How fragile was history?
Elsewhere (and better): 1, 2.
If one could go back in time and make a small difference in the past, would one expect it to effect dramatic changes to the future? Questions like these are fertile soil for fiction writers (generally writing under speculative or alternative history) but receive less attention in the historical academy, which tends to focus on explaining what in fact happened, rather than what could have been. Yet general questions of historical fragility (e.g. Are events in human history ‘generally’ fragile? In what areas is history particularly fragile? Are things getting more or less fragile over time?) are of particular interest to those interested in altering the course the long-run future by differences they make today.
The small but reasonably sophisticated historiographic literature in this area (e.g.) often uses analogies to chaos theory: whether minor perturbations in antecedent conditions can flow through into huge consequences, akin to the archetypal ‘butterfly flapping its wings causing a hurricane’ example. In defence of history being chaotic one can point to various events which seemed to have large consequences yet great uncertainty ex ante. Had Hitler been killed in the first world war, or was assassinated in 1944, it appears the subsequent course of the 20th century would have been very different. Although ‘great man’ views of history find little favour, many events seem to be contingent on the traits and foibles of particular individuals: whether the Nazi party would have risen to power, whether they would have prompted a world war, and the precise conduct of any such war seem to vary a lot depending on whether Hitler was alive and in charge.
The chaos of coital counterparts
I suggest chaotic views of history can draw considerable support from the sensitivity to initial conditions and large variation that attends any human conception. Conception involves one female gamete (which is replaced with a genetically distinct one on a monthly basis or so), and one male gamete from a pool of 100 million genetically distinct sperm (themselves taken from an even larger pool with roiling substitution). All manner of trivial contingencies scramble which two would fuse to conceive a child: if the parents decide to have sex another month, another day (or another minute), when they last had sex, ambient temperature, and so on and so forth.[1] Siblings give us an impression of the variability of (forgive me) coital counterparts, and they are far from identical.
Moreover (somewhat like chaos theory) replacing one person with their coital-counterpart seems to cascade onwards to even more scrambling: not only do all the coital-counterparts descendants differ, but there is horizontal contagion if the counterpart’s slightly different behaviour influences others around them to only minutely change the manner of them conceiving children. Maybe there would have been something like the Mongol empire without Genghis Khan, but the precise details seem likely to vary. Yet these precise details may be decisive into which pairings between potential parents occur, and especially for exactly when they conceive, and the ripples of this activity goes on to perturb further successors, and so on and so forth. It seems not too wild to suggest that none of us would be here if the parents of Genghis Khan decided to postpone having sex one night until the following morning.[2]
Constraints on chaos
Yet if whether an individual or a potential sibling exists is very fragile, many other parts of history seem fragile. Imagining the course of the 20th century where every individual is replaced with their sibling drawn from the genetic lottery seems nigh-impossible. Yet perhaps some very broad predictions (perhaps some technological developments, population growth, etc.) could still be confidently held.
One may speculate there is an issue of scale. Although maybe even very large historical events (e.g. a given empire, a world war) are seen as fragile, one could zoom out even further. If one was asked to summarise all of human history in a sentence, one may be able to include little more than, “Humans arose 300 000 years ago, and spent 290 000 years or so as hunter-gatherers. They developed agriculture 10000 years ago, and the industrial revolution happened about a hundred years ago.” These details seem pretty robust. There are broader contours of the human condition that constrain the course the eddies of possible people may take.
One may hypothesis more to the story than ‘stable over long run, very fragile over short run’. There seem some ‘meso-level’ factors which alter the balance of contingency versus necessity, and so constrain the fan of possibilities on particular matters. The technological completion conjecture might be one of these: although whether, which, and when great scientists are born may alter the precise timing of discoveries, they do get discovered not too much sooner or later: if Einstein was never born, the theory of relativity would have eventually been discovered.
By contrast, creative work does not have this property: if Dostoyevsky wasn’t born, The Brothers Karamazov would not have been written by someone else. Where ideas fall may vary between these: perhaps (e.g.) liberalism, democracy, and communism would have been proposed by an entirely different set of thinkers; the character of major world religions seems likely to be substantially different in worlds without (e.g.) Jesus, Mohammad, Confucius, or Buddha. To what degree these things vary is unclear, but perhaps not wholly intractable with careful study.
The goldilocks zone of future fragility
The value of efforts to shape the long-run future rely upon fragility being not too extreme either way. With very rigid laws governing history, the future is set regardless of what choices we make, and so directed efforts to change the future are as feckless as (on a Marxist view) a directed effort to prevent the eventual triumph of the proletariat. By contrast, ultra-fragile histories imply forecasting the effects of our actions on the future are effectively impossible, and concerted effort to achieve one end or another futile.
Intermediate views are not implausible: the future seems neither cast in stone nor flotsam borne by the vicissitudes of chance. Yet degrees and variations are productive to investigate: perhaps political history is too chaotic to steer, but technological development less so (or vice versa). I think these topics could reward further investigation.
[1] I leave other sources of ‘coital perturbation’ as a less-than-pleasantly-edifying exercise to the reader.
[2] A more proximal example is that whether a given person’s sibling counterpart is male or female is about a coin-flip. Given now and in the past there were incentives to have a child of a particular sex (consider the importance of having a male heir in many cultures), it is plausible (e.g.) a family’s male third child may never have been born if one of the first two was male instead of female.
As far as I’m aware, there’s no way to predict whether delaying the moment of conception will result in a better or worse person. I do not see why a completely unpredictable effect like this should affect our decision making.
For example, let’s say that I’ve decided I want to donate to the Against Malaria Foundation (AMF). My friend then reminds me that because AMF saves the lives of children today, it could affect the identity of children born a generation later. Without any way to predict whether this change in identity will be positive or negative, there is no reason for me to consider it in my decision.
It seems to me the more relevant question is how often actions have their intended effect as well as how often they have unintended effects that could have been predicted (and whether those unintended effects would be positive or negative).
In blackjack, a competent player wins against the house ~49.5% of the time, and the house wins ~50.5% of the time. If I was to record a string of 0s and 1s, where a 1 represents a win by a competent player and a 0 represents a win for the house, my string would look almost exactly like noise. If I sit and record 20 games, that’s 2^20 = 1048576 possible strings I could record. So you might naively think that there’s no opportunity for useful predictions here. But in fact, the house edge means that on expectation, the house is going to win money and a competent player (absent card counting) is going to lose it.
In the same way, it’s not the stochasticity of the system that matters so much as whether we can make forecasts. Blackjack has loads of stochasticity, but the ultimate financial outcome can still be usefully forecasted. Weather is also very stochastic and may exhibit chaotic properties (see butterfly example), but weather forecasts are still pretty useful. Etc. The issue for EA is that we are trying to make forecasts in domains where there isn’t necessarily a history of successful forecasting like there is for the weather. This is a hard problem to deal with, but I don’t think it’s completely intractable. I suspect the set of skills needed is similar to the ones you need to be a successful investor or run a successful hedge fund.
Genetics might be a constraint on ultra-fragility. If all of the most practically important traits are highly heritable, then one wouldn’t expect the contingency of conception to produce as much variation in outcomes as in the state of affairs in which the contingency of conception has a very large effect on average individual traits. While it is true that the individual born is a highly contingent matter, the traits of the individual produced might not be. If my parents had an argument on the blessed night of my conception but overcame their disagreement the next day, then there would be some reason to think that a one day older version of me would be writing this comment.
Chaos also doesn’t seem inimical to attempts to rationally steer the future. Notwithstanding the fact that the climate system is chaotic, pumping lots of greenhouse gases into it looks like a bad idea in expectation.
That seems surprising to me, given the natural model for the counterpart in the case you describe would be a sibling, and observed behaviour between sibs is pretty divergent. I grant your counterfactual sibling would be more likely than a random member of the population to be writing something similar to the parent comment, but the absolute likelihood remains very low.
The fairly intermediate heritabilities of things like intelligence, personality traits etc. also look pretty variable. Not least, there’s about a 0.5 chance your counterpart would be the opposite sex to you.
I agree even if history is chaotic in some respects, it is not chaotic to everything, and there can be forcing interventions (one can grab a double pendulum, etc), yet less overwhelming interventions may be pretty hard to fathom in the chaotic case (It’s too early to say whether the french revolution was good or bad, etc.)
Not that it’s obviously terribly important to the historical chaos discussion, but I think siblings aren’t a great natural model. Siblings differ by at least (usually more than) nine months, which you can imagine affecting them biologically, via the physiology of the mother during pregnancy, or via the medical / material conditions of their early life. They also differ in social context—after all, one of them has one more older sibling, while the other has one more younger one. Two agents interacting may exaggerate their differences over time, or perhaps they sequentially fill particular roles in the eyes of the parents, which leads to differences in treatment. So I think there are lots of sources of sibling difference that aren’t present in hypothetical genetic reshuffles.
(That said, the coinflip on sex seems pretty compelling.)
Yes my slightly flippant personal example doesn’t illustrate the case well. But the high heritability of relevant traits does suggest that there would be less variation in outcomes in aggregate than if heritability was very low and each person’s coital counterpart’s traits were a random draw out of the human population.
Which role would attractor states have in this thinking? Some thoughts:
If an attractor is strong/large, then many different starting points have the same end points. But if it is small, or if we are in between two (or more) attractors, our decisions could make all the difference in the world.
The technological completion conjecture conjects an attractor that we end up with if we are not caught by x-risk attractors.
Can we somehow affect the fragility of history so that we bring it into the center of the goldilocks zone?
Wow, this is a fascinating post, and so short! I think a write-up on the historiography on this would be really useful. This is really important for EAs, particularly those focused on the long term or systemic change, and it could use a detailed treatment.
Great read and interesting take on alternative considerations. A discussion about fundamental attribution error would be interesting here—or a closely related concept. Not applicable to existence vs. non-existence, but I’d imagine we have poor intuitions at knowing the effect of changes to perturbations in individual human characteristics, and I wonder if something similar is at play when we estimate the effect of our actions, personal choices or character. In a stochastic enough system with a large number of players, perhaps single changes become absorbed into the background chaos.