How Long Is A Generation?
Four generations of the Dukes of Richmond, painted in 1900, artist unknown. Seated front left is 6th Duke of Richmond and Gordon (1818-1903); seated right is the Earl of March and Kinrara (later the 7th Duke) (1845-1928); standing is Lord Settrington (later the 8th Duke) (1870-1935). The infant on the Duke’s knee is Hon. Charles Henry Gordon Lennox (the son of Lord Settrington) (1899-1919)..
Note that artistic works by unknown artists remain copyrighted until at least 1 January, 2040, unless they were published or exhibited publicly prior to 1 January 1953.
Source: ClanMacfarlaneGenealogy.info via Wikipedia Commons (image page)
Four generations in one image means that it has to encompass a span of three generations. The dates given suggest that those generations are roughly 82 years, which gives an answer to the central question of about 27 1/3 years. Make of that what you will…
Last week, I promised readers something completely different from the Economics in RPGs series, and even though this is a different article to the one I had in mind at the time, I still think it delivers – in classic Campaign Mastery style :)
While working on the Adventurer’s Club campaign this week (as I outline this article), I was prompted to ask myself this question.
“Been in the family for x generations,” is an entirely valid statement that expects the listener to be able to decode the message.
“…and lay undisturbed for generation after generation, until the events had long ago faded into myth and legend,” lays less emphasis on unpacking the units, but the GM still needs to have some vague idea of what they mean.
And it’s not as simple as it sounds.
Prior Engagements
In the past, I’ve skirted around this issue as much as I possibly could. Nevertheless, there are at least two prior articles when I could not have fully avoided it (the first, in two parts):
- Growing The Perfect Family Tree Part 1
- Growing The Perfect Family Tree Part 2
- Throw Me A Life-line: A Character Background Planning Tool
There may have been others, but those are the ones that come most readily to mind.
A fixed number – 20? 25? More? Less??
There have been all sorts of fixed numbers thrown around by various sources through the years and marginal agreement between them. Four claimants of the definition stand out, in my opinion.
25?
Twenty-five years was the number that I first associated with the term, I’m not entirely sure why – this was long before I got into gaming. For many years, it was my default go-to interpretation.
It’s a convenient subdivision of a century, for one thing, and long enough to encompass most viable analysis-based interpretations.
Over time, a subtle difference crept into my interpretation of the term, “generation” – rather than a strict 25 years, it became an ‘average’ of 25 years. This attempted to reconcile function-based applications of the term with the simplicity of a numeric definition.
It made the whole term fuzzier, and potentially more useful.
30?
When I raised the question with my fellow GMs – because I had most of this article written already – one offered this up as his go-to interpretation, while admitting that he had never really put any thought into the meaning of the term..
He was no more sure of where his number had come from than I was of mine, but he justified it in functional terms. There was a hint of a suggestion that the term may have originally had a functional definition that was rounded and approximated by later users, clouding the whole issue – sometimes, a user would have meant the strictly functional definition, and sometimes, the generalized interpretation – and sometimes, they would have simply tossed the term out as a vague “long time” with no significant thought invested in the meaning at all!
I don’t know about you, but I found this line of thought fairly compelling, even though it clouded the issue more than a little. But then I realized that it could operate in the other direction too – starting with a vague definition, to which people like myself had tried to apply functional refinements, only to find that none of them quite fit the sources (in fact, it would be an astounding coincidence if they did line up.
More cloudiness, less clarity – both theories have a ring of plausibility.
An online source then suggested 33 years, using exactly the same logic as for the proposed 30-year units – my impression was that this was compromising the unit to get a simple fraction of a century (well, of 99 years).
20?
Another number that I’ve seen seemingly plucked out of thin air from time to time is Twenty Years. In modern times, this would be a reasonable fit for the usual functional definition, perhaps rounded to a convenient number.
We all know people who consign everything that happened before they were born to this vast dumping ground of “Don’t know, don’t care, don’t know why you care, either”. These are the people who allegedly ask about wooden aircraft carriers, and why Knights didn’t use rifles instead of swords – though I find those to be caricatures, not entirely divorced from reality but not truly reflective of it, either.
I suspect that this value is a vague compromise between a lot of numbers that have some element of plausibility, bolstered by the modern ‘fit’ and the convenience of being a nice, simple, number.
But this raises another complicating factor to think about – the possibility that the numeric value interpretation of a definition has changed over time with improvements in medical knowledge and changes in society.
18?
A functional definition is the age at which a set of generic parents could be replaced – I’ll examine that definition with more rigor later in the article.
The first time I remember encountering it, it was being used to justify a generational ‘unit’ of 16-18 years. The line of logic used was that you could marry at 18, and have two children before you were 19, if you really worked at it.
Actually, the definition used marked the next generation as being the birth of the first child, perhaps for simplicity. Yet another complication!
What’s more, the fact that people used to marry younger – historically, 16, 14, and even 12 were not unheard of – lends added credence to this value.
My personal suspicion is that this is simply the youngest legal age in modern times at which marriage, and hence legal childbirth, is permitted. So you couldn’t have a generation be anything less, and anything more sat in the difference between a theoretical future potentiality and the current reality.
But this brings in a whole raft of new complications – social factors, legal factors, and the difference between theory and application in reality.
It’s a lazy definition, but an illuminating one.
Age Of Consent?
In particular, this raises the question of legal age of consent, and whether it’s anything more than an artificial line in the calendar that most people observe – at least officially.
It’s tempting to toss this issue aside as an irrelevant distraction, but it seems unlikely that social expectations and behavioral demands can be completely divorced from the question at hand.
These social restrictions would act as a cut-off filter, setting a minimum legal value which is only loosely related to the biological elements of the definition.
Biological capacity, in other words, sets a range of values that could be acceptably associated with the term “generation” but social restrictions limit which of those possible answers are considered “socially acceptable”.
At best, this is a secondary factor – something to be taken into account, but not the primary foundation of a definition.
But it does point to one more complication: Perceived Value vs Reality. Just what you – and I – needed….
Complication Scoreboard
It’s probably worth rounding up all those complicating factors and questions at this point, and putting them all in a list.
- Functional definitions may be compromised for simplicity.
- A simple fraction of a century is an obvious and attractive arbitrary value.
- Vague and arbitrary definitions may be compromised to fit a functional definition.
- Definitions may have changed over time, so you can never be sure of the intended interpretation in any specific text or reference.
- Modern interpretations can cloud and bias interpretations.
- Applied medical knowledge is an obvious factor in ‘modern interpretations’ and is therefore an implied factor for other historical periods / game settings.
- Social factors, especially those consequent to medical knowledge, are also an obvious factor in ‘modern interpretations’ and a factor to definitions in other historical periods / game settings.
- Even if a constant, consistent definition is assumed, the numeric value associated with that definition can and will change over time (as a consequence of 6 & 7 at the very least).
- The limiting end-points must be a part of any practical definition.
- Legal factors can apply a bias, a socially-acceptable limit, or both.
- There may be a disconnect between theoretical values and reality on the ground, because 7 and 9 have no impact on the biological reality.
- There may be a disconnect between perceived values and theoretical or actual numbers.
It suddenly seems completely UNsurprising that there is so much confusion surrounding such a “simple” question.
Functional Generational Replenishment?
It takes (for most species) two adults to create an offspring. So replacing those two adults is a valid measurement of a generation.
This builds in all sorts of social factors. It re-frames the question to “What is the average age of a couple at the time of the birth of their second child who survives to adulthood?”
It’s a simple answer to specify it as “Age Of Consent plus two” – assuming human gestation periods and social structure.
If the Age of Consent is 18, that takes us back to 20. But if marriage is permitted at a younger age, as was common in medieval times, we get a different answer.
But that whole “+2” is problematic. Infant mortality in medieval times was appalling – I’ve seen values of 70%, 80%, and even 90%, depending on who you ask. In fact, that was a contributing factor to the lower consent age – where “consent”.is construed as “Consent to Wed”.
Mortality Impact
I tried running calculations to determine what the “+2” should be at different mortality rates, but got bogged down in detail to the point where I was no longer confident of the results. For the record, at a 90% mortality, I ended up with a value of +24.5. That’s huge, but is it right? I’m not sure. I can equally see it being half of 1.5+24.5, or 13 years – because we’re looking for the average, not the certainty.
The only thing I can state with confidence is that it’s going to be a LOT higher than +2.
Other Mortality Factors
This is another very real factor that should be taken into account. If the mother dies during one of these childbirths, then (to continue having children) the father needs to remarry, and that means that we are no longer looking for the second surviving child, but the third, because three discrete adult individuals now have to be replaced.
A man with extremely bad luck or judgment might need five wives to have five children.
It works in the other direction, too – if the man gets killed in a battle someplace, the widow needs to remarry in order to continue having (legitimate) children.
It’s entirely possible that BOTH parents will perish before having two children who will survive to adulthood. What do we do then?
Fait Accompli
For all practical purposes, it’s far better to presume a fait accompli and work backwards.
Current Generation
Let’s say the current date is 1210, and the current person of interest is 32 years of age. That means that they will have been born in 1210-32 = 1178.
Prior Generation
How old was this current person’s father or mother when he was born? Subtract that from our running date and you get the year in which the parent was born.
Let’s say that he was 22 years of age at the time. That means that parent was born in 1178-22 = 1156.
Grandparent’s Generation and older
Repeat for as many generations as you need, or until you reach the critical date in question.
What we were working on was a treasure hidden just prior to the US civil war, with a starting date of 1938. We ended up going back 4 generations – and this is not some nebulous generic “generation”, it’s the term as it applies to this specific family.
Age Of Death
If you know the age of the parent at which the heir to the family was born, and the year of birth of the parent, it’s then a simple matter to add X years to that age to get the age at death of the parent.
X is important here because that’s the number of years that the parent and child co-existed. You are essentially constructing a family narrative while and anchoring it to actual dates in your chronology.
What we found, when applying this concept, was that it was best to start documenting the family history with the earliest significant member and work forward.
We tried arbitrarily saying someone died at age Y but found that Y rarely married up to the chronology of events within the family; in effect, it was putting the cart before the horse.
Complications
We assumed that only one direct line of descent was important, even though we knew that this was inaccurate.
Every child has at least 2 parents. Every parent has at least 2 parents, who are grandparents to the child. Every grandparent has at least two parents, who are great-grandparents to the child, and so on.
A more complete (and much more complicated and tedious) approach would track each of these family links back using the same technique.
Then we get to the implied question of siblings – uncles and aunts, great-uncles and great-aunts, and so on.
And then we get to the question of their descendants – cousins and the like.
The guiding principle should always be the lived experience of the ‘child’ at the center of your family narrative. If they didn’t know relative Z, there is no need for you to mention relative Z, let alone place them on your growing family tree.
Integrated Histories
I’ve touched on this already, but thought it was worth explicitly considering: no family history should exist in isolation. No family is immune to the big events in the world around them – such as the US Civil War.
There are two general approaches that you can take to such integration.
The Fast Approach
The first is to have a list of the critical dates, in sequence oldest to most recent, and simply incorporate them as you are constructing the family narrative.
Much of the time, this will work seamlessly with no problems. Trouble arises when – for whatever reason – you need to alter the birth and death dates to marry up to the narrative that you are creating. Suddenly, you can find that an event that was supposed to impact a particular family member no longer touches them, or that an event that you thought you could avoid is suddenly very much a part of the family story of the generation in question. Any mistake in your arithmetic can tear your entire chronology apart.
The Second Approach
The alternative is to take a little more time and effort and actually map out a chronology:
xxxx Richard Randall born
1832 Steven Randall is born
1834 Efram Randall is born
1835 Eliza Douglass is born
1853 Richard Randall dies
1858 Steven & Efram duel for Eliza’s hand
1858 Efram Randall survives but is disinherited
1858 Steven Randall marries Eliza Douglass
1861 Civil War begins
1861 Efram Randall joins the Confederate Army
1861 Steven Randall joins the Union Army
1862 Efram leads his unit on a raid on the family farm…
….and so on (note the careful use of indents!)
For the record, this example is completely fictitious and bears no resemblance to what was being developed for actual game use.
This shows that there is no need for the father, Richard, to be affected by the Civil War; he’s already dead and buried. But it shows the siblings, Steven and Efram, caught up within it as an extension of their pre-existing family feud.
If your narrative requires Richard to be (a) alive but (b) too old to serve in the Civil War, and so (c) in a position to repel the raid, or perhaps to be killed in the course of it, the date of his death will need to be altered, and possibly the date of his birth.
This can have ripple effects both up and down the timeline that are far more easily handled with a simple list like this – especially if you can use cut-and-paste to move a whole line and change the date.
If you stick to the principle of only listing those family members known or directly relevant to the current generation, you can invent and insert long-lost relatives and other chapters of the family history as you need them.
Lost To Living Memory
A similar technique can be used when you need to set events beyond living memory. I actually went into this in some detail in discussing the planning tool linked to above, so I’ll try not to belabor the point here.
- Generation 0 – directly involved.
- Generation 1 – first-hand accounts from parents
- Generation 2 – first-hand accounts from grandparents
- Generation 3 – possible first-hand accounts from great-grandparents,
more likely 2nd-hand accounts (some distortion) from parents and grandparents. - Generation 4 – 2nd and 3rd-hand accounts from grandparents and parents, respectively.
- Generation 5 – events become part of family mythology.
- About 1/4 will be more-or-less accurate, though circumstances may be wildly different than expected and details will be wrong.
- About 1/2 will be, at best, half-truths and potentially misleading.
- And about 1/4 will be outright inaccuracies or willful falsehoods that have been perpetuated through the family history.
The Rule Of Threes
From that point on, the significance of the event (whatever it was) fades in relevance.
I always work on the principle of the rule of threes – at any given time, there are normally three living generations of significance: subject, parents, grandparents.
Anyone older is likely to be deemed an unreliable source.
That’s why, in the chronology-by-generation listed above, it is Generation 3 where the first-hand accounts stop. Generations 5 or 6 are where the second-hand accounts stop, and the story becomes a family legend. Somewhere around generations 8 or 9, that legend will be so vague and unsubstantiated that it is completely unreliable, if it’s remembered at all.
For example, one of my Grandfathers was killed in the Second World War; I never knew him. My father’s middle name commemorates him, and I was named for my father’s middle name, so I am concatenated directly to my grandfather’s life – but know very little about it beyond the simple fact of his service. He is a family legend to me.
A great-grandparent by marriage was in the first world war, but survived; I have personal memories of him and his stories of service. He survived the Gallipoly landing, for example. I can remember him telling me of the sense of wonder he felt as a child when reports of the Wright Brothers flight reached Australia. But going any further back? There are just fogs and mists. My living memory extends to second-hand reports of those early-20th-century events.
Application
One of the truths that I have gleaned from Who Do You Think You Are? is that the facts about the way the earliest actually-experienced generation were – personality etc – are known, but the reasons rarely are, and anything beyond that wall of time are lost, at best preserved in myth, folklore and rumor. In some cases, that earliest-generation is a parent; in some cases it’s a grandparent; and in only a few cases is it a great-grandparent.
Using this, and the preceding section, as a guide, you can determine how many generations forward you need to go before an event becomes an almost forgotten legend, or even gets lost entirely in the sands of history.
Once you know that, you can start assembling the family history from that moment forward, stopping when you get to the present day. Or you can determine that it was nothing more than a myth for X years and start your narrative from that point.
What stories were those who are alive today, told as a child? Who was around to pass on those whispers and murmurs?
Error
If you list four family tales from beyond that point of personal second-hand knowledge – stories from those who were actually there – you can break them down as follows:
It’s usually helpful to the GM to at least have a vague outline of what the truth was, and then to apply these ratios to the stories known to the current generation.
Non-humans and Hi-Tech
Everything written above applies to humans and those who live on a time-scale that is something resembling that of humans. There clearly have to be modifiers applied to such considerations when you are talking about non-humans, and other modifiers that have to be applied to take into account the penchant of medical advance to meddle in the ‘natural’ state of affairs.
Fortunately, it’s not all that hard.
- Start with the natural time-span. Derive a multiplier that can be used to transform a lifespan into human terms, or vice-versa.
- Think about the life-cycle of the species, as modified by medical science. This can be considered to shorten some stages of life, lengthen others, and make still others more rigid (treating any variation as a potential ‘medical issue’ that needs to be ‘treated’). Derive an appropriate set of multipliers for these factors and apply to both the human scale and the non-human scale.
- Think about the social structure of the species, especially in light of medical advance and the life-cycle impacts already defined. Modify accordingly, and add social restrictions on life-stage transitions.
- Use the resulting modified human scale to sketch out the foundations of a family tree, ignoring anything not of direct relevance to the core of the modern family. This works because it’s easier for us to think about such things on the human scale, even a modified one – it takes away one more point of confusion. The answer will be in years relative to the birth of the focal character of the current generation. Go back one further generation than you think you have to.
- Apply the human-to-alien scaling factor to get ‘real experienced years’ on the alien timescale. Translate the relative dating into actual dating using whatever the protocol is for such in your game world. Generate a list of the individuals and the dates of their births and deaths, and a timeline which lists those events in chronological sequence.
- Starting from the earliest date on your timeline, work forwards through time, looking for key dates in the campaign background to define generational transitions and life-altering events experienced by the population of the family tree. Add each to the overall chronological sequence, and add each event to the “bio” of the characters that experienced it. As you do so, make a brief note as to the consequences / impact on the individual. NB: I always start with a snapshot of the ‘status quo’ at the time of the earliest family member.
- When you reach the modern day, you have compiled a set of ancestors and milestones experienced. Some of these may be important enough to expand into a fuller biography – in particular anyone of special significance to the focal character, and anyone still surviving.
There’s a fair amount of work there, but none of it is especially difficult.
NB: you can also use the same technique to generate ‘histories’ of Kingdoms, of multi-generational businesses, of towns – anything you want. The use of a consistent campaign background creates the functionality of a checklist of important events. After you’ve done a few of these for the one game setting, that ‘checklist’ will start to make the whole process even faster and easier.
The same technique works in Sci-Fi, in Fantasy, in Steampunk – in fact, in any genre that you care to apply it to.
Wrap-up
Family histories are not always necessary, they’re not even useful a lot of the time. But, when they are relevant, they can provide a background narrative that makes a character more substantial, or create an adventure in uncovering the past or which has it’s roots in a part of history intimately and directly connected or a specific PC.
This article isn’t about when to create such histories; that’s best left to each GM and the circumstances of their campaigns. The purpose here is to offer a practical answer to the impossibly-vague question of how long a span of years comprises a generation, and how to employ that when it’s useful to do so.
Postscript Sidebar: Adventure style, tone, and sub-genre
Every player has a particular set of preferences and dislikes. One of the foundational players in my development as a GM loved Sherlock Holmes stories – and hated being ground zero of a mystery plotline. One of my current players hates “Big Cosmic” adventures – but loves Space Opera as a Sci-Fi sub-genre (just not in his RPGs, thank you very much). But it’s not just negative preferences – there are some “zones of subject matter” that are certain to bring certain players to life whenever they are encountered. Some players love major plot twists and surprises, some hate them.
More than at any other time, these preferences should be taken into account when constructing a plotline around the background of a particular PC. It does you no good to make a character the focal point of an adventure that won’t interest them – not unless that is the foundation of the whole adventure, at least!
On the other hand, setting such a background-based adventure in a genre that the player likes and enjoys boosts their interest levels, and makes both character and campaign more appealing to them – and, vicariously, to everyone else.
Bear those facts in mind and don’t be too clever for your own good :)
Comments Off on How Long Is A Generation?
