It is well known that there has been a sharp fall in our marriage rate and a high and still rising rate of births outside marriage - but there has been relatively little analysis of these two phenomena.
To what extent is the drop of over two-fifths in the marriage rate during the last quarter of a century a function of a gradual postponement of marriage - or an abandonment of this institution? And what explains the huge increase in non-marital births - from 1.5 to 27 per cent of births since the mid-1960s?
Neither question has ready answers, but a close look at the evolution of our marriage and birth figures since the 1970s offers some clues as to what may lie behind these dramatic demographic changes.
The latest year for which a detailed breakdown of birth figures have been published is 1994, when there were 9,200 non-marital births. Since that year there has been a further increase of 5,000, or over half, in the number of non-marital births, and this has pushed the previously declining total number of births back up by one-eighth, to almost 53,000.
First of all, it is clear that few people now marry young. Although the number of women aged 19-25 had risen slightly between 1981 and 1994, the number of married women in this age bracket fell by almost two-thirds. And as the birth rate for married women in this age bracket dropped by a further one-third, the number of babies born to them decreased by over threequarters.
In the age group 26-33 the number of women also rose, but here again there was a drop in the number of married women, and a one-third reduction in births.
The consequence of all this was an increase of a half in the number of women of child-bearing age who are single. In the age group 26-33 there were 120 per cent more single women in 1994 than in 1981.
All this is relevant to the increase in non-marital births during this period. For the trebling of the number of non-marital births reflected the combined impact of the 50% increase in the number of single women and of a doubling of the overall non-marital birth rate.
It is worth noting that the non-marital birth rate for women in the older child-bearing age group, 34-45, rose almost six-fold between 1981 and 1994. In 1981 most non-marital births had been to very young women - half were to teenagers. But by 1994 a half of such births were to women aged over 24. This development seems to have reflected the emergence of a significant group of women in stable relationships who have postponed child-bearing and who, even when they do start to have children, do not marry their partner.
In due course, when the detailed data on births between 1995 and 1998 become available, it will be interesting to see to what extent there may have been a further development of this new trend.
A consequence of these changes in both marriage and birth rates has been a significant upward shift in the typical age of the mother giving birth. In 1981 births peaked in the 26-28 age group. By 1994 the commonest age to give birth was in the 29-31 age bracket - an upward shift of as much as three years. On the Continent only the Netherlands has as high an average age of mother giving birth as Ireland, viz. 30 the highest age recorded in Europe. However, we have for long been near the top of the European league in this respect. In part this reflects the fact that family size here remains slightly higher than in most other parts of Europe - several countries have a slightly higher average age of mother for first births, but as in these countries fewer children tend to be born to mothers, the overall average age of mothers giving birth is somewhat lower.
It has to be said that this pattern of late births is clearly not a good situation from the point of view of the future general health of our community.
What of the drop in the marriage rate?
The proportion of women aged 25 or less who are married is now almost two-thirds below what it was in 1981 - just over 20 per cent now as against an astonishing 60 per cent then. But by age 35 this differential is down to three percentage points - 82 per cent now, as against 85 per cent in 1981.
That seems to suggest that women in their twenties have been postponing marriage rather than abandoning the institution. Nevertheless, whereas less than 10 per cent of women born in the 1940s never married, it seems likely that for later generations this proportion will turn out to be somewhat higher.
We lack the important statistic of the proportion of younger people who are now cohabiting - although cohabitation before marriage is now a common phenomenon and may indeed have become a normal preliminary to marriage.
But we know little about it. We have no hard data on its prevalence, and little early prospect of getting a fix on this. We know nothing of the stability or otherwise of these relationships - how many survive, or end, and after what duration. We can guess that these arrangements commonly evolve into marriage, either because of a decision to have a child or because a child is conceived - but we have no firm information about this either.
It seems clear that a common factor both in the postponement of marriage and in cohabitation is a reluctance of young people to enter into lifelong commitments. In some cases this clearly reflects experience of an unsatisfactory parental relationship - but this must be a minority experience, for the parents of most young people have stable marriages, very many of which are indeed, by any reasonable standard, a success.
Why are the children of such successful marriages often just as reluctant as those with a bad experience to enter into long-term commitments with a partner? Is it really the case, as has often been suggested, that young people are now all intimidated by the fact that they have a longer expectation of life today than in the past?
I find this hard to believe for two reasons: first because between one generation and the next the increase in expectation of life is marginal; second, because for young people the future has always seemed long.
In most areas of public policy it is possible to base decisions on reasonably solid data. But in this area governments seem to have almost nothing to go on. Of course, I know that some people may regard this area of personal relationships as being outside the proper frame of reference of public policy, but I am quite unconvinced by that argument.
For governments have the responsibility of securing and maintaining social conditions that are optimal for the development of society, not merely by such actions as infrastructural investment and transfers of resources, but also my maximising the chances of a good and healthy life for future generations. That is why governments concern themselves with such matters as education, child care, health, marriage, divorce, fostering, adoption, provision of holiday time and so on.
Now the optimal conditions for bringing up children, and thus for the future of civil society, include a secure and stable two-parent home. Governments are not merely entitled to take action to favour such a social structure, but have a duty to do so - by encouragement and assistance, although not, of course, by constraint or penalisation.
For example, if a government can improve the health of future generations by encouraging earlier maternity, it should do so. If it can help women who want to look after young children at home, it should do so - while at the time facilitating mothers who want to work by ensuring good child care for the children of couples who both work. And if it can encourage the stabilisation of unions by encouraging marriage, it should do so.
These are all legitimate objects of public policy, even if some of them are not currently favoured by some element of contemporary opinion formation.