Past is not prologue but the past contextually influences the present

An interesting conjunction of articles. Much of conversation today around various issues is clouded by confusion between determinism and probability, between determinists and probablists. In simple systems, the differing perspective has little implication.

For a determinist, the sun will rise tomorrow. For the probablist, there is a 99.999999999999999999999% chance that the sun will rise tomorrow. For practical purposes, they share the same outlook.

The differences arise when you are dealing with complex, multicausal systems which covers virtually every biological system, and more particularly, every human system. Poverty, education attainment, obesity, morbidity and mortality, commerce, income, etc. They are all complex multicausal systems.

Determinists will look at something and draw a direct and invariable connection between an originating condition and an outcome: "Born in poverty, stay in poverty." Probablists look at it differently and observe "Born in the bottom income quintile, only a 7% chance of achieving the top income quintile." With complex multicausal systems, the difference between the two perspectives becomes much more noticeable because their interpretations lead to dramatically different possible solutions. In the above instance, a determinist might easily conclude that the only solution is to redistribute income. The probablist might more likely focus on those other causal elements that shape attainment of top quintile status (how do we improve productivity, how do we impart more skills, how do we instill better behaviors, etc.).

Determinists carry the heavier burden of proof because their claims tend to be more extreme. Determinists come in all sorts. There are biological determinists, genetic determinists, class determinists, technology determinists, economic determinists, behavioral determinists, geographical determinists, etc.

For example, someone wrote an interesting book recently which hypothesized a connection between technology adoption and current day wealth and income of nations. If you moved from the Stone Age to the Bronze Age 500 years before most others, you would expect to see higher income today. Likewise with a range of other technology transitions. An interesting conjecture and the author had good supporting empirical evidence. It's a dramatic claim which requires dramatic support. My recollection is that the author had good affirmative supporting data but did not address alternative explanations and therefore the overall argument remained somewhat weak.

At the heart of the argument though was the idea of event persistence - To what degree do events in the past influence outcomes today? There is interesting evidence and arguments on both sides, those arguing that past events are determinative of present outcomes and those arguing that the distant past and even to some extent recent past are minor contributors to current outcomes.

An intriguing piece of information supporting the idea that there is a long persistence in the effect of human capital comes from The Persistence of Human Capital by Robin Grier. The italics are the post author's quotation of the original paper on which they are commenting.

“After the international ban on slave trade in 1850, and in the midst of a massive inflow of European immigrants to Brazil, immigrants with relatively more education were channeled into specific localities through deliberate government policies. In the end of the 19th century and beginning of the 20th century, public authorities established a number of official settlement colonies throughout the state of São Paulo. This policy had goals involving occupation of territory, food production, “whitening” of the population, and was driven by a centralized decision at the state level. The settlements were established typically near previously existing rural villages and were occupied by relatively high- skill European immigrants of various nationalities.”

Amongst other things, one of the reasons this is so interesting is that the state-sponsored settlements were not appreciably different from other regions except for the fact that they had higher levels of schooling on average.

So what did they find?

First, “in 1920, a few years after the establishment of the last settlements, the literacy rate in settlement municipalities was 8 percentage points (or 27 percent) higher than elsewhere in the state, despite an only marginally higher share of immigrants.”

Second, one century later, people living in those regions had an average of more than half a year of schooling, and more than 15% higher average per-capita incomes, than people living in other municipalities. That’s pretty amazing after 100 years.

That lends credence to the determinist suggestion that there is event-persistence.

A second article cites evidence from a the northern hemisphere of the Americas. From Canada’s History of Violence by Pascual Restrepo. Emphasis added.

Steven Pinker, in his book “The Better Angels of Our Nature,” hypothesized that the differences [between violence in the USA and Canada] are deeply rooted in culture and history. In the 19th century, he wrote, the Royal Canadian Mounted Police — the Mounties — got to the Canadian Western frontier “before the settlers and spared them from having to cultivate a violent code of honor.”

During the settlement of the American “Wild West,” in contrast, there was no centralized authority. Plunder and feuding were the rule, and settlers often resorted to violence to protect their lives and property. Violent codes of honor, revenge and self-justice were second nature for early settlers and were transmitted from parents and society to children.

Before the settlement of the Canadian West, which I date from 1896 to 1921, the Mounties established a series of forts. That’s where they exercised authority, enforced contracts and protected the property of settlers. Where Mounties were present, self-justice was rare. Canadians on the whole developed a less violent culture.

In recent research, I tested Mr. Pinker’s explanation by focusing on the settlement of the Canadian frontier — modern-day Alberta, Manitoba and Saskatchewan. Because it is Canada, I also looked at recent N.H.L. players from those areas to see if they carried a cultural baggage of violence to the rink.

To demonstrate the role of the Mounties, I compared settlements that in the late 1890s were near Mountie forts with those that were not. There are no homicide statistics for that period, but the 1911 census reveals male mortality patterns. Settlements far from the Mounties’ reach had more widows than widowers, suggesting unusually high adult male death rates. In fact, remote Canadian settlements during this period looked a lot like those of the Wild West. We do not know for certain why male death rates in these communities were high, but homicide is the prime suspect. After all, men kill other men more often than they kill women.

Even a century later, the violence in these areas continues. In 2014, communities at least 62 miles from former Mountie forts during their settlement had 45 percent more homicides and 55 percent more violent crimes per capita than communities closer to former forts. The distinction holds even when we take into account differences in population size and the level of urbanization. Given that the authority represented by the Mounties long ago expanded into every corner of the Canadian prairies, the persistence of this difference is surprising. Apparently in some remote and lawless areas, the Mounties arrived too late to prevent the development of a culture of violence.

That is interesting data and there is more later in the article, including

The players share a common environment in the ice rink, but those who were born in areas historically outside the reach of the Mounties were penalized more often — an average of about 1.4 minutes per game — than those who were not — an average of about 1 minute per game. That 0.4 minute difference actually amounts to about 100 additional penalty minutes over a player’s career.

So yet more data indicating that there is event persistence that needs to be taken into account.

I suspect that event persistence is a real phenomenon within particular contexts but that determinists overstate their case. Yes there is a founder effect, but you have to take into account subsequent events as well which might swamp the founder effect.

For example, there is a common claim among Social Justice Warriors out of the postmodernist schools (deconstructionism, critical theory, critical race theory, critical legal theory, postcolonial theory, third wave feminism, etc.) that the socioeconomic status of African Americans today can be determinatively explained by their ancestral experience of slavery. Almost certainly there are some elements of event persistence. But which ones and with what effect size?

Thomas Sowell argues convincingly that there is little determinism between the institution of slavery 150 years ago and current socioeconomic outcomes. His evidence is that for the period 1880-1960, African American socioeconomic trends were converging on those of whites or exceeding them. Such trends as family formation, falling single parenthood, education attainment, income increase, improving morbidity and mortality, etc. His argument is that with this long period of socioeconomic convergence obviates a strong event persistence between slavery pre-1865 and current socioeconomic gaps today. Sowell's other evidence is the dramatically higher performance of immigrant groups who have achieved more in the modern era even though they share a ancestral slavery experience (such as Jamaicans and Haitians).

As is often the case, there is not a binary resolution. It is not a choice between determinists and probablists. It is a matter of testing the validity of the respective explanations and integrating the best of both. In some instances, it seems likely that there can indeed be exceptionally strong event persistence but that event persistence is not in itself deterministic but subject to other contextual pressures and events.