avec le catalogue

En plein texte
avec Google

Recherche avancée

Tous les ouvrages
numérisés de cette
bibliothèque sont
disponibles en trois
formats de fichiers :
Word (.doc),

Pour une liste
complète des auteurs
de la bibliothèque,
en fichier Excel,
cliquer ici.

Collection « Les sciences sociales contemporaines »

Une édition électronique réalisée à partir du texte de M. Jean Benoist et Jean-Luc Bonniol, “Social values and population structure: phenotype and genealogies in La Désirade (West Indies)”. Un article publié dans la revue INTERNATIONAL JOURNAL OF ANTHROPOLOGY, Vol. 4, no 1-2, 1989, pp. 103-111. [Autorisation formelle accordée par MM. Benoist et Bonniol, respectivement le 17 juillet 2007 et le 28 septembre 2007 de diffuser, dans Les Classiques des sciences sociales, toutes leurs publications.]

Jean Benoist et Jean-Luc Bonniol 

Social values and population structure :
phenotype and genealogies in La Désirade
(West Indies)

Un article publié dans la revue INTERNATIONAL JOURNAL OF ANTHROPOLOGY, Vol. 4, no 1-2, 1989, pp. 103-111.

1. The context of the observation
1.1. Colour and society in the Caribbean
1.2. The case of Désirade
1.3. White endogamy in the Caribbean islands
2. Approach and results
2.1. The contribution of genealogies
2.2. Anthroposcopic observations
3. Discussion
Table 1Contribution of the different founders categories to the successive generations (0%).
Figure 1Contribution of the different founders categories to the successive generations [%].
Figure 2Anthroposcopical characteristics following admixture




Key words : Carib, anthroposcopy, race, genetic structure, assortative mating, skin pigmentation, genealogies 

In a small population (La Désirade, West Indies) with few internal socio-economic differences, the « white » segment survive on the basis of long-term assortative mating. This study compares the genealogies for one hundred and forty years, with phenotypical markers used in the population to define « racial » identity of individuals. Physical appearance, acting as an isolated factor, is able to maintain endogamy and to ensure the survival of a segment of the population without any black admixture. The social values do not produce the same effect among people of African origin. Genetic structure reflects these values and the genealogical pattern is able to assert their impact on the mating pattern.



Isolation can occur in human populations as a result of different mechanisms. Some are purely ecological, some are of social origin. In the West Indies, the genetic barriers are generally based on a social, economic and political contrast between the « White » and the « Black » segments of the population. In the small island of « La Désirade », close to Guadeloupe, the genetic barrier originates only in the perception of phenotypical differences, inside an homogeneous socio-economic group. It is very difficult to observe any kind of clear racial discrimination in the island. White colour, and some of the phenotypical characteristics connected with « white » people are attributed a strong positive value, independent of any correlation with the social status. 

In « La Désirade », we are very close to a model in which color has a great autonomy with respect to social inequalities, commonly associated in the Caribbean with racial status. In such a situation, the phenotype is not only the marker of identity, but it is also the main support for the biological persistance of a small group of « white » people. Operating through the cognitive level, mating choice, as reflected in the genealogies, enables the survival of an informal endogamous group. 

What is the impact of such a situation on Desirade's genetic structure ? How is the cognitive level reflected both in genealogies and at genetic level ? Does a social value attached to a phenotype correlate with the genetic basis of a phenotype, in such a way that « the biological and social dimension are so intertwined as to be practically inseparable » (HARRISON, 1985, p. 14) ? It is our hope that a combination of biological and social data will help to explain their interaction.


1. The context of the observation


1.1 Colour and society in the Caribbean


The intrusion of « race » in the social order seems fundamental for the genesis of West Indian societies. This is an essential characteristic, at the core of the historical process that they share with the other plantation areas, founded on the African Slave Trade. Social stratification has been reinforced by racial barriers, which appeared in the period of Slavery and have been maintained even since. After the initial correlation between social stratification and phenotype, color and physical features have been used as labels for the recognition of groups, or as markers to align individuals. 

« Race » is based on a certain number of discriminatory characteristics, involving a whole system of values. First of all, it appears as a sign in which a social signifié and a biological signifiant superimpose themselves, a sign which allows us to recognize clearly the social position of the individuals. These societies are extremely aware of physical type and of the transmission of features ; they are based on biological thinking, as far as phenotypical diversity and heredity are used as conditions for social differenciation. Such differenciation is transmitted by the perception of the phenotype. This perception is related to a whole set of representations, concerning not only the somatic variation but equally its transmission. This cognitive aspect finds its origin at the cultural level. For this reason we may assert that « race » escapes of a strict socio-economic definition. As a matter of fact, when we look at the Caribbean societies, one understands that the variation in racial relations seems to be linked to the cultural character of each of the colonial nations involved (H. HOETINK 1967).


1.2 The case of Désirade


The island of Désirade, « dépendance » of Guadeloupe, did not have sugar plantations. It was occupied in the eighteenth century by « poor white » settlers. They started to grow cotton and some foodstuffs, helped by a small number of slaves. So the island is peripheral to the dominant economic and social form of the area. Nevertheless, it appears partly as a small-scale model of larger islands. However the growing conditions were too uncertain for these micro-plantations to survive the consequences of the 1848 Emancipation, which gave rise to a traditional economy, based on commercial shipping (and occasional fishing) and on subsistence agriculture. In this economic structure, the white man and the son of slaves would be relatively equal. The white group, in order to maintain a racial system, in which its superiority was safe, practised a severe endogamous marriage control. 

Here we note the action of racial ideology on the biological history of the group, as on the « sugar islands ». This situation persisted, long after the juridical criterion with which it coincided had disappeared, and where the original social-economic divisions seemed to progressively vanish...


1.3 White endogamy in the Caribbean islands


Caribbean peoples are preeminently crossbred populations still receiving gene flow from various sources. There are also a number of endogamous subsections. Among them the « white » groups are often preserved from any introduction from the miscegenated mass. On some islands, like Martinique, the « white » group (the « Békés ») based its endogamy on genealogies and not only on morphological features. Homogamy is very rigid, and any individuals of mixed blood are left out of the group. In other islands, like Terre-de-Haut des Saintes (BONNIOL, 1980), the white group is more inclusive. Biological admixture is socially accepted, and many « whites » are in fact people of mixed blood. Identity is rooted in territory rather than in physical features or in genealogies. Everybody, even of mixed blood, is locally qualified as « white ». In Saint-Barthélémy, « Whites » are a large majority. They can exclude any « mixed » people, as the Békés are doing. But, as genealogies are often unknown, some admixture is possible (BENOIST, 1965). 

Contemporary Désirade exhibits a very different pattern. There is no « white » group as such. Some people are « white », even some families, but it is a criterion for individual identity, not for group identity. Esthetic values, and probably a remnant of the ancient ideology of racial hierarchy is the only explicit basis for the survival of a « white » « quasigroup ». As we shall see later, these values have up to now operated strongly enough to maintain an endogamous circle of people of pure European origin and to put up barriers against gene flow. However, recent changes have been breaking up these barriers as the whole population has become more integrated. 

Different social ways of organizing the survival of a « white » group result in different genetic structures. Social history appears to be the more important factor to examine if we want to understand the genetic differences between white groups and the process of their recent evolution.


2. Approach and results


2.1. The contribution of genealogies


Indeed, Creole societies were obsessed by the genealogy of their members, through their attention to the heredity of racial differences. If these differences have been occulted nowadays, one nevertheless notes that individual physical appearances still show the outcome of different genealogical trajectories, often symbolized by transmission of patronymes. And even if genealogical awareness seems to be rather superficial, an official « memory » remains, which is that contained, in an explicit manner, in the island's parish and civil records. 

The constitution of an exhaustive population file based on these registers has allowed us to reconstruct such genealogical networks. This enterprise was founded on the methodological advantages obtained in a small isolated population. It became possible to follow the transmission of genes from one generation to another, that is to say to define the biological flow active in the population, basing our research on the hypothesis of a correspondence between genealogical networks and real genetic links. 

An essential document, dated 1848, the Registre d'Inscription des Nouveaux Citoyens, allows us to obtain a precise idea, individual by individual, of the components of the population. This enables us to distinguish between former landowners and former slaves, who had recently been freed and, by this document, recieved an official patronym. From now on family names can serve as clues for the analysis of genealogical networks, whose intersectioning may be recognized in the descent of the two founding groups. 

Consequently, one may define initial segments which may be split up as follows : 

  • freed from 1848 and recently emancipated : 412 founders
  • individuals with names of original European landowners : 229 founders
  • « undetermined » individuals : 122 founders 

From these social positions, one may set forth the hypothesis that the first segment is nearer to an African origin (we will therefore refer to the « black » pole of the population) and the second segment to a European origin (the population's « white » pole). 

Using a program to calculate the probability of gene origin, it becomes possible to evaluate the total of these segments' genetical contribution to successive generations. This provides us with a first approach to the dynamics of population admixture. One observes in the table and in the graphic representation of the phenomenon (Table 1, Figure 1) that the « black » stock, the largest part of the initial generation (over 50%) progressively decreases and ends up by contributing around 30% of the last generation. 

Contribution of the different founders categories
to the successive generations (0%).









Initial Founders

« Whites »






« Blacks »






« Undetermine »

























generation 1






generation 2






generation 3






generation 4


















Unknown fathers

generation 0






generation 1






generation 2






generation 3






generation 4













The « white » stock, seems much more stable and in spite of a slight downward tendency maintains its contribution at the same level, passing from 29% for the initial population to 25% for the last generation. 

For each individual, we calculated his probability of descent from one of the initial segments of the population - having estimated the whole list of his fore-fathers. Each individual is then given a probability coefficient of « white » descent and a probability coefficient of « black » descent. These may range from 1 to 0 : if it is 1, it means that the individual has a perfectly homogeneous origin. Thus a certain number of individuals have probability 1 of being descended from the « white » segment (because all their founders ancestors are part of it). Thus they make up a white group which has remained untouched by all admixture throughout the generations. In putting up a barrier, the white sector has refused the penetration of outside genes. We may follow the fluctuating number of this sector from generation to generation : 229 individuals in the initial « white » segment ; 236 for the following generation, then 146, 89, and finally 86. This is certainly an obvious erosion, but the group has managed, nevertheless, to maintain its individuality and even to preserve a stable number, over the two last generations. This is a conscious strategy, influenced by the structuring pattern of racial ideology.

Figure 1.

Contribution of the different founders categories
to the successive generations [%]. 


However one may always assert that chance might have come up with the same result. The a contrario example of the black sector's destiny proves the contrary. To begin with a much larger initial segment (412 individuals), a light development to 431 individuals having a descent probability of 1 related to « black » segment for the next generation, followed by a drastic decrease : 152 for the following generation, 24 and finally 1 individual for the last generation. This is to underline how quickly a group may dissolve in a general population when it has no conscious strategy of survival. The term « Black » can only mark individuals who are in one way or another already of mixed blood. 

On the other side of the barrier encircling the white group we notice a strong mixing of « white » and « black » elements. The « white » genes do not stay inside the « white » group, but spread into the rest of the island population. The barrier appears to be permeable in one sense but not in the other : this hemipermeability means that the flow can only move from the white to the Coloured population, which will thus be in continuous evolution. Such a representation of reality implies a certain redistribution of the social order deriving from the identity processes : the children of mixed couples are in fact part of one of the two parent's populations, which will move perpetually while the other stays stable.


2.2. Anthroposcopic observations


The purpose of anthroposcopic observations presented in this paper is to account for the perception of morphological characteristics in the everyday life of the Désirade population. It is also to approximate the link between these perceptions and the mating pattern as it appears in the genealogies. Is a barrier to gene flow mostly ideological as in Terre-de-Haut, or more rooted in biological differences ? Is the endogamy appearing in the genealogies clearly related to physical features ? 

From the social point of view, the more significant physical traits are : skin colour, hair type, hair colour, nose shape and eye color. A difficult situation arises in Désirade and in Guadeloupe : due to political circumstances, any research explicitly mentioning racial differences is quite impossible. We were not able to use any kind of measurement (anthropometry, skin reflectance) directly related to such a problem without serious difficulties in particular for the sampling. As our purpose was to reflect the subjective criteria as they are defined in local society, it was still possible to use anthroposcopic observation. We used the same scales as we did previously in Martinique (BENOIST, 1963). Skin colour is evaluated with a scale derived from von Luschan's scale. Hair type and colour and eye colour have been observed in the same way as indicated by BENOIST (1963). All observations were made in daylight by the same observer. As previously mentioned, any kind of facial measurement was impossible and nose shape is not envisaged in this study. 

The results are presented in Table 2 and 3 and Figure 2. We present three groups of people regarding the probability of Black admixture according to genealogical records : group A has no black ancestors ; probability for people included in group B is between 6,25% and 43,75% ; in group C, the probability varies from 56,25% to 100%. The differences for the morphological markers of identity show a clear contrast between groups, for any characteristic, mostly for skin colour and for hair colour. In this preliminary analysis, it was not necessary to use more sophisticated statistical analysis ; the results are clear enough to give an unambigous answer to the main question of this paper : The white « quasi-group » (in sociological terms) is still a phenotypical group. The congruence between genealogies and morphological features gives way to a more complex analysis, in which genealogies, morphoscopy and genetic markers have to be confronted.

3. Discussion


« Whiteness » appears as a biosocial reality. It is a social one, because the colour is perceived according to social criteria, and will play different roles according to social values. But colour is also a biological fact. The phenotype is mostly the result of genes, and any decision based on phenotype interacts, directly or indirectly, with the genetic level. The morphological criteria are the object of assortative mating, contributing to a reduction of chance exchanges and to the subdivision of the whole population between semiopen sub-groups. In a subsequent paper we will present the results of different genetic markers following the rate of admixture in these different sub-groups. 

Proceeding from a social to a biological level, we can explain certain biological features of the population. But social reality is changing. Values are related to the larger scale organization. In Désirade, as in Guadeloupe, the old models have been profoundly altered in recent years. The situation depends on the historical period involved and on the different external societies of reference concerned. 

Figure 2.
Anthroposcopical characteristics following admixture. 


 At historical level, it is possible to distinguish three periods : the first before 1848 when a racial system emerges from a social system and from phenotypical diversity ; a second after 1848 when the latter behaviours are still present (and the biological structures Concerned) without being related as earlier to social positions ; a third since 1948, when this racial system vanished and where social practices and associated biological phenomena disappeared. Finally the very existence of white group, maintained until then in Désirade, is threatened. The current ideological atmosphere is characterised by a certain official anti-racial ethical code, which tries to minimize differences of colour. Such an evolution is partly due to internal maturation, but also to external variables. 

We cannot understand the Désirade situation at different times without referring to different social spheres into which the island has been integrated. The first one is the sphere of Guadeloupe. There we notice an evolution which is characterized on the one hand by the conservation of a certain socio-racial system (from the correlation between classes and colour categories) and on the other hand by the progressive accession to power, essentially at the symbolic level of representation, of the Coloured masses. The Guadeloupean identity, from now on, passes through a certain valorization of blackness. The second is the sphere of France. There one observes, especially from the Third Republic onwards, a strong ideological promotion of racial equality. But this new deal must be tempered by practices which to a varying degree tend to marginalize Coloured people in the French social structure. One should remember that a White may always live down his West Indian origin if he leaves the Caribbean. Colour then appears as a de facto barrier to social elevation in French society as a whole. Insertion into French society reinforces the emphasis on equality, but also the principle of white domination, which underlies the ambiguity of the system.




It seems useful to dismantle the mechanisms of this phenomenon - an extremely complicated series of continuous interactions and retroactions between the biological sphere and the social sphere. At the beginning, there was a biological difference between social sectors, caused by historical hazard. This difference is interpreted culturally from the most visible feature, that is to say a certain number of contrasting phenotypical characteristics : colour of the skin, face features, hair shape. Discrimination based on these characteristics, through positive or negative values given to physical types, is evidently of an ideological nature. This kind of ideology, however, depends on biological reality in which the original phenotypical differences have reappeared generation after generation. This reproduction has been socially conditioned by whether people meet or not, that is to say how they marry or procreate. The barriers have come from this system which canalized the transmission of a collective genetic pool. These societies then permanently manage a biological phenomenon, namely the transmission of a certain number of discriminating characteristics, in the same way as others manage the transmission of heritage within family lines.




BENOIST J., 1965. Saint-Barthelemy : Physical Anthropology of an Isolate. Am. J. Phys. Anthrop., 22 : 473-488. 

BENOIST J., 1963 Les Martiniquais, anthropologie d'une population métissée. Bull. Mem. Soc. Anthrop. Paris, 4, XI : 241-432. 

BENOIST J., 1971, Population Structure in the Caribbean Area. In : F.M. Salzano, (ed.), The On-going Evolution of Latin American Populations, C.C. Thomas, Springfield, pp. 221-249. 

BONNIOL J.L. 1980. Terre-de-Haut de Saintes. Contraintes insulaires et particularisme ethnique dans la Caraïbe. Paris, Editions Caribéennes. 

HARRISON G.A. 1985. Human Biology in Britain. Anthropology Today, 1, 2 : 13-14. 

HOETINK H., 1967. The Two Variants in Caribbean Race Relations. A Contribution to the Sociology of Segmented Societies, Oxford. 

NATSH JULIA, 1972. Race and Rank in a Caribbean Island : Désirade, Dact, London. 

Received : July 4, 1987 ; Accepted : June 18, 1988.

Retour au texte de l'auteur: Jean-Marc Fontan, sociologue, UQAM Dernière mise à jour de cette page le dimanche 30 septembre 2007 14:31
Par Jean-Marie Tremblay, sociologue
professeur de sociologie au Cégep de Chicoutimi.

Saguenay - Lac-Saint-Jean, Québec
La vie des Classiques des sciences sociales
dans Facebook.
Membre Crossref