Theory of Time

I have a theory of time that underlies how timelines work and how history is changed -- or isn't. This is sort of fundamental to Timelines and I'm kind of surprised I haven't ever written it down before.

1. You cannot change fundamental events. Big, huge, epochal events. You can't stop 9/11. You can't kill Hitler. You can't prevent the Kennedy assassination. And so forth. What will happen, always, is if you go back with intent to make those changes, you find yourself in a situation where you cannot affect that history, e.g., you show up thousands of miles away, or you show up after the event, or both -- the upshot being, you can never get close enough in time or space to make a difference. However, see exception below, following 2).
2. You can change minor events that make no significant difference, because either they are low-probability events that split off a short-lived timeline that folds back into the main timeline, or they are high-probability events that spin off a longer-lived timeline that might survive on its own (this is what leads to supra- or sub-lines of a numbered timeline, e.g., Timeline 1287 Left Sub 6).
3. Exception: A fundamental event that has 50/50 (or close to 50/50) probability MAY cause a major timeline shift, to the left or right. Such "split" events create separate, numbered timelines that then go on to develop their own future history. (Timeline Zero and Timeline 1 Right, for instance.) Such events might include
   • Kennedy either is or is not shot (or survives being shot).
   • Baby Hitler isn't killed but Baby Woodrow Wilson is. (I love that meme.)
   • Hitler is killed and someone much worse takes over, which is the current consensus among a lot of strategists.
   • The 9/11 hijackers are thwarted before they ever get on the planes, or by US Air Force shooting the planes down, or by passengers Flight 73-ing the other three planes, or whatever.
   • Joe Biden gets a second term and it ends in a major regional nuclear war in East Asia. (Timeline 240 Left Supra 2)

There are a couple of other things.

1. If a timeline is instantiated in "test mode", or, crudely, "NPC mode" (see The Lion of God), it is created with a complete history going forward and observers can move freely up and down the timeline. However it's important to note that a "test mode" timeline does come to an end at some point, unless external input is provided to create randomness so the line can generate future events. See (yes, sigh) The Lion of God.
2. A regularly-instantiated timeline supports time travel only into the immediate past. That is to say, a person in Timeline Zero 2026 cannot travel forward ("uptime") of that date, but only backwards ("downtime"). And yes, in The Lion of God, I got those reversed. Someday I'll go back and publish a revised version with the corrections. However, note that someone from the future can come downtime and take people from downtime uptime to their current time hack. See BBESP #4 for a specific example, though I already implied this in The Lion of God. The reason for this is that you can't determine the correct main line from its beginning because you have no clue as to the probabilities that create the timeline going forward. And indeed, if someone comes downtime to take you uptime, you have no real way to know if they are from the main line or from a sub-line that's split off of it at some point in your future...but the odds are that they're from the main line if they actually made it back to pick you up. See BBESP #4.
3. A causal loop (a type of temporal paradox) trumps just about everything, since the nature of the loop is such that you're already past the downtime part of it when you realize you're going to start one. See The Cross-Time Kamaitachi, in particular the epilogue.
4. Timelines that manifest to the left of Timeline Zero generally are bad outcomes; timelines that manifest to the right of Timeline Zero generally are good outcomes. There is otherwise no real rhyme or reason for the placement of a given timeline; they branch and are thus created randomly, so you don't get groups of timelines where there are only superficial differences and it's otherwise difficult to tell the difference between them. (Timelines with superficial differences tend to be clumped as the main line with supra- and sub-lines clustered around them as noted in #2 of the theory section, above.) Also, things don't get better or worse depending on the increasing numbers. One could argue Timeline 1287 Left Sub 6 has a better outcome than Timeline 240 Left Supra 2, for instance.