth century missionary activity had stressed the universality of the Christian faith, by the early nineteenth century, Christianity had become for 'Portuguese' both an exclusive religion and a marker of the boundary between themselves and other Africans. As Amanda Sackur observes:

80 Abbe P.-D. Boilat, Esquisses Senegalaises (Paris, I853), I IO. 81 A mission was re-established at Joal in I850. 82 Boilat, Esquisses Sene'galaise, i io. Boilat was from the Petite Cote. 83 I refer, inter alia, to Amselle, Logiques me'tisses; see also Amselle and M'Bokolo, Au

coeur de l'ethnie: ethnies, tribalisme et Etat en Afrique, (Paris, I985). 84 The label 'Portuguese' had direct implications for the broader, socially constructed

category of 'race', into which the group was to be situated. Groups that do not fit comfortably into any single category or whose members may lay claim to one category while the outside society assigns them to a different one are a focal point for particularly intense contestation. The existence of such ambiguous or liminal groups may call into question the entire categorical structure.




'creole identity ... required definition against two competing 'others': Europeans and Africans '.85 I would argue that until the eighteenth century, this was not the case, but Sackur is quite correct for the nineteenth century. The need to redefine themselves in a manner that emphasized their distance from other African societies led 'Portuguese' to an exclusivist approach to their religion.

When the Abbe Boilat visited the fishing village of Joal in I846, he found that the villagers defined themselves as Christians even though local religious practices were now an amalgam of Christian prayer (the two elders who led communal prayers knew only the 'Ave Maria') and local rituals.86 One powerful shrine, baton, attracted clientele from St Louis, Goree and the Gambia, while the spirit of another powerful shrine, Mamanguey, had appeared in broad daylight in the guise of a uniformed naval officer.87

Despite their Africanized religious practices, however, the people of Joal were careful to distinguish themselves religiously from the surrounding populations. Villagers who wished to be baptized and whose parents were Muslims or idolatres had to pay the price of one iron bar.88 Outside of Joal the distinction between Christian and non-Christian was carefully maintained; no Christian parent would permit their offspring to marry a non-Christian.

In practise, none of the inhabitants of Joal would ever ... allow their children to marry the children of the neighboring village, because the latter are fetishists [sic]. They would never permit a Muslim or a fetishist to be buried on their land.89

Ultimately, it was by means of the creation of ritual prohibitions against marriage with non-Christians and against the burial of outsiders in the local cemetery that the people of Joal were able to give concrete expression to their identity. Christianity was central to that identity, based upon an oppositional relation between Christians and everybody else, including both Muslims and idolatres. Boilat, himself a missionary, also observed that this attitude led the people of Joal to oppose missionary activity among their non-Christian Serer neighbors.90 The Christians of Joal thus exhibited an exclusivist approach both to religion and to identity, an approach diametrically opposed to that of the early Luso-Africans.

85 TI'his excellent essay is, unfortunately, unpublished; Amanda Sackur, 'Religion and the construction of creole identity in precolonial Senegambia', paper presented at the African Studies Association of the United Kingdom, September 1992, 2.

86 Boilat's description is preserved in manuscript at the Bibliotheque Nationale, Departement des Cartes et Plans, Carton Bo-Bon (I09), Boilat, 'Voyage a Joal, I846. ' For a later and somewhat different account see Boilat's Esquisses S6negalaises, especially 20-I17. 87 Ibid. 88 Ibid. 89 Ibid.

90 Ibid. I 17; 'as for the Serers, the priest must never maintain relations with them, and he most assuredly must not offer them [religious] instruction' (author's trans). See also Sackur, 'Religion and the construction of creole identity', 6. I wish to thank Sackur for calling these passages to my attention. Jean Boulegue, Les Luso-Africains en Se'negambie, (Lisbon, I989), 85, writes: 'The weakening of their cultural identity did not prevent the Luso-Africans of Joal from vigorously asserting their uniqueness with regard to the neighboring Serer population. They based this distinctiveness on religion' (author's trans.).



I90 PETER MARK

LUSO-AFRICANS ON THE EVE OF THE COLONIAL ERA

A somewhat different attitude towards religion prevailed among nineteenth century 'Portuguese' of the Casamance, just south of the Gambia. Like the inhabitants of Joal, the 'Portuguese' of Ziguinchor, a trading center on the south bank of the Casamance River, asserted a Christian identity. But they also continued to maintain religious interaction with their non-Christian Floup neighbors.91 In the mid-nineteenth century, 'Portuguese' and Floup traded charms, reminiscent of the commercial role of Muslim marabouts:

All the inhabitants call themselves Christians, although they fulfill none of the obligations of this religion.... They deeply venerate images, medallions, and [medals of] Christ, to which they ascribe the power to protect them from all accidents.92

Since Floups also trusted in the power of Christian medals, Portuguese carried out a thriving trade, exchanging crucifixes for slaves. The 'Portuguese' community of Ziguinchor, more numerous and less isolated from other 'Portuguese' than the Luso-African settlement of Joal, could maintain such interaction with their neighbors without losing their cultural identity.

It is significant that, while French and English narrators questioned 'Portuguese' identity, the members of African communities continued to accept Luso-Africans' older terms of self-identification. In their interactions with African traders, nineteenth century Luso-Africans continued to be identified by profession, religion, and, to a lesser degree, material culture. Muslim merchants from as far away as the Mande heartland understood that their 'Portuguese' counterparts maintained their own religion.93 To Muslim traders, Christianity was a primary marker of 'Portuguese' identity, just as Islam was of Juula. Mungo Park reported that he met a wealthy salt trader along the Niger 'who had traveled to Rio Grande and spoke very highly of the Christians '. Although Europeans were often skeptical of Luso- Africans' religious affiliation, Muslims with whom they did business identified the 'Portuguese' primarily as Christians.

To some African trading partners, material culture also appears to have remained one of the markers of 'Portuguese' identity. In i850, the French

91 By the early twentieth century, colonial administrators had classified the Floup, along with several other groups who spoke dialects (not always mutually comprehensible) of the same language, under the category of Diola (Joola). This is a clear instance of the creation of ethnicity by French colonial administrators, at least one of whom, Dr Maclaud, was also an ethnologist.

92 Hyacinthe Hecquard, Voyage sur la Co6te et dans l'Interieur de l'Afrique Occidentale (Paris, I854), IIO. Hecquard derives much of his information from Bertrand-Bocande, the French factor at Carabane for twenty years.

93 Juula merchants may have perceived that their 'Portuguese' counterparts derived professional advantage from their religion. Christianity, as Sackur observes, provided Luso-African traders with similar advantages to those that Islam provided Juula. Shared religious orientation provided a sense of common values and a foundation for trust and cooperation, important for merchants living in diaspora; Sackur, 'Religion and the construction of creole identity'.

94 Mungo Park, Travels in the Interior Districts of Africa (London, I977), 215 ff. In view of the similar discourse on identity and color among the Maures and the 'Portuguese', it is worth noting that the salt trader cited by Park was himself a Maure.



THE EVOLUTION OF 'PORTUGUESE IDENTITY I9I

explorer Hecquard described a Portuguese Christian who was a respected merchant and architect in the Futa Jalon. The man, named Wolli, had constructed the interior of the mosque at Timbo.95 That this 'Portuguese' Christian, who lived for many years in the Futa, was entrusted with such a project, reflects the respect with which he was regarded by his Muslim peers.96 Eighteenth and nineteenth century African trading partners thus continued to view Luso-Africans according to long established parameters of identity that were common to both the Upper Guinea Coast and the Mande diaspora.

CONCLUSION

Historical sources for Senegambia and Guinea-Bissau afford a unique opportunity to trace 250 years of changing discourse on 'Portuguese' identity. Initially defining themselves on the basis of economic, linguistic and religious criteria, they were subsequently challenged by Europeans' denial of their identity as 'Portuguese', Christian and 'white'. By the eighteenth century, Luso-Africans were forced to redefine themselves on a priori physical grounds that led them to abandon local discourse, marked by flexible and non-exclusivist identities, for a more rigid paradigm based on an oppositional and exclusivist understanding of identity. This historical process constitutes the earliest documented confrontation between West African and European models of identity. It also serves as prologue to the widespread imposition of more rigid ethnic categories during the colonial period.

SUMMARY

In the fifteenth and sixteenth century, Portuguese traders settled in the Cape Verde Islands and along the West African coast from Senegal to Sierra Leone. Descendants of these traders and of local African women formed the nucleus of a Luso-African community that soon developed a distinctive culture, joining elements of European and local African culture. These Luso-Africans, or 'Portuguese' as they called themselves, were commercial middlemen, distinguished by their language (Portuguese and, later, Crioulo), architecture and Christian religion. As each of these characteristics could be shared by members of adjacent African communities, identity transformations in both directions were relatively common. 'Portuguese' identity remained both fluid and contextually defined through the seventeenth century. During the eighteenth and nineteenth century, however, 'Portuguese' were drawn increasingly into a European discourse on identity, one based upon a priori characteristics, primarily skin color. Forced to respond to this imposed identity, Luso-Africans continued to maintain that they were 'Portuguese'; now, however, they also began to define themselves negatively by reference to what they were not.

9 Hecquard, Voyage sur la Codte, 283. 96 J5id.



The Evolution of 'Portuguese' Identity: Luso-Africans on the Upper Guinea Coast from the Sixteenth...

Documents

Transcript of The Evolution of 'Portuguese' Identity: Luso-Africans on the Upper Guinea Coast from the Sixteenth...