Raymond T. Smith

Copyright 2000:   All Rights Reserved

Go To Site Map

CHAPTER IV

COMPOSITION OF THE HOUSEHOLD GROUP

n the literature on the West Indian family, writers have frequently mentioned the fact that the household is an important social unit and they have sometimes tried to indicate the nature of these units.  Professor and Mrs.  Herskovits writing of a village in Trinidad, state explicitly that:–

‘The definition of the Toco family, in any functioning sense, must give a prominent place to the individual household’ (Herskovits, M. J. and S. F. 1947: 164).

Having said this, they then proceed to base their discussion on the case of one family group only, though they do give some very rough figures on the approximate number of children per household in a sample of 106 households.  The figures were compiled from the statements ‘of third persons concerning their neighbours and friends’ and consequently are not very reliable.  The Herskovitses show considerable insight into the nature of the problem involved in studying the family system in this area, but their predominating interest in ‘African survivals’ soon leads them away from it, and the greater part of the chapter on ‘The Functioning Family’ is taken up with a description of the customs surrounding childbirth, child rearing and courtship.  The preliminary insight into the basic importance of the household as a social unit is never developed, and the formulations concerning the nature of the household group are never given any precision.

Simey is similarly vague on this subject, though he is writing from a much more general point of view, and he does quote certain information collected by Mr. Lewis Davidson in Jamaica.  He tells us that:—

The family group is, indeed, one which is brought together in a very casual way, and this obtains for the Christian as well as the Disintegrate family.  Mr. Davidson discovered the striking fact that amongst all the 270 families he studied in Jamaica, not a single one consisted only of parents and their children (Simey 1946: 84).

Since we do not know what Mr. Davidson’s object was in studying these families nor the methods he used it is difficult to evaluate this piece of information, but it does seem that there is at least an implied comparison with what is conceived as the ideal type of European family group consisting of the nuclear family only.  Madeline Kerr, also writing of Jamaica, states that:—

The members of a Jamaican household must not be identified with the family.  Sometimes the household does consist of mother, father and own children only, but more often it contains a collection of people tied by kinship or sometimes only by proximity (1952: 56).

Again there is the implied comparison with the ideal European type ‘family’ and we are left with a very vague impression of what actually does constitute a household group in this society, and if there are variations we certainly don’t know the frequency of their occurrence.

Henriques has a more detailed discussion of the domestic group and proposes a four-fold classification according to the type of conjugal union on which the domestic group is based, with one residual category to take care of those groups where there is no male head present (1953).  Simey has proposed a similar type of classification based on the work of Mr. Lewis Davidson (1946).  Henriques is well aware that his four types of domestic group do not constitute mutually exclusive categories, and he specifically states that:–

‘This classification is not rigid as a domestic group can during its history experience several or all of these forms’ (1953: 105–6).

Not surprisingly he runs into some difficulty in imposing his fourfold classification, especially when he finds it necessary to include households with a male head under his category of ‘Maternal or Grandmother Family’.  It is difficult to evaluate the significance of his scheme since he presents very little in the way of case material, and practically no distribution figures apart from statistics taken from the Jamaica Census, which deals in different categories.  His ‘rough estimate’ of the incidence of different types of domestic group is very unsatisfying, for it is a matter of some importance to know the incidence of each type more precisely when one is dealing with such apparently heterogeneous data.

In British Guiana Negro communities, individual households do appear to exhibit a wide variation in the categories of persons who make up their membership, and a cursory examination of August Town was sufficient to make it quite clear that there was no superficially apparent norm of domestic grouping.  Over 300 houses in the village seemed to contain a somewhat bewildering array of occupants, and although one could draw up a list of various ‘types’ of grouping it seemed necessary to make a comprehensive survey of all households in order to introduce an element of certainty into any discussion of what is ‘normal’ for this community.  Fortes has demonstrated the difficulty of arriving at definitions of normality in domestic groupings in modern Ashanti, and shown how the application of simple statistical techniques can serve to illuminate structural features which are otherwise obscured by the application of blanket terms such as ‘matrilocality’ and ‘patrilocality’ (Fortes 1949b).  The same general method of approach was applied in British Guiana, and indeed it would seem to be the only logical method to adopt under the circumstances and for this particular kind of study.[1]

We begin then by considering the composition of the household group in terms of the relationship of its members to the person we have defined as the ‘household head’.  Our first distinction will be between households with a male head and those with a female head, for these categories form convenient points of reference for a synchronic enumeration and they are broad enough to prevent us making prematurely specific classifications.  This avoids for the time being any preconceived notions as to the genesis of household groups, and we take particular care not to introduce terms such as ‘matriarchal’ or ‘patriarchal’ at this stage.

For the purposes of classification it has been necessary to attach labels to persons to indicate their conjugal status, and the following terminology which was used in the field has been consistently adopted throughout this book.  The terminology has its limitations but taken in conjunction with case histories it does provide a convenient set of working definitions.

1.      Married.—Legally married.

2.      Common-law married.—Where a person is living in the same house as their partner without being legally married.  This category definitely implies cohabitation.

3.      Single mother or single father.—Where a person is the biological mother or father of a child or children which he or she recognizes, but has never lived with a member of the opposite sex in a marital or common-law union.

4.      Single.—Where a person has never had any children and never lived with a member of the opposite sex in a marital or common-law union.

5.      Widower.

6.      Widow.

7.      Common-law widower.

8.      Common-law widow.

9.      Separated.

10.  Common-law separated.

11.  Divorced.—Only refers to those cases where a divorce has been granted by the courts.

It is realized that some individuals may fall into more than one of the above categories if their whole life history is taken into account, though of course it is the status of the person at the time of the survey which is of primary interest to us here.  The term ‘spouse’ has been used in a rather unorthodox way to include not only marital partners of either sex, but also common-law partners of either sex, so that wherever the term is used no distinction is made between marriage and common-law marriage.  The term ‘common-law marriage’ has been adopted not only because of its wide use in the West Indies by government agencies, but also because it does convey the idea that such a union is almost the same as a marital union apart from the legal and religious sanction.  The term ‘concubinage’ carries more of the suggestion of extra-marital relationship, and will only be used in contexts where its meaning is made quite clear.

In the following tables the membership of household groups in terms of relationship to the head is enumerated in detail for all three villages.  In the case of August Town and Perseverance the figures are based on complete censuses of the villages, whilst in Better Hope they are based on a random sample of eighty-four households.  For all three villages certain cases have been excluded, including those where a man or woman lives completely alone of course.  It must be remembered that these are simple distribution figures of persons standing in a definite relationship to the head of the household and have a limited value in that they cannot tell us anything of the particular reasons for the configuration of any specific household.  Tables VIIIa and IXa present a more detailed breakdown of the categories enumerated in Tables VIII and IX.

Tables VIII and VIIIa show the composition of those households with a male head in all three villages.  It is immediately apparent that by far the greater number of households with a male head also have the head’s wife or common-law wife present, and this confirms the view that all households irrespective of headship tend to be matrifocal.  In all three villages there is a uniformity in the percentages of different kinds of kin present in the household, and in each village there is a larger proportion of persons in this type of household who are related to the head’s spouse than to the head himself.  Apart from a few adopted children, we do not find any members of these households who are not kin of either the head or his spouse, so that we can immediately characterize the household group as a kinship group.  There are one or two exceptions where we find persons boarding with the head but these are very marginal cases.

It is also clear that the majority of male household heads in all three villages are married, and even where the head has a common-law wife he is definitely supposed to provide for her and their children, and their relationship cannot be considered as any frivolous arrangement.  It implies quite definite reciprocal rights, duties and obligations, and is a socially recognized form of union.  Categories 3 to 8 inclusive represent children of the head and his spouse and comprise the largest single category of kin in these households, representing 57.91 per cent of all members (excluding the heads themselves) in August Town, 62.79 per cent in Perseverance and  62.78 per cent in Better Hope.  The great majority of these children are the offspring of both the head and his spouse, but the presence of a few children of either the head or his spouse alone shows that couples entering a conjugal union can bring children by other liaisons to live in the household, and they are treated in exactly the same way as the other children.  In August Town there is a tendency for these children to be found more frequently in households where the  couple are in a common-law union, and this bears out the observation that women only enter common-law unions in this village when they are already mothers.  Of the children, relatively few are over 18 years of age, and there are more adult daughters than sons.  So far as August Town is concerned, this is correlated with the fact that many young men are away working at the bauxite mines and these have not been included as resident members of the household.  In Perseverance young men who make regular seasonal migrations to the sugar estates have been included, and only those who are permanently away or away on extended working trips have been excluded.  This balances the numbers somewhat in the case of Perseverance.

In Better Hope the numbers are almost equal, and this reflects the more stable marriage pattern in this village, as well as the fact that there are almost equal opportunities for both young men and young women to find work in Georgetown.

Households with male heads are primarily two generation groups devoted to the rearing of children.  They are also predominantly based upon a conjugal union of some kind.  Categories 9 to 20 inclusive are the grandchildren of the head and his spouse (with an almost negligible number of great-grandchildren which only represent one case in August Town, where an old man was titular head of a household containing his granddaughter and her husband).  These grandchildren are predominantly daughters’ children and are found more frequently in August Town than in either Perseverance or Better Hope, being relatively few in the latter village.

Categories 21 to 25 inclusive are children’s spouses and in keeping with the rule that a man sets up his own household on entering a conjugal relationship there are very few persons in this category.  The figures are not large enough to have any significance for the formulation of a statement as to whether marriage or common-law marriage is viri-local or uxori-local in cases such as this, and the people themselves have no explicit rule.  Where a child’s spouse is living in the household in these cases it will usually be for some specific reason.  The young couple may be waiting to finish building their own house, or in the case of Perseverance the young husband may be working away for most of the time.  In one case in August Town, a daughter’s common-law husband was a stranger to the village and the girl had lived with him on the East Coast of Demerara before they both came to settle in August Town.  He was in a very precarious position in the household and whenever a quarrel broke out, his common-law wife would threaten him with expulsion.  Such quarrels were particularly frequent when there was no work available on the sugar estates and the man would have to sit around all day doing nothing.  He had no land in the village and absolutely no means of providing anything for his common-law wife and their child, and on one occasion he resorted to stealing cassava from someone else’s farm, only to be discovered and branded as a ‘tief-man’.  Whenever a quarrel broke out, he would remove himself to a neighbouring yard and sit there until the storm had passed.

Categories 26 to 29 inclusive, and 30 to 35 inclusive, make an interesting comparison, because it is clear that collateral kin of the head’s spouse predominate over his own collateral kin and it is particularly children of the spouse’s sister that get incorporated into the household.  This helps to confirm the fact that the female spouse is the real focus of the group, but it also illustrates a point about the relations between sisters which is discussed more fully in Chapter VII.  This is much less true for Perseverance and Better Hope, and in Perseverance there is actually a greater number of the head’s collaterals than those of his spouse.  However, the figures alone conceal certain important variables.  In Perseverance there is one case where a young man and his common-law wife are sharing a house with his sister and her common-law husband, and this one case accounts for a large proportion of the numbers involved in the whole category.  An arrangement such as this is a temporary one and were it not for the fact that the two couples cook one pot we should have treated them as two separate households, which in many respects they actually are.

Fathers and mothers of the heads and their spouses are very few, and in these cases there is no doubt that the old persons are not the heads of the households.  They have been taken in by their children, who are caring for them, and although they are respected and exercise considerable influence in the household they do not control it.

Adopted children are adopted in the full sense of the term, not being regarded as servants or domestic help.  They grow up as children of the person adopting them, and often inherit property in the same way they would if they were natural heirs.  Childless couples often adopt one or more children, and in one case in August Town, a Negro couple had adopted East Indian as well as Negro children.

Other kin of head and his spouse are persons not covered by the above categories such as a mother’s sister’s child or grandchild and the persons listed as non-kin are paying boarders, such as young men working at the near-by Post Office, or in the Perseverance case, three young male schoolteachers who were boarding with the head-teacher and his wife.

COMPOSITION OF HOUSEHOLD GROUPS WITH A FEMALE HEAD

Tables IX and IXa show the categories of persons present in households with a female head, excluding the heads themselves, and once again it is clear that these household groups are in fact kin groups, there being very few non-kin present at all.  Persons on very short visits were not included, where the period of the visit did not extend beyond a few weeks.  Persons away from the villages on short visits of not more than a few weeks were included, as were men working away for short periods of time.  Some households in August Town which at first appeared to have a female head, have been included in the male-head group where it was a case of the head being out of the village working, but regularly sending money home to his spouse, and coming back to the village whenever he could.  In these cases, the female was only head of the household by virtue of the absence of her spouse at the time of the study.  Conjugal relations were maintained, and the couples were definitely not ‘separated’ in the technical sense of the term.

Households with female heads are predominantly three-generation groups and this can be seen from an examination of the tables.  Categories 1 and 2 of Table IXa are children of the head and they comprise 36.58 per cent of all household members (excluding the heads) in August Town, 35.71 per cent in Perseverance and 49.12 per cent in Better Hope.  However, unlike the case of households with a male head, they are not the largest single category throughout, and the majority of them are over 18 years of age.  They are predominantly daughters.  The largest single category is that of grandchildren and great-grandchildren, comprising categories 3 to 12 inclusive in Table IXa.  Better Hope is excluded from this generalization, for there are fewer unmarried mothers and fewer women taking care of their daughters’ children.  The reasons for these variations as between the three villages will be taken up later, but the point which concerns us most at the moment is that in these households the children under 18 years of age are mainly the head’s grandchildren or great-grandchildren, or her nieces and nephews.  Categories 13 to 16 inclusive are siblings and siblings’ children and here again we see quite clearly the predominance of sisters’ children, particularly in August Town.

The percentage of spouses of the two descending generations from the head is greater than in the case of households with a male head, and the percentage is highest in Perseverance, where housing is bad, and the age at which persons enter conjugal unions is lower.

At this point, a deficiency in the field-work must be noted which may have resulted in our placing some cases in categories 4 and 6 which should properly be in category 9.  It is likely that information concerning the mothers of some children did not include whether they were dead or alive, though it did include whether they were married, common-law married or single.

Category 28 calls for some explanation.  In these cases there was no doubt that the woman was head of the household and that the man was in a subordinate position in the household group.  In most of the cases, the woman had her grown-up children around her, and had no children in common with the man.

CONCLUSION

These tables present us with an over-all picture of the situation as it existed in the three villages at the time of the study, and if we so desired we could go on to develop a synchronic classificatory scheme in much more detail.  Synchronic in this sense would mean at that point in time at which the field-study was carried out, or the phenomena observed.  This is essentially the process followed by the writers who have drawn up lists of family types.  The limit to the number of such types would be determined only by the range of variability of our data, and the range of variability is fairly extensive.  Thus we could have elementary or nuclear families based on marriage and on common-law marriage; three generation families with a female head; two generation families with a female head; families including collateral kin of either the male head or his spouse; and so on until we had exhausted our range of variations.  This procedure would be no more valueless than squeezing all our ‘types’ into three or four procrustean categories and leaving it at that.  In either event we should have left out the vital dimension of time, and it is only by trying to arrange the various types of household grouping along a time axis, that we shall be able to gain a real insight into the nature of the structural principles, which are at work in determining the composition of any particular household group at the time of the field-study.  Our synchronic unit then becomes not the arbitrary period of time spent in the field but the period of time represented by the lives of the three generations normally existing at the time of the study.  The actual cases we observe represent various stages the life process of individuals and of household groups and if w allow for factors of change which have taken place in the recent past we should get a clear picture of the normal cyclical growth an decay processes involved.  In fact, there has been very little real change during the past fifty years or so in any of the three villages, and such changes as there have been, have not affected the family system to any significant degree.  This general type of analysis has been documented by Fortes and the present study confirms his view that structure is to be seen as ‘an arrangement of parts brought about by the operation, through a period of time, of principles of social organization which have general validity in a particular society’ (9).  Our only modification to this view would be that the ‘principles of social organization which have general validity’, may vary as between certain groups or social classes within the same total society, and account will have to be taken of this new social dimension.  Thus in British Guiana, these principles will vary as between different status groups in the colour/class hierarchy, and this has certain repercussions on our three villages in so far as they exhibit a greater or lesser tendency towards the norms of status groups higher than themselves.  Thus Better Hope approximates more closely towards the ‘middle-class’ norms than do August Town and Perseverance.  There may exist separate, though overlapping and cross-cutting value systems, and structure will vary correlatively with the intensity of application of these as well as along a time dimension.

In the next chapter we shall deal with the analysis of variation over time, and in Section III we discuss some aspects of the variations to be detected in association with the variability of cultural and class norms.

Return to TOP

Go to Chapter V