This is the first part of the third part of our series (I, II, IIIa, IIIb, IVa, IVb, IVc, IVd,IVe, V) 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.
And before we launch in, I suppose I ought to give some warning for this week’s post and next, that we’re going to be talking about societies where marriage and child-bearing begin substantially younger than today, often distressingly so, a valuable reminder that just because things were normal in the past does not mean they were good.
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
No, I will not apologize for this joke.
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.
Elite marriage patterns do not always match non-elite marriage patterns in a given society at a given time.
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.
- See fn. 1 on the previous post on this point.
- Societies with polyandry exist, but are, to my understanding, very rare.
- Frier, op. cit., 799.
- In particular, monogamy was the rule in ancient Greek culture from at least the Archaic if not earlier, but some Greek-speaking cultures (Epirus, Macedon) allowed for polygamy for the king and, so far as we can tell, no one else. So Philip II or Alexander III could (and did) have multiple wives, but even Alexander’s top generals were expected to be monogamous. Interestingly, the extinction of the Argead royal house in the Wars of the Diadochoi also largely ends Macedonian royal polygamy with it.
- R. Saller, Patriarchy, Property and Death in the Roman Family (1994), 36-8.
- In particular, as we’ll note below, I am wary of John Hajnal‘s ‘Western European marriage pattern’ and ‘Mediterranean marriage pattern’ dichotomy, not the least of which because, as we’re about to see, there’s more than one Mediterranean marriage pattern.
- Used correctly to mean demanding a counterpart in an argument to concede the main issue of the argument at its outset
- My very tentative critique here is that it seems to me to often be a racially inflected assumption, with scholars actually using data often correcting assumptions of pre-modern ‘white’ populations AAFM younger, while outside of European populations it seems like the ‘early’ pattern is assumed even when data doesn’t exist.
- E.g. the otherwise very useful T.W. Gallant, Risk and Survival in Ancient Greece (1991)
- S. Pomeroy, Families in Classical and Hellenistic Greece (1997).
- S. Blundell, Women in Ancient Greece (1995), 120-1.
- See Saller, op. cit. and also R. Saller, “Men’s Age at Marriage and its Consequences in the Roman Family,” CP 82 (1987) along with B. Shaw, “Age of Roman Girls at Marriage: Some Reconsiderations” JRS 87 (1987). Cf. K. Hopkins, “The Age of Roman Girls at Marriage,” Population Studies 18 (1964/5), which posited something closer to the ‘early’ model for the Romans; Saller and Shaw demonstrate this is inconsistent with the epigraphic evidence. Contra Saller and Shaw, note Lelis, Percy and Verstraete, The Age of Marriage in Ancient Rome (2003), which instead essentially posits an ‘early’ model for the Romans; I don’t find the author’s arguments ironclad here and infact neither do they (“the overall shape of the available evidence, even though it does not allow us to fix the ages of first marriage with extreme precision, strongly suggest…” (emphasis mine) p. 100, is a pretty wild uncertainty caveat for a book that otherwise presents itself as decisively overturning Saller and Shaw’s model), the thrust of the argument that anything other than an early AAFM regime makes no sense for the Romans seems belied simply by the existence of an extremely well documented late AAFM regime in early modern (and likely late medieval) Europe. If a ‘late’ AAFM model works under similar mortality conditions to the Romans (and it does) it is hard to so casually rule out an intermediate model, which is what the epigraphy suggests.
- So Roman men and Greek men marry around the same age, but Roman women appear to marry around half a decade later than Greek women, somewhat narrowing the age gap in the typical family.
- Do not snip out this caveat, it is a crucial caveat.
- As an aside, I think this data also weakens the case that Lelis, Percy and Verstraete are making for an early Roman marriage pattern because it serves to demonstrate that not every Mediterranean society necessarily had an AAFM in the mid-teens. Consequently, the Roman pattern teased out by Saller is not some strange outlier exception to a ‘Mediterranean marriage type,’ but rather a point on a continuum of Mediterranean marriage types.
- On the more complex Eastern European evidence, see Szoltysek, “Spatial construction of European family and household systems” Continuity and Change 27.1 (2012), while on the greater variety of household and marriage structures and the ‘bright line’ divider would admit, see R. Wall, “Characteristics of European Family and Household Systems” Historical Social Research 23 (1998) and “European family and household systems” in Historiens et populations (1991), both discussed by Szoltysek.
- G.S. Nathan, The Family in Late Antiquity (2000), 116-133.
- Some data on this in R. Woods, “Ancient and Early Modern Mortality: Experience and Understanding” Economic History Review 60.2 (2007)
- Both languages have a way to say ‘adult female’ – femina in Latin and ἡ ἄνθρωπος (‘human’ with a feminine article indicating that a human female is intended) – without implying a marriage state but neither is a particularly polite way to describe an adult woman. They both have the same sense of “a mere adult female, rather than a woman” and so might be used dismissively against individuals of low status or with intended contempt, though they are sometimes used neutrally.
- Also remember that, because people in the past generally believed their own religion, the superiority of a human ‘chosen-family’ over a divinely chosen family – for most of these cultures believe these matters to be ordered by God or the gods – would hardly have been obvious.
- Though with the complexities of Roman family law, a wife might be under the patria potestas of her father or her husband or – if one of the two was dead – legally independent, while being married. Again, we really need to do a compare-contrast of Greek and Roman family law at some point.
- As Crone, op. cit., 111-2 notes.
Can we have a reference for footnote 4? From my reading on the Ptolemies and Seleucids, Royal polygamy seems to have been the norm rather than the exception. Possibly as official concubines rather than wives, but that difference seems purely semantic. I’d love to know exactly what Bret means here, or if there is further reading on the matter.
On a similar note, there’s a typo in ft. 9: it should be “Risk and Survival in Ancient Greece (1991)” rather than “Rick and Survival […]”. Although that might make a good Rick and Morty episode lol.
1. Dr. Deveraux said that polygamy was royal, and you also say polygamy is royal, this doesn’t seem to be a contradiction? The point was that non-royals are (expected to be) monogamous.
2. The Ptolemic and Seleucid Dynasties were ethnically Macedonian (which is already questionably “greek” depending on who you ask), but neither ruled over actual Greece, and probably got a lot of their customs from the places they ruled over – Egypt and Persia, basically.
The last sentence of footnote 4 says: “The extinction of the Argead royal house in the Wars of the Diadochoi also largely ends Macedonian royal polygamy with it.” That’s the bit I’d like more detail on.
Maybe (as you mention) he means it ended in geographical Macedonia rather than in all ethnically Macedonian royal lines. Maybe his “largely” carries a lot of weight. Maybe he was drawing a distinction between wife and concubine that I wasn’t. Maybe my knowledge of the topic is just wrong. Maybe it’s something else. Either way, that’s precisely why I asked for a reference: to find out more.
Yeah, the difference between “Marriage+concubines with lesser rights” vs. “Marriage with several wives but the first/primary wife having more rights” seems largely semantic.
In any employment situation there is a material difference between seniority and a job title distinction. It may not be a legible difference to us now, but I’m sure it was meaningful to the people involved at the time.
Whether it’s an illustrative comparison or not…
Think of Versailles-era France, where a royal mistress could have enormous influence but the position had no legal rights.
Societies with polyandry exist, but are, to my understanding, very rare.
They are rare, but precisely because they are rare, they’re of special interest to anthropologists (and before that, to other types of intellectuals). Looking at where they exist, why, and what forms they take helps us test and refine our theories about marriage and mating more generally.
The rarity suggests this isn’t a particularly successful pattern. one example, a Himalayan tribe I believe, several brothers marry one woman in order to jointly inheritance and work marginal land and reduce the next generation of heirs.
Truly excellent article!
I was about to indulge in pedantry about the point that “sex outside/before marriage was intensely discouraged” in pre-modern societies, but in the end it was promised that the topic of “sexual outlets for young men who were not yet of marriageable age” would be covered in the next part.
I’ve also read that (at least pre-Hellenistic) Egyptian society was freer than average for women; I wonder if we know anything about marriage ages in earlier eras of that civilisation?
Additionally, life must indeed have differed a lot by social class, given that Philostratus could write that the sophist Aelian “avoided child-rearing by never marrying. Whether this is a blessing or a curse it is not the right time to consider”!
I was about to indulge in pedantry about the point that “sex outside/before marriage was intensely discouraged” in pre-modern societies,
Yes, my response to that would be it depends *heavily* on which premodern society you’re looking at. Pre-modern Catholic Ireland is going to look very different than the pre-modern Trobriand Islands.
Although as a description specifically of the pre-industrial Christian world (and to a certain extent the Islamic world as well) I think that’s more or less right.
Even in nominally Christian societies, there are lots of hints that women’s lives in Merovingian Gaul were more “LA Gossip Column” than “monogamous marriage with the occasional secret affair.” Ancient Egyptians don’t seem to have cared about sex outside marriage except infidelity. Its very hard to tell about barbarian Europe because most of our sources are post-Augustus Romans, many of them Christians, who are more interested in moralizing than describing.
I’m not sure we have too much information about the pre-modern Trobriand Islands, but in regard to pre-modern Catholic Ireland, we know that it was a polygynous society, with the law texts specifying varying grades of concubinage that included women who were visited by a man without the consent of the woman’s kin, and that most women did not have a legal personality and had a very limited ability to testify in court, although women could divorce their husbands for injury or failing to perform conjugal duties, and senior wives could void disadvantageous contracts entered into by their husbands. A senior wife could also freely injure secondary wives/concubines for a period of up to three days following the discovery of the said secondary wife/concubine.
Where women did own property in their own right, their children would not generally have been their heirs – rather their property would be inherited by their male kin (presumably uncles/cousins, though conceivably they might have a younger brother if they had been trained in their father’s profession and their father went on to have a son late in life).
A young man who lived in his father’s house had a restricted legal personality. I think they might have been able to enter into advantageous contracts like marrying someone of the same social rank without their father’s permission, but I’m not sure. (Marrying someone of higher station was a disadvantageous contract, and would definitely have required their father’s permission). IIRC, men who lived in their own house on their father’s land did have a full legal personality, and the ability to freely enter into contracts based on their honour price.
Everything I’ve encountered on the topic states that ancient Egypt was early/early, and needed to avoid the fertility delays that Bret mentions because it had a much lower life expectancy than a lot of these other societies. While the upper classes had lifespans similar to the northern Mediterranean cultures, based on skeletal evidence the lower classes typically only lived to the age of 30-40 if they made it to adulthood. Marriages thus tended to be around age 14 for women and 20 for men. Waiting longer would have tended to drop the net fertility rate below replacement level, between child mortality, maternal mortality, and the low average lifespan overall.
Please check Bagnall and Frier, Demography of Roman Egypt. Like summary tables in the end.
The censi is snapshot – combination of snapshots but still snapshot. Recording everybody from baby to elder, both sexes. The rich are included but they´re minority – just 26 out of 167 households have any slaves, and the slaves themselves are included.
Table A pages 348-350 gives people with preserved exact ages – 710 out of 1084. These are papyri with age preserved – the rest are papyri where ages surely were once written but which are damaged at the place of age (while other stuff like name or relationship may be legible). Note how there is a modest number of people with known age but unknown sex – damage in that other place.
Summing up, I get:
0-9: 170 people
10-19: 132 people
20-29: 139 people
30-39: 110 people
40-49: 74 people
50-59: 48 people
60-69: 26 people
70-78: 20 people
80: 1 person
above 80 – nobody
Sounds the Egyptians who lived in their 20s had above even chance of also living in their 40s.
At the risk of overstating the obvious, Roman Egypt is not pre-Hellenistic.
“such [sexual] outlets were mostly available to the wealthy or urban, not to the peasant in a small village”
Didn’t small villages have donkeys, sheeps or goats ?
Was social control enough to prevent horny teenagers to have a moment alone with the animals ?
Also, was it considered shameful and thus mocked if caught on the act, or actually vilified (so risks were more severe) ?
This is in a medieval context, but I imagine it applies in many ways across many cultures: https://going-medieval.com/2020/01/20/no-bestiality-was-never-ok-you-absolute-rabid-weirdo/.
I imagine a lot of horny teenagers then do what horny teenagers do now and that’s to find another horny teenager.
Thanks !
But your link makes it abundantly clear in its prose that the condemnation of bestiality in official texts is massively tied to their christian approach.
And Devereaux made clear that finding another teenager was not really an socially accepted option.
Bestiality is like many other unusual sexual preferences more common than you might think (as much as a few % of people have engaged in it) and also generally considered *very* distasteful and not just by Abrahamic faiths. Eg the Hittities it was one of the very few things that could get you the death penalty. A fairly common category of ethnic insult is to imply those people are having carnal relations with their livestock.
Beastiality is clearly A) One of the things that happen and B) One of the things that is just about always HEAVILY stigmatized, and usually outright banned and persecuted.
And we’re not talking low numbers here. People getting executed for beastiality is semi-normal.
“And Devereaux made clear that finding another teenager was not really an socially accepted option.”
Still strange that your first thought is of sheep or donkeys, when situational homosexuality would have been just as hard to prove and not morally abhorrent to a modern person.
The one thing that would for certain have gotten a male peasant teen in trouble would have been impregnating a female peasant teen who was still unmarried. (If he could find one. With girls being forced into marriage as soon as they entered puberty, finding an unmarried horny female teen would be hard.)
Of course, adultery would also have been an option, especially with all those young girls already in forced marriages to much older men, and no genetic testing to prove paternity.
Risky, sure, but not moreso than bestiality. If you are going to risk execution, then why not for actual sex rather than tormenting an animal?
(I also cannot understand what a man would get out of molesting a donkey, but then, I’m a woman, and female bestiality mostly only exists in stories invented by men, so I guess the male sex drive must be very different.)
Fair point.
Thing is, this practice is always reviled, but depending on societies it can be reviled either as an abomination (as apparently it was for the Hittites, and also in the Bible) or as a social shame that pursue the author, making him the butt of the jokes in the village.
Basically, what I was asking is how these two conceptions were distributed in the rural societies the Pedant¹ talks about.
Which includes, yes, in which of those homosexuality was more reviled than bestiality and in which it is the reverse.
¹Aside note, I just finished Jaworski’s “To the victor goes the spoil” adaptation, so mentioning Pedants is a bit weird
“Girls being forced into marriage as soon as they hit puberty”…Er, there’s a cool blog post I read recently about how that’s not actually the predominant marriage pattern…
In theory, one potential factor is that given the mores of the time, it’s the married (young) woman who would have faced far more severe consequences. I already mentioned in an earlier thread that according to The Faithful Executioner: Life and Death, Honor and Shame in the Turbulent Sixteenth Century (a book adapted from the diary of Nuremberg’s longtime executioner Franz Schmidt, who served for 45 years and executed approximately one person every 6 weeks, to a total of 361), one of his cases involved a woman who slept with both a father and a son (I think all three were unmarried at the time, but it’s been a long time and I currently lack a copy.) This was apparently considered incest by proxy, but she was the only one Schmidt was ordered to execute, while I believe the men got away with corporal punishment. (His diary recorded something like “Well, she definitely had to have known what she was doing, while the two may have believed each one was exclusive to her, so it must be fair.”) It seems quite likely that the knowledge of such legal approaches would substantially lower girls/women’s risk tolerance relative to their male peers, and make it harder to obtain consent.
P.S. About that last paragraph…
https://en.wikipedia.org/wiki/Zoophilia#Extent_of_occurrence
Girls wanted to be married. They didn’t have the romantic notions of modern teens but they had the burgeoning hormones. More importantly a girl’s life didn’t begin until she married. Mistress of a household and mother of a family was the closest thing to power and freedom available.
In the Cuban poet Reinaldo Arenas’ autobiography, he provides an account of his rural, mud-hut, no-electricity upbringing as a peasant in rural Cuba as including after-school episodes of grouping up with all of the other boys to escort a mare to a large rock, which they would stand on top of while taking turns having their way with her.
The cases of fornication brought up before the priest and manorial court suggest societal norms didn’t stop people from indulging.
“Both languages have a way to say ‘adult female’ – femina in Latin and ἡ ἄνθρωπος (‘human’ with a feminine article indicating that a human female is intended) – without implying a marriage state but neither is a particularly polite way to describe an adult woman.”
The more things change, the more things stay the same. I think its interesting that modern English has a very similar linguistic usage for females vs women, with the former being faintly or blatantly derogatory depending on context, with an overtone of misogynist politics(yes I know military people use it differently, I am commenting on common parlance). I am not a historian, so I can’t really comment on this, or honestly read anything into it other than ‘huh, thats interesting’, but it jumped out to me.
What kind of data or estimates do we have in what % of arranged marriages were somewhat consensual? As in “we’re going to try to match you to this guy/girl you know and approve of (because everyone in this village knows each other)” vs “Surprise, you’re marrying an horrible person you don’t know, but I got a fat cow for a dowry!”
That sounds like a really tough thing to get estimates of. Not saying it’s impossible, but it would require very detailed “visibility” into the specifics of the marriage negotiations and how the marriage went afterwards. We might have that kind of thing in some cases in early modern Europe or in China and a few other relatively literate, documented societies whose records have mostly survived… but even then it’s hard to figure out if we’re looking at typical or atypical cases.
We might have that kind of thing in some cases in early modern Europe or in China and a few other relatively literate, documented societies whose records have mostly survived… but even then it’s hard to figure out if we’re looking at typical or atypical cases.
Another alternative is to look at the work that 20th century (and maybe even some early 21st century) anthropologists did with surviving “premodern” or “semi-modern” people, and use them to extrapolate what we think premodern people might have been like. There are some obvious problems with that approach too, of course, since no society is really a perfect proxy for any other.
Yeah, one of the challenges is ‘arranged marriage’ can rapidly shade into not arranged marriage. At least in modern times you often get the full spectrum from forced marriage to stranger with no concern about compatibility at one end to on the other end parents suggesting some sensible dating options or facilitating a love match at the other. Conversely parents often had or have considerable ability to veto a match even if match-finding is largely a self-run romantic effort.
Yeah, one of the challenges is ‘arranged marriage’ can rapidly shade into not arranged marriage.
Correct. I’d also add that ‘consent” is kind of flexible concept because to a certain extent, we consent to what our culture shapes us to think of as normal and expected. This goes for modern American society too, of course.
That’s one of the problems, actually: One of the reason Hajnal was so revolutionary with his late/late family pattern thing was that earlier scholars had assumed that early-modern and medieval western europeans married like russian serfs. (who tended to marry earlier, for various reasons) and he could show, with evidence, that was not the case. (at least in some areas, etc. as many revolutionary scholars he arguably over-egged the pudding a bit, and the geographical terms is a bit wierd since IIRC; even Hajnal noted there are patches of early marriage within “Western Europe” etc.)
Do keep in mind that in return for the fat cow, the father gets a horrible son in law who doubtless comes from a horrible family, who will now be relatives. I suspect that even among peasant families, marriages were arranged to secure useful alliances.
While I don’t have data backing me up, I’d imagine that it would lean towards the “we’re marrying you to someone that we think you’ll get along with” end of things when possible.
Leaving aside emotional reasons for doing so (the whole “my child’s happiness matters to me because I love them” thing), a marriage where both parties are broadly content is going to do a better job at maintaining the all-important horizontal social bonds that stop people from starving in bad years than one where one of the parties is actively miserable.
I don’t think “female” vs. “woman” in English is really the same thing. Neither word says anything about her marital status. I think Devereaux is saying it was rude in Greece and Rome to not call her a wife.
Even in English language, calling someone a “woman” is a bit rude. Compare the three sentences:
*”There is a female asking for you in the lobby.”
*”There is a woman asking for you in the lobby.”
*”There is a lady asking for you in the lobby.”
The first is so weird that it is not even an insult but a linguistic error demonstrating infamiliarity with English vocabulary. The second one is not really an insult, but you would not use it if the woman in question is a social peer.
“When it comes to females, Cosmo ain’t got nothin’ to do with my selection.”
50 years ago, I think #2 would have always been considered rude. Today, it’s at least as common as #3 in the US overall, and not considered rude at all in most of the US. Indeed using #3 can be now considered as indicating a conservative / patriarchal mindset.
#1 is still asking for trouble, though.
On the subject of “females” vs. “women”, one thing I hold by is to listen for consistency. Men and women? Great. Males and females? Sure, whatever. Men and females? Problem.
The female being considered derogatory is interesting, because it’s still very much a twitter/bluesky taboo that hasn’t made it over to ‘IRL’ yet.
Random examples from google: a few days ago there was an article from The Guardian titled “Michelle Doughtery made first female UK astronomer royal”, while Columbia University had a recent news article titled “Columbia’s First Female Academic Was An Astronomer”. That sort of wording would get you serious side eye on, say, reddit, but it apparently doesn’t raise any eyebrow even in firmly progressive spaces offline.
I think “female” is perfectly still acceptable as an adjective- @Ian was referring to it being used as a noun, where (outside of some marginal contexts, like police blotter reports) it definitely sounds…..off.
I think that’s because referring to someone as “a female” (or “a male”) leaves open the possibility you mean another species instead of human.
How´d you handle an ambiguous/marginal case as to whether to specify a female as “a girl” or “a woman”, or whether to specify a male as “a boy” or “a man”?
“How´d you handle an ambiguous/marginal case as to whether to specify a female as “a girl” or “a woman”, or whether to specify a male as “a boy” or “a man”?”
The terms are vague enough to allow a bit of error. And there are other terms like “teenager” or “young woman”/”young man” which bridge the gap nicely. If you really don’t know – for example, you’re designing a swimming pool and you want to talk about the changing rooms – you can just use it as an adjective and say “female customers will go through this door, male customers go down the corridor and turn left”.
We manage.
And there are other terms like “teenager” or “young woman”/”young man” which bridge the gap nicely.
Right. and even in the rare instances where you might really want to differentiate the two, it’s not like it takes *that* much work to type “women and girls” rather than “females”.
I slightly disagree here, but only slightly, because there’s a reason that Ferengi references are common when talking about people saying “females”. The Ferengi have been very clearly awful people who are also super sexist since they first appeared in Star Trek. (Although it became much more obviously highlighted in the 1990s with Deep Space 9 having a Ferengi ensemble member.)
No, the thing that actual progressives are generally concerned with (related to this use of ‘female’) is known as “nouning an adjective.” This is an intentional, derogatory habit of speech common to many conservative conversations, where sustained/consistent use dehumanizes the group or individual but discrete instances are deniable microaggressions.
This intersects with the derogatory use of “female” in complicated ways, because the adjective (and adverb) use of ‘female’ was created through noun-to-adjective (-adverb, too) conversion–what might be termed ‘adjectiving a noun.’
‘nouning an adjective’
Also adjectiving a noun, like the right-wing push to use ‘Democrat’ in place of ‘Democratic’.
The Romans and Greeks were “relentlessly monogamous” in legal sense – but not in the social sense.
Please remember to account for slaves.
Bagnall and Frier says that out of 95 known households in villages of Roman Egypt, 11 report slaves. The total 42 includes 36 females and just 6 males (page 158), and a lot of the slaves are children.
Some female slaves were in family relationships with slave men. But very often there is no other father in sight and very probably the master of the household begat both his wife´s children and his slave´s.
Census schedules from Egypt report data for poll tax only, not also for land tax, and thus do not combine family data with any income they might have got from leasing land out to free tenants. But a householder who owns just one adult woman, burdened by her small children, is not very rich by slaveowning alone. Chances are he might have to work with his own hands to feed himself, his wife AND his slaves.
In Classical Athens, freed slaves were still metics. In Rome, they were citizens, nor did Roman Egypt bar freed slaves from taking up the ex-owner´s citizenship. The evidence from Rome and Roman Egypt was that freed slaves were demographically important. At several % of population being currently slaves in countryside of Roman Egypt, and several % being ex-slaves, the well-off peasants who could afford to own a sex slave and then manumit her were demographically something to consider.
A peasant would not be able to afford a “sex slave”. Sex is not the primary purpose of slavery. Even today, if you read the stories of women enslaved by the ISIS, the main occupation of slaves was diverse menial work. Rapes were a side show: an important part of the utter degradation and lack of dignity, but still, most of the time, the enslaved women were doing something else than being raped. One horror of slavery is the fact that all slavery includes the element of sexual abuse. Adding “sex” in front of “slave” is redundant and only serves titillation.
On the other hand, a farming community, with small houses, tight interaction and very little privacy meant that most likely, many married men who actually had slaves probably refrained from utilising the legal opportunity, if only to maintain domestic peace. “It is better to dwell in a corner of on a household than with a brawling woman and in a wide house.” (E.g. Jacob in the Bible takes slaves as concubines only after his wives explicitly ask for it.)
One horror of slavery is the fact that all slavery includes the element of sexual abuse. Adding “sex” in front of “slave” is redundant and only serves titillation.
@FinnishReader,
This may be a minor exception, and I’m not sure how well enforced it was, but wasn’t sexual intercourse between slave-owner and slave explicitly forbidden in the Crusader Kingdom of Jerusalem?
“On the other hand, a farming community, with small houses, tight interaction and very little privacy meant that most likely, many married men who actually had slaves probably refrained from utilising the legal opportunity, if only to maintain domestic peace. “It is better to dwell in a corner of on a household than with a brawling woman and in a wide house.” (E.g. Jacob in the Bible takes slaves as concubines only after his wives explicitly ask for it.)”
The ideal of a husband having sexually monogamous relationship with his wife and refusing to have sex with dependent household females is also attested in Ancient Greece by start of records – in Odyssey. Laertes has a wife Anticleia and a well-treated maid named Eurycleia (Odysseus ends up having 50 maids, Eurycleia is in a leading role in household, pointing fingers at disloyal maids to hang), and expressly refuses to sleep with Eurycleia because he respects Anticleia. When Odysseus lives for seven years with Calypso and has sex with her, the narration points out that he is doing so as unwilling captive of a powerful goddess.
But while the ideal was there, Laertes and Odysseus may be presented as unusually faithful husbands, not average ones.
For sexual relationships between masters and maids, one treatment concentrating at 18th century France but with views towards 16th and 19th is this:
https://muse.jhu.edu/book/71470
Domestic Enemies: Servants and Their Masters in Old Regime France
This book reports two waves of movement against polygamy, and I can add two more between Antiquity and Middle Ages:
1) Spread of Christianity, 4th-5th century. Express disapproval of sex outside marriage, mentioning sex between master and slave under disapproval.
2) Retreat of domestic slavery and replacement by free maids. Of course sex by maids, consensual and otherwise, went on happening. So did pregnancies of maids. But the incentives of master and mistress changed. If the maid was a short term employee free to leave (to another master, to marry or to return to her birth home) without compensation then a pregnant maid was a liability (her working capacity was temporarily impaired, she might have a standing to sue somebody for the loss). When the maid was a slave, she was a long term investment, and so were her children – the wife could hope to see her legitimate children own their half-siblings in due time, while if the maid was free, all she got was her husband paying child support out of family budget.
Now the ones discussed in Domestic Enemies:
3) Tightening of Christian sexual morality as a result of Reformation and Counterreformation in 16th…17th century. In Middle Ages, up to 16th century, rich men could still keep a mistress and bastards in a household alongside wife and legitimate children. By end of 17th century, this made a public scandal – the wife, her family, community and authorities felt empowered/willing to protest. By 18th century, a married man who kept a mistress usually kept and visited her outside his and wife´s marital home. Sex between husband and maid happened but if the wife found out, the maid left (whether in disgrace if the husband did not care, or with private compensation and possibly continuing the affair in her new place if he did care).
Domestic Enemies discusses various sexual options available to maids (besides celibacy). Manservants, masters, non-master family members of masters… There are some statistics available because in 18th century, French authorities inquired into out of wedlock pregnancies. And the results indicate that master/servant sexual relationships were more common in small, rural households, compared to servant/nonmaster sexual relationships. In towns and the richer households, both the master and the servant had more alternatives. The master could more easily have sex outside household – prostitutes, kept mistresses outside household etc. Likewise, the richer masters were more likely to keep nonservant mistresses than have sex with servants.
4) End of 18th…19th century. The spread of romantic nuclear family further clamped down on sexual relationships with servants – even though the presence of maidservants in households remained ubiquitous into 20th century.
And now look at the countryside of Roman Egypt of 2nd century, and compare it with say countryside of 20th century Finland (early 20th that is).
Yes. (Free) farmhands and maids of 20th century Finland did mainly various menial work. So must the slaves of 2nd century Egypt. The people who could afford to hire 1…2 farmhands in 20th century Finland were peasants who had to work with their own hands alongside their slaves, and probably so were the people who could afford to own 1 maid (and her children) in 2nd century Egypt.
The farmhands and maids of 20th century Finland were expected to be involuntarily celibate. Pregnancy as a maid was a misfortune, and she was expected to point at and sue the guilty male. Because the normal approved conduct of a maid was to date chastely, without sex someone of suitable status, and then leave employment as a maid to marry and become housewife of a newly founded household. And there were a lot of farmhands around.
Compare the female slaves of 2nd century Egypt. They were fertile. Female slaves with children but no father given were routine.
And since male slaves were rare in Egyptian countryside – they were concentrated in towns – the fathers of slave children must have been free.
Who? Three options:
1) outside household
2) non-master men or boys of the household
3) master
There must have been cases of 1) and 2), of course. Like a household with children but no men at all – an elderly lady and her two maids with their children. It´s possible that the father/s of the children have recently left, or the children recently moved in, but another option is that the fathers were never in household.
But looking at the plausible factors:
1) ubiquity of slave children
2) rarity of slave men
3) that it was before the above mentioned crackdowns on master/maid relations
my guess is that a lot of slave children were begotten by (married!) masters after all.
“20th century Finland were peasants who had to work with their own hands alongside their slaves”
Sorry, cannot edit posts here once posted. Of course the 20th century farmhands in Finland were not slaves!
(Their agitators in 1918, however, pointed at the similarities and argued that a freedom to move to a different master without the practical opportunity to become a householder/housewife was not quite freedom.)
My understanding is by the early modern period at the latest most marriages in NW Europe were self-arranged. NW Europe was odd long before the Industrial Revolution. Arguably a lot of our ‘modern’ values are regional traditional values spread by cultural hegemony. Female marriage in 20s, low rates of cousin marriage, self-arranged marriages trends that well pre-date the Industrial Revolution. Marriage age for women was pretty much flat from 1550-1950, and probably well before then, though we lack the data. They may be a factor in the Industrial Revolution happening or merely accidently caught up in it, but they cannot be an *effect* of it. It wasn’t even all of Western Europe, eg clear into the early 1900s arranged and teen marriage was quite common in southern Italy. England in 1300 may have been more ‘advanced’ by modern sensibilities than Mezzogiorno in 1900.
Incidentally, I do think it is quite possible there is a causal factor where later age of marriage helped economic development. Unmarried women were much more likely than married to engage in wage labor. This pre-industrialization encourages monetization of the economy. Once you industrialize it provides labor. In the early 1800s close to half of *all* industrial workers were unmarried women, mostly in their teens and early 20s.
The theory I remember had it the other end: Neolocal family patterns require people to set up new households. This means accumulating resources (be it money, just stuff, etc.) this in turns spurs demand for various goods and services. (these new households are going to need various household goods, not all of which can be produced in the household, etc.) this creates incentive for something of a cash economy and opportunities for people to provide those goods.
I’ve read some speculation (unfortunately I don’t remember if it was more than an obvious guess) that the three odd practices perhaps co-originate. These being:
– manorialism (easy enough to contrast with e.g. Roman landownership ideas, with centurionation as the obvious example of “we just wiped the slate of previous ownership, what is the simple approximation of the ideal arrangement”);
– exogamy (the unusually wide interpretation of consanguinity, and particularly the choice to treat the marriage itself (as well as other church affinities) as links, thus forbidding rather than mandating levirate/sororate marriage);
– and the late/late+neolocal marriage pattern.
There is an interesting note that the high point of marriage as an institution in the West was the 1960’s. (IE: It was the point in time where the largest % of the population was married)
“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.”
I object to this fairly strongly. It’s common for non-monogamous relationships to be frowned upon (though not nearly as common as people believe), but IN PRACTICE the most common arrangement appears to be “Pretend to be monogamous but have discreet partners on the side.” In our culture this was so common that they had to stop allowing students to test their own blood type; too many cases of “Wait, he’s not my dad?!” There are also anatomical features that support this–in apes testes size is inversely related to monogamy, and we have fairly large testicles relative to body size. I’m not much for evo-psych (behavior rarely is preserved), but the common explanation is that the most common arrangement is to marry for stability and have affairs to maximize potential gains. The history of marriage also supports this–the whole reason the Church got involved in marriage in the first place (the originally blessed the union, but it wasn’t a sacrament) was to keep records because previous versions allowed far too many questions to creep in. The Church ran brothels for a while and priesthoods occasionally passed from father to son; they knew their ideals were not being adhered to, and profited by it (in fact, this caused a major controversy in the early Church, over whether a priest had to be a good person to perform their duties. They opted for the practical answer of “No, within limits.”).
Then there’s war. Not to belabor the point, in the past rape was used as a weapon of war, to the point where NOT raping victims was considered notable. And there were relationships one had along the way that were far more amicable–“Every man is a bachelor beyond Gibraltar”, the women you’ve discussed joining the foraging parties, and all that. And it wasn’t just the wealthy who had slaves. Peasants did as well, at least when they could, and the idea that large numbers of people would own other people without violating their bodily autonomy in this particular way is nonsense (it’s horrible that this is true, but it remains true none the less).
And let’s be clear as well about another point: There was a LOT of “Pretend to be normal” going on. Fertility was critically important, which would inevitably lead some infertile people to make arrangements. Homosexuality was banned, so those who were would have married to hide it by getting married and then, once socially safe, being discreet. Not marrying wasn’t really an option–“leading apes in Hell”–so they made other arrangements.
This matters for two reasons. First, we should be taking an honest look at the past, including the unpleasant bits. Understanding the ideal is important, but we should explore where those ideals break down, because those breakdowns, while not as common, still had significant impacts on society (see kin selection). Second, the myth that humans are by nature heterosexual and monogamous is currently being used to attack and oppress people. And to be clear, I’m defending relationships where everyone is aware of and consenting to the situation–open relationships, poly relationships, marriage between homosexuals, and the like, among consenting adults.
Fidelity / faithfulness / genetic ancestry is not the same as monogamy / polygamy for discussing social structures.
Regardless of who is having it off with who, in a monogamous society the father is only held responsible, legally and socially, for “providing” food, care, etc for the children of one wife at a time. Only the children of that one wife (at a time) have any claim to inherit a share of the household property. That’s very important in a pre-industrial agricultural society.
Where polygamous relationships are recognised (legally), there may or may not be some kind of ranking system, but all the children have a claim to inheritance etc.
Yes there’s plenty of sex outside these monogamous structures happening, but if it doesn’t change who gets to inherit what, I’m fine with leaving them out of a blog post as not in scope.
I think it’s a bit like the way that this blog article is addressing points like “yes, these societies were patriarchal.” That’s not strictly relevant to analyzing the birth, death, and demographics of premodern societies, but it’s important to know because it influences how we contextualize things we learn (or think we already know) about them.
If we are aware that there was a great deal of technically illegal and socially disapproved sexual intercourse, a lot of children whose official parents were not their biological parents, a lot of couples that were officially married but in practice… complicated…, then knowing that helps us to understand the picture we are looking at, rather than compressing that picture down into a cartoon shaped by the propaganda we’re given in our own society. Propaganda that tries to command us to interpret the past in a certain way, whether it really was that way or not, is not the friend of truth.
So, I am not any kind of historian or demographer, but I do occasionally read old census data while doing research for my roleplaying hobby, and I suspect this sort of thing might have been less common in premodernity than we might think.
Swedish census data, for example, shows a fairly clear upward trend in rates of bastardy from 3-ish % in the late 18th century to 9-ish % in the mid-19th century, which suggests that this sort of biological but not social polygamy might be more prevalent in industrial societies than agrarian societies.
Obvious caveats:
– Census data only exists for late modernity, and our host notes that even within a given period, regional differences could be very significant; using 1780 Sweden as a stand-in to represent agrarian societies in general sounds like the sort of generalisation that will get history professors angry at me
– Since census officials don’t have supernatural insight into every private infidelity that goes on in a society, it might be that the higher recorded rates in the 19th century are because of changes in methodologies or social structure, not because of an actual increase in children born out of wedlock
– Did I mention that I am not any kind of historian or demographer, and thus not qualified to assess this data?
That being said, though, I do agree with your overall point that this is something that would be interesting to read about. The blog post lays out how family formation and human reproduction in premodern agrarian societies were *supposed* to work… so did it actually work that way? Since we’re discussing *premodern* societies, does the evidence even exist to make inferences either way and, if so, what sort of evidence? How did societies and cultures deal with the reality that at least occasionally, things wouldn’t have worked the way they were supposed to?
Aside from elite carryings-on (Carolingian illegitimate children were common enough that there was a distinct recognised naming pattern for them), English/Scottish custom was tolerant of sex between affianced persons, on the understanding that any children would be legitimised by marriage later (‘brought under the pall’). Not popular with the church, of course, but sufficiently common that birth records show quite a lot of births outside or just after marriage.
I tend to agree with you here because when I hear people talking about this purported infidelity in agrarian peasant communities my first reaction is have none of you ever lived in a small town? These villages were VERY small towns, and so many people were related to you…they are not swinging places. Let alone in winter when everyone was living on gruel and whatever fauna you could poach. And where were these great assignations to happen? When you worked in the very same place as your spouse?
And if you were a women, your husband, uncle, brother, father would then be socially sanctioned to beat you? Violently. Do you think there is a lot of infidelity in Indian villages where your brothers can kill you for having an honest romance with intent to marry (if they disapprove)? Yeah that attitude isn’t that different back then. One could argue, there might be a lot of rape. But rape serve a totally different social function. It is not related to monogamy or polygamy, or a society’s value for it. Both have proven themselves to be just as hysterical about women’s virginity.
Also a note, in slave holding societies, it is only a “benefit” to a monogamist female head of household if the male head of household fathers slaves if slave status is a)inherited and b) comes by the mother. That’s quite an assumption.
But even in societies where slavery was inherited and from either parent (quite unusual) i.e. the Americas–slave owning women didn’t tend to show pleasure in this habit. One slave owners wife reportedly plucked the eyes out of object of her husband’s affections and put them in his soup. Quite surprisingly Shakespearean of her.
So perhaps we can not talk about slaves and their offspring as if they were just widgets, weighing cost/benefit analysis. Everyone involved were people and as such these were complex and conflict driven human relationships
‘where were these great assignations to happen? When you worked in the very same place as your spouse?’
Well you didn’t work in the same place, entirely. Women more in the home, men more in fields for plowing and such. There’s also stuff like sneaking off to the forest, or sneaking out at night. (Or, by the Decameron, sometimes sneaking _in_ at night.) Women going off to bathe or do laundry or food prep together, and maybe someone makes a detour and shows up late. Men going off to market, or drafted for lord’s labor, or war, leaving ‘their’ women less attended. Women going to the market, and finding someplace in town to sneak off to. Priests having more trust and access because they’re a priest, and abusing that trust.
There are difficulties, but horny people can be creative in overcoming them.
“Regardless of who is having it off with who, in a monogamous society the father is only held responsible, legally and socially, for “providing” food, care, etc for the children of one wife at a time.”
Actually not true. Early modern Europe had and still has widespread practice of paternity/bastardy lawsuits, which DID establish legal responsibility for providing for bastards.
“Only the children of that one wife (at a time) have any claim to inherit a share of the household property. ”
Legal claim. With several buts.
One is the child support. Under a widespread legitimacy/bastardy scheme, a father with both legitimate and bastard children was legally liable for providing food/child support to both while they were minors; but when the children had grown up and the father died, the residue would be inherited by legitimate children only. Note how this makes the poor LESS monogamous than the rich! When child support was set according to standard poverty line but the wealth of the father was widely unequal, a rich father could afford to pay the legal minimum to a number of bastards and this would have minimal direct effect to the bulk of the estate reserved to his legitimate wife and children. Whereas a poor father would have little left over after providing for his children and his old age, and therefore his being legally obliged to pay child support to bastards distracted a lot from the wealth of his wife and her children.
And another is the social/emotional obligations. Married men who had women and children besides the wife might optionally transfer wealth to these other women and children. Including the value of the mother and children themselves – in societies where manumission was common but optional for the master, a common motivation to manumit was to manumit master´s children and sweethearts.
“Yes there’s plenty of sex outside these monogamous structures happening, but if it doesn’t change who gets to inherit what, I’m fine with leaving them out of a blog post as not in scope.”
The question was how the peasant society got reproduced. And since the discussion was around the margins between stable population or population growth, a few % of all adult women like presence or absence of role of “spinster” or a few years of delaying first marriage is in scope.
Then it also IS in scope as to whether there is a social role for bastards, or for freedpeople.
Roman Egypt had both. We have appreciable numbers of people reported as “apator” – apparently accepted in free society. Like a mother who was living with her 5 bastard children aged 14 to 30 – 3 of them married themselves.
And we have slaves and freedpeople. Some slaves with identified slave husbands, most single mothers. Freedpeople were accepted and absorbed in free community of Roman Egypt
If single mothers (free and slave) make up a group relevant to track their contribution to reproducing the society, it is also relevant to ask exactly who actually begat the children of these single mothers – and whether they actually behaved towards the single mothers and bastards as if they cared about their prosperity.
The fathers of bastards might have been young men who were socially debarred from formal marriage due to being regarded as “too young”, or lack of property. Or they might have been men who were legally debarred from marriage by enforced legal celibacy (Roman Army, Roman Church…). Or they might have been men who were legally debarred from marriage because they were already married.
It is relevant to ask who the fathers of the bastards were – how many of which group. Because this is relevant to how the under the table transfers fathers to bastards go.
I think part of the problem with this discussion is the lack of clarity of both terms and time and inheritance. Medieval peasants mores aren’t going to have much too say to 15-17th century elite, let alone the mores of the elites in the long Roman empire or the Greeks which were quite different–or slaves. Slaves by definition are socially dead (Orlando Patterson, read it before you debate that point with me please)–how much and how – shook out for their children(It often didn’t) was very different from one society to another, even one time to another. Slavery in the US was quite different even here, depending on the time.
I also wish as does Prof Devereaux says we spoke as if people believed their religion. This also effected mores and actions.
I object to this fairly strongly. It’s common for non-monogamous relationships to be frowned upon (though not nearly as common as people believe), but IN PRACTICE the most common arrangement appears to be “Pretend to be monogamous but have discreet partners on the side.”
Thanks for bringing that up, I was considering mentioning that myself. Evolutionary biologists, and those who study human behaviour and society using that lens, typically distinguish between ‘social monogamy’ vs. ‘sexual monogamy’. If human beings are naturally anything (well, I think we’re really quite flexible, but if you had to pick a ‘normal’ arrangement) it woul be social monogamy, with a bit of (overt or covert, depending on the society) occasional “extra pair copulations”, as the behavioral biologists say.
And serial monogamy at that — as shown by the repeated mentions of divorce and the emphasis on first marriages.
The Church ran brothels for a while and priesthoods occasionally passed from father to son; they knew their ideals were not being adhered to, and profited by it
Priesthood (and even bishoprics/patriarchs) sometimes ended up being hereditary even in societies that had celibate bishops, the inheritance there would be uncle to nephew instead of father to son. As late as 1920, the Assyrian Church chose a 12 year old boy as its supreme patriarch, with him essentially inheriting the office from his uncle.
https://en.wikipedia.org/wiki/Shimun_XXIII_Eshai
Montenegro being one of the more famous examples of this dynamic! The tribal confederation of Old Montenegro elected their bishop as leader in 1516. From 1697 the Petrović-Njegoš dynasty established hereditary rule over the Prince-Bishopric, passing from uncle to nephew. Finally they secularized in 1851.
I would also like to ask when and where did the Church (apparently, Catholic or Orthodox Church) run brothels. I mean – such practice has never been openly accepted. There have been times and places (e.g. 18th century Venice) where some nunneries were very lax in discipline and apparently it was possible for sisters to prostitute themselves. (This might not be just immorality but a result of poverty. Economics of female cloisters were really problematic, as most avenues of income available for male monasteries were impossible.) Such cases were, though, absolutely against the civil and Canon law, and against the doctrine. Usually, sooner or later, some prudent bishop usually suppressed the deviation. Many such rumours can be dismissed mainly as exaggerations against political or religious adversaries, though. For example, beguine women were often accused of immorality.
Having brothels openly run by the hierarchy was something that, even in the darkest Middle Ages, I don’t know to have happened – and I have studied ecclestiastical history academically. I would be very happy to get my perspective expanded, though.
“I would also like to ask when and where did the Church (apparently, Catholic or Orthodox Church) run brothels. I mean – such practice has never been openly accepted. ”
“Having brothels openly run by the hierarchy was something that, even in the darkest Middle Ages, I don’t know to have happened – and I have studied ecclestiastical history academically. I would be very happy to get my perspective expanded, though.”
A high profile example is Liberty of the Clink.
Otherwise called Manor of Southwark, but there were several manors in Southwark. Basically, a jurisdictionally autonomous suburb of London – one of several such suburbs. It was common for medieval cities to have jurisdictionally autonomous suburbs – commonly controlled by powerful church lords (or ladies).
Liberty of the Clink licenced prostitutes. Licenced in the sense of permitted as the secular authority – the lord of the Liberty of the Clink was not going to arrest prostitutes and lock them up to prevent them from leading immoral life, or beat them up etc., nor do the aforesaid to their customers. The church still “disapproved” of prostitution, and active prostitutes who died were buried in unconsecrated graveyard.
The thing with the Clink is that the secular lord who licenced prostitutes to be free from arrest etc. while alive was the same Bishop of Winchester who refused them Christian burial when they died.
Another thing which highlighted the role of Church in running brothels in Clink is that while the Bishop of Winchester licenced prostitutes in his liberty, his secular neighbours around the London conurbation such as the commune of City of London wouldn´t.
So that´s one high profile example. Any others?
Thank you. I had not known about the case, and it is an interesting one, albeit a bit non special circumstance (case of feudal overlordship and a ecclestiastical authority doing civil administration).
Would that really count as “the Church running a brothel” though? Like, was the Bishop owning the brothel in virtue of his position as a bishop, or in virtue of his role as a political authority?
It’s really unclear if the Bishop is even running a brothel there or just letting one be run (and potentially collecting the fees) which was a much more common arrangement.
While this is a really interesting example, glossing it as the Church *running a brothel* is patently absurd.
In response to the comments minimizing the role of the Bishop in this activity – I think it’s worth noting that the contemporary Londoners really didn’t seem to share that opinion, given that they specifically associated them with his office. After all, the nickname was “Winchester geese”, not “Southwark geese” – and it was definitely not in reference to geography, since both the city of Winchester itself and the Winchester Cathedral in Hampshire happen to be ~100 km away from Southwark.
Moreover, while my searches have not yet found a webpage explicitly detailing every arrangement, this Smithonian article certainly appears to suggest that between the Bishop and the British Crown, it was the latter who was far more opposed to the practice and more eager to shut it down. From that, it seems only a small step to presume that the successive Bishops were actively stymieing the Crown’s efforts with every culturally acceptable means at their disposal.
https://www.smithsonianmag.com/history/london-graveyard-s-become-memorial-citys-seedier-past-180953104/
The last part is certainly the strangest to modern sensibilities. However, I would really like to point out that it is the King who writes a rule which explicitly bars nuns from joining the trade – or from it taking place during the religious holidays. This very much implies that: a) such cases have already happened by then, repeatedly; b) the Bishop(s) knew about them, yet apparently did not move to prohibit them under their own authority.
*This reference appears somewhat sloppy, since syphilis is not believed to have reached Europe until Columbus – and by that point, “the stews” would apparently only exist for an additional century or so. Gonorrhea, though, was a whole different manner – in the same year of 1161, a related law was intended to curb the spread of “the perilous infirmity of burning”, now believed to be gonorrhea.
I believe the bishop of Winchester in England was the landlord of brothers in Southwark London but he didn’t run them
IIRC from my now fairly distant BSc Psych degree, humans are intermediate in testes size between chimpanzees and gorillas. So more sexually monogamous than chimps, but less so than gorillas.
At least from an evolutionary perspective, considering that the environments most of our major physical traits evolved in was really quite distant from your typical agricultural society in some pretty significant ways.
“Not to belabor the point, in the past rape was used as a weapon of war, to the point where NOT raping victims was considered notable.”
If you think rape is only a weapon of war in the past, you’re not paying attention to current affairs.
Rape is absolutely still a weapon of war, from Ukraine to Sudan to Myanmar and elsewhere.
” IN PRACTICE the most common arrangement appears to be “Pretend to be monogamous but have discreet partners on the side.” In our culture this was so common that they had to stop allowing students to test their own blood type; too many cases of “Wait, he’s not my dad?!” ”
I admit I don’t know what culture you’re from, but in mine, and in almost all human cultures, the rate of extra-pair paternity is around 1-2%, both currently and historically https://pmc.ncbi.nlm.nih.gov/articles/PMC7030936/. There are myths about studies finding very high rates (the myth I’ve heard most often is 24% in “a tower block in Liverpool” which says a lot about our stereotypes of the Merseyside poor, and says it really quite loudly) but they are either not representative or just false (though it does seem to be true that poor people living in dense urban areas have higher rates of EPP https://onesearch.cumbria.ac.uk/discovery/fulldisplay?docid=cdi_proquest_miscellaneous_2315524814&context=PC&vid=44UOC_INST:44UOC_VU1&
lang=en&adaptor=Primo%20Central&tab=Everything&query=creator,exact,%20Calafell,%20
Francesc,AND&facet=creator,exact,%20Calafell,%20Francesc&mode=advanced&offset=0).
The first linked paper describes a rare case of 48% EPP in a Namibian pastoralist culture.
Your paper relies on a conclusion drawn from a handful of citations that all rely on limited samples (e.g. 23 families in South Africa in one paper). While these results are interesting, they do not point at an absolute constant of human ecology as you seem to imply:
-the 1-2% reported is a lower bound. It’s not hard to see why bastard offspring of unfaithful women would have a lower chance of passing their genes to the present day
-the samples are really small. Rarely more than a couple dozen families
-there is most likely a cohort bias, a kind of effect (demographic factors such as class) that could depress the rate within the people that accepted to be samples. Well known issue in genetic biobanks
-the paper you link to explicitly points to significant exceptions to the pattern, complicating the picture
Either of these would be enough to solve the discrepancy between the purported rates of infidelity *as reported by the populations themselves* or anthropologists referred to in the paper, instead of a blanket “no, they are all wrong and genetics prove it”.
The reality is likely more complex – people cheat for all kinds of reasons and they are more or less repressed from doing so in myriad ways as well, it’s hard to see how that would aggregate in a stable background rate.
Those are fair points, and the paper you link is really interesting! The Himba (the group with 48% EPP that you mention) are actually famous for their “liberal” sexual morality, so that’s totally consistent with them having a really high EPP. It’s also interesting that this isn’t really ‘cheating’ in the strict sense since the Himba themselves seem to be totally aware what the real rate of EPP is, and to accept it as just the way things are.
I would just note though that 1-2% of EPP is still enough to cause some social disruption, or not, depending on how seriously society takes it. If you have about 300 people in your social circle, then you’re quite probably going to know several people whose father is not the ‘father of record’, and I’d say that’s definitely enough to make it a bad idea for school kids to test their blood type. I definitely know at least one person who got their genetic ancestry tested and was surprised to learn that her ‘father of record’ wasn’t really her father.
You keep emphasizing the “patriarchal” part, which while technically accurate is also misleading.
Possibly because the meaning of that word as commonly used today has changed so much.
A son with a living male ancestor *really doesn’t have much more power than a woman*. Because all the legal power there rests with the older male. Until your older male ancestor dies, you still don’t get much in the way of choices.
For Romans at least I’m pretty sure you couldn’t even run away to join the army easily, since you joined with the others of your village.
As far as how much power and choice the male had, I think the fact that almost always the male was older than the female explains at least some of it. You point out that AAFM correlates positively with relative power/choice for females across societies. Why wouldn’t AAFM for men also correlate? Later age = more power/choice about it.
For that matter, this is true even today. What do we think about people who get married young? “Oh, they don’t know their own minds”. OTOH, get married in your thirties and “Well, they’re adults, even if I think it will be a disaster”
“For Romans at least I’m pretty sure you couldn’t even run away to join the army easily, since you joined with the others of your village.”
No. Recall the Roman dilectus.
Before calling the names from the list, the magistrate reading the dilectus made a call for volunteers.
If a Roman young man stepped forward to volunteer at dilectus and the magistrate had no obvious grounds to refuse, could a father (in front of thousands of his peers and superiors) embarrass himself to speak out to veto? Even if he did, would it be followed, or would the magistrate side with the son? And the father might likely be a senior, not physically invited to dilectus because not an emergency situation. Seems a Roman boy who preferred volunteering to army to obeying his father would have had easy practical opportunity to present his father with fait accompli.
I suspect that having to provide your own armor and gear would be a check on this. If dad (or grandad) can’t or won’t provide your kit you probably aren’t going.
Note that the armor is likely kept in the same house the son lived, thus opening up the question of stealing the armor to serve.
Also, is the son’s declaration at the dilectus legally binding? I mean, the father may not be happy that his son volunteered, but he most certainly does not want him to die because of the lack of proper armor/weapons, so he might reluctantly give him the kit kept at home, just to keep him safe.
My understanding? Depends. If the father is saying “No, I need my son to work my fields.” He’d likely get to keep him (at least if it was voluntary)
there’s a bunch of points where doing the opposite (IE: Letting your sons go to war despite needing them at home) is seen as especially praiseworthy.
I think Bret more or less already addressed this point, didn’t he? Did you miss the part where he emphasized that “patriarchy” specifically means “rule by *fathers*”?
No I didn’t miss that.
Yes he did explain it. Though there are a large number of “remember this is patriarchal” vs. one or maybe two mentions of “by the way this doesn’t mean what you think it means”.
Bret is better than other history professors at explaining where you could get tripped up by word meaning changes. Which however is a low bar. Possibly because those other history professors don’t actually know that the meaning has changed.
It requires a lot of emphasis and repetition, since the Humpty-Dumptys (“a word means what I say it means”) out there, and they will trip you up.
I feel there is a limit to the number of times you should expect to be reminded of what the same word means in the same essay.
I think that to the contrary, most of Dr Deveraux’s audience understands the distinction just fine, and that we are getting bent out of shape here over a hypothetical secondary audience that may as well not even exist.
“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)”
This should probably be reworded, lol.
I assume it’s more than *likely* that a woman’s father is older than her. Maybe something more like “One of whom is likely to be older than her and the other certainly so”.
Sometimes it’s important to write the rules in such a way that they forbid dogs from playing basketball.
Well, with adoption, you could technically end up with a situation where the legal male elder is younger than his adoptive daughter. In thousands of years of history, you never know.
This is a good point, and adult adoption was extremely common in Roman society, though I imagine the vast majority of examples were of men being adopted, not women.
“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”
I can imagine rather more!
The problem here is that the available fertile adult women are capable of producing a bit more children than needed. This the reserve fertility to expand to cover up losses from wars, plagues, famines etc. Now the war is over, the losses have been made up – how do you put the women into reserve rather than pump out surplus children who would starve and starve their families?
Counting the options:
1) Control fertility within marriage by direct birth control.
Then several options that come under “involuntary celibacy”…
2) Delay first marriage – meaning that it is most women who are involuntarily celibate for a few years in their youth.
That´s as far as Bret counted – and stated these were basically the only ones. But keeping counting!
3) Have a group of old maids/nuns who are involuntarily celibate for their whole fertile age
4) Have appreciable numbers of women who become widows before menopause – and then do not remarry. They are thus involuntarily celibate in the later years of their fertile age
5) Have women who are divorced and do not remarry, again before menopause. Same effect, but the ways to influence divorce and widowhood differ.
And then there are things that probably have indirect effect on the fertility inside marriage…
6) In even 30 years ago, elders tended to not be dependent as long as now. Women have a clear menopause around 50. Men don´t, and can now and always have had some children in their 80s (both Ecclestone and Masinissa are examples). But still, do elder men have an appreciable decline of fertility? This combines with 4). If a 40 year old woman´s husband is not a 40 year old man but likely either dead or 60 year old and sickly, even the latter will lower her chances of pregnancy.
7) Polygamy. A husband of several fertile wives would have more children than HE would have by one wife only – but THEY would between them have fewer children than they would have if they had a husband for each.
Between these 7 things, which are the likely contributors?
Don’t forget 8: post-partum birth control, aka abandoning/exposing children you couldn’t (currently) afford to rear.
I really wondered about number 6). At least in Finland, it is considered a common wisdom that about everyone spends their last two years in need of care, and this doesn’t really depend on the age when you die.
Of course, this is partially caused by the fact that nowadays, people die of chronic illnesses. So, regardless whether it is cancer, Parkinson’s or Altzheimer that takes you, you are going to be ill for some considerable time before the death liberates you (often via a fatal inection).
Was the situation particularly different earlier, in a pre-industrial society? The fatal infection came earlier, but even then, a person would be suffering from chronic illnesses before their death: various maladies caused by lots of hard physical labour, parasite infections, etc. A person dying of “old age” would probably be suffering from something serious that decreased their resistance to routine infections – and decreased their capability of being a fully capable member of the household. If I read the Finnish hymnal, I find several old hymns that acknowledge the coming death as a liberator from the immensely bad physical health that the singer is suffering from. One old hymn literally says, “My body withers and the tired head is laying on the pillow. Soon I may sleep from my worries. When I already, every now and then, prayed from the Lord, only of security and release from my pain, He has now answered me and leads me to home.” These are the words of a chronically ill person. The fact that they were selected for hymnal shows quite well that there was a perceived need for a hymn that discusses the feelings of a person dying after a lengthy maladies. (I think that the point was to sing these hymns to the person lying on their deathbed, to help them maintain a suitably salutory state of soul while dying and to give the grieving loved ones something constructive to do.)
I think that the “optimal” point of dying with minimal care was exactly about 50-30 years ago. At that point, the heart diseases were a prominent way of dying, as many lesser maladies were conquered by medicine but cardiological medicine was still heavily developing. (Widespread smoking didn’t help.) It was very normal to live almost healthy and then drop dead of a sudden heart attack. Now, if you have proper routine medical attention and access to a good hospital, you survive your heart disease and then die of something much slower sometime later.
The poem basically refers to being run down due to not being able to recover from work as quickly due to age. Since hard work is the lot of pretty much anyone who isn’t quite rich in that society, this is something that gets you if you survive long enough.
There are a number of ways that disease can kill you fairly quickly even if you’re not old.
E.g. Calvin Coolidge Jr.’s death in 1924 from sepsis developed when he got a blister on his toe due to playing tennis. It burst and the wound got infected. Dead in a week.
We don’t appreciate just how lethal infection can be, due to antibiotics.
Prof. Brad DeLong often cites Nathan Mayer Rothschild, the richest man in the world in 1836, who died in his 50s from an abscess on his butt.
I mean, I do note that not all of this celibacy is going to be involuntary: Not only are there genuine religious reasons, but given the risks of childbearing at least a subset of women were presumably fine with celibacy after having “done thier time in the trenches” so to speak. (you can even see that in some of the 19th/20th century literature)
Oh, definitely. And not just because of the dangers of childbirth.
I always say, had I lived in the Middle Ages, I would very likely have become a nun.
And I am not religious at all.
Having a spouse you do not love, but just regard as business partner does not sound at all appealing to me. No idea how other women do it, but I couldn’t have sex with someone I’m not attracted to.
(Or rather, I would be traumatized by it. Severely so.)
I’d have volunteered for celibacy for sure.
I guess choosing the least bad alternative isn’t really fully voluntary (seeing as I never felt the slightest inclination to become a nun in the present day), but if you check what manuscripts survived, I bet women became nuns much more voluntarily than they became prostitutes. (I think we can safely assume that the number of noblewomen who ran away from an arranged marriage to become prostitutes was zero.)
The main chunk of those forced into celibacy explicitly against their will would have been men – either directly by being pressured into priesthood or a monastery, or indirectly by there not being any women left.
“I bet women became nuns much more voluntarily than they became prostitutes. (I think we can safely assume that the number of noblewomen who ran away from an arranged marriage to become prostitutes was zero.)”
Check a comment about Winchester geese.
There was a need for a rule to forbid nuns from volunteering to be prostitutes.
This came across as a very much a “set-up” post in the series. There was a lot of talk about the facts as we understand them for AAFM, but very little about the whys and what factors make it shift from early to middle to late. It’s about birth control in some way, but exactly how it all works is “to be discussed in later parts” I suppose.
People didn’t generally select their own marriage partners, I get that, but then how were those marriage partners selected? If the patriarch of a peasant household is preparing to marry off his daughter, is he hoping to get something out of it? Or does he have to pay the other party to take her off his hands. If a father is looking to marry off his son, what factors is he looking for in “a good match”?
Since we are discussing peasant marriages, not aristocratic marriages where is a lot of property at stake, aren’t we almost looping back around to… if not love, then at least compatibility as a primary concern? If there’s no money or property to be transferred because no one has any in the first place, what other factor is there for a marriage than “do I think these two will get along with each other”?
I’d say marriage partners were selected on the basis of social capital. A family would prefer their daughter to marry the son of a respected family with good plot of land, over a son of a family with low reputation. Even if the land would not change hands, the connection is still valuable and could be something to fall back on during hard times.
Of course the reputation of the individuals involved also mattered. If the daughter in question had a reputation for sloth or immoral behaviour or some other negative trait, this would make it more difficult to marry into a good family.
> If the patriarch of a peasant household is preparing to marry off his daughter, is he hoping to get something out of it? Or does he have to pay the other party to take her off his hands.
It depends. If the husband has to pay the father, that’s called a brideprice. If the father has to pay the husband, that’s called a dowry.
As I understand it, you get one or the other depending on how the economic value of a woman’s labor compares to the cost of supporting her. And this varies depending on economic conditions and what kinds of work the culture considers women’s work.
There are three possible situations, though they don’t always separate if the new couple is not neolocal.
– Brideprice: husband pays to the father of the bride. Commonly associated with polygyny and hoe agriculture.
– Groomprice: wife’s father pays to the husband’s father. Rare but extant, to the point that people try to get out of having to pay it by using violence instead.
– Dowry (proper): much as the husband’s father contributes the, oh, the whole damn eventual inheritance of the newly formed couple’s household, the wife’s father contributes a chunk of moveable wealth to the new household (rather than to the groom’s father).
Patrilocality in particular and patrilineal inheritance in general can cause the latter two to merge together and to be mistaken for each other even when they are conceptually distinct.
It should be noted that in a lot of western europe (though it varies) the Wife’s dowry was… even if she wasn’t actually able to dispose of it, in some sense legally tied to her (the reasoning often was that it was supposed to support her in widowhood) sometimes it passed directly to her children if she died, and it could get really complicated if there was a divorce/annullment.
It also varies with the availability of potential marriage partners. There is a documented shift from dowry to bride-price in Italy in the 10th/11th centuries thought to relate to an imbalance in the sex ratio.
More generally, In England (and elsewhere?) there was a strong communal prejudice against much older men marrying young women, so much so that there was a traditional form of protest against it. Again, in England the peasant norm was to marry in the 20s, when the parents would be in their late 40s. Most people died in their 50s, and the parents would ‘retire’ to a cottage, leaving the main house to the new household, and do light work (kitchen garden, child-minding). There was also the option of a monastery.
I think your assumption that there is no money or property to be transferred is mistaken. The value of a small subsistence farm may be a tiny fraction of the value of an aristocratic estate, but *to the farmer* that is his livelihood and probably ancestral home. My assumption is that the peasant farmer would care just as much as the aristocrat about finding a suitable heir to inherit *all his property*.
For comparison, I assume that you care about your personal house even though there are millionaires or billionaires with much more expensive houses. If a billionaire sniffed that you couldn’t care much about who inherited your house because the house is practically worthless, you’d be seriously offended.
It really, really depends. It’s one of those things that is very culturally dependant.
That said, I do think it’s important to note that for a peasant *the most important economic value is probably the spouse themselves* in terms of their labour capacity and skills. Someone being good at a particular skill like weaving or spinning, or being known to be a hard worker was a big plus.
Every piece of cloth, from towels and diapers to sails for ships was spun, and on spindles and woven on warp weighted looms from Roman times to the 19th Century in some parts of Northern Europe. Every woman could spin and weave. If a rich man wanted more cloth, he got more women. Great wheels and flyer wheels, and long warp looms with treadle opened sheds were also harder to build than back strap or frame looms. If a man wanted to have clothes, his wife made them from fibers to finished clothes. And men trivalized spinning but if it wasn’t done, they didn’t have any sails for the boat to go trading.
Great post, looking forward to the rest of the series!
I was late to ask something that bugged me about part II, but fortunately the point is referenced here again: the footnote about pre-modern farmers being the majority of humans who ever lived. Just as the world population was much higher in the agricultural era than in pre-agricultural hunting and gathering times, so the world population exploded in the industrial era – from 1 billion in 1800 to 8 billion today. I’m sure this doesn’t tip the final conclusion that “pre-modern farmers represent almost certainly a majority – probably a large majority – of all humans who have ever lived” but I wondered if this has been factored in / the math has been done?
As we’re currently in the vertical phase of an exponential / S-shaped curve, yes there’s a reasonable chance that in the not too distant future (end of 21st C?) the pre-modern farmers will no longer be a majority of all the humans who have ever lived.
Pre-20thC world population estimates, even allowing for the massive uncertainty involved, haven’t changed too much: I’ve got almost identical tables in popular science books from 1990 and 2025.
There’s more discussion and some sources in a previous ACOUP post:
https://acoup.blog/2023/12/22/collections-how-many-people-ancient-demography/
I wonder if there is some kind of a causal link between Roman men marrying somewhat late and them being – to my understanding – more likely than men in many other cultures to spend a lot of their early adult years in military service.
A man marrying and possibly having a child before then getting killed in the next campaign season adds two new mouths for the household to feed – one of whom is not going to be of much help in running the household in the near future – while removing one productive member from it.
“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)”
In the minimally polygynous state of those polygynous marriages just involving two wives, that would still mean that half of all wives are in polygynous marriages. If some of those polygynous marriages involved more than two wives, MOST wives would be in polygynous marriages. Most people would not experience the “average marriage”.
Remember each woman is a separate marriage. So a third of marriages, but a rather smaller proportion of husbands. My understanding is few societies have an outright majority of women in a polygamous marriage.
I always forget if it’s a Australian Aboriginel society or a Polynesian society, but there’s at least one society I’ve read about where they had a relatively ‘egalitarian’ model of polygamy, with a extremely high age of marriage for men. Status was strongly linked to age, so if you reached an advanced age you were more or less guaranteed to have multiple wives, while younger men would have none. (And since a lot of men died before marriagable age, that reinforced the level of polygamy). So in that society, at least, the typical woman would experience a polygamous rather than monogamous marriage.
So you have an existence proof that there’s at least one society that doesn’t follow what OGH was saying.
He’s talking about the *majority* of people who ever lived.
This is like pointing to modern day SF tech communes and saying “See, there’s a society with a high level of polygamy”.
Although even there I’m doubtful of its overall level of “egalitarianism”.
Yeah, people are weird. Given enough time and separation, you’ll get some society doing pretty much anything you can think of, and some things you can’t until you see them. How functional was that society? Did it die out?
How functional was that society? Did it die out?
@Mark Magagna,
I think whether or not a culture “dies out” (or at least is changed fundamentally) depends on so many contingencies that it’s hard to draw normative conclusions from whether or not the culture happens to have survived to the present. Especially when the contingency involved is, like, “the English/Spanish/French/Germans showed up in the late 19th c and imposed their religion and values”.
Like i wouldn’t look at the fall of Byzantium in 1453 and conclude “oh hey, Islam must be the true religion after all” although i’m sure many people did.
Anthropologist Kim Hill has the following to say on male reproductive trajectories ( https://books.openbookpublishers.com/10.11647/obp.0251/ch4.xhtml ):
“… human male reproductive trajectories [is dichotomous] with the end of reproduction in the early fifties for most males who remain monogamously pair-bonded to post-menopausal females (e.g., Hill and Hurtado, 1996, figure 9.9; Kaplan et al., 2010), but significant fertility for a smaller fraction of males from their fifties to their seventies. While few chimpanzee males produce offspring after their mid-thirties and no chimpanzee male has been observed to father an offspring after age forty-three (Emery Thompson and Sabbi, this volume) human males in traditional societies often reproduce after age fifty because of late adult income peak (Kaplan et al., 2000; Koster et al., 2018), and the ability to accumulate resources and political alliances over the lifespan. Gurven (personal communication) has found a wide range of variation in the fraction of expected male fertility achieved after age fifty in small scale societies, ranging from only 1.8 % in the Piro to 3.6% in the !Kung and Tsimane, to 14.3% in Forest Ache, 14.8% in Yanamamo, and then 31.4% in Gambia. Clearly, there are socioecological conditions that allow human male reproduction far later than is ever achieved in apes (even when lifespan is considered).”
For people interested in this topic, the entire chapter is well worth reading imho.
Gurven (personal communication) has found a wide range of variation in the fraction of expected male fertility achieved after age fifty in small scale societies, ranging from only 1.8 % in the Piro to 3.6% in the !Kung and Tsimane, to 14.3% in Forest Ache, 14.8% in Yanamamo, and then 31.4% in Gambia.
That is really interesting, and yes, I should definitely read that chapter!
One factor favoring polygyny in societies like Gambia, in the modern period, is that since about 1900 the population has been growing quite rapidly (and continues to be growing, although TFR is steadily declining there as it is nearly everywhere else). If you assume there’s generally an age difference between spouses with the man being older, there are going to be substantially *more* 20 year old women than 35 year old men.
Tiwi Islanders? Young men married older women, creating ties which (they hoped ) would result in their receiving young brides when older. Very patriarchal – but a lot of infidelity (young women making out with young men)
@PeterT,
I think that might be right!
Though if I remember correctly, this particular culture also allowed older widowed women to marry young men who hadn’t yet accumulated enough status/wealth to afford a young wife.
You may be referring to the Tiwi of Melville and Bathurst islands Australia who had a fascinating marriage and family structure based on the fact that women were the main food providers through foraging and the more women supporting you the better. Basically there was a thriving market in women, starting in infancy and men achieved affluence and status through controlling women as wives and the marriage of daughters.
Basically a father, defined as the husband of the mother, would bestow the first marriage of his infant daughter in a man he regarded as a good ally, often his own age or older. Only rarely would a promising young man get a bride bestowed on him. Most men started their careers by marrying a much older widow who could keep him fed and had all kinds of connections. Young girls were pawns but after her first widowhood as her senior menfolk died off women became players in their own right. A widow with daughters was very desirable mate.
Normally a man wouldn’t get his first young wife till his thirties at the earliest .
“Remember each woman is a separate marriage.”
That’s obviously not the way I was interpreting it, but it would hardly be the first time I have been wrong.
“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.”
I wonder how this fits in with the idea of life-cycle service. On the face of it, western Europeans were finding their children positions, rather than spouses.
I’ve tried to respond many ways to this statement but I think my kindest take is sometimes things seem more illuminating in our head than when they get to other people. This was one of them.
Any opinion that says there wasn’t monogamy because of the ubiquity of wartime rape and then turns around and says we must be honest about this because of the great oppression of Brooklyn polycules…well. I mean well. Yeah, I know that’s the snarky way to re-phrase it.
And again as Prof Devereaux points out, the math is not on your side. In places where polygamy is well known, it is absolutely one sided and distributed by wealth. You pay for a bride in Africa for example and it is no small thing for a man to be married at all there. And even in post-medieval Europe where a sort of unofficial polygamy could occur, mistresses were not cheap.
A bit more charitably, I do think there’s a valid argument for trying to engage with the reality that no, it is not the case that 100% or “almost” 100% of people before the present generation were in perfectly faithful monogamous marriages established for purposes of producing children in the most vanilla missionary manner possible.
It is worth being aware, for instance, that if you talk about who your male-line ancestor was ten or fifteen generations ago, you are accepting a very risky bet.
Or that ‘traditional values’ both in the distant past and in the present often acted as cover for various ghastly and abusive treatment of the people who were expected to be the silent support structure on the bottom layers of society’s pyramid.
Or that people feeling romantic attraction to members of the same sex is not some kind of recent aberration created by television brainwashing, but a thing that a small but noticeable minority of humans have been doing throughout history, and the only real question is what our society chooses to do with that fact.
Many of these facts are relevant in multiple contexts, or subtly change the way we interpret many other things about the past. It is worth at least knowing them and being aware.
a thing that a small but noticeable minority of humans have been doing throughout history
I honestly wonder about the “small but noticeable” qualifier, given the difficulties with gathering statistics on a behavior/identity/however-your-culture-defines-it that is generally discriminated against.
It also depends on what you’re counting – a study that only counts people who exclusively engaged in homosexual behavior for a significant period of time is going to give very different numbers than one that counts anyone who has had at least one homosexual encounter. If you believe the Kinsey Report’s numbers, that’s the difference between ~10% of the men they polled and ~37% of the men they polled – that’s not an insignificant difference!
I wouldn’t trust Kinsey at all. Considering the populations studied were a) prisoners and b) sailors. They were the ones he could get answers from after all.
Extrapolating that to the general population, most of which have never been either, is naïve or deceitful.
Especially as those are groups somewhat notorious for lots of homosexual sex! Don’t forget the male prostitutes in the report also, which is I’m sure a group very reflective of the general male population! But in general my understanding is that is thought to be a couple times too high.
On the other hand, Kinsey was also doing his study at a time when massive numbers of people would probably lie on a survey asking such questions. And follow-up surveys in more recent eras do tend to bear out the idea that the percentage of the population that’s LBGT is, in fact, up around 10% or maybe higher, once you adjust for all the people who are deeply closeted, frequently due to social pressure.
Who cares about whether our culture happens to discriminate against it? Well, a lot of people do, but other than them, who cares?
By which I mean, you already know the list: ancient Greeks, boarding schools, etcetera. Clearly a significant subset of young men are more horny than selective, and will f* whatever is broadly socially acceptable, or for that matter not broadly acceptable but locally customary.
Compare how in this post the professor is quite neutrally talking about the early/early and the early/late marriage patterns, without any reference to the fact that contemporary American culture has constructed itself an absurdity whereby these arrangements would be treated as inherently immoral.
“It is worth being aware, for instance, that if you talk about who your male-line ancestor was ten or fifteen generations ago, you are accepting a very risky bet.”
As noted above: not particularly, no. Assume the historically standard 2% extra pair paternity and go back ten generations, and there is an 81% chance that your great [etc] grandfather is who the records say he is. Fifteen generations and it’s still 74% and at that point I would think that the main reason for being wrong is that the records are wrong, incomplete or missing, not that great [etc] grandma was sleeping around. I’m lucky enough to come from a country where birth records are unusually complete and so I can trace my male-line ancestry back to around 1250 AD, call it 32 generations, and even at that point I can still be more confident than not (52%) that the guy in question is actually my male-line ancestor.
Your reasoning seems to entirely rely on a single specific paper that admits to underreporting and then retroject that underreported state of affairs to an unchanging and pristine past. Not specifically wrong per se but not especially solid either
A single paper that cites a lot of other papers as sources. But happy to change my mind if presented with evidence that EPP is generally much higher!
On the contrary, heterosexuality and homosexuality are BOTH recent inventions. Most cultures regarded lust for the opposite sex and for the same sex to be potentialities existing within every individual, not within a specific kind of person, the heterosexual and the homosexual. Plato wrote a dialogue where Socrates coaches a teenage boy on the correct way to go about romantically and sexually courting one of his fellows at his wrestling school, and another where all of the male characters present remark on how hot and desirable Alcibiades was.
Sexuality is like hunger; it exists as a lack and a frustration first and foremost. We are not born knowing that hunger means we need to eat, we just cry and scream and have reflexes in our brain stem that make us latch on to any soft object (a nipple, a baby bottle, a pacifier, a thumb) inserted into our mouths. We have to develop over time the connection between consuming food and satisfying hunger in order to make us desire food when hungry. The reason British people are more likely to be spicy-averse and Indian people are not has nothing to do with biology and everything to do with environmental, social, and cultural factors.
Similarly, modern human sexuality functions the same way as did the sexuality of men in Ancient Greece; that the normative rate of bisexuality in men was approaching 100% was not some racial idiosyncrasy but instead simply an artifact of the way they socialized male children to direct their sexual desires. Normative heterosexuality in the modern day is of the same origin. These cultural processes aren’t perfect; aberrations happen, as in the case of Socrates’ asexuality, or in modern day homosexuals.
Was this meant to be a reply to another comment? It seems odd standing alone under the main post.
I believe the above was meant in reply to Dinwar’s comment. Nothing else posted on here fits.
https://acoup.blog/2025/08/01/collections-life-work-death-and-the-peasant-part-iiia-family-formation/#comment-85671
It is interesting that the age at which women are married does not greatly affect their life expectancy. Particularly since women that start having children earlier start risking maternal mortality earlier, but also because the risk of maternal mortality usually follows a J-shaped curve with the low point being around 25 years old (this is at least the case for women today, and is not universally true for every country).
Interesting indeed.
You’d expect life expectancy to have been quite a bit lower for women in Ancient Greece than in Ancient Rome, for example.
(European nobility would have lived longer than peasants even while being married earlier for a variety of reasons, of course.)
Perhaps women who were married early, to old men, became widows in their mid-thirties, so avoided the risks of giving birth at the end of their reproductive years?
I’m thinking this could also be a matter of statistics and life expectancy at birth vs life expectancy at marriage.
If 50% of children die by age 10, and 40% die around age 50, then the 10% of women that die giving birth may not shift the needle significantly, even if there’s a 10 year difference.
I.e. 10% of 10 years is only 1 year. If the 10% of women dying of birth complications now live 10 years longer, that only shifts life expectancy at birth by 1 year. This effect is further reduced if the 10% number is of women that survived childhood (to 5% of the total and half of a year extra on average).
Divorce in Rome very much depended on the form of marriage. It was extremely difficult and rare (though not impossible) with a conferratio marriage than with other forms (at least one of which compelled the woman to reside outside the household for three days a year playing canasta with friends or relatives).
From what I remember of my Roman history, conferratio was rare because it was expensive and a general pain to go through with.
So most didn’t do it.
This is another case where we don’t really have any “lived experience”. There’s pretty much a single form of marriage and the legal system recognizes that. Divorce is through the legal system; whether you consider yourself divorced religiously the judge doesn’t rule on.
Heck 3 states in the US have a Covenant Marriage status which is harder to get out of (though still easier to exit than regular marriage was until 50 or so years ago). But is also less than 1% of marriages in those states.
@Hastings,
I was quite interested, from a sympathetic viewpoint, in ‘covenant marriges’ when they first arrived on the scene, and I remember there being a lot of enthusiasm for them in Christian circles. It’s interesting to see where the phenomenon has gone (i.e. nowhere, really- just 1% of marriages in three states) since then. It looks like a lot of people strongly believe that tougher divorce laws should exist, but don’t really want them for themselves. (Which is fine, I think a lot of us have views like that to some extent, including me).
I would note that Nordic countries do, in social and anthropological sense, have two official forms of marriage: there is the traditional official marriage and cohabitation, which is also legally recognised. Cohabitors are, in sense of taxes and social security, equal to married persons in most cases, there are limited legal provisions for “divorce” and inheritance also for cohabitors. Even the paternity of children, while not automatically conferred, is recognised in a very simple proceeding. (Essentially, you start cohabitation by moving together with someone and reporting the same address. After that, you are considered a couple – and if you wish to contest it with the authorities, you actually need to do some paperwork.)
So, it is somewhat easy for me to understand how Roman law could integrate several legal types of marriage.
At that point, are there two separate marriage forms? Or does it come down to how much fuss the couple makes about the marriage?
Is a religious ceremony treated legally differently, or is it the difference between “Bridezilla” marriages vs. running off to Las Vegas?
“Conferratio” was very different from the other forms.
Here, cohabitation is not marriage, legally, but it is marriage socially. It is quite bad form to consider cohabitation somehow lesser or problematic.
The definition of cohabitation is different in different laws, but for taxes and social purposes, it happens automatically: for social security purposes, you are cohabiting with an adult with whom you share a dwelling. (If you want to argue otherwise, you need to show the non-marriage-likeness of the arrangement.) For tax purposes, you are a cohabitor if you live together and have a common child. For survivor’s pension purposes, you need to have five years of cohabitation and a common child. For the purposes of “cohabitation divorce”, the defining issues are five years of cohabitation or a common child. For paternity purposes, the cohabitation, regardless of time, allows the father to use a very simple procedure for confessing paternity. This is because the legislation around cohabitation has been built somewhat piecemeal. Naturally, there is no ceremony, except simply reporting a move into the same address with another person to the national population database.
Marriage, on the other hand, is a legal marriage, regardless of whether the ceremony is civil or religious, and it has fixed effects, regardless of time or common children. The most important differences to cohabitation are the more equitable law on the division of assets in divorce and much better protection of the widow after spouse’s intestate death. And naturally, the husband is by default the father of all his wife’s children. (Mutatis mutandis for same-sex marriages.)
The custom is that nowadays, marriage without at least a year’s cohabitation is considered somewhat hasty and irresponsible. On the other hand, marrying only after a long cohabitation tends to signal problems in the relationship and increases the risk of subsequent divorce considerably, according to statistics.
Under Australian law a marriage de facto (sharing premises, property, costs, children) is treated very similarly to formal marriage. One can even by formally married and at the same time in a de facto relationship.
Beyond “cohabitation” in Europe, as noted, there are some other examples of countries today with different “types” of marriage- often in postcolonial countries that had to incorporate different legal and cultural traditions regarding marriage.
In India for example, Muslims have a different ‘personal law’ system than other groups regarding marriage (although there have been recent legal changes that make it….somewhat less different). I know less about South Africa, but it appears they have distinct “civil marriage” and “customary marriage” institutions there. Customary marrige allows polygyny but not gay marriages, with civil marriage it’s the reverse.
It should be made clear what “monogamous” means in this context: only a single legal sexual partner, or only a single wife (as many societies insisting on monogamy did not consider concubines “wives”).
The Romans clearly had non-legal sexual partners, and could have sec with slaves.
” (the radical breakpoint in standards of living is the industrial revolution in the 19th century, not the arrival of ‘modernity’ in the 16th). ”
the concept of extending ‘modernity’ to the 16th century always feels like a polite courtesy to the self-sentiments of the ‘Early Moderns’. “Yes, yes Divine Right of Kings theorists you’re modern too, don’t worry your little head” [snickers in the direction of a boilermaker]
Western Europe was already a very odd place long before the Industrial Revolution. It may have pulled ahead of even the wealthier parts of China and India by the 1500s. Some of the more prosperous areas like the Netherlands and England had their last peacetime famine in the 1600s. As I mentioned in an earlier post, by the 1500s at the latest NW Europe basically had among the common folk the ‘modern’ marriage patterns that would last until the Sexual Revolution. Fairly rapid scientific and technological development and increase in financial sophistication was happening well before the Industrial Revolution.
I remember Adam Smith’s comparison of British North America, Britain itself, and China, in the 1770s. The colonies were the poorest of the three in terms of per-capita wealth, but had the highest wages. China was the wealthiest but had wages so low that people would swim in the harbor to scavenge rotten food that European sailors had thrown overboard.
I’d be interested to see if there’s been any analyses of wealth distribution/inequality in these societies. There’s some interesting economic theories floating around about present-day wage stagnation vs asset price inflation (and thus living standards)as it relates to wealth inequality.
There is analysis of income inequality, collected in Branko Milanovic’s work. He blogs these charts and you can look there, or you can get Global Inequality.
The short version is that Early Modern inequality was high, close to the inequality frontier, the highest level sustainable for a given GDP per capita if one assumes worker wages are all near subsistence.
Except that, rather typically, Smith failed to note the poor of Edinburgh scavenging anything saleable from the sewers …
He failed to note that because Edinburgh famously didn’t have any sewers as we would recognise them today.
I think this is one of the cases where military history dictates periodization and everyone else just deals with it. Pike&shot, artillery, trace italienne, very large (theoretically-) standing armies, sudden existence of competent literate centralized bureaucracy. But see also, oh, Gutenberg and reformation and exploration (the Portuguese built their Indian Ocean empire how fast?!?).
It’s a little in the weeds, but the Divine Right of Kings thing was a clear forward step in having a philosophical justification for the social structure, driven by people feeling the need to have one. I mean, the “normal” situation is that you live in a polis surrounded by other poleis, or on a manor surrounded by other manors, and the far-away people who organize their society differently are barbarians you don’t need to care about. It is only after you manage to establish long-distance trade, campaigning, governance, etc. that you have a problem with religious syncretism and its analogs, namely: how do you explain to your children that your customs are correct and to be followed, in the face of not vague rumors that inaccessible barbarians do things differently, but easily observable evidence that your neighbors do and seem better off for it? Now you need to rationally justify the existing social order. A few generations later, this will allow people to invert that relationship, and have rational discussions (“the marketplace of ideas”) drive changes to the social order (joint-stock companies, English Civil War, etc.).
Part of the problem is that a lot of things change around 1500 (much of it as a result of europeans going abroad and killing a whole lot of people, taking their stuff and bringing it all over the world.
I’m confused about what you mean about 90% marriage rates. Given that the median age at death was 10, it’s clear that you don’t mean 90% of people born into the society were eventually married (or that 90% of people alive at one time were married at that moment). Would 90% be the number of people who eventually got married, divided by the number of people who lived long enough to reach the legal minimum age for marriage?
Yes I think he means “90% of the people who survived to the age where marriage was legally possible”
“with scholars actually using data often correcting assumptions of pre-modern ‘white’ populations AAFM younger, while outside of European populations it seems like the ‘early’ pattern is assumed even when data doesn’t exist.”
I’m having a hard time parsing this sentence, particularly the first half. Do you mean that the scholars who actually use data tend to show that the AAFM is younger than assumed, or higher?
Seems clear to me. Scholars using data say that the AAFM for pre-modern white populations is younger than commonly assumed.
Hmm, how should we then treat the case of Allia Potestas? Now, it’s true she apparently wasn’t married to either one of the two men she lived with, but then, the quoted paragraph explicitly includes concubines in its definition of polygamy. I also don’t think a freedwoman “with hard hands” from constantly working with wool would count as a “high elite” – although the existence of a marble gravestone in the first place does the suggest that the collective wealth of the entire “family unit” was somewhat above the median.
https://en.wikipedia.org/wiki/Allia_Potestas
Seems like fairly normal widow behavior.
Erm…while the euphemistically called “widows’ fire” might be well-attested to (the link below, with its “existentialist phenomenological approach”, is quite something),
living with two younger men at the same time can hardly be remotely “normal” for a widow for purely numerical reasons alone!
https://www.researchgate.net/publication/380264250_Young_widowhood_A_qualitative_study_of_sexuality_after_partner_loss
Besides, there does not appear to be any explicit evidence suggesting she had ever been formally married before this arrangement began, let alone widowed. While one can imagine that a lover of a widow paying for her tombstone might want to leave out that detail entirely, I am not sure the Roman culture would have allowed him to bend the truth to such an extent! From my understanding of our host’s writings, this would have both been seen to undermine the communal ties (in a society much more collectivist than (most) industrialized ones) and could well have risked offending the di manes which the departed husband belonged to – not something which people believing in their own religion would have wanted to trifle with!
I think it’s worth noting that Tales of Times Forgotten, recommended on here the other week, disagrees with this somewhat.
https://talesoftimesforgotten.com/2023/02/28/at-what-age-did-ancient-greek-women-typically-marry/
Followed by.
Granted, the above would simply shift things from “AAFM…around 15” to “AAFM…around 16”. I suspect the more important suggestion n this context is that all these ranges are again restricted to only the Ancient Greek elites. If Pomeroy’s work truly was referenced to Oikonomikos alone in this regard, then it would lack evidence for the lower classes. Thus, the aforementioned caveat
Might well be equally applicable to Ancient Greek poleis as well.
I would also like to linger on this point a bit more.
So, how much evidence do we actually have for poleis not named Athens or Sparta regarding this manner? After all, not many other “Greek law codes” have survived to the present day, have they? I.e. the Wikipedia link provided in the article for epikleros is a lot more cautious than The Pedant’s phrasing (“in most if not all other Greek poleis”), noting that most of our remaining evidence comes from the Athenians’ writings about Sparta as well as…Gortyn (Crete) and Rhegium (Calabria) – far from the first places that come to mind when one considers poleis! Clearly, very little legal material survives from elsewhere – so apparently, we are just assuming their systems to be equivalent to the Athenian one unless stated otherwise?
I’ll admit that in this case, “evidence from absence” would be congruent with Occam’s Razor, following the assumption that if a polei had a comparatively progressive approach, then our Athenian sources would have mentioned it (the way Aristotle apparently mentioned the opposite for Troizen). Nevertheless, this blog has already frequently mentioned the tendency of ancient sources to disregard commoners’ lives. If so, wouldn’t a considerable variation between polei in this manner be plausible in practice? Even some of the elite evidence might complicate this narrative – i.e. is everything we know about the life of Sappho really congruent with the idea Ancient Mytilene’s approach to women was just as “instrumental” as that of the Athens?
I am particularly curious about (pre-Roman) Corinth and Thebes here, due to their position as the next most powerful poleis after Athens and Sparta. Corinth is also atypical in another way, owing its power to sitting atop the isthmus and hosting a port on either side of it. Hence, a massive number of sailors passed through the city – and predictably, the demand for prostitution was extremely high. Now, as to be expected of that period, a large fraction of that demand was met by slaves, current or former (that is, those who were freed after they were bought and groomed into the profession as children.) Yet, it seems worth noting that once freed, they had the right to perpetuate the cycle on their own – while someone like Nikarete would nowadays probably be best compared to Ghislaine Maxwell, it certainly appears that she did hold property, alongside possessing other rights her Athenian contemporaries may not have had?
https://en.wikipedia.org/wiki/Nikarete_of_Corinth
In theory, Corinth owing a larger-than-usual fraction of its economic activity to these women could be expected to have had some effect on the rest of society. Now, Athens itself also had significant prostitution – but the manner in which it was instituted in the society was more than a little different. Let’s just say that Solon, the legendary city-founder, bulk-purchasing slaves to then establish artificially cheap brothels as a public good sounds more than a little reminiscent to certain ideas emanating from the mano/incelsphere in recent years – and it seems at least plausible to argue that certain aspects of Athenian law and social structure were “downstream” of the attitudes enabled by that custom, rather than being ubiquitous across Ancient Greece.
bulk-purchasing slaves to then establish artificially cheap brothels as a public good sounds more than a little reminiscent to certain ideas emanating from the mano/incelsphere in recent years
The incels themselves are not in favor of sex work, since that’s a vaguely “liberal coded” issue in America rn and the incels/manosphere are very strongly “conservative/republican” coded. To the extent I’ve heard proposals like that it has not been from the incelsphere or anything like them, it’s from people on the other side of the aisle wondering “would this make these people less insane and antisocial”.
Also, when people in modern industrialized societies discuss the legal or social status of sex work, we’re generally discussing free (adult) women exerting agency and making choices, so I think it’s more than a little different than sex work in a slave society (though you do also correctly note that there’s a certain gray area with freed *former* slaves).
“The incels themselves are not in favor of sex work”
If they refuse to pay for something they could buy, are they really going without it involuntarily?
Smart remarks aside, perhaps the thing they really miss is an emotional connection that can’t be paid for, because the very act of requiring payment proves that it isn’t there.
@ad9,
I mean, you’d have to ask them.
But to be clear, when I (and i think most people these days) talk about “incels”, it doesn’t literally mean “everyone who wishes they could be having sex/romance/relationship and isn’t”, which I bet includes a large porton of the population, it refers to people who actually self identify that way, who are a…..very distinct and specific ideological/political movement.
One other thing worth noting is not associating romance with marriage is not necessarily the same thing as not having a marriage with romantic love. My rough impression of royal marriages is at least a quarter of marriages seem to have ended up what we would consider a romantic love successful marriage. The balance is roughly evenly split between couples that broadly get along fine and couples that don’t. The lower sort would have the advantages of compatibility being more likely to be considered and avoiding culture shock issues. Seriously a pretty large share of queens and princesses in the Early Modern Period seem to have been kind of miserable from culture shock in their country of marriage.
By comparison in the US about 25% of marriages are rocky at any given point and about 40% of first marriages end in divorce. Of course even if success rates are not necessarily broadly wildly different, it still is a very different dynamic. Western culture expects the spouse to be the primary person meeting the other’s social needs and is very feelings focused. While traditional societies it is more of a partnership dynamic (albeit often a fairly unbalanced one), more pragmatic, and perfectly normal if same-gender relatives/friends/colleagues are more important for meeting social needs.
Like there are a lot of arranged marriages where it is pretty clear the parties loved each other and not just in the sense people are attached to their siblings: Steven and Matilda; Edward I and both his wives, Henry VII and Elizabeth, heck Henry VIII and Catherine of Aragon for much of their marriage (not that this prevented him having affairs).
To be clear Stephen’s wife Matilda, the one who led an army after he was captured, not his cousin/rival Matilda. Medieval Europeans were not big on name diversity…
Matilda vs Matilda?
Please tell me they met in battle.
I’m imagining the orchestra strikes up _The Blue Danube_, but then again I’m an antipodean.
Yep they did! After Stephen was captured at the Battle of Lincoln Empress Matilda took London and then proceeded to alienate the Londoners. Meanwhile Stephen’s wife Queen Matilda of Boulogne assembled an army of levies from her county, mercenaries, and defectors. The Empress was forced to abandon London and her army was ambushed and nearly annihilated by the Queen’s forces in the Rout of Winchester. The Empress escaped, but her army commander/bastard brother Robert was captured. Robert’s wife had custody of Stephen, so the Queen negotiated a prisoner swap with her. The Anarchy would last for another 12 years…
If only Robert were married to another Matilda the story would be perfect…
The Anarchy is a surprisingly underused period of english history. I think the only fiction I’ve seen about is the Cadfael series, which is set during the period.
“The Anarchy is a surprisingly underused period of english history.”
I dunno about actual historical fiction, but GRRM’s “the Dance of Dragons”, Targaryen civil war, is basically the Anarchy with dragons, AIUI. Or at least very obviously inspired.
Arilou, I always rather liked George Shipway’s 1969 novel Knight in Anarchy.
“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.”
Question, also belatedly applying to the funerary inscriptions regarding children: why would we assume that they are a representative (as opposed to thoroughly biased) sample in this dimension? By which I mean, even if it were somehow the case that only a minority of men loved their wives/toddlers, that minority (wealth permitting) would leave touching funerary inscriptions, while the majority who did not largely would not, thus the pool of funerary inscriptions would vastly overrepresent the former minority. As a result, reading the surviving inscriptions, we would not be able to differentiate this society from one in which such affection was a large majority, at least by the composition of the inscriptions, as opposed to their quantity (“demography and wealth distribution suggests there should be rather more inscriptions than we see”).
Do we have something of the latter class of evidence, or is there a surviving representative sample of a smaller part of society, e.g. was it perhaps customary for rich Romans to give inscriptions to every family member, even those who weren’t much liked by the people commissioning the inscription?
Strictly speaking the data is rarely there for non-elites. But documentation goes up a lot in the Medieval and Early Modern Period. We have a decent idea which English royal marriages were loving, fine, and terrible. By the Early Modern Period there is a meaningful number of people we have diaries or extensive collections of letters (people in the 1700s and early 1800s *really*, *really* liked writing letters).
This is an excellent and highly informative post. I do, however, have three comments/corrections regarding Athenian women.
First, while it is certainly true that examples of ancient Greek girls being married off as young as fourteen are attested and Sarah Pomeroy does claim that around fourteen or fifteen was the typical age at first marriage for Athenian girls, Pomeroy’s case relies heavily on Xenophon’s Oikonomikos 7.4–5, in which Iskhomakhos says that he married his wife when she was “not yet fifteen.” But the reason why Iskhomakhos states his wife’s age when he married her in that passage is specifically to emphasize how young and inexperienced she was, which suggests that she is somewhat younger than the typical age for an Athenian bride.
I have argued in this blog post I wrote back in 2023 that a more holistic view of the ancient Greek primary evidence (taking into account statements about the ideal age of first marriage for women from writers like Hesiodos and Aristotle) suggests that the typical age at first marriage for Greek women was between roughly fifteen and eighteen, which is certainly still “early,” but not as early as scholars since Pomeroy have tended to assume. This puts the AAFM for ancient Greek women closer to the AAFM for women in Greco-Roman Egypt.
Second, although it is true that the bride’s consent does not seem to have been necessary for marriage in Athens, Athenian women could, in fact, divorce their husbands. The catch was that, in order to divorce her husband, a woman needed the permission of her closest living male relative (usually her father or, if he was dead, her brother) and she was required to appear in person before the arkhon to file for divorce.
The best-known account of an Athenian woman attempting to divorce her husband occurs in Ploutarkhos’s Life of Alkibiades 8.3–5. According to Ploutarkhos (who, of course, was writing centuries later, but had access to sources from the time that have since been lost), Hipparete (Alkibiades’s wife) attempted to divorce her husband because she did not like that he frequently consorted with hetairai. With her brother’s support, Hipparete went to file for divorce, but Alkibiades seized her when she was on her way and forcibly carried her back home to stop her from divorcing him. Ploutarkhos approves of this abduction, saying that it seems to him that the reason why the law required a woman to file for divorce in person was precisely so that her husband would have a chance to capture her and stop her. Whether the specific story Ploutarkhos tells is historical or apocryphal, his assumption is that woman-initiated divorce did exist in Classical Athens, although the conditions may have made it hard for women to obtain it in practice. There may be other evidence of woman-initiated divorce in Classical Athens, but this is the only example I know off the top of my head.
Third, although it is true that Athenian women could not own real property or inherit property in their own names, Demosthenes’s speech Against Spoudias (written c. 340 BCE) suggests that, in practice, women often controlled or managed property that was nominally owned by their husbands or male relatives. Notably, Demosthenes describes Polyeuktos’s widow as having loaned a huge sum of eighteen minas to her son-in-law Spoudias after her husband’s death. He also describes Spoudias’s wife (Polyeuktos’s daughter) as having attended the reading of her father’s will while Spoudias himself was absent, even though Spoudias himself was nominally to inherit a share of Polyeuktos’s property, because he trusted his wife to handle to matter for him. Meanwhile, in Xenophon’s Oikonomikos, Iskhomakhos entrusts the entire management of household resources and accounting of goods brought in and out of the household to his young wife.
wow. Bret got out-pedantized.
This is just one example of the problem of source interpretation in antiquity when the material is so sparse: if we just have a handful of attestations of something (eg, female age at marriage), should we assume this represents the norm or the exception?
Both interpretations have reasonable arguments in their favour, and yet they’re opposite. Which means that we can’t really draw either conclusion with any degree of certainty.
I’m not sure that the ability of Athenian women to divorce with the agreement of a male blood relative is evidence that their consent was necessary for the marriage. If anything, it reinforces the notion that their family’s consent is what’s important, not their own personal opinion on the matter.
I think you are misunderstanding what I have argued above. I explicitly say in my comment that a woman’s consent to marriage does not seem to have been important in the Greek world. Bret is right about this and I am not disagreeing with him. The point I make in my comment above is simply that, contrary to Bret’s claim in the article above that Athenian women could not divorce their husbands, there is, in fact, evidence that Athenian women could divorce their husbands if they had the support of a male relative. Thus, legally speaking, woman-initiated divorce did exist in Classical Athens, but the conditions for it would have made it fairly difficult for a woman to divorce her husband in practice.
Fair, I mistakenly conflated two of your points together.
It’s funny how I linked that exact post a little earlier, and now the author personally shows up with even more detail! Either way, it’s great to see you here! Is there any chance I could see you attempt to answer the questions I left in the comment linked below? (Perhaps you would prefer to do this in a full post of your own, rather than here. Up to you, I suppose.)
https://acoup.blog/2025/08/01/collections-life-work-death-and-the-peasant-part-iiia-family-formation/#comment-85726
Oh yeah, one more thing: The fact that all this data is about *first* marriages can conceal quite a bit. I did a course on historical demography at uni and one of the things they pointed out (apart from regular stuff like “Church records record baptisms not births” was that for at least parts of northern Sweden there seemed to have been a pattern of… Well, basically young men would marry older propertied widows, they’d then outlive them, at this point they’d be “older”, they’d then marry young women, who who would in turn outlive their old husbands becoming the next generation of old widows that married younger men…
Now this seems to have been constrained certain areas, but it’s a point of note: First Marriage ages does not neccessarily say much about the age gaps within marriages absent other data!
@Arilou,
That sounds really fascinating! Do you have a citation for the northern Sweden cultural practice, so I can learn more?
And when the propertied widows died, was their property usually inherited by their children, or by their surviving husband, or divided among both?
Alas, I don’t. They’d be in swedish anyhow, but this was something I read at uni like 10 years ago.
Northern Sweden is a bit weird in general: Very sparsely populated, and withs ome differences in terms of patterns of agriculture (and famously, basically no nobles, thanks to the aforementioned sparse population)
Ah OK, got it. How were the patterns of agriculture different?
More transhumance than in southern sweden , different crops to some extent (less wheat, more rye and barley)
Fascinating pattern. Everybody gets both money and a younger sex partner just at different times. Everybody happy.
Just to pop off about those ancient Spartans, who were puzzling to other Greek states because their women tended to marry late (c. age 17) under the laws of “Lycurgus” while men tended to marry in their 20s (but were not free to live in their own household until age 30, from what we are told); and most astounding to other Greeks, Spartan women could own land and become independently wealthy, which was simply baffling to contemporaries but makes them seem almost enlightened to us moderns. Much closer to the Romans in that way.
The Spartans also had problems with a gerontocracy, just like us moderns!
What were the kinship systems in Latin and Ancient Greek? Modern European languages use Eskimo kinship, in which there are coverterms “uncle,” “aunt,” “grandfather,” “grandmother,” and “cousin.” In contrast, modern Arabic, Hindustani, Chinese, and Japanese use Sudanese kindship, which breaks these up – there’s no single word for “uncle” or for “aunt,” but one must specify maternal or paternal, and sometimes whether the uncle/aunt is related by blood or only by marriage. Usually Sudanese kinship correlates with patrilocality and Eskimo kinship with neolocality (Eskimo kinship uses coverterms for extended family but distinct descriptive terms for nuclear family). Checking on Wiktionary gives me, for both Latin and Ancient Greek, distinct terms for a paternal and a maternal uncle but also a coverterm for both from which the modern words descend, and I don’t know what was more in use in practice.
That varies even by language in europe: English just uses “uncle” and “aunt” while swedish specificies (“”Father’s brother” vs. “Mother’s Brother”) and same thing with paternal and maternal grandparents.
Yeah, I should have made a parenthetical that Swedish reworked its kinship system to be affixal and this made it more descriptive (“farfar,” “moster”). But that’s unusual in a bunch of ways, such as the ability to say farsyskon and morsyskon (ungendered terms for uncles and aunts are more or less nonexistent, even in languages like Malay, which have ungendered words for siblings).
Right, a lot of languages have more complex terms for kinship than English does. In my own ethnic group the term for “my uncle” is going to depend both on whether you’re an uncle on my paternal or maternal side, as well as whether you’re an older or younger sibling of my parent.
In Malagasy, which like many Austronesian languages doesn’t natively have much concept of “gender”, there are no distinct terms for ‘brother’ or ‘sister’, but there are separate terms for ‘older sibling’ or ‘younger sibling’. So if you want to say ‘my brother’, you have to automatically express whether he’s an older or younger sibling of yours, and then add ‘male’ or ‘female’ to specify the gender.
I think at this point “beg the question” has evolved to two usages, one with an object and one without.
“X begs the question.” -> traditional usage: X commits the logical fallacy of assuming the conclusion
“X begs the question, Y” -> new usage: X raises the obvious question Y
I don’t think there’s any reason we need to continue rear-guard actions against a battle that has already been lost. Both usages are perfectly acceptable and mean different things.
I agree with you, but also Dr. Devereaux is right to explain what he means, since most people aren’t familiar with the meaning he’s using.
Oh god no. Neither of these cases are the colloquial use of ‘to beg.’ If the path of transformation was simply to interpolate a common usage of ‘beg’ in place of an obscure one, “begging the question” would mean (in use) ‘demanding something from a question.’
What has actually happened in the language is much more interesting: confronted with a use of “beg” that does not scan in their personal dialects, people have conflated the use in the phrase with a different but similar sounding archaic word. In everyday use people are not misusing “begs the question,” they are *mispronouncing* (and consequently misspelling) “begats the question.”
English, with its extremely large and forgiving vocabulary, is notably prone to this particular process of malapropism, that is the version of malapropism which produces a term that is both comprehensible and useful to the audience to a degree that it is then widely adopted in speech.
I find it unlikely that an ordinary person would hear an ordinary present-day word (“beg”) and decide it’s actually an archaic word they’re barely aware of (“beget”). I figure the new meaning of “begging the question” comes from people thinking it means “begging *for* the question”.
Yep, that seems far more likely. (source: That’s what I thought it meant when I heard it the first time, as a someone who isn’t a native english speaker)
I’m not even sure it’s that. I think it’s simpler.
1. Expert bod uses the term ‘begs the question’ to mean ‘this presupposes a question that it is wrong to presuppose’, in a way that would be obvious to someone who understands the topic at hand.
2. Layman bod who isn’t as steeped in the topic hears this and understands the ‘this presupposes a question’ part but misses the implicit subtext of ‘…and it is wrong to presuppose that question’.
3. Layman uses the term in its simpler usage with other laymen, and the trajectories of the two phrases depart.
“Mawwiage, mawwiage is what brings us together, today.
No, I will not apologize for this joke.”
Nor should you.
Thanks for this intersting blog post.
However, I stumbled upon your notion that polygamy was banned in medieval Europe. I am no histroian, but from what I’ve learned in law school, there were three legally permissible forms of marriage in early medieval (pre-Christian or Germanic) marriage law: Muntehe, Friedelehe and Kebsehe (concubinage), of which only Muntehe was monogamous, see also https://en.wikipedia.org/wiki/Friedelehe
One further comment: although the demographic observations are undoubtedly important and interesting, I would have liked to read a little more about the cultural significance of weddings and the associated ceremonies. After all, paintings such as Pieter Bruegel’s famous Peasant Wedding clearly show that weddings have always been more than just legal matters and economic necessities; they also brought variety and joy to rural life.
> Muntehe, Friedelehe and Kebsehe (concubinage),
Your wikipedia page is pretty skeptical about whether Friedelehe really existed.
The ideas about what the old Germanic law (or, for that matter, archaic Roman law) really included tell usually at least as much about their presenters than about their subject. The evidence about pre-Christian or semi-Christian Germanic law is in law codes written by Christian monks – or in epic poems similarly written down by Christians. Then, in the 19th century, these were seriously studied by people who sincerely believed that these books contained the unsullied core of the national spirit.
For my part, I would say that the most significant mistake is the assumption that there was a clear, nice systematic law with well-defined concepts. In reality, there was a judge who was a local big man who knew, or was supposed to know, the law by rote, and the people assembled who would need agree at least so much they did not to riot when they heard the judgement. (The traditional image is the elected king or tribal duke sitting under an ancient oaktree delivering justice to the popular assembly of warriors armed to their teeth, demontrating their approval by hitting their shields with swords. You may note that it fits nicely with the self-conception of the German nationalism of the 19th century.)
Obviously, people claimed – and surely, sincerely believed, that they were applying the fathers’ customs. (It is easy for a man to believe things he needs to believe.) However, oral law is somewhat fluid, concepts vary according to the judge and the situation, and one concept may be suddenly two different things or vice versa, depending on each person views it – and needs to view it in a particular situation. It is not like the lawyers of the past were not competent enough to twist the law if it suited their purposes.
I can’t say too much about old Germanic law, but I know that in early mediaeval Ireland the law was very much a serious profession, and, given that 8th Century law texts set out things like the correct procedure for seizing the property of someone in debt to you, and the fines for not following the correct procedures, it seems likely that following those procedures was well worth the 25% fee your professionally qualified lawyer would charge for assistance in debt recovery.
“…we are talking about peasants. Remember peasants? This is a post about peasants.”
I’ve been writing this post for twenty-five minutes. I’m not proud. Or tired.
Nice to meet a fellow member of the Group W Bench
Always glad when someone gets the reference.
Really odd for you to claim the “RETVRN” types don’t understand patriarchy as the typical misunderstanding (rule by men) is way more common in the political left, especially amongst feminists.The right, due to being far more Christian put more emphasis on fatherhood and understand the traditional practice more.
Without concrete examples it’s hard to really argue about, for or against whatever you report “the left” as saying. For what it’s worth, my experience of discussions about patriarchy doesn’t point at anyone from any side caring very much about ancient concepts like Roman potestas, or the fact that many grown men didn’t have full rights because the family patriarch was still alive. Modern political discussions are really about controlling the narrative and asserting power (over self, or others), with “the way things were” only thrown a fig leaf in order to maximize emotional appeal with little care for historical accuracy.
The understanding of patriarchy as “rule by men” is common on both sides of the political spectrum, because our post-industrial family structures strongly push us towards equating the two. Very few of us live in households that encompass more than two generations, and more than two individuals per generation, so in most post-industrial Western homes, the patriarch is very likely to be the only adult man of a household. Something which was definitely *not* the case in pre-industrial societies.
So it is not a case of misunderstanding, but a case of applying a perfectly accurate term to the same phenomenon existing in two wildly different societal configurations.
Why is the AAFM for men usually higher than for women? Is it something along the lines that a men has to be more “established” to marry vs the women who did not (since the bride would join the groom’s household)? Or is it more something along the lines that an out of marriage pregnancy would be problematic and so women were arranged to marry quicker than men?
Why not both?
AAFM is higher for men than for women even today. It’s partly because of the things you say, but also because boys mature more slowly than girls, and reach peak attractiveness to the opposite sex a few years later.
I’ll add to your doubts about AAFM regional patterns: in South Italy, in the 18th century, you had peasants in the Amalfi Coast marrying very late, in their 30s, while in Apulia peasants married in their teens (source: “Gli italiani prima dell’Italia”, Carlo Capra). Why? because in the Amalfi Coast the economy was based on small patches of lands, still owned by very tiny land owners, where to cultivate lemons and other fruits to sell at the market. There was no space for divide even more the land between multiple children, the land wasn’t enough to sustain more than one family, the second sons went to Naples in search of fortune. Apulia, on the contrary, was more of a classic rural world with large farms, owned by large landowners, in a big plain territory.
I wonder if perhaps the Black Death was the event that led to the development of the late/late marriage pattern in Western Europe?
The Black Death famously minted a LOT of corpses, and likely affected older people worse; this probably killed off a high percentage of heads of household and left a lot of widows, widowers, and orphans who were now free to choose their own spouses and also had the opportunity to create independent households on freshly vacated land.
That means that, post-plague, the percentage of married couples who had selected their own spouse and were living independent of their own parents would have been MUCH higher than it had been pre-plague. I can see how this could lead to a lasting change in the marriage pattern: if people decided that the new norm was better than the old norm, they well well have stuck with it even after vacant land was no longer abundant, creating the unusual late/late marriage pattern described here.