Hajnal’s essay was published in 1965. The opening sentences ran as follows:

Hajnal then gave statistical substance to his assertion by demonstrating the scale and consistency of the contrast in marriage patterns between the countries of western and northern Europe on the one hand and those in eastern Europe in 1900, a date by which the spread of census taking provided reliable statistics showing the proportion of men and women single in the age groups 20–24, 25–29, and 45–49. Table 1.1 shows that the contrast was stark. Indeed the table may understate the contrast. In each grouping there were marginal cases (Spain in the western group and Greece in the eastern group, for instance) so that the starkness of the contrast between the core area in each group is greater than the table suggests.

Hajnal then made use of data for a range of countries in Asia and Africa to show that in extra-European countries at dates chiefly in the first half of the twentieth century, marriage patterns were very similar to those in eastern Europe. In particular, only roughly a quarter of women were still single in the age group 20–24, and in the 45–49 age group the comparable figure was only 2 or 3 per cent.

The bulk of the subsequent text of the essay was concerned with the question of when the ‘western’ pattern emerged. Hajnal considered in turn evidence relating to the eighteenth century, the middle ages, and the ancient world before turning to non-statistical evidence. He came to the tentative conclusion that, in medieval times, marriage patterns in western Europe did not differ greatly from those elsewhere and that it was probably in the sixteenth century that change took place, initially among elite groups and only later in the general population. Hajnal was very careful to emphasise the fragility and uncertainty of the evidence available to him. In a sense, the chief message he delivered was that it was both desirable and probably feasible to settle conclusively many questions which he was obliged to leave open.