Calibrating Stardates
There should be no problem about calibrating stardates. As I pointed out earlier, if 1,000 stardates is equal to an Earth year, then they should be fitted into the system by assigning them in the same way as the “Star Trek Chronology” does. Unfortunately, that raises some problems. Occasionally, a hint was dropped about the day of the week, the year, a festival or a specific date in our conventional calendar. I’ve compiled the following table:
Stardate Date
41986 2364
44390.1 Diwali
48579.1 2371
49334.53 24th of the month
49364 14th of the month
49419 14th of the same month (!)
49655.2 Tuesday
51268.4 20th May
51449.2 Monday
51471.3 Tuesday
51658.2 Tuesday
51715.2 Tuesday
51721.3 Friday
51948.3 Sunday
54868.6 Soon after 5th April 2378

It’s not comprehensive, but it does cover most of the specific date references made in the 24th century. Although most of the dates will sort of fit if you apply the “Okuda system” of stardates, there are problems. The biggest is VOY “Homestead”, when the date is very specifically fixed as being the 315th anniversary of First Contact, or Wednesday 5th April, 2378. Extremely unfortunately, if stardates are calibrated exactly to calendar years, then the stardates in the last season of “Star Trek: Voyager” all fall in 2377. If you “bump up” by a year, then it invalidates earlier date references. Then there’s the original series. The stardates will fall conveniently over five years up to sometime in 2269, but the earliest stardates are in 2264, when the references in the “Star Trek Chronology” and the bottle of champagne at the start of “Star Trek: Generations” suggest that 2265 is the starting year. Worse, if the end of the mission is fixed as being in 2270 in VOY “Q2” then it begins to stretch inconveniently beyond the “five years” it’s supposed to last.

My Calibration

What if you start with the 2378 reference in VOY “Homestead” and work backwards? This was the decision that really convinced me to adopt this idea. Naturally, the 2364 reference from TNG “The Neutral Zone” becomes unworkable, but I discovered that the original series stardates all now fell in the years 2265 to 2269. Amazingly, S.D. 44390.1 from TNG “Data’s Day” was very close to the date of Diwali in 2367. Naturally, for every date that works, there’s another one (or two, maybe three or four) that don’t. There’s no way around that, at least that I can think of.

At this point, I began to run into problems. Simply “winding back” stardates from the zero-stardate for five-figure stardates creates a few problems. If you count forwards from TOS “The Trouble with Tribbles” then DS9 “Trials & Tribble-ations” ends up in the wrong season of “Star Trek: Deep Space Nine”. There’s a similar problem with the two parts of “Star Trek: Generations”; the story makes it plain that there’s 78.2 years between them, but the stardates run into problems when you do that.
The underlying problem is that the length of 1,000 stardates comes out as being just less than a year, more like 364 or 363 days when you use those figures. Naturally, getting that to work with a base unit of 24 hours is incredibly difficult. Getting a 365-day year is bad enough. The only solution that I could come up with is a reset point. It’s not ideal, but it does address another niggling point: if stardates are 24 hours long and not reset at some point, then when TOS “The Trouble with Tribbles” is on a Friday, DS9 “In the Pale Moonlight” can’t be.

Setting the zero date for “23rd century” stardates was fairly straightforward. Provided I ignored the only major clue, that TOS “Charlie X” is at Thanksgiving, then I could move backwards from a sensible date for DS9 “Trials & Tribble-ations” to place TOS “The Trouble with Tribbles” on a suitable Friday, meanwhile trying to get the opening of “Star Trek II: The Wrath of Khan” to fall in early January, matching the date for Kirk’s birthday established in the “reboot” films.

Constructing a System

How does all this fit together? More importantly, why would the people of the “Star Trek” future scrap the calendar we have now in favour of my creation?

“Star Trek: Enterprise” makes it pretty clear that conventional dates will still work for people whizzing about at warp speeds, but do they work well enough? Time begins to behave quite strangely when you travel at very high speeds, and simply knowing that it’s Tuesday the 4th may no longer be sufficient. At some point, the problems (whatever they are) must have reached the point where something had to be done. There is a hint that the Vulcans might have used something similar based on their calendar, but it’s clear that the stardating system in “Star Trek” is based on the calendar used on Earth.

Since it’s my idea, I’ve assumed that the original version of stardates was intended to run alongside conventional dates, and not act as a replacement. This would have addressed the main criticism of my system; that a stardate refers to several possible days, not just one. Obviously, that’s not what happened. All the same, the initial system of stardates included an “extra” 370-day “year” every 40 years, to accommodate leap days. This would serve to keep the stardates approximately in step with the conventional calendar, at least for the first hundred years or so.

Purely for convenience, I’ve assumed this first version of stardates with an average “year” of 365.25 days ran from an arbitrary zero date I’ve put at 00:00:00 (UTC) on Wednesday 10th December, 2223.

Evidently, it didn’t operate as intended. I’m assuming that the “conventional” dates weren’t included, and that the “leap day correction” simply made calculating the difference between one stardate and another unnecessarily complicated. By omitting them, the length of an “average” year is exactly 365 days over any length of time. In addition, by having an extra digit, the stardates themselves will only repeat once every century, rather than every ten years, as with the “four digit” stardates of the earlier system.

In the end, the advantages of carrying on with the “original” system were outweighed by the benefits of changing. I’ve set the changeover at 00:00:00 (UTC) on Sunday 17th June, 2323. That’s near the end of the last cycle of four-digit dates, but not so close that confusion could arise between the two systems. It also “offsets” the stardate “year” from the conventional calendar year by about six months.

It’s a very flimsy argument, but it could explain why people keep assuming that the calendar year and the stardate are about the same. The most problematic example is when Data says it’s 2364 in TNG “The Neutral Zone”, but there are others. If the map-across was quite close, as it was for the “four-digit” system, then assuming the two were the same would just about work. The offset in the “five-digit” system is much greater, and using the stardate to guess at the year would be much less straightforward.

A Last Thought

There are still a lot of stardates that don’t work at all, or don’t work well. One of the really annoying things about stardates is the way they sort of hover on the edge of making sense. If only I could twiddle the numbers just the right way, it would all fall into place! Part of the problem is that people like things to work to a pattern, even when they don’t. Most of the difficulty is that you can get closer and closer, but never actually solve everything, all the time. Eventually, you just have to accept that the only way stardates could work consistently is if they had been calculated according to a system, right from the beginning. And if they’d done that, there’d be a “canon” way of doing it. I have been looking at all this closely for quite a while now, and I have never seen anything that suggests there was ever a formal system for working out stardates. The surprise is that they seem as plausible as they do, since they’re based on “back of an envelope” rough guesses.

Not having a system at all means never having to say that a date doesn’t work, but it also means that you can never say “this is the stardate, so it happened then.” Having a system at least means you can tell what stardate it’s meant to be, and allows the stardate to at least have some influence on the placing of an event into a timeline. That’s my approach, and it’ll have to stand or fall on its own merits.

Introduction to Stardates

by StrauchiusStrauchius on 24 Oct 2012 14:42, last updated on 09 Jul 2017 09:24