Sigurd Skirbekk:

Family structure and functionality


IV - The difference between marriage and cohabitation

The method which is closest at hand for testing the status of cohabitation involves comparing its stability with that of marriage. However, there are a number of problems with such a comparison. In the first place, what is termed cohabitation is not just meant to be an alternative to marriage. It also purports to be an alternative to engagement, which has traditionally been easier to dissolve than marriage. In the second place, cohabitation is such a new phenomenon that we do not have enough experience in analyzing its long-term effects. In the third place, both marriage and cohabitation are set in cultural contexts. This means that in societies in which marriage clearly predominates, cohabitation will assume the form of those expectations which characterize a marriage-oriented culture. In societies where cohabitation is customary and accepted, this will also affect people’s attitudes toward each other in their marriages. Among other things, this means that a cohabitation union at the end of the 20th century is not completely the same as a cohabitation of mid-20th century.

The change in these relationships over time may, on the other hand, represent a kind of important bit of information, if we wish to evaluate what these changes predispose people for. From the Nordic countries we can find different kinds of statistics that point out these changes. Already by the 1980s it was possible to show, in all the Nordic countries, that the marriage rates for childless women living out of wedlock declined markedly for each new birth cohort during the period from 1936-40 to 1956-60. In Norway, the chances that a relationship would develop into marriage dropped by about two-thirds, from about 90% to 31%. The figures for the development of registered cohabitation in Sweden and Denmark were even more dramatic 21].

By working with data from the Norwegian Bureau of Statistics fertility study from the end of the 1970s, demographer Randi Selmer was able to show that the probability of a cohabitation relationship ending in marriage had declined even in cases where children were involved. She found that out of those who were born between 1941 and 1945 and who had begun living together as 15-19-year-olds, and who had not had children or gotten married, only 12% had not continued their relationship two years after having entered into it. The corresponding frequency of dissolution had risen to 26% for the birth cohort 1956-59. For those who had begun living together as 20-24-year-olds, the corresponding frequency of dissolution was 10% for the birth cohort 1941-45 and 24% for the birth cohort 1951-55 22].

The increasing instability over time among cohabitors has been succeeded by an increasing number of divorces among married couples 23]. Nevertheless, marriages have consistently represented the most stable category of cohabitation 24]. This is not just because cohabitation has often been regarded as a kind of trial marriage. Even where the couple has had children together - which should indicate that they considered the relationship to be a lasting one - cohabitors have a dissolution rate as much as three times higher than that for married couples with children 25] .

These calculations are not only relevant in a Nordic context. Studies from other countries also confirm that non-institutionalized cohabitation is more fragile than marital cohabitation 26].

One of the most frequently used arguments for legitimatizing cohabitation is that it acts in the capacity of a trial marriage, in which the parties get a chance to become acquainted with each other before they make an earnest commitment. This was supposed to lead to a selecting of couples that were compatible, which in turn was expected to make for more stable marriages. To a large extent, cohabitation unions are also regarded as less binding. Those who are willing to bind themselves tend to get married 27]. In Germany it has been claimed that the inherent instability of non-marital cohabitation can be traced back to a personal selection. It is those with the fewest predispositions for stable family forming who wish to avoid marriage. Still, this line of reasoning is somewhat incomplete when it is broadly used to legitimatize trial marriage. The proportion of "risk candidates" is not constant from country to country. Sweden, a country that has gone very far in legitimatizing non-marital cohabitation, also has the highest percentage of such risk candidates. American data does not suggest that those who have spent some time in a cohabitor relationship are prone to form more stable marriages than those who have chosen the more traditional route to marriage. As for having been through a cohabitor relationship with a partner other than the one with whom one has children, the data seems to suggest that the chances increase for the dissolution of the family further on down the line 28].

Different studies tend to suggest that changes in the pattern of cohabitation has been determined by factors other than functional considerations for the formation of families. These changes can largely be explained on the basis of changes in the framework of meaning and cultural references. On an individual level, it can be shown that people who live in a "common law" relationship generally have more liberal attitudes than marriage people, particularly with regard to divorce 29]. Nevertheless, it would be wrong to perceive cohabitation as something that two people choose individually on the basis of a specific prioritizing of values. In the Norwegian family and vocational survey from 1988, we see that while 18% of a selection of 4019 women stated that they were living with a cohabitor, only 2% supported the claim that "cohabitation is always preferable to marriage" 30]. What people do is one thing; what they would have done if circumstances had been more propitious for a different choice, is another matter.

A Swedish study shows that 53% of a selection of cohabitors stated that both parties wanted to get married eventually, while 33% stated that they themselves, but not their partner, wanted to get married 31]. The lack of determination with respect to marriage was explained in terms of a weak norm pressure in Swedish society for choices of this kind. This type of data encourages us to take a closer look at the question of whether the development can be explained in simple equations as the liberation of the individual from institutions. There are several reasons for asking whether women as a whole have been well served by the changes in question.


21 Østby, Lars (1986): Første skritt i familiedannelse i Norge, sammenliknet med Danmark og Sverige, Tab 2 - Demografisk avdeling, Stockholms Universitet.

22 Article 146, Norwegian Bureau of Statistics, Oslo, Tab. 11 and 13.

23 Noack, Turid and Øystein Kravdal (1988): Divorces in Norway 1965-1985, Statistisk Sentralbyrås rapportserie 88/6.

24 Cf. Noack, Turid and Nico Keilman (1993): "Familie og husholdning", chpt. 8.1 in Sosialt Utsyn 1993, Norwegian Bureau of Statistics, Kongsvinger, Norway.

25 The comparison is based on calculations made by Øystein Kravdal on the basis of data from the Norwegian family and the vocational study from 1988 of dissolution frequencies. Cf. Kravdal, Øystein (1997): Wanting a Child but not a Firm Commitment to the Partner: Interpretations and Implications of a Common Behaviour Among Norwegian Cohabitors. Memo from Department of Economics, University of Oslo. No. 19, May. - Also Swedish data suggests a significantly higher frequency of dissolution among cohabitors with children compared with married parents: Cf. Hoem, Britta and Jan M. (1992): "The disruption of marital and non-marital unions in contemporary Sweden". Published in Trussel, J. R. Harkinson, J. Tilton (Ed.): Demographic Applications of Event History Analysis. Clarendon Press, Oxford; pp 61-93.

26 Bumpass, L.L. and J. Sweet (1989): "National estimates of cohabitation. Cohort levels and union stability." Demography 26: 615-25. Teachman, J.D. and K.A. Polonko (1990): "Cohabitation and marital stability in the United States". Social Forces 69: 207-20.

27 Figures from the beginning of the 1990s show that of all cohabitors who started their relationships at 23 years of age and who continued to be together after five years, 79% had gotten married, while only 21% continued to live as cohabitors. - Sosialt Utsyn, SSB 1993, p. 293.

28 Maris, Alfred de and William MacDonald (1993): "Premarital cohabitation and marital instability: A test of the unconventional hypothesis". Journal of Marriage and the Fajily; pp 399-407. - Cf. also Saunders, Janic M. and John N. Edwards (1984): "Extramarital Sexuality: A Predictive Model of Permissive Attitudes". JMF Nov.; Fergusson, D.M., Horwood, L.J. and D.E. Dimond (1985): "A Survival Analysis of Childhood Family History", JMF May; Bennett, Neil G., Ann Klimas Blanc and David E. Bloom: "Commitment and the moern union: Assessing the link between premarital cohabitation and subsequent marital stability"; American Sociological Review 53 (1): 127-139.

29 Blom, Svein (1986): Famileverdier i Norge. Rapport 22, INSS, Trondheim.

30 NOS B 959; Tab 5.5.4.

31 Levin, Bo (1982): "Unmarried Cohabitation: A Marriage Form in a Changing Society", Journal of Marriage and the Family, pp 763-773.