34

I've frequently heard it stated that "people used to marry at much younger ages" historically. Recently, these kinds of statements have tended to show up in sociological discussions about young people choosing to get married later. Many such studies tend only to trace data back through the 20th century, often stopping at 1960 or so, when age at first marriage was particularly low.

In historical studies of marriage, the more common statistic seems to be the "age of consent" for marriage, which was often lower than today. But those numbers tell us little about the average age that most people married.

I've come across a few references here and there which have tried to measure actual marriage ages historically, and they often seem to come up with numbers that are perhaps surprisingly higher than we might expect. For instance, in 17th-century England Wikipedia notes that a survey of 1000 marriage certificates showed the average age for brides was 24 years and 27 years for the grooms. I've seen a number of other sources which have indicated that first marriage ages were often in the early to mid-20s in parts of Western Europe in the past few centuries. (Obviously, this can vary significantly by region as well as time period.)

Based on data like this, my sense is that our perception of young marriage ages has been skewed by selection bias: we know more about aristocratic families, and they tended to arrange marriages at particularly young ages for political reasons.

Moreover, at least for the U.S., it seems we may have met a local minimum of median age at first marriage in the 1950s, if these statistics derived from census data are to be believed.

Data like this has made me wonder if our perception of younger marriages on average is a historical myth (or at least needs significant qualification), perhaps based only on our perception that parents and grandparents may have married younger on average than current generations. But were the mid-1900s actually historical outliers, or at least part of a more complex picture?

IN SUM: Prior to the last century or so, what evidence do we have for the statement "people used to marry at much younger ages," in general? Some historians have asserted that people even used to marry basically upon reaching puberty; is there evidence to support or debunk such claims? Obviously there are plenty of historical examples of marriage at a young age, but does that also relate directly to a similar trend in median marriage ages (or not)?

Since marriage records, legal issues, and customs in Western marriage have varied by time and region over the past 1000 years or so (i.e., since the beginning of some regulation and standardization of Western marriage), do we even have enough information to speculate on such general trends over time? How much variance is in the data? And if we can observe any broad trends, what are they?

(I realize the answers here may vary significantly by region. I'm particularly interested in Europe, as well as the U.S./Canada, but data on other regions could be interesting as well.)


EDIT: I've edited the question somewhat to try to make clear what I'm asking for here. I do NOT expect a general history of world marriage ages over the past 1000 years. (However, if anyone can point to a reputable external resource that surveys such information, that would obviously be a superior contribution toward answering this question.) I began the question with a common historical claim about an overall marriage trend, and I'm interested in whether actual historical data supports that claim in general -- OR whether there are other general trends, or whether it's all so inconsistent that we can't make any useful general observations.

  • 2
    +1 for own survey and research. I don't like this topic, but appreciate and wish other questions be so well written – Voitcus May 21 '15 at 19:14
  • 1
    @SamuelRussell - I appreciate the recommendation, and I hope you might provide further insight if you have information on that particular region. However, I would note that your comment is actually part of the question, i.e., do we have reasonable estimates? Presumably if we do, they are based on studies of particular regions, but I don't know those ahead of time. In any case, I'm interested in the broader question of how much variance there is historically (or whether ages haven't varied as much as we think), rather than the exact range of ages in region X during decade Y in century Z. – Athanasius May 22 '15 at 1:12
  • 1
    @Semaphore - frankly, I'm mostly wondering if we have sufficient evidence to debunk the "marry upon puberty" (or at least much younger) theories. But I'm also interested in overall variance, e.g., you mention our current "anomalous" average, but the Wiki passage I cited mentions average marriage ages that are more like the mid-20s in the 1600s, which might not be quite like our modern situation, but is also significantly older than, say, the 1950s generation. – Athanasius May 22 '15 at 13:46
  • 1
    @Semaphore - thanks for your help. I'm really trying to figure out how to ask this question, which seems pretty simple. If I asked a question like "How has the population of Europe generally trended and varied over the past millennium?" the answer would be: "It mostly went up -- here's the lowest number, here's the highest number, and there were probably some dips around a few big plagues and wars." I don't see how you could answer a question about whether there's a general trend in marriage stats without noting how much variance there is, particularly given the difficulty of sampling here. – Athanasius May 22 '15 at 14:31
  • 2
    As life span increases, it becomes less necessary to marry early. – Canadian Coder May 22 '15 at 17:07
37

Not really.

Generally speaking, most European women since married in their early to mid twenties, to men in their mid to late twenties. The age gap for the commoners, i.e. the vast majority of the population, were typically not large. Unfortunately the question declined to define how much younger is "much younger" supposed to mean, but most Europeans married well after the onset of puberty.

Overall, there does seem to be an upward trend in marriage ages. However, we have little statistical evidence prior to the 17th century. The paucity of records makes claims of trends over the whole millennium rather hazardous.


Both spouses married late in Europe during the Early Middle Ages. Citing Carolingian survey data, the late David Herlihy argues[1] that prior to 1000 or so, barbarian marriage customs - marrying in late twenties to similarly aged spouse - predominated in Western Europe. From about A.D. 1,000, however, the value of women appears to have declined. Rather than receiving a bridal price from the husband, families now paid dowries to unload daughters much earlier. The age of first marriage for women thus plummeted to their late teens, but largely left that of men unaffected.

For reasons that remain unclear, the situation began to be reversed at some point during the High and Late Middle Ages. This gave rise to the curious nuptial phenomenon known as the northwestern European pattern, which has dominated Western Civilisation to this date. Proposed in his highly influential 1965 work[2] by John Hajnal, this paints a picture where both spouses married late and established their own households, independent of their parents. Another feature is that significant proportions of both men and women abstained from marriage completely. Under Hajnal's classification, this system prevailed west of an imaginary line running from Trieste to St Petersburg.

Hajnal's pattern is sometimes thought to originate from the value of retaining a daughter's labour on Late Medieval farms of Western Europe. Later on, the habit of young women and men to work in other households also delayed marriages. This contrasts with the Mediterranean situation, where domestic servants were more likely to be married and widowed. Other arguments propose that the need for financial security (due to the habit of relocating away from home upon marriage) forced delays.


Data from the Middle Ages are scarce, the earliest statistical records from the Late Medieval and Early Modern periods demonstrates a relatively high, and increasing, age at first marriage. By the Late Middle Ages Dijonese women were known[3] to marry at 20. This rose to 21 during the 16th century, and everywhere in France the mean age of first marriage seemed to have climbed to about 25 by the 18th.

Similarly, in most German regions, women married in their twenties - averaging between 22.7 to 28.5 in one study[4]. Demographic data from the late 17th century[5] reveal that commoner women from Giessen and Heuchelheim on average first married when just over 24, although Mainz's average was much lower at 21.3.

Likewise, Medieval English couples are thought to have married during their their early to mid twenties[6]. By the Early Modern period, the average 17th century English women were marrying when 25.6-26.2 to men 28.1 years of age, although it declined slightly subsequently.

enter image description here

In the Netherlands, by the middle of the period, the mean ages at first marriages for women were estimated to be about 20-21 at mid 16th century Leiden, and 23.5-25 at late 16th century Amsterdam. Both groups married husbands who were on average 1-2 years older. These numbers further increased after the 17th century.

Overall, the evidence is that European marriage patterns resembles that of the 20th century.


Not all of Europe followed the same pattern. Southern Europeans women were more likely to marry young to older men, although ages were generally still around 20. A landmark study[7] of 1427 Tuscany reveal the mean age of first marriage there to be 19 for women, but 28 for men.

Subsequent studies[8] on 15th and 16th century Florence confirms that all women married when 18 to 19, to men between 27.7 and 31.2. However, men with higher socioeconomic status tended to marry older, a trend not reflected in women's marriage patterns.

While the Florentine situation is often regarded as unusual, it is not unique. Another study[9] of 15th century Ragusa showed that women were on average betrothed at 18, but gave birth to their first child when 22. From this the authors surmised that Ragusan couples consummated their marriages when the women were 21 and men 36. In this case, local cultural norms seemed to be the main culprit.


Nuptial patterns in colonial North America were also different from the colonists' Western European motherland. A lack of eligible women relative to available bachelors resulted in fierce competition for potential brides[10]. This led to a reduction of women's age at first marriage in the 17th century, though it gradually caught up to European norms as the colonies grew over the following centuries.

However, few colonial couples marry as young as earlier writers had once assumed [11]. In early English colonies, the average age at first marriage for women were late teens to very early twenties[12], roughly five years lower than that of England. In Massachusetts[13], women married around 19 to 20 in the early 17h century. Maryland women married even younger at 17 to 18, while for Virginians it was closer to 21[14].

The difference during the early colonial period is much smaller for men, who married mid to late twenties in the colonies. This was only a couple of years lower than that of English men. Mirroring developments in England, the gap in ages between spouse closed overtime. Women's age at first marriage climbed back up to almost 24 by the 19th century, while men's dropped slightly to around 25-26. In both cases, the mean age of different colonies evened out over time.


Many cultures elsewhere in the world did have lower marriage ages than contemporary Europeans. For instance, Song China at the start of this period had legal minimum ages of marriage set at 16 for men and 14 for women. A survey[15]. of tomb inscriptions found on average, women married when slightly over 18 to men slightly over 23. Similarly, in Japan during the early modern period, women were found[4] to have married around 16.7 to 22.7. By the late 18th and 19th centuries, especially in areas of high commercial development, women's mean age of marriage had rose to around 22-25[16].


References:

[1] Herlihy, David. Medieval Households. Harvard University Press, 1985.

[2] Hajnal, John. "European Marriage Patterns in Perspective." (1965): 101-43.

[3] Rossiaud, Jacques. "Prostitution, jeunesse et société dans les villes du Sud-Est au XVe siècle." Annales (1976): 289-325.

[4] Murayama, Satoshi. "Regional Standardization in the Age at Marriage: A Comparative Study of Pre-industrial Germany and Japan." The History of the Family 6.2 (2001): 303-324.

[5] Hurwich, Judith J. Noble Strategies: Marriage and Sexuality in the Zimmern Chronicle. Vol. 75. Truman State Univ Press, 2006.

[6] McSheffrey, Shannon. Marriage, Sex, and Civic Culture in Late Medieval London. University of Pennsylvania Press, 2006.

[7] Herlihy, David, and Christiane Klapisch-Zuber. "[Tuscans and their families: a study of the Florentine catasto of 1427]." Editions de lEcole des Hautes Etudes en Sciences Sociales Ouvrage 8 (1985).

[8] Siegmund, Stefanie Beth. The Medici state and the Ghetto of Florence: the construction of an early modern Jewish community. Stanford University Press, 2006.

[9] Rheubottom, David B. "“Sisters First”: Betrothal Order and Age At Marriage in Fifteenth-Century Ragusa." Journal of Family History 13.4 (1988): 359-376.

[10] Haines, Michael R., and Richard H. Steckel, eds. A Population History of North America. Cambridge University Press, 2000.

[11] Lancaster, Jane Beckman, and Beatrix A. Hamburg, eds. School-age Pregnancy and Parenthood: Bisocial Dimensions. Transaction Publishers, 1986.

[12] Smith, Daniel Scott. "The Demographic History of Colonial New England." The Journal of Economic History 32.01 (1972): 165-183.

[13] Demos, John. "Notes on life in Plymouth Colony." The William and Mary Quarterly: A Magazine of Early American History (1965): 264-286.

[14] Wells, Robert V. "The population of England's colonies in America: Old English or new Americans?." Population Studies 46.1 (1992): 85-102.

[15] 〈宋代婚姻禮俗考述〉方建新《文史》第24輯158頁, 1985 April

[16] Saito, Osamu. "The Third Pattern of Marriage and Remarriage: Japan in Eurasian Comparative Perspectives." Marriage and the Family in Eurasia: Perspectives on the Hajnal Hypothesis (2005): 165-193.

  • 3
    Thanks so much for this. And thanks for the references. – Athanasius May 22 '15 at 19:14
  • Did more single man go out then single ladies in the "early colonial period", hence did the 1st generation immigrant man take the ladies that were 2nd generation immigrants? – Ian Ringrose Oct 27 '16 at 9:46
  • In the early middle ages (at least) it was not uncommon for nobles and royalty to marry very young. For example, William the Atheling's wife Matilda of Anjou was no more than 12 when they were married in 1119. and Edward I's son was due to be married around his thirteenth birthday but died a few weeks before the nuptials. Eleanor of Castile was about 13 when she married Edward I, who was 15 at the time. – Lars Bosteen Aug 28 '17 at 7:11
  • @LarsBosteen Matilda could not be less than 12 - most sources puts her at 14 in 1119. In any case, such marriages were mostly restricted to the high nobility and royalty due to political needs. While you can find plenty of those examples, they are not representative of the population at large, nor even of the general nobility. Further, even if the marriage was consumated early, regular cohabitation likely did not occur until their late teens. Hence, "the marked tendency of most English and French queens to bear children only in their late teens or early twenties" - Eisenbichler (2002) – Semaphore Aug 28 '17 at 9:54
  • @Semaphore. I should have been clearer - my comment was to point out an exception relating to nobility & royalty, not to contradict anything in the answer. Concerning Matilda of Anjou's age, my source is C. Warren Hollister 'Henry I' (Yale University Press, 2003). I would be interested in other sources relating to William the Atheling as I am currently working on a documentary and am looking to be as balanced / accurate as possible. – Lars Bosteen Aug 28 '17 at 13:03
5

Others have already provided excellent information and cites. There are a couple other things to look at.

There may actually be a proxy that you can use to fill in data that you can't directly obtain: the number of children a woman bears should be related to her marriage age. The larger the family size, the younger the marriage age. Another proxy might be the length of one generation (which would indicate the average age of the mother when having any of her children). Finally, a proxy you could use is when property (farms etc.) was passed through the generation. In many regions, they were passed on only from father to the firstborn son, which would give you a good indication of the age of the father when he had his first child, and thus indirectly of the marriage age of men (of course, daughters as first children would be a confounding factor here!)

In central Europe, you also will have a hard time going back 1000 years with your research, because the 30 years war (1609-1639) destroyed most relevant records, if they were even ever collected.


The definition of marriage itself has changed multiple times over the last millenia.

Marriage wasn't always the formal recorded matter it is today, and in some cases it may not even have been one-man, one-woman.

Based on data like this, my sense is that our perception of young marriage ages has been skewed by selection bias: we know more about aristocratic families, and they tended to arrange marriages at particularly young ages for political reasons.

I think you may actually be subject to another selection bias of your own: most marriages weren't recorded until, IIRC, around the 16th to 19th century, depending on the region. In medieval Europe, what we today would call "shacking up" was the very definition of marriage - you were married when one partner moved in with the other, and maybe your family or the church held festivities for the occasion.


The purpose of marriage has changed multiple times.

Marriage could be for love.

Marriage could be for political reasons (not just in the higher levels, but potentially even at the village level).

Marriage could be for procreation.

Marriage could be for social security.

Marriage could be for mutual protection.

Marriage could be for division of labor.

Different purposes would lead to different optimum marriage ages.


Biological factors play a major role.

People, and in particular women, have a limited age range when they can procreate. When maximizing procreating (whether for its own sake, or to have many children providing social security) was the goal of marriage, that would argue for an earlier marriage age.

Infant mortality would also call for women having more time for plenty of pregnancies.

Maternal mortality would probably call for higher average marriage age. Very young mothers would be at higher risk (and people would have known that).

Today, we are probably near the upper end of the age range where marriage for procreation purposes is feasible at all, and then only with very small family sizes. That would support the notion that historically, people did marry younger (although it does not say how much younger).

Marrying late is, in a way, a luxury. During times of turmoil or disease, people would have children (and thus marry) as early as possible. During times of peace, prosperity, and longevity, people could afford to wait longer. Incidentally, it seems that @Semaphore's data also follows the same general pattern: a higher average marriage age in wealthy Florence, for instance, and it seems that generally the higher marriage ages seem to correlate with peaceful periods.

Based on all of that, you will probably find the following patterns:

  • Average marriage age varied with conditions, both up and down.
  • Average marriage age among unrecorded marriages are likely to be lower than among recorded marriages (because ordinary people had less security and more need for many children).
  • Today's average marriage age is likely near the historic peak.

Incidentally, one way to validate this is to look at international comparisons today. Today's developing countries may have their own issues, but in many ways, especially when it comes to the basics of humanity, very much resemble medieval Europe.

  • Another possible factor that may postpone marriage age would be if marriage is acceptable only if the couple could more or less provide for themselves plus their offspring. This would mean that at least the husband would need to have reached a certain experience in his profession but probably also for the wife to be experienced enough to manage the household side of the business or farm (and somehow I imagine this would include servants as they couldn't risk their position too early). Again, this may be more important (or more sensible to expect) in peace time. – cbeleites supports Monica Apr 4 '18 at 14:14
  • @cbeleites Excellent point. The age when people became self-sufficent like that has also changed dramatically. 500 years ago, you could be self-sufficient while being illiterate and, by today's standards, under age. Today, at least, you need a high school degree, if not a college degree. – Kevin Keane Apr 5 '18 at 16:08
  • Not so sure about age of being self-suffient: I have in the back of my mind that "professional" hunters and gatherers of stone-age-like cultures reach their maximum productivity around age 40 after maximum physical power at age 20 due to gain in experience. I wouldn't be too sure that a, say, couple of 15-year-olds would have been considered to have reached a level of self-sufficiency. And as for today: I'm living in a European wellfare state (Germany) so this is basically a non-question at the level of a higher living standard than upper middle class had 60 years ago. Still, that aside, ... – cbeleites supports Monica Apr 5 '18 at 18:00
  • ... if instead of attending university the couple would each learn a trade nowadays, they can feed their family in their early 20s. And at the time the urban academics graduate, such a say, nurse + HVAC couple in a rural area may be quite a bit further towards owning their house than the urban academic elite ever gets... (but that's politics, not history...) – cbeleites supports Monica Apr 5 '18 at 18:06
  • Another possible counterfactual, from a community of European origins faced with isolation, low population and scarce ressources: the consequences on women's average age of mariage were dramatic en.wikipedia.org/wiki/… – Evargalo May 30 '18 at 14:48
2

I'm surprised to hear of women marrying so late, since having children late could be an issue. But maybe they married late due to the fear of death in childbirth. I would also like to know what the differences could be in the different classes. Surely the age of marriage for the aristocracy could well be different to those of the peasant class. I don't think this topic should ignore that issue.

There are a lot of parish records in England which could be examined for this topic, and a lot of those are available online.

In my own family history research, which of course is only a small sample, I have noticed in Southern England among the lower classes, in the 18th and early 19th centuries that marriages were at about age 20 and both male and female about the same age. Usually they married when the woman was already pregnant (this does actually appear to be very common). I do have just one marriage from the 16th century, where the man was in his late 20s and the woman in her mid teens, but I don't know the class of that marriage, and yes, it is just one marriage stat.

  • a lot of parish records show the mirage date to be VERY close to the date of the first child..... So maybe the records are not showing what most people think of as marriage, but are recording a marriage that happened some time before. – Ian Ringrose Oct 27 '16 at 9:59
  • 2
    Marrying late is not so surprising when you realize that it's an efficient way to reduce the number of children when households had a limited ability to raise them or to provide means of living (mainly land) for each of them as adults. – Pere Jan 9 '17 at 23:26
  • Also: maybe it wasn't as late as it seems to us: the onset of puberty has become considerably earlier (e.g. mum.org/menarage.htm has graphs and references). Average age at onset of puberty for girls around 1830 in (northern) Europe was almost 17, end of puberty would then have been at about 20. – cbeleites supports Monica Apr 4 '18 at 14:23

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.