Emre Kongar'ın Resmi* İnternet Sitesi


Green Bullet Makaleler

Green Bullet Articles in English

Sürekli Yazılar


Emre Kongar, “A survey of familial change in two Turkish gecekondu areas”, in J. G. Peristiany, ed., Mediterranean Family Structures, (Cambridge University Press, 1976), pp. 205-218.


This paper makes a comparative study of the size, structure, kin relations and relation to formal organizations of the gecekondu family. The intention is to explore some of the changes that a rural family undergoes when it moves to the city.[] The data used have been gathered through two community studies, the first being the study of the family in Izmir, in which the gecekondu family has been made a separate category,[] and the second being the study of the Altindag gecekondu area in Ankara.[]


THE CONCEPT OF gecekondu

One of the most acute social problems in Turkey is the mass migration from rural areas because cities lack the facilities to absorb the incoming population. The result of the migration is the formation of gecekondu areas where a rural population wrestles with the problems of adapting itself to urban life.

The gecekondu is officially defined as ‘the dwelling unit on somebody else’s site which was built without obtaining the approval of the landowner and built in a way which is not approved by the general legal provisions for buildings and construction’.[] It is usually constructed out of second-hand material to a very low standard;[] it lacks utilities and, by urban standards, constitutes a health hazard.[] Gecekondu areas grow very rapidly and the lack, or weak enforcement, of master plans in metropolitan areas is the major contributory factor to such a development. At election time the rate of growth of gecekondus rises as vote-seeking politicians pave the way for building new gecekondus by legalizing old ones and by providing utilities for them in order to get the votes of the population arriving from the rural areas.

Gecekondu areas are located around large cities like Istanbul, Ankara and 59.2 per cent of Ankara’s population, 45 per cent of Istanbul’s population[]


Year Interval

Annual total

population increase


Annual urban

population increase (%)























and 33.4 per cent of Izmir’s population[] live in gecekondu areas. These figures seem high when compared to Lima, for instance, where the barrida population (which corresponds to the gecekondu) was only 20 per cent of the city population in 1964. Again, in Rio de Janeiro the favela population was about 16 per cent of the city population in 1964.[] Though there are other contributory factors, there is no doubt that the main force behind the gecekondu phenomenon is the urbanization rate of Turkey. The increase in the urban population (see Table 1) sheds some light on the problem.[]

As Table 1 shows, urbanization in Turkey gained momentum in the 1950s, and since then the annual urban population increase has been twice that of the increase of total population. This rate of urbanization is very high in comparison to Latin American figures.[]

The inhabitants of gecekondu areas come mostly from rural areas, although some come from small towns and villages. Direct migration from the place of birth to the gecekondu area is usual and incomers start by taking up urban-type jobs.[] With regard to their clothing, diet, and daily habits it seems that they have assimilated some urban patterns.[] Though Yasa asserts that some rural characteristics, like traditional wedding ceremonies, bride price, etc. still exist among the gecekondu population,[] it is my belief that these characteristics do not exist in the gecekondu population itself but can be observed in the relations of gecekondu population who maintain relations with the rural communities still observe rural traditions, it does not mean that rural traditions (such as bride price, etc.) are followed by the gecekondu population as a whole. Furthermore, though some people maintain contact with their rural communities, hardly any of them want to go back. Gecekondu areas have their own social stratification. The upper class consists, roughly, of traders, shopkeepers, etc.; the middle class of qualified workers, artisans, etc.; and the lower class of non-qualified workers, porters, janitors, peddlers and the unemployed.

As Turkish cities lack the facilities and other advantages which constitute the ‘pull factors’ of city life, it seems that the disadvantages of rural life, or ‘push factors’, are the predominating reasons for migration. Lack of land and mechanization of agriculture are the main impulses behind the ‘push factors’. Migration is, therefore, not by free choice. People are in a sense forced to move.


Izmir is the second largest industrial city in Turkey. It has a population of about 1.5 million people, 600,000 of whom live within the municipality. The gecekondu population of the city consists of about 200,000, most of whom moved into the city more than ten years ago (61 per cent; the comparable figure for Izmir as a whole is 71 per cent). The newcomers to gecekondu areas (who moved in 5 years ago or less) form about one-sixth of the population (16 per cent; 12 per cent for Izmir). Ownership of houses is high: 76.5 per cent of the gecekondu population are families living in their own houses, whereas in the city of Izmir the figure is only 65.8 per cent.

Income is rather low. The mode of the income distribution among gecekondu population is between 250 and 500 TL (9 TL = U.S. $1.00 in 1968) per month per household. The mode for Izmir as a whole is between 500 and 1,000 TL.[]



The second community to have been studied is the oldest gecekondu area in Ankara, the capital of turkey, in which the service sector is predominant. This is Altindag gecekondu area first settled in the late 1940s. It consists of about 5,000 households with a population about 27,000. A majority of the population (61 per cent) moved to the city more than 10 years ago (the figure for all the Ankara gecekondu areas is 53 per cent).[] The newcomers (those who moved in 5 years ago or less) form about one-fifth of the population (almost the same rate as for all gecekondu areas in the city).[] More than half of the Altindag population (57.3 per cent).[] Mod value of income distribution in Altindag is between 501 and 1,000 TL per month per household. (Yasa and others found an average about 400 TL.)[]


As stated before, the data used in this paper have been gathered through two community studies (Izmir and Altindag) and structured interviewing was used in both cases, therefore any limitations are the result of survey techniques. Secondly, the communities studied are not similar. The Altindag gecekondu area has a population of about 5,000 households. It is the oldest gecekondu area and is located in a city in which the service sector is dominant. The Izmir study, on the other hand, covers the whole city which is an industrial one. Therefore, not only do the universities differ, but the characteristics are also dissimilar. One is old-established, the other is more mixed: one is located around a city in which the service sector is dominant, the other is in an industrial city. So these differences must be borne in mind when evaluating comparisons.

A third limitation is the result of treating the gecekondu phenomenon as homogeneous and uniform all over Turkey. There is no doubt that it is the outcome o urbanization processes and that, therefore, the population in gecekondu areas is in transition, but there the similarities end. In a paper which studies a subject like the family, the differences between the various gecekondu areas cannot be taken into account as they should be. To begin with we do not have enough comparative data on the family in the various gecekondu areas. Then the characteristics of the different gecekondu areas have not been studied according to objective criteria. Moreover, we have no overall vision of the gecekondu family.

A fourth limitation is due to the fact that there are not enough studies of the gecekondu phenomenon in Turkey itself. The fifth and last limitation is the lack of family studies in Turkey as a whole. Consequently, the findings of this paper must be treated with caution: the writer has used them merely as clues towards further research.


The average size of the family in various parts of Turkey, according to the various studies, is indicated in Table 2. In this paper ‘family’ is defined as ‘household’: I.e. relatives living under the same roof whose food is cooked in common. First of all it should be noted that our findings with regard to the gecekondu family size in Izmir are very close to the figure given by the Ministry of Reconstruction and resettlement. If the small family size is taken as an indicator of urbanization,[] it could be said that the gecekondu family in Izmir and Istanbul is quite urbanized. But in Ankara the case is rather different. An average family size of 5.5 is what Yasa found in all gecekondu areas of Ankara. He claims, therefore, that such a high average could be taken as indicating the rural character of the gecekondu family.[] On the other hand, Kiray says that the average size of 5.2 of the family in Eregli (which is a Black Sea coastal town) approximates to the urban family.[] This may be due to the context in which the figures are considered. Yasa is studying the gecekondu families in Ankara, which is an urban area, while Kiray studies families in Eregli, which is a semi-urban area. Consequently Yasa thinks 5.5 is more than the city average of 4.63 and Kiray thinks 5.2 is less than the village average of 6.16. I, myself, think Table 2 solves the problem by showing that the average of towns like Eregli is 5.2 which is very close to what Kiray found in Eregli. This figure also demonstrates the intermediate character of Turkish towns. Table 2 shows that the average gecekondu family size in Ankara is greater than that in Izmir and Istanbul. This may be due to the fact that both Izmir and Istanbul are more

TABLE 2 The average number of household members

Turkeya 5.7

Citiesa 4.6

Townsa 5.2

Villagesa 6.2

Central Anatolia(villages)b 7.2

Eregli (villages)c 6.3

Elbasid 6.0

Gecekondus in Ankarae 5.6

Half 5.5

Hasanoglang 5.4

Eregli (town)c 5.4

Gecekondus in Istanbulh 5.2

Izmir Cityk 4.7

Gecekondus in Izmirl 4.4

Sindelm 4.1

Gecekondus in Altindag 5.1

Gecekondus in Izmir 4.7



a Census of Population, 1965, pp. 672-5, Tables, 53a, 53b, 53c.

b Sahinkaya, R. Orta Anadolu Koylerinde Aile Strukturu (The Family

Structure in the Central Anatolian Villages), Ankara Universitesi Ziraat

Fakultesi Yayinlari, 1966, p. 34.

c Kiray, p. 201 and p. 115.

d Stirling, p. 37.

e Yasa, Ankara’da gecekondu Aileleri, p. 108.

f Erdentug, pp. 31-2.

g Yasa, Hasanoglan, p. 79, Table 20.

h Istanbul gecekondulari.

k Kongar, p. 68.

l Gecekondus in Izmir.

m Yasa Sindel, pp. 34-5.

industrialized than Ankara, which would affect the rate of change in family size and structure. Our Altindag finding, which is less than the Ankara gecekondu average, could be due to the fact that Altindag is the oldest gecekondu area in Ankara. Because the area is older, the population has had more chance to assimilate urban values and way of life which, in turn, affects the family size. It could be said, therefore, that, in terms of size, the gecekondu family in Turkey is quite urbanized, and that this urbanization is probably affected by the level of industrialization in the city in which the family lives.

As for the structure of the gecekondu family, the comparative percentages are shown in Table 3. Before we evaluate Table 3 it should be remembered that in sociological literature certain structures tend to be associated with certain functions. For instance, extended forms of household are thought of as families of ‘great functionality’ and nuclear forms of household as families of

TABLE 3 Percentages of nuclear and proliferated families in various places in Turkey



Villages in Central
















































Sources: a Sahinkaya, p.34. e Kongar, p. 72.

b Stirling, p. 38. f Kongar, p. 64.

c Yasa, Hasanoglan, p.79. g Yasa, Gecekondu Aileleri, p. 104.

d Kiray, p. 114. h Yasa, Sindel, p. 35.

‘little functionality’.[] But I, myself, do not agree with this and especially not when transitional societies are in question. Even if such an approach were true for ‘modern societies which have completed their industrialization, or for traditional societies which have not yet started their process of change, it is certainly not acceptable for transitional societies in which the rate of change in structure and function is not necessarily the same. The best example of a different rate of change with regard to the structure and functions of the family is given by Yasa. In one Turkish village he found that nuclear family-type household members have extended family-type interactions. For this he invented the term ‘narrow family’[] to describe the nuclear household forms in which the father has the rights and privileges of the family head of extended family-type households. In this paper, however, we follow Levy’s classification of ideal family structures without giving them a functional content.[]

When, with the above considerations in mind, we look at Table 3, we could say that the gecekondu family in Turkey has a modern nuclear family structure as far as the composition of the members of the household is concerned. It is interesting to observe, however, that the percentage of nuclear families in the Ankara gecekondu area is quite high. This may be due to the fact that Yasa classifies one-member households and households in which there is no married couple as nuclear, whereas I classify them as broken or incomplete.[] The table also indicates that the Turkish family structure is changing from proliferated to nuclear. If we differentiate between vertical and horizontal proliferations, the transition can be seen more clearly. The percentage and characteristics of the familie souche in Izmir and Altindag are also given in Table 3. It is interesting to observe that the horizontal proliferation of the nuclear family is practically non-existent. It could, therefore, be said that even the proliferated families in the gecekondu areas have lost their traditional character and have become transitional.

Table 4 shows s once more that the main characteristic of the Turkish family is patrilineal, Kiray found that the same tendency is to be seen in Eregli.[] Patrilineal families form 27.5 per cent of the whole, whereas matrilineal families for only 7.8 per cent. Similar findings were reported by Yasa for the gecekondu areas in Ankara. According to his figures, there are five


TABLE 4 Percentages of family structures in Izmir and Altindag

Family Types

Izmir City

Izmir gecekondu

Altindag gecekondu





Famille souche





Famille souche









Broken or incomplete









as many patrilineal as there are matrilineal families.[] The same tendency is observed by Erdentug in two villages in the province of Elazig in the eastern part of Turkey.[] The patrilineal character of the proliferated nuclear families is also reported for the village near Ankara,[] and can be seen quite clearly in Stirling’s study of a Turkish village.[] But such a patrilineage should be evaluated in the context of dominant family structures. In communities in which the nuclear structure seems dominant, patrilineages would mean sons are taking care of their parents. For instance, in Izmir the head of the household belongs to the younger generation in 65 per cent of the patrilineal famille souche structures. One must, therefore, be very careful about patrilineages in Turkey, especially where the family cycle is concerned.

In order to find out how the head of the household sees his authority within the family, I asked questions about attitudes towards children’s free choice of a mate and a profession. Tables 5 and 6 show us the results. Unfortunately, our data in Izmir would not allow us to differentiate

TABLE 5 The percentages of heads of households who will

let their children decide whom to marry


Izmir City

Izmir gecekondu

Altindag gecekondu

For girls For boys



44.5 75.1

TABLE 6 The percentages of heads of households who will

let their children decide whom to marry

Izmir City

Izmir gecekondu

Altindag gecekondu

For girls For boys



25.2 87.6

between boys and girls. The results from the tables can be summarized as follows: (1) in Izmir, the heads of households in the gecekondu areas are more likely to direct their children on both questions than average household heads in the whole of the city of Izmir. This would suggest that they have not yet completely absorbed urban values. (2) The Altindag gecekondu family seems much more lenient than the Izmir gecekondus family when it comes to children’s choices. This might be due partly to the fact that the wording of the questions has been changed slightly and partly to the fact that the Altindag area has a somewhat higher socio-economic status. (3) There is a clear-cut differentiation between boys and girls in favor of the boys; understandably, families are more likely to direct the girls. (4) Most household heads seem authoritarian towards their children’s choice of mate and profession. (5) This authoritarian attitude reminds us of the family structure, which Yasa has called ‘narrow’. (6) When dominant authoritarian tendencies are seen to go hand-in-hand with the nuclear family structure, it re-enforces our views about the different rates of change between structures and functions.


Physical proximity to relatives is shown in Table 7 and relationships with them in Table 8. The relatives whose interactions were studied are ‘first degree’ relatives, I.e. parent, children, siblings, aunts, uncles of both spouses and their children. Interaction between them for the purpose of lending and borrowing goods and money was investigated. Table 7 shows us that more than half of the families have close relatives living in the next neighborhood or even nearer. On the other hand, Table 8 shows us that the majority of the families tend not to combine their resources with

TABLE 7 Percentage of families who have close relatives in

the next neighborhood or nearer

Izmir City

Izmir gecekondu

Altindag gecekondu





TABLE 8 Percentage of families who borrow and lend goods and money with relatives and with neighbors




Borrow Lend







Borrow Lend










Altindag gecekondu

With relatives

30.0 18.3


32.5 12.3



With neighbors

11.9 8.7


17.7 10.6



those of relatives for the purpose of lending and borrowing goods and money. This picture suggests that the gecekondu family is fairly isolated from its relatives. The suggestion is supported by the fact that the majority of the households (67 per cent) in the Izmir gecekondu area do not have a reciprocal helping relationship for the purpose of housework with anybody. Such an isolation seems quite significant as the physical conditions (i.e. proximity) for interaction with relatives exist. This certainly suggests that the idea of separate nuclear families under separate roofs functioning as one family under the dominance of the paternal head is not a valid concept in the gecekondu areas. It means that neither joint family structure, suggested for rural areas, nor modified extended family structure, suggested for industrial societies,[] is likely to be found in gecekondu areas.

Yasa has also pointed out the weakening of kinship ties. He thinks that because those who find jobs in Ankara through their relatives, kinsmen and acquaintances constitute a low percentage (17 per cent), this is an indication of weakened ties.[] In our Altindag survey the figure is only 12 per cent and it is my opinion that it indicates that kinship ties of the gecekondu family are already weakened rather than that the passage of time weakens them.

Table 8 also shows us that the lending and borrowing of goods is more frequent among neighbors than it is among relatives. In addition to the figures shown in the table, my additional data on reciprocal help in housework reveals the same tendency. The number of families in Izmir gecekondus who have a reciprocal helping relationship with neighbors is greater (17 per cent) than the number of the families who have this relationship with their relatives (14 per cent). Though the difference between the figures is not great enough to allow any meaningful comparative analysis, we can observe a tendency to interact more with the neighbors than the relatives. There is no logical explanation for such a tendency at the moment, except to say that the neighbor hood can be seen as a primary group in gecekondus functioning as a substitute for relatives.[]

The family’s interaction with formal organizations in selected areas of life is shown in Table 9. The interactions with formal organizations are important as they shed some light on the

TABLE 9 Percentage of families who have interaction with formal organizations in some areas of life

Areas of interaction

Izmir City Izmir gecekondu Altindag gecekondu

Visiting with friends at work

Benefits from the visit

Contacts with school and teacher

Benefit from the contacts

Helping children with their courses

40.3 38.7 31.1

48.9 48.5 57.8

67.0 62.2 52.5

90.6 88.6 76.8

63.9 61.35 42.3


adaptation processes of the family migrating into the urban environment. As can be seen in Table 9 there is considerable contact between gecekondu families and formal organization, especially over education, and, according to the respondents, such contacts prove useful. Most of the families who have contacted schools or teachers think that this has benefited their children, which suggests that the families believe in co-operating with schools. This gives us one clue about how the gecekondu family functions in an urban setting.

A second finding is the intensity of interaction with work life. About one-third of the families are on visiting terms with friends at places of work. This means that one-third of gecekondu families support their working members by a primary type relationship. About half of the visiting families think that such visits are a help to working members of the family with regard to their life at work. This could be another clue about the urbanizing tendencies of gecekondu families.


I have tried in this paper to explore some of the changes that a rural family undergoes when it moves to the city. As stated before, the paper has several limitations. Bearing these in mind, the major point is that it seems as if the gecekondu family adapts itself to an urban environment very quickly and ceases to rely on kin. As the Turkish rural family is always thought to rely heavily on its kin, the process of adaptation as well as the independence from the kin means that the gecekondu family has undergone a rapid and drastic change. The following factors may account for the change.

1. Most of the families moved to the city because they had no other choice. The majority said they moved because they could not make a living in the village, and could not benefit from their ties at home. Consequently, they came to the city with hopes for the future and a willingness to change and adapt. Hardly any of them want to return to their villages.

2. The move from village to city means moving from gemeinschaft to gesellschaft. This means, in general, that:

(a) their socio-economic position has change. Yesterday’s farmer has become today’s worker, trader or shopkeeper. About a quarter of the heads of Ankara gecekondu families have acquired the skills that they need for their new occupation after they moved to the city, e.g. as mechanics, carpenters, tailors, masons, etc.;[]

(b) the reference group and the social controls to which they are subjected have changed. They have established primary types of relations with friends at work, contacts with schools and teachers and have joined voluntary associations and trade unions. About one-third of the heads of families in the Ankara gecekondu area belong to voluntary associations.[] One-fifth of the family heads in the Altindag gecekondu area belong to trade unions. The population of gecekondu is fully aware of formal and impersonal rules and regulations, which have been learned through the difficult experience of building homes. The inhabitants are, therefore, conscious of the change in their milieu.

3. The gecekondu dwellers have rising expectations. A majority of them expect better jobs and higher income in the future, and most think that their children will live elsewhere than in the gecekondu area. Those who do not own a house expect to own one in the future.

Though convincing up to a point, these factors hardly seem sufficient to explain such rapid and drastic changes. We have, therefore, to look for additional explanations. First of all, it should be noted that to some extent there are substitutes for kin relations in gecekondu areas. Formal organizations, friends at work and neighbors are the substitute groups. The gecekondu family, by transferring primary relations from relatives to other groups, may easily find substitutes for kin reliance. Secondly, as the migrating families come mainly from the rural areas in which agricultural mechanization, cash-cropping, etc. are already established facts, it may be that the ideal type of kin-relied rural family as the origin of the gecekondu family is no more than a myth.[] It is unfortunate that we do not have enough information about the structure of the rural family in Turkey. Until the appearance of further studies our suggestions must be considered as no more than tentative conclusions. It can tentatively be said that the main reason why the rural family adapts itself so smoothly to urban conditions in gecekondu areas is that the process of change has already started in the village.




E-Mail to Emre Kongar

E-Mail to SysAdmin