Field Notes

Posted by Bret Devereaux

https://acoup.blog/2025/08/01/collections-life-work-death-and-the-peasant-part-iiia-family-formation/?utm_source=rss&utm_medium=rss&utm_campaign=collections-life-work-death-and-the-peasant-part-iiia-family-formation

https://acoup.blog/?p=30696

This is the first part of the third part of our series (I, II) discussing the patterns of life of the pre-modern peasants who made up the great majority of all humans who lived in our agrarian past and indeed a majority of all humans who have ever lived.¹ Last week, we looked at death, examining the brutal mortality regime of pre-modern societies, typified by extremely high (c. 50%) infant and child mortality, very high maternal mortality and often high male military mortality, which kept life expectancy at birth as low as the mid twenties, while life expectancy at adulthood was better – around 50 – but still very low by modern standards.

This week and next, we’ll start working out some of the consequences of this mortality regime, looking at family formation which in these pre-modern agrarian societies means marriage. While the intense variability of mortality meant that peasant households came in a variety of single- and multi-family forms, pre-modern agrarian societies generally had strict and rigid expectations for marriage: in nearly all of these societies everyone got married and was expected to get around to having children because the community required them rather than necessarily because they wanted to.

So this week we’re going to look at marriage patterns, particularly the question of age at first marriage. Then next week, we’re going to turn to the implications those patterns have for child-bearing and child-rearing. The family and the household were the fundamental institutions of everyday life for pre-modern people, so understanding their structures and assumptions is crucial for understanding the rest of life in these past societies.

But first, if you like what you are reading, please share it and if you really like it, you can support this project on Patreon! While I do teach as the academic equivalent of a tenant farmer, tilling the Big Man’s classes, this project is my little plot of freeheld land which enables me to keep working as a writers and scholar. And if you want updates whenever a new post appears, you can click below for email updates or follow me on Twitter and Bluesky and (less frequently) Mastodon (@bretdevereaux@historians.social) for updates when posts go live and my general musings; I have largely shifted over to Bluesky (I maintain some de minimis presence on Twitter), given that it has become a much better place for historical discussion than Twitter.

Marriage, Marriage is What Brings Us Together, Today

We begin with marriage as the first step in family formation (though not necessarily household formation, as we’ll see). Whereas the pre-modern mortality regime is broadly consistent over different cultures, marriage patterns (nuptiality) vary significantly. Very nearly all human cultures practice something we can identify as marriage, the mostly-permanent pair-bonding of individuals to create a new family (but not necessarily household) unit into which new children are born. Different cultures, even in the pre-modern world, differ notably on the rate of marriage (though it is, in all cases, by modern standards very high, for reasons which will become clear), its timing, and the presence or absence of polygamy.

Before we get into those variables though, we need to make a very important point: we are talking about peasants. Remember peasants? This is a post about peasants.

The marriage patterns of high elites in a society are often quite different from the marriage patterns of most of the society. The classic example of this is to note that students are often mislead by European aristocrats in the medieval and early modern periods marrying very young and so they assume that everyone in medieval Europe married very young, but in fact, as we’ll see in a moment, medieval western Europe is notable for very late (mid-twenties for women, late twenties for men) typical age at first marriage among the general population. The very wealthy do not marry under the same economic constraints and incentives as the enormous majority (upwards of 90%) of the population living as peasant farmers or even the smaller subset working in cities or having specialized trades or so on. Indeed it is very common for elites in pre-industrial societies to marry much younger than non-elites, because of the different pressures (family alliances, the need for heirs, the lack of direct economic pressure) placed on those marriage decisions.

And this has an immediate implication for us when it comes to another one of our three variables: polygamy – or more correctly polygyny, since we are effectively always in this context referring to the practice of one man having multiple wives, not the other way around.² Polygamy occurs as a social practice in quite a lot of societies (though it is somewhat scarcer in agricultural societies than non-agricultural societies), but within societies, it is a practice generally restricted to the wealthy, who have the resources to keep multiple families. Even in polygamous societies, most families and households are monogamous, for what should be fairly obvious reasons. There are, after all, a roughly even number of men and women, so each polygamous marriage means another male who cannot marry and societies that generate massive numbers of unmarriagable young men with no prospects don’t tend to be very stable (especially if they also need the labor of those men in the fields). So even in societies where polygamy was highly normalized, it would represent only a minority subset of marriages (to judge from modern statistics, ‘about a third’ is a decent rule of thumb) and in many societies where polygamy was accepted it was rarer still. Often this is in the form of societies where rulers or high elites might take concubines or secondary spouses, but not the common folk. And then, of course, you have societies where polygamy was not accepted, which includes my own study of the Romans, who were, as Bruce Frier puts it, “relentlessly monogamous.”³

The impact of class thus provides us a useful simplification: we can focus on monogamous marriages. After all, in many of our societies, monogamy is the only game in town and in the rest only the richest peasants are likely to have multiple wives and so on either event, the modal peasant family is monogamous. And, to be frank, that’s also useful for me because I study the Romans – again, “relentlessly monogamous” – and have better grounding in medieval Europe (where polygamy was banned) and ancient Greece (where it was extremely rare).⁴ If we wanted to get into family patterns for polygamous households, we’d probably need to bring in someone who specializes in those cultures.

But overall, I want to stress that it is a mistake to assume that the marriage behaviors of highly historically visible people – like monarchs, high nobles, senators and so on – are indicative of the marriage patterns of regular people. Instead, we need evidence of the marriage patterns of our social stratum – the bottom 90-odd percent of people – which can create some challenges, because those people are not generally historically visible to us.

With that said, on to…

Marriage Patterns

In terms of demographics, when we talk about marriage patterns, we’re thinking in terms of a few key variables: age at first marriage (AAFM) for both females and males (the focus tends to be on the former) and the percentage of people (again, the focus tends to be on women) married at a given age, which also ties into the existence or non-existence of a ‘spinster’ or ‘bachelor’ class (women and men who simply never marry). Culturally, I’d add to this calculation the acceptability and prevalence of divorce.

The variables differ substantially (but within a range) from one culture to another. To understand why, we need to return to mortality as our ‘forcing function’ on social organization. We’ll be coming back to some of these points when we talk about fertility next week, but the mortality rate for pre-modern societies is very high, thus necessitating a lot of births, but it is not so high that societies need to approach a maximum ‘natural fertility’ (the birth rate using absolutely no means of birth control) to hit replacement and slow growth. But these are also peasant households with significantly constrained resources. So the question becomes how to restrain fertility to a high, but not maximum level and there are basically two options: either control fertility within marriage or delay the age of first marriage (for females). Naturally these two strategies are not mutually exclusive and could be combined to an extent (so long as it doesn’t push the birth rate below the high level required for replacement).

We can think of these strategies by breaking them into simple models, but before we do that, I want to hit the necessary caveat that we’re talking here about statistical averages not individual families. If we say the mean AAFM for women is 20, that doesn’t mean every 21-year-old is married or that every 19-year-old is single; we’re representing a range. And we’re going to start with women and loop back around to men.

Richard Saller proposed three such models – he termed them ‘early’ ‘middle’ and ‘late’ – as a way of testing the data from Roman inscriptions and I think they provide a decent window into the potential different models here and also have the happy advantage of separating the names of the patterns from assumptions about where those patterns existed (though we’ll talk about that too).⁵ I also really like calling these models ‘early’ ‘middle’ and ‘late’ because it gets us some distance from geographic naming models which may not, as we’ll see, be accurate in all periods.⁶

I should note, because you will encounter these terms elsewhere that the late male/early female marriage pattern that we see in Greece is termed by John Hajnal the ‘Mediterranean type,’ which is awkward because it is not clear all Mediterranean societies in the past followed it, while an early male/early female model (in which ‘early male’ here means mean AAFM around 20, not around 15, so there is still an age gap) is termed the ‘eastern type.’ There is also a ‘western European marriage pattern’ we’ll get to in a moment – but I am avoiding these geographic labels because they strike me as a form of question begging⁷ since we do not know the historical marriage patterns in all of these places and assuming they follow a simplistic three-part geographic model is perilous.

Now my expertise in this sort of historical demographics is really restricted to the broader Mediterranean world (to include western and central Europe) but fortunately all three patterns occur in that geographic space, so we can have an example of each, but I want to be clear that I probably cannot tell you with any confidence which pattern is common in any given place and time outside of that broader Mediterranean world. No one can know everything.

An Early Pattern: Ancient Greece

My sense is that Saller’s ‘early’ pattern is often thought to be the most common among pre-modern peasant populations and sometimes forms the default assumption of those societies. I am simply not familiar enough with the evidence for pre-modern marriage patterns outside of Europe and the Mediterranean to offer a view as to if this ‘default’ assumption is correct.⁸ In the early pattern, females begin marrying almost immediately after menarche, in their early to late teens. The mean age of first marriage for females in this model tends to be around 16 and marriage rates for societies under this model are generally very high: virtually all women marry. Fertility control (discussed next week) thus has to happen within marriage, not by delaying its onset.

In the Mediterranean, this ‘early’ marriage pattern seems to have been common in the poleis of ancient Greece, although we must note that the evidence here is quite limited. While older scholarship,⁹ tended to assume that we could use the Roman marriage pattern for Greece, work in the mid-1990s and onward, particularly by Sarah Pomeroy has tended to show an AAFM for ancient Greek girls around 15.¹⁰ By contrast, male AAFM is substantially older, roughly thirty. Marriage in ancient Greece was functionally invariably arranged: women were legally incapable of arranging their own marriages and Greek marriage rituals do not appear to include even superficial nods to bridal consent.¹¹ My sense is that this extreme dearth of legally and socially recognized female agency is typical of societies with very early marriage patterns.

An Intermediate Pattern? Rome

The next pattern is Saller’s intermediate pattern, where female AAFM begin marrying in their mid-to-late teens with an average AAFM around 20. Although there remains some lingering uncertainty and debate on this point, Saller argues that the Romans followed this intermediate pattern and my sense is that this remains the consensus view, though substantial uncertainty still exists.¹² While Roman law permitted marriage very young – the legal minimum age at marriage was 12 for girls and 14 for boys – the observed marriage pattern outside of the elite (who married significantly younger) seems to have been marriages starting around 14 or 15, but with the mean AAFM for females close to 20. For men, the evidence suggests even later marriage ages, with only a few men marrying in their teens but the bulk of men marrying in the mid-to-late 20s and a AAFM for males around thirty.¹³

While this isn’t the place to get into all of the details, one reason I find the concept of an intermediate model for the Romans distinctly plausible is that, as pre-modern patriarchal societies go,¹⁴ Roman women occupied, from a legal and social standpoint, a remarkably favorable position. While intense social pressure must have meant that a bride’s consent was generally a formality (‘non-objection’ was taken as consent), bridal consent was legally required in Roman marriage in a way that we have no sense it was so required in Greek marriage. Likewise, Roman women, at least by the second century BC, had the right to initiate divorce on a ‘no fault’ legal basis (that is, for any reason or no reason). Roman women were legal persons in Roman law, in a way that at least Athenian women (and we usually assume Greek women generally) appear not to have been under Greek law codes. Roman women could and did hold property, something that, for instance, Athenian women could not do. All of that seems consistent with a social regime which, while still very patriarchal by modern standards, had a less instrumental approach to its women, who in turn had somewhat more control over their own lives, and thus might modestly delay female AAFM.

Interestingly, the data from Roman Egypt, seems to suggest a marriage regime that might have fit between the Roman intermediate and Greek early models. The evidence for Roman Egypt is meaningfully better and Bagnall and Frier’s (op. cit.) data suggests a female AAFM of 17-18, with marriages starting very young (as early as 12), rising steeply in the late teens and being nearly universal by 30 – a bit older than Greece and a bit younger than Saller’s model for marriage in Roman Italy.¹⁵

Which leaves…

A Late Pattern: Early Modern (and Late Medieval?) Western Europe

What remains is a ‘late’ female marriage pattern, with an average AAFM for women in the mid-twenties. To my knowledge, the only instance of that pattern before the industrial revolution is in Europe. Termed the ‘Western European marriage pattern‘ and advanced as a theory by John Hajnal, it has attracted attention because of course any way in which Western Europe was unusual in the early modern period attracts attention as part of the ‘Why Europe?’ question which dominates the early modern and modern periods. The ‘Western European marriage pattern’ is, in effect a late/late pattern (that is, late for both males and females) in which female AAFM is around 25 and male AAFM is around 30.

Some caveats are immediately necessary. While Hajnal proposed this marriage pattern to prevail over most of Europe west of the ‘Hajnal line’ (which cuts through what is today Poland, the Czech Republic, Austria to hit the Adriatic in northern Italy), to my knowledge the pattern is clearest in Britain (but not Ireland) and along the Channel (Northern France and the Low Countries) with a lot more variation in the rest of Western Europe. Meanwhile, my understanding is that the ‘bright line’ distinction implied by the ‘Hajnal line’ in the East has been substantially eroded by more detailed scholarship, with the case for a single, clear and consistent ‘Eastern European type’ of marriage and family formation coming apart as it fails to meet the complexity of the evidence.¹⁶ All of which makes me – quite far outside of my specialty – very hesitant to hold forth on the geographic extent of the pattern or variation within it.

Nevertheless, this late pattern clearly existed, particularly in Britain and the Netherlands: the early modern period provides significantly more robust evidence to make that assessment, with much of the uncertainty of the previous sections melting away under the weight of detailed records. The marriage pattern here comes alongside a bunch of other notable differences. First, under this marriage pattern, a significant percentage of both men and women never marry, on the order of a quarter or a third, compared to the 90+% marriage rates under the other marriage patterns.

The second has to do with household formation: whereas in most pre-modern agrarian family patterns, the assumption is that a newly married couple remains a part of an existing household (usually the groom’s) after marriage, the late/late pattern is associated with newly married couples immediately forming a new household. The term for that is neolocal residence, the cultural pattern and assumption, likely familiar to most readers, that a newly married couple moves out of their parents’ houses into a new dwelling and a new household of their own.

Now, as noted this pattern is really well documented for early modern Britain and the Low Countries, but naturally that raises (not begs) the question of how far back that pattern goes and – because it is such an unusual pattern – what caused it. It is difficult not to see at least some of the pattern, particularly the increased prevalence of never-married individuals, as at least partially connected to Christian teachings; Geoffrey S. Nathan notes the fairly clear early connection in Late Antiquity between Christianity and both an increased status for women to remain widows and not remarry if their husband died or, in some cases (far less than in the late/late pattern) lifelong celibacy.¹⁷ Delayed marriage also would obviously function as a form of fertility control in the contexts of societies – and this is true of nearly all agrarian pre-industrial societies – where sex outside/before marriage was intensely discouraged. As we’re going to see next time, even under the pre-modern mortality regime, some form of fertility control was both possible and also clearly practiced. In the context of a culture perhaps unwilling to practice fertility control within the context of marriage, delaying marriage may have allowed for the same outcome. Finally, it is worth noting – this was in the footnotes of the previous part – it is possible that the mortality regime in early modern Europe was somewhat less harsh.¹⁸ If that was the case, households looking to avoid expanding too rapidly might delay marriage for the same fertility control reasons.

But then how far back does this pattern go? After all, if this is purely a modern pattern, we might dismiss it, though that would raise some significant secondary questions, since the early centuries of the modern period were not all that different in terms of agricultural production or medical technology than the pre-modern period (the radical breakpoint in standards of living is the industrial revolution in the 19th century, not the arrival of ‘modernity’ in the 16th). Now the challenge of course is that the evidence for the Middle Ages is much weaker, getting dramatically weaker the further back you go. Now medieval demography is not my field, but my sense is that at least in England, we can see evidence of the early modern late/late marriage pattern pushing back at least to 1500 and quite possibly as early as 1300, suggesting that the ‘western European’ model may, in fact, project back in some form into the late Middle Ages.

Marriage and the Individual

To recap all of that, one way we can classify marriage systems is by the typical age at first marriage (AAFM), with the common combinations in the pre-modern world being (expressed as female/male): early/early, early/late (as we see in Greece), and intermediate/late (as in Rome). The oddball is the pattern in parts of early modern and perhaps late medieval western Europe, which was a late/late pattern, which came with its own quirks in terms of household formation. We’re going to set the late/late pattern aside for right now, though we will discuss its fertility implications next week.

But I do want to note something about the age ranges here, which is that what we see is not a set number of highly distinct systems, but rather a range of marriage ages that overlap between ‘types.’ Some Roman girls were entering their first marriage in their early or mid-teens, much as would, so far as we can tell, have been typical for Greek girls, while some Roman women only married in their twenties, close to the typical ages for the early modern late/late pattern. That doesn’t make average AAFM a meaningless statistic – what was normal or typical in a society matters – but it is important to keep in mind we’re dealing with something like a continuum of practice rather than a clearly distinct set of buckets (arguably with the exception of the late/late outlier pattern). I think this actually comes out fairly clearly comparing the evidence for Classical Greece, the imperial Roman Italy and Roman Egypt, which despite all being Mediterranean societies with similar agrarian economies, sit at different places on a sliding scale of average AAFM between 14 and 20 – 14-15 for Greece, 17-18 for Egypt, c. 20 for Roman Italy.

But I want to close talking about some of the cultural assumptions embedded in these different models. Now marriage is a culturally specific institution, so a discussion of it in some general terms is going to smudge over significant differences from one culture to the next. One of these days we’ll need to loop back and discuss Roman families and family law to get more into the weeds on a single specific culture. But there are some things we can say in general.

First, as you will recall these societies are substantially less individualist than most modern societies: they understand individuals primarily as filling a role within a larger community, as parts of a whole, gears in a machine rather than ends to themselves. Thus it neatly follows that they do not understand marriage as an expression of individual love, but rather as an institution important for its communal role. At Athens, the legal formula for marriage specified that it was “for the production of legitimate children” – as both heirs for the father’s family but also as contributing citizen-members of the state; the marriage served the community by creating children, not the individuals being married. It isn’t that these societies have no concept of romantic love, to be clear – they certainly understand both love and lust – but that marriage, as an institution, was at best incidental to those feelings. Instead marriages, in particular first marriages, were almost always arranged by the families of the betrothed.

In that context, marriage became, particularly for women and girls, a key and functionally mandatory stage of progression through society. In Latin, marriage is the dividing line between the puella (‘girl’) and the mulier (‘woman’), uxor (‘wife’) or matrona (‘matron’); and in Greek it is the transition from κόρη (girl, with a strong implication of virginity) or παρθένος (‘maiden’) to γυνή (‘woman’ ‘wife’).¹⁹ Marriage was somewhat less defining for grooms but only somewhat less, often a capstone on successful early adulthood rather than the entry to it. In most of these societies, men were not much more in control of their (first) marriages than women: those decisions were made by parents, family and community and intended to serve the interests of the community, rather than the individual.

In both cases, marriage was expected rather than a personal choice; functionally everyone (who lived long enough) got married and then proceeded to have children, if they were able. Doing so was part of the human condition, an essential part of the role of being a son or daughter, a member of the community.

That is not to say all or even most pre-modern marriages were loveless. The texts these societies produce, particularly funerary texts, are full all over of expressions of deep affection for spouses. That may seem strange given that for the most part these people didn’t choose their spouses, but I imagine, dear reader, that you likely care deeply for your parents, siblings or other relatives and you didn’t choose them either.²⁰ While these societies mostly did not expect romantic love within a marriage, there was an expectation of the sort of affection that comes from living and working together with someone to whom your interests and future is tied. Wholly uncaring, callous or negligent spouses were understood in these contexts as both an aberration and also a moral failing: spouses might not have love but they had duties to each other (again, consider the relationship between parents and children or between siblings).

Patriarchy

These marriage structures were also invariably patriarchal – by which we mean power in the household was concentrated in the male head of household, almost invariably the eldest surviving father – and we should be open about that. Now the ‘RETVRN’ crowd seems to imagine that such arrangements meant that men were ‘in charge,’ but that is a misreading of how these societies are structured: remember these are radically less individualistic societies. The male peasant head of household isn’t the master of his own fate any more than any other member of his family is: he is a cog in a communal machine, bound to obey the dictates of his elders and fill his role in the community. Even once his elders pass away, he remains under the thumb of his social superiors, who like him are also bound by strong social expectations of conduct (and strong social claims by relatives and other connections on his time and resources) that most moderns would find intensely stultifying. Very few people in these societies, male or female, would feel very much in charge of anything. I am struck, for instance, for that a medieval European Christian, be they humble or noble, deciding to take holy orders and become a monk, it was probably the first, last and only true life-choice that person ever made for themselves about the role they would fill in society. In a patriarchal society, males wielded more power, but everyone alike was born into a role they were expected to perform quite regardless of their own wishes, talents or abilities.

All that said, power in the household was concentrated, not in the men generally, but in the male head of household, who owned all of the property and controlled all of the people in the household. Patriarchy, after all, does not mean ‘rule by men’ but ‘rule by fathers‘ and the title is apt for the families formed here.

Which is not to say there was no variation at all. A brief return to the brief comparison of Greek and Roman marriage and family-law customs (I really do need to write this up as a blog post at some point) can serve to demonstrate the range, as the poleis of Greece were some of the most restrictive patriarchies in the ancient Mediterranean and the Romans some of the least restrictive patriarchy (while still very much being a patriarchy). Women in Athens were not legal persons, their consent was, so far as we can tell, neither sought nor necessary to form a marriage. They could not inherit or hold property and indeed there was at Athens (and, it seems, in most if not all other Greek poleis) an office with the power to compel a woman who was the only valid heir to property to remarry (entirely irrespective of her wishes) in order to generate a valid male heir to the property. As far as we can tell, Athenian women could not initiate divorce (but Athenian men could).

By contrast, Roman women were legal persons. A fig-leaf of consent was part of the Roman marriage ritual, although social pressure and the fact that silence constituted consent must have meant the bride’s opinion was rarely decisive. Roman women could and did inherit property and could, upon the deaths of their fathers or husbands (both of whom are likely to be older than them) become legally independent. A Roman widow could not be compelled to remarry and indeed it seems like second marriages – common at Rome, both due to the death of spouses, but also fairly frequent divorce – were often at the discretion of the couple. Roman women could, from at least the second century BC, initiate divorce, taking their dowry and any personal property with them when they did.

Roman society was still very much a patriarchy – the male head of household had patria potestas (‘the fatherly power’), total legal control over the members of his household, controlled the property and exercised tremendous power – to the point of being legally able to kill them – over his children.²¹ But even a casual glance tells us Roman society allowed remarkably greater latitude for wives. Perhaps unsurprisingly, later average age at first marriage for women seems to correlate with greater freedom for women in that society and from what we’ve seen the Romans largely fit this pattern: as one of the most gender-liberal ancient societies (again, still a patriarchy, we are judging against a very low bar here), they also have one of the latest apparent average female AAFMs in antiquity, sliding into the ‘intermediate’ range above.

Of course brides are only half a marriage and so we should also give some attention to the male marriage patterns implied here. Male average age at first marriage is almost always older than that for females. Even in an early/early pattern, that implies a female average AAFM around 14, 15 or 16, but a male average AAFM generally around 20. In a late-male-marriage pattern, the male AAFM might be as old as thirty. And notably, the common patterns above are (female/male), early/early, early/late, intermediate/late and late/late. Which is to say under all of these pre-modern marriage patterns, grooms will generally be older than brides at first marriage, sometimes much older (notably, this age gap does seem in some societies to narrow for second marriages, again speaking to a situation where first marriages are for the community while second marriages were for the spouses, but of course most individuals only married once). That has its own implications for the structure of power in a household, of course, further reinforcing the patriarchal nature of the household.

But at the same time, that pattern also speaks to how even men are instrumental rather than individual, within a patriarchy because of course we have two models for men: an early model where boys marry while their parents still live and have little if any choice in the matter and a late model where men are made, for social reasons, to delay marriage until fairly late in life, likely also against their actual wishes. I plan to talk about the differential attitudes these societies have to male and female chastity in the next post covering children, but I’ll note that while on the one hand it was common for these societies to have sexual outlets for young men who were not yet of marriageable age, such outlets were mostly available to the wealthy or urban, not to the peasant in a small village.²² Instead, young men chafing, sometimes violently, against family structures which denied them the ability to start households until later in the lives are a common feature of these societies. In Greek literature, for instance, sharp, sometimes violent conflict between fathers and sons is a frequent motif and Greek law with a concern in particular over sons killing their fathers which starts to make a bit more sense when you think about how a late marriage pattern that demands a son delay marriage and household formation until well into adulthood or the death of his father might create intense resentment and anger.

Of course, marriage is only half of the story in family formation: the other half is children. And indeed, as we’ll see, pre-modern peasant societies generally understood these two as parts of a whole, to the point that a marriage without children might not be much of a marriage at all (barrenness, almost always blamed on the woman, was often a valid reason for divorce even in cultures that otherwise did not accept divorce). So now that we have our marriage pattern, next week we’ll begin looking at childbirth and child rearing.

https://acoup.blog/2025/08/01/collections-life-work-death-and-the-peasant-part-iiia-family-formation/?utm_source=rss&utm_medium=rss&utm_campaign=collections-life-work-death-and-the-peasant-part-iiia-family-formation

https://acoup.blog/?p=30696