This is why I think this thread is unnecessary, because there isn't and can never be a "theory" on butterfly effect. It's not a scientific phenomenon, it's a literary trope. Alternate history is not a science, it's a literature genre. And to literature, literary laws apply, which don't necessary follow the scientific method.
You're right, but I should have qualified my original post. I want to talk about butterflies - shorthand for the changes resulting from a specific point of departure in a timeline -
philosophically. Not from a literary perspective, because we all know how authors use butterflies to either get to the world they want to create or, alternately, use butterflies to let the story write itself. If it was a scientific phenomenon, such as a world in which the multiverse theory has been proven true and they have the capability to monitor other timelines, what might they find? And since we don't have that kind of data, our conversation would need to be of the philosophic variety.
I like to think of it as the POD causes a gradually expanding series of ripples myself
I agree with galveston that the best, or at least my preferred, way of looking at it is as a series of expanding ripples. The ripples of change expand not so much geographically but conceptually. The more closely related an event is to the POD the faster the ripple travels there, and vice versa.
As for yuor question about people being butterflied away, I don't think there's a single answer to that. There certainly isn't a set time limit after which you can state that people will be different. I suppose in part that gets down to a nurture vs. nature dilemma. However, I think of a POD's effect on people as re-rolling the dice. Sure, in this case there are a hell of a lot of dice, but that doesn't mean you're assured of a different result. Ultimately, it comes down to authorial fiat and the timeline author's ability to construct a scenario where both he divergences and convergences seem reasonable.
I like the gradually expanding ripples approach, and it's one that I think would be accurate. I'm trying to tie down how big those ripples would be from a given POD. Obviously, different PODs would have different impacts, so maybe we could have a sliding scale of PODs. Alpha, Delta, Epsilon, Sigma, Omega, in ascending order of impact. But that would involve determining whether or not any Omega PODs are actually Alpha PODs that spiraled quickly; but if they spiraled quickly enough, would it really matter?
Unless Author deliberately create Butterfly net, any conception after Alpha POD would be different person.
Remember origin quote : flap of butterfly wings could create storm in faraway places.
so if a Nevenk hunter in Siberia killed by meteor in 01 January 1501AD; his non-existence would cause all kind of different movement : running rabbit stop running, birds in area didn't fly because human presence, his body cause splash when fall to ground etc; so 'butterfly' already flap, there will be small change everywhere in the world, a breeze that blow in Missisippi, storm cease a minute shorter in Bengal, calm instead breeze in Kyoto; since there are millions of sperm and even very small change affect who 'win'; anybody who born in 02 January 1501AD would be different person from OTL.
I disagree on a number of points. Those kids born on 2 January 1501 would have already had their genetic structure determined; but if you meant that any person
conceived after 2 Jan 1501 would be a different person, I disagree as well, but less so. As far as I know, a child born to the same parents would have the same genetic structure, the father giving Aa and the mother giving Bb. That results in the possibility of a child being born as AB, Ab, aB, ab. So that could change, which parts of the genome they got from which parent. So not a completely different person but different enough, like a sibling to the OTL person. But in the above theory, that we can name Theory of Exponential Change, that these changes take time and certain changes could make zero impacts on the timeline. If your Nevenk hunter dies in Siberia, we can extrapolate certain weather patterns - for a while - but if - and I stress if - a certain region has more rainfall because of the butterfly flapping, how much could that really affect things? Sure, it may rain on a battlefield, giving the OTL losing side the advantage, but how many battles were actually the turning points for wars out of all the battles fought? It's like 95 percent more likely to rain on a skirmish that means nothing in the grand scheme of things. But how long would it take a single butterfly flap to create a storm? Over how far would this change take?
If we approach this scientifically, there needs to be a threshold for change. A random person dying earlier that sets off a chain of successively bigger changes resulting in the fall of Rome a hundred years later than in OTL, which is the actual POD? On the one hand, it's the random person dying - but in the event we were given an alternate timeline, there's almost zero change we could trace such a chain back to the original Alpha event, and we - as scientists studying an alternate timeline - would claim the POD to be Rome not falling, since that's the big event from us looking in from the outside.
I'm pretty strict with my butterflies. Yes, general trends are going to stay the same for some time (with minor changes 9), but the particular details will be different. This is especially true with human conception - anyone born nine months after the POD is going to, by nature, be a somewhat different person.
Somewhat is the thing I'm trying to tie down. Siblings - which the OTL person and ATL person would genetically be, unless I'm mistaken - can be incredibly different, or completely the same. Parents are likely to name the child the same thing, raise the child the same way.
I prefer a linkage theory, any changes that happen happens as a result of the change affecting the world. Whether physically, verbally etc. An African kingdom rising in the centre of the continent isn't going to change Genghis Khan being born nor conquering the way he did. For it to affect his birth it has to affect what his parents do up to it from that moment on and with an inability of the news or the people of the kingdom affecting them somehow their lives won't change. They'll go hunting the same, they'll eat the same, they'll sleep the same and the smoosh booties them same. That applies to his conquest he shall never know nor be affected by anything happening in that area, he'll never even be able to talk about it delaying or increasing his time of doing anything in any area.
Of course, you could dream up a series of possible happenings that occur, human or otherwise that do end up changing the Mongol conquests but each event added to the chain only increases the MTTH - Mean Time to Happen. The longer the MTTH is the more time for that ripple to start to affect change and from there people's births can/will be changed but the strength of the ripple is what changes events. Even if an event chain is created to affect Genghis' birth his conquest may stay the same under the same chain. Of course, all of this is decided by what the events on the chain are along with when they start affecting change on the Genghis' and his life.
All this means to me that when deciding:
A. Who is born? Think about your POD and what it changes, when the person should have been conceived about and whether either any news of the change or an event chain could reach their parents between the time of divergence and conception.
B. What events occur? Think about whether an event chain from your POD to any action or event of a person and what if any changes such would actually have on them i.e the early death of a Chinese farmer really does shouldn't change the actions of Henry VIII within any period short of 6-7 years.
TL;DR - The formation of an Amazonian South American native kingdom shouldn't prevent the birth nor death of William III of Sicily if it happened within a certain time period, in this case, 15 - 20 years.
Right. To change the rest of the timeline (and I mean
actually change the timeline, not just cause different weather on the other side of the world) there has to be a connection. To make a man impregnate his wife at a different time than in OTL, there has to be a reason for them to change their sexy time. Maybe they see a news report of the POD and it puts them off another half hour. Maybe a battle happens in a different place, leading the random farmer to worry more that particular night and their son is born differently, marries a different woman because of his change in preference, which eventually leads to a thousand different marriages and births. But if a tenth century Native American in the American South West (somehow) unites all the tribes in the region into one cohesive group and they expand eastward and turn North America into a great Confederacy of Tribes, it still would have zero effect on anything going on in Europe.
Since what we're talking about here is fiction, I'd say it's whatever you want it to be. There is no "science of butterflies," no correct opinion on who would or would not be born if X happened and Y didn't Z days after the fact and so on.
What interests me in alt-history is the logic of the counterfactual: If X happens instead of Y, what changes? Whose lives are changed? How do the people of the time react to it? It's about reasoning from A to B to C, and showing your work by explaining how one might plausibly cause the next. Yes, as time goes on the ripples of the POD become impossible to string together like that and you must settle for interpreting broad historical changes and inventing characters to see them through, but I see no reason to bring on the chaos prematurely by random happenstance right after the POD with such explanations as "a gust of wind" and "the vagaries of human conception." I mean, there's nothing wrong with that if that's your jam, but to me it strays from the original thought experiment of "if X instead of Y, then what?"
I agree completely, and I really don't want to get into "random gusts of win changes all history" because that's counterproductive. We can physically see the link between Thomas Jackson surviving Chancellorsville and a different Gettysburg, but a gust of wind in Canada in Jan 1863 traveling down to Chancellorsville in May 1863 and causing Thomas Jackson to take an extra moment to grab a blanket or something, thus setting the chain of events off that means he's in a different place than OTL is just silly and far too ridiculous.
Even the tiniest butterfly can have the most enormous consequences. If one of us were magically transported back to ancient Rome, looked around for a few minutes, and then came back to 2017, the world to which we would return would be unrecognizable. The simple act of standing there for a few minutes would cause ripples, simply due to the weather (the course of the idea of the butterfly effect). After all, a human being is between five and six feet tall on average, so the air would move slightly different in the surrounding space. This, in turn, would have an extremely tiny impact on the way the wind blows through a small area of space, which in turn would affect air flow in a gradually increasing space of both time and area. Wind patterns dictate weather events. So the philosopher was right when he said that a butterfly flapping its wings in New York would cause a storm in China (or whatever the actual quote is).
I received an email about a year ago by a person who had read Shattered Nation, remonstrating with me that the weather I described on a particular day in September was historically inaccurate. However, the POD for the Shattered Nation timeline takes place on July 17, which is more than enough time for ripples caused by the butterfly effect to accumulate to such a degree that the weather will bear no resemblance to what is historically was. (And thank goodness. I would hate to spend time researching what the weather was on each given day of a story.)
Genetics and the circumstances of conception and birth obviously dictate much about what a person becomes, including if the person exists at all. The statistical chances of any two people meeting and conceiving a child together are incredibly tiny, which means that the chance of any particular person (pick a historical figure at random) is tinier still, since not only do their parents have to meet, but so do their grandparents, great-grandparents, and every set of ancestors literally going back to the beginning of life on Earth. It's easy to pick out particular instances of genetics impacting history (what if Henry VIII and Catherine of Aragon had a boy instead of a girl?) but the easiest POD is simply deleting the existence of critically important figures by having some set of their ancestors fail to met and conceive. And, as already pointed out, the child's parents would have to conceive in precisely the same manner they did historically. Just having Aristotle's parents meet isn't enough to guarantee the existence of Aristotle.
I'm not so sure I agree. Even if it storms on a different day in China, that doesn't necessarily change things - from a macro perspective. Sure, it could possibly result in an entire region of babies being born completely different, because their parents conceived at different times, but even then, if that village is sufficiently isolated, what changes could it bring to the timeline as a whole?
Like, say a prominent general is born in an isolated general in 1200. It rains the day he was going to be conceived, and his parents give birth to a completely different person. Time goes on and nothing changes, because nothing the boy does before the age of 30 had any effect on the timeline. So in 1230, changes start to be noticed. Small changes, little things here and there, but nothing that changes the timeline outside of that particular region. The guy that gets promoted to general in place of our OTL general happens to do things the same way so the large events - whatever battles our OTL general won to make him so prominent - happen the same way as OTL leading to the same outcomes as OTL. Nothing has changed in the timeline - on a macro perspective - than OTL.
The opposite is also true, a completely different general takes the position our OTL general and does things completely differently, resulting in losing those battles and changing the timeline completely.