# Posts Tagged daylight savings time

### World Building Question: How Long is an Hour?

Has anyone else been thinking about the nature of time lately? Two weekends ago we lost an hour, an entire hour, as Daylight Savings Time began. Today is the vernal equinox, the transition point where days are once again longer than nights. I know I must think about time around this point of the year, because it was almost exactly a year ago that I started asking these World Building Questions, and started by asking What Time Is It? I’m going to move back to my questions about how time works on earth. So we lost an hour recently, but what is an hour?

Cesium. For all your time telling needs. Can you see it performing hyperfine level transitions?

Alright, that’s an easy one. An hour is sixty minutes long. Each of those minutes is sixty seconds long. So an hour is 3600 seconds long. We know the length of a second because it was defined at the 13th General Conference on Weights and Measures, an occasional meeting that sets things like the international standards for metric measurements, and just how many yoctometers are in a yottameter (a hell of a lot). A second, per this SI definition, is “the duration of 9,192,631,770 periods of the radiation corresponding to the transition between the two hyperfine levels of the ground state of the [cesium] 133 atom.” Whipping out a calculator to do the math, an hour is thus 33,093,474,372,000 of…whatever the hell they’re talking about with transitions and hyperfine levels. I was never that great at the applied sciences. As I’ll one day demonstrate when I crowd source some physics questions for an upcoming novel.

This is clearly a modern definition. An extremely modern definition. A definition less than 50 years old with a clarification less than 25 years old that “[t]his definition refers to a [cesium] atom at rest at a temperature of 0 K.” Cesium itself wasn’t discovered until 1860. And yet the human race has had hours for far longer than this definition has been around. For most of the history of the hour it was defined as 1/24 of a day-night cycle. Actually, this is a subtle lie. For much of the history of the hour, it was defined as 1/12 of the period between sunup and sundown. The period between sundown and sunup didn’t get hours because no one was doing anything during them, so who the hell needed to know what time it was? Night was divided, if at all, into watches.

Why 12? I covered that one when I asked what time is it? For those who don’t want to do the back reading, the short answer is “the Egyptians.”

As a side note: back in that post I said “Once a society becomes more advanced, it needs more granular time,” without really going much into it. I’m not really proud of that statement, as my wanderings through history in the last year have smacked down notions of societies as more or less “advanced,” which is really a modern ethnographic concept. To employ better phrasing, the use of granulated time, which flows from broad chunks of the day to hours to half and quarter hours, relates less to the “advancement” of society as it is to the ability to easily know what time it is while on the go. If you’re bustling around Rome and not stopping to check the nearest sundial, all that matters are the periods before noon, noon itself, and after noon. When you stop at a sundial, you could see where the shadow fell and know more broadly what time it was, but the same could be gauged by looking overhead. It’s not until clocks are visible from multiple points in the city, and the advent of portable clocks we call “watches,” that the person on the go could quickly have a better notion of the actual time.

So, alright, an hour is 1/12 of the time from sunup to sundown. This was thanks to those sundials, which started working every morning when a shadow first appeared and stopped working every night when the last shadows slipped into total darkness. There’s a problem with this definition, however. The period from sunup to sundown is not a constant. Here in the DC area the winter solstice produces just 9 hours and 26 minutes (by modern reckoning using the cesium atom) of sunlight. During the summer solstice DC gets 14 hours and 54 minutes (cesium) of sun. Dividing each of those by twelve, the length of a classical hour here in DC would be 47.2 modern minutes on the winter solstice, 74.5 minutes on the summer solstice. This is a significant swing. Playing this out over a modern eight “hour” work day, this would mean working just 377 minutes in late December but 596 minutes during late June.

Which is why it’s fitting to talk about this today, one of the two equinoxes, when the modern definition of an hour is as close to the classical definition as it gets during the year. Actually, this isn’t quite right as most of the world is today experiencing a day of 727 minutes, not 720, but it’s as good as you’re going to get. This is less the day where we get equal amounts of day and night, and more the day that everyone gets an equal amount of day.

Later hours were defined not as 1/12 of the daylight, but as 1/24 of the period from sunup to sunup. This wouldn’t result in 45 minute swings in time seen by the old definition of an hour but would still drift ever so slightly and require clocks to be reset a little each morning. Using DC from March 20, 2013 to March 21 as an example, on the 20th the sun rose at 7:11, on the 21st it will rise at 7:10, resulting in a day that is a minute short, and hours that are each about 2.5 seconds off. Which doesn’t sound like much, until you turn that into 22,981,579,425 of the cesium things.

So when did hours become equal in length? Not until the 1400s. Why? It took that long for the combination of accurate time keeping and a willingness to move away from the previous sunup-to-sunup definition. Technology fighting the momentum of “that’s how we’ve always done it,” just as it will throughout human history on oh so many issues. This technology continues to push forward, and we now understand that what we once defined as an hour isn’t necessarily a constant. But I’d rather not get into that, because I’m far more likely to say something entirely wrong. Or more wrong than the wrong things I’m sure already litter this post.

I like to bring these posts back to world building, so let’s give it a shot. We’ve seen how the hour evolved on earth, and how it was originally tracked thanks to sundials. The sundial is an intuitive piece of technology, by which I mean it emerged independently at several points in human history. So let’s take it out of the picture. How does time keeping evolve on a planet where the light is diffuse, say through a constant cloud layer? There would be a clear morning and dusk as light grew and diminished in intensity, midday could be intuited as the midway point between the two, but no clear progress of shadows would be seen. How would hours be divided by a species who has a sleep cycle that doesn’t line up with the planet’s day/night cycle? Would two clocks develop? What if a culture developed under more extreme cycles, such as those seen in Iceland where the length of a day swings from 21 hours to 3 hours?

There’s one more bit I wanted to go into, but I’m already over 1200 words so I’ll hopefully get to it later in the week. It’s the related, but interesting question: What time is noon?