They call this day the 11th of March

[melannen]

Who wants to talk with me about interplanetary calendars?

I just finished the old book that's been on my to-finish for ages about the history of calendar reform, and it was surprisingly interesting for a book that consists in large part of quotations from Hansard and League of Nations committee reports (the one about cross-cultural perceptions of time is still on the to-finish for awhile yet, sorry.) They both ended up on that pile after the first time I read Machineries of Empire and was curious about how the calendar-based magic system in that might actually worked. (Mind you I don't think the MoE system can really be specifically calendar-based in its actual mechanics, also the farther I got in the calendar reform book, the less reasonable Cheris's quick-change reform seemed, and it didn't seem that reasonable even to start with.)

But! Calendars in worldbuilding are an interesting thing to think about, and I don't think I've ever seen a system used in a multi-planetary polity that really made sense. You have systems like stardates (that was a deliberate throwing up of hands at the idea of coming up with a workable one) and ones that just copy Gregorian time-words without really putting much thought into it, and not much else. (It's amazing how rare it is to see a SF story that doesn't have a very-nearly-24-hour day cycle, even. Where is the spacelag of shifting from a 30-hour day to a 20-hour day??)

So.

Suggested preliminary reading: Paul Krugman's The Theory of Interstellar Trade. Wikipedia Calendar (and subsequent rabbit hole.) How To Observe Judaism In Outer Space (and subsequent rabbi hole.) Falsehoods programmers believe about time

So there's like basically three things calendars do, right?

--Be accurately aware of an abstract concept of passage of time
--Keep track of natural cycles, or other recurring time spans like growth and speed
--Allow people in different places to do things at the same time as each other

Pretty much every human society has some way of tracking recurring cycles/times, but it doesn't have to be in a way that's obviously recognizable as a calendar. Like, I can tell you how far into Spring we are right now just by which wildflowers are blooming in mowed lawns, and I could probably be accurate based just on wildflower blooms within two weeks from about February to October most years; paying some attention to day/night length just in case, and that would probably give me all the timekeeping I would need for basic agricultural/foraging stuff, and I would remember years based on "The year we had the very cold February and I made maple syrup" and "the year of the earthquake-hurricane" and "The plague year" and so on, which is basically the way I remember years anyway. Even in societies that have a formally-kept calendar, probably most humans have used that kind of timekeeping for most things most of the time anyway.

But keeping an accurate count is also useful, and there are calendars that have been verifiably keeping an unbroken count of days for at least 2500-3000 years (and that's just as far back as we can verify.) It's useful when you really want to be more certain of when to expect something than "Around the time the buttercups bloom", and it means that if for some reason you miss the buttercups one year, you're still good. It's also really necessary for bureaucracy (which is at least as old as written history) - if you want to be able to set deadlines for things like taxes, and enforce them in a way that at least appears fair, you need to be able to accurately give the date. And then eventually of course you need it for science.

And, of course, buttercup-based timekeeping only works in a hyperlocal system. The buttercups will bloom earlier or later even a few days' walk in any direction from here. If I want to be able to say "meet me at the river-mouth when the buttercups bloom," we need to both be able to convert our local buttercup time to the time they bloom at the river-mouth. Even astronomical-based systems can have issues with this, even across fairly small distances - apparently there were court cases in England well into the modern period for rent payments due 'on the day after the full moon is seen' where people claimed they could pay late if it was overcast at their house, even if it wasn't at the landlord's.

Most widespread modern, and recorded premodern, calendar systems combine all of those - they will have a count of time, but will try to connect it to natural cycles, and generally use the count of days and some kind of astronomical cycles as cross-checks on each other. And they will have some kind of correction factors to align the two, and make it work across long distances and different contexts.

There is on top of this, of course, the cultural factor, which is partly that humans put all kinds of extra significance into dates and cycles and calendar-counts; and also partly that once you've committed to a certain way of doing dates, bureaucratically it's just a lot of work to change it even if it's terrible.

SO, thinking in interstellar terms, it seems to me like a human society that was trying to maintain some kind of bureaucracy/science/cultural alignment across light-year distances, or even just on different planets, you would almost definitely end up with at least two timekeeping systems in general use: a local calendar based on local cycles that are important to human life; and a universal system that uses an arbitrary counting-tick that lets two different systems figure out when something happened at the 'same time'.

The universal system would have to have some way of dealing with relativity issues; over interplanetary or interstellar distances you *can't* just keep a count of days, or even atomic clock ticks. It is too late at night for me to think about it too hard, but you would have to have *something* to serve that purpose. Maybe you would need some kind of interstellar landmark like a quasar, where anyone in the galaxy could observe it, know what time that observation shows, and then calculate their distance from it via redshift for a very accurate speed-of-light correction? Or just give up on having any kind of science or bureaucracy that is easily compatible between planetary systems; ships just arrive when they arrive and you treat each system as an isolated island in space-time.

The whole question of timekeeping for spacecraft is big and complicated and already having major consequence for everyday human tech - your smartphone has a relativity correction built into the GPS software.

The local calendars would be based on the same things Earth calendars use: dark/light cycles (lunar? Solar? something more complicated?); weather/temperature cycles (seasonal); water cycles (tides, melts, flood seasons); biological cycles (lifespans, circadian rhythms, reproductive cycles, of humans or other life-forms); astronomical cycles (either tracking the planet's motion in the sky, or detectable movement or changes of other astronomical bodies.) You could imagine other important cycles as well - i.e., a moon where tidal heating caused a regular cycle of geological activity. What cycles would you track in the calednar if you were living on something like a Jovian moon, with a lot more relevant things to track than on relatively simple Earth? How about on a planet with no seasons to delineate a year and no moons to delineate a month - you could still track orbital revolutions via the stars, but would you?

There would also be the question of people living in artificial habitats where any cycles were also at least somewhat artificial. Do most space habitats use the same calendar - is there a standard calendar based on logical math and human biological cycles that is used in deep-space habitats (and presumably planetary ones where there are no local cycles in the right scale for humans)? What cycles other than sleep/wake cycles would be important enough to track? What would we use for a 'year' if we had no real need to use an Earth year? Or is a more-or-less Earth-based one still used as 'standard' when there is no local cycle that can be adapted? Or do they each make up their own based on somebody's private crackpot calendar reform?

If there is a 'standard' calendar in use for artificial habitats and planets with 300-year-days, you'd have to assume parts of that calendar would also become used at least in the vernacular even on planets that had a local-day-based calendar as well. So at that point, most cosmopolitan places would be juggling at least three calendars - the 'standard' space calendar, the universal count of time, and the most important local calendar. Then you'd have to assume there would be at least one or two 'cultural' calendars that were based on some entirely *different* planet's cycles that some people still insisted on using for whatever reasons, and you're up to at least four.

(Four calendars may seem like Too Many, but then I remember as a white american christian, I am regularly dealing with the Gregorian calendar; my work's accounting calendar, which is a steady count of fourteen-day pay periods with a New Year in the summer; and the Christian liturgical calendar, which mostly lines up with the Gregorian except when it doesn't. Also there's the academic year, with all its hyperlocal variations, that still shapes my years even long after graduation; I have enough Jewish acquaintances I try to at least track their holidays (even if I'm super awkward at doing anything with that knowledge); I tend to time my vacations to lunar/tide cycles; and every so often I do something ridic in Excel and fall back on Julian days. So that's more than four. But at least they all use the same length of day/night count.)

Anyway, basically, this is my plea to work that kind of calendrical and timekeeping nonsense into more of your SF. It's fun! I swear!

Comments

[hannah]

A friend of mine worked out a Cardassian calendar with weeks that were thirteen day long, with months varying in length based on the cycles of the planet's three moons. Me being me, I asked her about the names of the days of the week and which ones are cultural holdovers which hold little relevance to the present society, as we've done it in most of the English-speaking world.

I'm so here for all this, basically.

[melannen]

Did she work out a day-length relative to Earth days? That's what always trips me up.

I guess it's not completely unreasonable to postulate that most Earthlike planets will have days not too far off ours - the Martian sol is very close to an Earth day, and IIRC Venus's originally was when the planet was more Earthlike (Venus's day length apparently varies by half an hour or so depending on the weather???)

But we don't really have any good reason to assume they would, either, and it always seemed to me that for something like Deep Space Nine or a starship, it would make sense to run wildly varying shifts based on different species' natural rest needs, instead of the alpha/beta/gamma 8x3-hour days Starfleet seems to generally use, at least on Human-led missions. There would be other issues with scheduling for that, of course, but figuring out the compromises would be fascinating. Starfleet has catpeople! You wouldn't try to put catpeople on 8-hour work shifts, surely!

Varying month lengths is also fascinating! It's one of those things that's hard for me to wrap my head around, even though some Earth calendars do it (and some users of Earth calendars that don't have variable month lengths would point out that the Gregorian *does*....)

[ilyena_sylph]

David Weber actually goes on about this in exhaustive David Weber detail in his Honorverse books!

A Loooot of computing power is put into conversions to appropriate local time-and-date on various planets, I enjoy it a lot.

[melannen]

Oh excellent! Someday I will get to that in my to-reads.

What do they use for ship's time?

[beatrice_otter]

I was going to mention this if nobody else had :D

[naraht]

This is a fantastic post. I've been thinking about this sort of thing myself, not so much calendars but all the different holidays that humans will bring out into the stars with them, plus all the new holidays that will arise naturally in the course of interstellar life. I'm trying to get some of that across in the SF novel that I'm writing at the moment, that sense of overlapping human cultural traditions. I hope you'll forgive me for sharing this snippet of a scene where the crew of my interstellar scout are celebrating the biggest holiday for spacefarers: the anniversary of Apollo 8.

But there was one holiday that every spacefarer celebrated without exception, on the evening of the 24th of December in the secular Old Earth calendar. They put up the decorations together in the crew lounge: a tree to represent the Earth and lights to represent the stars.
"Did you know that this was another solstice holiday on Earth originally?" said Sedna, tongue protruding just a little as he threaded the lights. "I don't know why it's not actually on the solstice. But all of these traditions, it just took them over from another holiday..."
"Nah, it's an anniversary," insisted Jay. "It actually did happen at this time of year."
"That's what they want you to think," said Rhodos.
True, there wasn't much left from the mess that Earth had made of its very early digital records. But there was enough to know that much.
"Really?" said Sedna. "But the tree..."
"Isn't the tree something that came from Christmas?" said Lewis.
"New Year," said Jay. "Some people still celebrate Christmas in Karelia. The Russians do. But you put up a tree for the New Year, not Christmas."
When he was little, his mother had a hell of a time explaining to him why Tu BiShvat, the New Year of Trees, wasn't on the 24th of December – or on the 1st of January either – even though that was the only time of the year that they had a tree in their apartment. All in all, it had been pretty confusing, but that was cultural traditions for you.
"We used to cut our own tree," he added, studying the disappointing polymer specimen that was all they could manage on board. The lights were pretty anyway.
"But the reading," added Lewis. "The reading is definitely..."
"It's Jewish," said Jay and Eisenhart in unison.
"Isn't it just what the crew read down?" said Sedna, baffled.
"But they had to get it from somewhere, Sed, keep up."
Arguing while you put up the tree was practically a tradition in itself. That was the way Jay's family did things, anyway.
And then it had given him chills every year, sitting by the light of the tree listening to message after message being read down from the stars. Good night, good luck, and God bless all of you, all of you on the good Earth.

I think it's inevitable that there are going to be a lot of calendars in the future, a lot of calendars. I mean, we currently have three Christian liturgical calendars, not just one: Western, Eastern using the new calendar and Eastern using the Old Calendar. We don't stand a chance!

As a bonus, one mission controller jocularly correcting another on time tags during Apollo 11. "It's GMT, babe!" https://apolloinrealtime.org/11/?t=136:29:16&ch=20

[duskpeterson]

I love it that the rabbis are trying to work out a solution to the problems of people in space, who cross the International Date Line, and who visit the Arctic. Because, of course they are.

(I was saddened, though, to read about Ilan Ramon.)

With my alternate history series, I didn't have to deal with the wonderful complications you mention, but I did take into account that the characters have a ternary alphanumberic system, which in turn is based upon their crop practices or their religious beliefs, depending on who you believe.

Their ternary calendar, and how it relates to the alphanumeric system.

More about their calendar.

However, as one character points out, their three-season system doesn't make sense from a natural point of view, because where they live (their version of the Chesapeake Bay), the seafood harvesting is mainly divided into two seasons.

[melannen]

At least four Christian liturgical calendars, if you count the Coptic! And I think there are still a few Western groups that run the Western calendar but on Julian, too.

That excerpt is really fun!

[pyraxis]

I love this!

[pedanther]

Vernor Vinge has some stuff about calendars and timekeeping in an interstellar society in A Deepness in the Sky, but it's been long enough since I last read it that I've forgotten most of the details.

One bit I do remember is the anecdote about the spacers' timekeeping system which is based on counting the number of elapsed seconds since some forgotten event in the distant past. One of the theories is that it's counting from the moment a human first set foot on another astronomical body, which makes sense to them as something to have started counting from. The implication is that they've inherited the Unix Epoch, which is used by many present-day computer programs and is based on counting the number of elapsed seconds since (arbitrarily) the beginning of the first day of the year 1970 -- so the people who think it's counting from the moon landing are off by a dozen or so megaseconds, but considering the timescale they're looking back from that's practically not worth mentioning.

[melannen]

Oh, that's really cool! And makes a lot of sense. From now on I shall pretend Unix time is dated to the first New Year after the moon landing. :D

(At least until the epoch ends and our civilization falls in 2038.)

[mindstalk]

Yeah, that's fun. IIRC they *know* that the count is six months (X megaseconds) off from the moon landing and are puzzled by it (which is pretty good memory for 8000 years in the future). It's so logical, counting from that epochal event, but... off. And the geeky readers do the math and have a laugh.

[lannamichaels]

One way that multiple calendar systems fucks with your head is that I'll be outside, look at the moon, and go "oh, it's the middle of the month" or similar, as if I can divine the date from the moon. Which I can. In one calendar system. But not the main one I use. ;)

Anyway, yes. This is great in SF to do.

[melannen]

Growing up under a solar calendar, in a place that both has a ton of light pollution (so you can't really see moonlight) and a lot of tall trees (so you can only see the actual moon a few hours a night, if you're lucky, most places) meant it took me a long time to realize just how truly practical a lunar calendar is when you can actually see the moon and use the moonlight. One of the things that I learned new in that book is that really up to the start of the 20th century, even Christian Britain still had a lot of contracts that were timed to observations of moon phase!

I don't think I've even seen someone practically work out what might be important to a lunar calendar in a place with multiple moons though. People will run the math for phases but they're less likely to think about what amounts of light would be useful for doing things, or what would be easiest to observe and calculate on the fly.

[ursula]

I classified a fun article for work recently that was about Greenwich Mean Time and railways. I guess railways spread the common time, but they also had this pesky habit of creating vibrations that disturbed the vats of mercury astronomers were using for their actual observations at the Greenwich observatories?

[melannen]

Yes! I guess railways were really the first time we had something that was both fast and accurate enough and likely enough to crash into each other if off-schedule that having different places' clocks synchronized to each other really mattered.

[superborb]

I also wonder at how easy it will be for humans to adjust to a different length of day. I often feel that I naturally have a 25 hr circadian rhythm, but I bet the natural variance in this centers strongly around 24 hrs, so how will humans deal with different lengths of day? Maybe a few thousand (Earth) years on, humans in different planets would evolve to have different natural circadian rhythms...

[melannen]

The circadian thing is interesting! There was a long-held datum based on experiments that isolated people from all external time-cues and let them just "free-run" that set the average natural human cycle at about 25 hours, and that's also what a lot of completely blind people end up on.... But if you instead stick people on an artificial isolated rhythm (like a 28-hour day) and measure things like body temperature instead of just sleep times, you get a cycle very close to 24 for nearly everyone, even as it gets farther off their sleep/light cycles.

Which makes me wonder if they're measuring two different things- if sleep cycle naturally lags our other circadian cycles for some reason. I haven't found anything where they tested other circadian cycles on people who were freerunning on sleep, or specifically tested whether those other cycles are as effected by light exposure as sleep is.

I used to think I had a longer-than-24 sleep cycle, but if I'm careful about artificial light exposure, that effect seems to mostly go away (though that might also be related to getting older.)

Anyway, yeah, it's interesting to think about at what point off a 24-hour day humans would just give up on working with planetary day!

Some NASA teams that work closely with the rovers have run long-term on Mars time 24-hours-and-almost-forty-minutes, and it has not worked very well, but also they seem to have treated it more like shift work than actually adapting to a different day.

I also have to wonder if people with off-kilter sleep cycles would be disproportionately likely to volunteer for that kind of mission! The only thing is that a seriously off-kilter sleep cycle tends to make it a lot harder to be the kind of person who wins slots on space missions....

[azurelunatic]

In Lois McMaster Bujold's Books, she once said you can tell which planets she likes because they get longer day-cycles (for more sleep).

And I was thinking about cicada cycles, which are multi-year. I wonder how you confuse a multi-year cicada's clock.

[melannen]

I had missed that there was any of that in Barrayar! I guess it's never super-relevant to the plot. Now that you mention it, I do remember something at least about Jackson's Whole having a notably shorter day. And do I recall right that Komarr doesn't always keep time with the planetary day?

I had had the thought that on a planet with absolutely no consistent seasonal cycles, or ones that are too long for human scale, something like the cicada cycle might be the closest they had to a year. Imagine a culture where the timekeeping was based on the intersections of 13 and 17-year cicada clocks! Though you might not even have something like that on a planet with no seasons. I think everywhere on Earth, even right on the tropics, has some kind of yearly seasonal cycles, even if it's just chances in the water of rivers/currents running into them from nontropical areas.

The cicadas do get confused sometimes though - iirc there's a bunch of the local DC 17-year brood that somehow got off-cycle and hatched early a few years ago. I don't know if they ever figured out why.

[brownbetty]

(I thought this post was going to be about YHL's Machineries of Empire, from your opening.)

I like your thought about quasars, and kind of want to get a physicist to explain to me why it wouldn't work because I know physicists love to explain why actually relativity makes everything post-modern and contingent, but it feels like it would work?

[melannen]

It kind of is, in that I keep wanting to actually work out a calendrical maths system that works with Machineries of Empire. But as far as I can tell, that system is less based on timekeeping per se, and more based on performances of sacrifices, or at least expressions of strong emotions, which Kujen's system uses torture to evoke; to keep the interplanetary system running, lots of people in many different places have to be performing the sacrifices at the same time, and so you need a calendar for that? And then the forces that are powered by the sacrifices ebb and flow in different ways like currents, and you can use that to figure out how other sacrifices performed at other certain times might alter that flow.

So certainly timekeeping is important; they've gotta have some kind of universal time-tick they can track even through mothdrive and exotic effects. And presumably the way celestial bodies move through local space also effects those flows, there's some of that in the Candle Arc story, so someone doing the calendrical work needs to do that kind of basic astronomical cycle tracking too.

But the calendrical hacking itself seems to be more to do with fluid mechanics than calendars per se? In particular stuff like the hacking Cheris did in Glass Cannon involved finding exactly the right time to perform a certain act to change the local flows, but it didn't seem to require any sort of repetition or record-keeping or chopping time into specific divisions or even communication with others in order to be effective, and those are things I generally associate with calendars.

Something like the quasar thing would probably work but I probably missed an important bit in my explanation! It's based on how astronomers figure times over relativistic distances now. But they keep finding things that mess it up and mean it's not as accurate as they thought, so. And it might require the existence of FTL communication in order to know what you were correlating your observations to. But if you get far enough into relativity you start wondering if the concept of "the same time" has any meaning in a non-FTL-equipped universe anyway. But the Krugman article makes the mathematical case that you could have a functioning interstellar trade society without needing a concept of simultaneity - it doesn't actually matter how far off two systems' calendars are from each other if the only transit between them is relativistic.

[primeideal]

I feel like my engagement with this starts and ends with a couple throwaway lines in, like, Star Wars fic lampshading it.

“We’re on Hoth,” she pointed out. “‘Month’ means lunar month, here, and this moon takes much longer to complete a revolution than a standard Coruscanti month.”

Ilinc wondered whether this kinds of pointless standardizing among all the different star systems was why the Republic had fallen in the first place.

*

It took six Coruscanti months before the assassins showed up.

*

"Your birthday. The K-2 models were first released this week, according to the Coruscanti Revised calendar, so I thought maybe--"

"The only thing you know to about the Coruscanti Revised calendar is how to fudge dates to postpone your maintenance appointments."

*

And anyone who'd been to RK's Floater more than once or twice could tell you that its target clientele were Imperials who kept odd hours, holoconferencing with colleagues on Eadu or Coruscant without regard for what time it was on the quickly-rotating gas giant of Tchie Gele.

And so on and so forth. :D

[lannamichaels]

This is excellent :D

[mindstalk]

Psychohistorical Crisis had a lot about units and measurement, I forget if time/calendar was included.

[coyotegestalt]

For my current d&d setting I deliberately haven't made a calendar, but calendars in general are a recurring background plot point - because it's a part of the world where there isn't a broadly accepted universal calendar, every government and cultural group and religion has their own. So lawyer-astrologers who specialize in calculating differences between different calendars are an important profession who are a part of most major legal and business proceedings, to make sure that translations of treaties and contracts are in agreement about when they are in effect.

(There was an attempt by a colonial empire to impose a universal calendar several hundred years ago, and once they got rid of the empire all future attempts at calendar reform carry unfortunate political implications.)

So for my players, everything is just logged in days/years before the present; and in-world, there are periodic arguments over what day it is.

This was partly because I thought it would be an interesting cultural trait, and partly because it was a way to stop myself from sitting down and doing all the math to set up a calendar for the world when I should be doing something more relevant, like writing the actual plot.