The vast majority of 21st century marriages are preceded by a period of cohabitation, sometimes extremely lengthy. But was this the case for our ancestors? Did couples live together before marrying in past centuries?

In answering this it is useful to work backwards from the systematic collection of data in the 20th century to the evidence that can be used for the 19th century. It was only in the 1980s that pre-marital cohabitation became established as the norm. By the end of that decade couples were more likely than not to live together before marrying; in the late 1970s, however, fewer than one fifth of couples marrying for the first time had shared a home before marriage. Data for previous decades comes from retrospective surveys (in which people were asked to recall their behaviour earlier in their lives), and suggests that the proportion of couples cohabiting before marriage before the 1970s was even lower, perhaps as low as 1% for those marrying in the 1950s.

This might surprise those family historians who possess a marriage certificate for their Victorian ancestors, apparently recording that they were living at the same address at the date of the wedding. Any perusal of a marriage register will show that it was not uncommon for couples to give the same address – particularly, but not exclusively, in large cities. Taking this information at face value, one might jump to the conclusion that Victorian England had relatively high levels of pre-marital cohabitation, as high indeed as in the 1980s. Further investigation, however, indicates that addresses given in the marriage certificate simply cannot be relied upon, either as an indication of pre-marital cohabitation or even as evidence of actual residence, as the London-based case study in the box shows.

Nor was it only in big cities that one finds such patterns. A study of a very different population – a sample taken from those marrying in a range of parishes across rural Dorset in 1891 – found similar results. Of the 70 couples in the sample, 28 (40%) were claiming to be living at the same address when they married, but on investigating the census returns this proved true of only seven (10%), and only four of them were passing themselves off as husband and wife. Of the 42 who claimed to be living at a different address, only one couple turned out to be living at the same address.

Were couples waiting for proof of fertility before marrying, perhaps? Working out just how many brides were pregnant on their wedding day is no easy task. Some cases are clear-cut, for example where the birth of a child occurred within days or weeks of the wedding; more difficult are those cases where the interval between marriage and a first child is eight to ten months. How many of the births that occurred under nine months after the wedding can be explained by premature birth (more common if the mother is engaged in physical labour, is undernourished, or is seriously underweight) rather than premarital sex? How many baptisms that occurred more than nine months after the wedding conceal a pre-marital conception because of the time lapse between birth and baptism? How many first children were conceived before marriage but died unbaptized? And how many women lost a pre-marital pregnancy to a miscarriage, leaving only a subsequent post-marital conception to be recorded?

Much more research is needed before we can answer these questions with certainty, but an exploratory study suggests that some of the claims that have been made about the extent of bridal pregnancy may need to be revisited. Of the 832 couples who married across Northamptonshire in 1745, a subsequent baptism has been traced for 424.

The graph shows the interval between the wedding and the baptism. Intriguingly, in all of the cases where a couple had borne a child before their marriage, and in most of the cases where the baptism was recorded within two months of the marriage, the couple had married in the parish where they were resident. In such cases, one can infer that it was important for them to show the community that they were truly married, or perhaps that this was a wedding instigated by the parish, if one party had proved reluctant. There are also a handful of cases where the child was brought to be baptised within three to four months of the wedding, and it would appear that these couples had married in a parish to which neither belonged in order to conceal the fact that the bride was heavily pregnant.

Overall, however, the most striking point finding from Northamptonshire is that the majority of births occurred more than a year after the wedding. This evidence is more consistent with couples delaying a sexual relationship until after the wedding than it is with the commonly-held belief that it was necessary to wait for evidence of fertility before embarking on marriage. The family historian who has found a baptism record for a couple’s first child and is looking for the marriage can generally expect to find it between ten months and two years earlier.