Wednesday, January 31, 2018

Don't hold back!

Tuesday, January 30, 2018

trust, institutions, renown, etc.

There are many actions and policies an individual can use to preserve and enhance trust in a relationship. Similarly, there are many actions and policies an organization can use to preserve and enhance general trust in the organization's complicated web of relationships, both relationships within the organization and relationships between the organization and outsiders. Some of these actions are related to minimizing the dependence on trust in a particular episode --- choosing actions that demonstrate "see, you can tell you are not getting cheated" rather than "haha, you will have to trust me". This has been in the news recently with the FBI announcing the destruction of five months of text messaging from Strzok (following episodes such as stonewalling Congress for months about other months of Strzok messages that weren't destroyed, and Strzok being deeply involved in the exoneration of Hillary Clinton for among other things mishandling and destroying a lot of records related to her tenure as Secretary of State). But it's just a recent episode of what I see as a *very* long-running soap opera.

E.g., consider a chapter somewhere nearer the middle of the run on the soap opera... Someday I expect to write a proper recommendation for Macaulay's _History of England_, but meanwhile, one of the many things I find interesting about that old book is that it is full of events that bear on individual and institutional trust, loss of trust, and consequences thereof. Many of those events even echo today, although not everyone knows where the echoes come from. For example, the book describes an event in the late 17th century that helps the reader understand what the US Founders would have thought of the significance of the right to petition publicly for redress of grievances.

Maintaining a high level of trust seems to be significantly harder in big organizations than in smaller ones. And conversely, consistently overriding the need for trust with various kinds of brute force --- e.g., maintaining power by hunting down and killing dissidents --- seems to be considerably more stable in large organizations. Of course violent intolerance can happen even in small groups, but as far as I know, that is more common in brief spasms, things like riots or power struggles. To my knowledge it is unusual for a system of intolerant despotism to remain stable for generations except in quite large societies, not just tribes or city states but larger kingdoms or empires.

This tendency of larger kingdoms is only one the interesting transitions which seem to occur as organizations --- notably governments, and subunits of governments such as the official church, or the body of officers commanding ships in the navy --- grow from somewhat below Dunbar's number to far above it. Some of those tendencies tend to give relatively smooth forward progress --- an increase in bureaucratic friction and cost, for example. The tendency to move away from trust to subjugation doesn't seem to yield such smooth progress, but instead yields a process of fits and starts, with interesting possibilities of sidetracking back to investing in trust-enhancing mechanisms. Perhaps one reason is that low-trust arrangements tend to be relatively inefficient, so an oppressive central authority tends to need a really impressive advantage in size in order to maintain enough of an advantage in power to crush a dissident group.

Some of the practices involved in maintaining trust are pretty obvious. (Don't get caught lying, even if that requires you lie less often than seems convenient. Don't destroy evidence when people might reasonably conclude that you did something wrong, duh.) An example of an obvious informal policy that helps maintain trust is to favor fashionable upper-class virtues that are at least somewhat admirable and/or challenging, e.g. speaking a diplomatic language, hunting dangerous animals, or playing a musical instrument. (Instead of inbred box-ticking that even lazy incompetents can "do" pretty easily, because they are fundamentally easy and/or because only the deplorable outgroup would be so vulgar as to measure the fashionable performance, e.g. cultivating a fashionable cuisine or accent or vice, keeping up with the latest shibboleths and in-groups of the dominant faction, pimping one's relatives to powerful people, conspicuously spending face time on pilgrimages to prestigious shrines or enrollment in exquisitely fashionable and politically correct institutions or at camps devoted to sacraments --- totally effective ones that only the vulgar would attempt to measure --- such as training each other to sponge spilled oil from baby seals.)

There are also a lot of practices that are at least clever, and sometimes subtle enough to be less than obvious to many people in many times and places. An example of something that is somehat clever, but nonetheless probably easy to reinvent, is "I cut and you choose" for dividing something messy (a haunch of meat, e.g.) with a comrade who doesn't completely trust you. Modern freedom of the press (and the older petitioning publicly for the redress of grievances) is an example of something clever that's probably harder to fully appreciate unless one inherits it (or something like it) and sees it in action. Its virtues don't seem to be obvious to everyone, but in can serve as a somewhat effective antidote for various kinds of corruption, and as a very effective antidote for various systems of imposing preference falsification on the subjects of the regime. Other old formal norms like trial by jury, or indeed more basic things like rule of law and public trials, also involve some cleverness and some deep consequences that are hard to appreciate until one sees them in action. Informal norms that are somewhat nonobvious include many "trust flows both ways" expectations in hierarchies, notably in the military. Anecdotally, it seems to be fairly difficult for people in low-trust societies (and, perhaps, in those subcultures of our society that have rejected high-trust practices) to appreciate how man-for-man effective a military organization with high-trust norms can be against a more corrupt one.

In the West, some of these practices seem to have been inherited from deep enough in the Dark Ages that we don't have a clear record of where they came from, but a lot of them might plausibly descend directly from the tribal organizations that succeeded in those periods.

As mentioned above, when political units are reacting to conditions of increased scale and increased population density, there is a tendency for low-trust arrangements to become increasingly stable. I think a large part of the reason is that even if low-trust organizations have trouble being as effective man for man, high-trust organizations are tricky to scale up to huge size, and huge numerical advantages are hard to beat in war (especially in ancient warfare on land, less so at sea, and perhaps also less so in modern mechanized and computerized warfare). But in Western history, there've been two notable episodes of a powerful current running the opposite direction, so that even as states got bigger, a fairly high level of openness and trust (and of man for man efficiency vs. other societies) was sustained. One episode was the development of movable type, which (in combination with related technologies like cheap paper and a phonetic alphabet) not only helped touch off a great avalanche of technical knowledge, it also seems to have fundamentally altered what governments (and official churches) could get away with at a given scale, and fundamentally made it easier for high-trust rival societies to scale up without their high-trust institutions falling apart too badly. Representative government is much older than movable type, but AFAIK stable representative government of countries the size of England or Holland or the USA only became historically important after Gutenberg. And Martin Luther's criticism of the Catholic Church was strongly foreshadowed by similar dissent for centuries before, but the Church knew its history, and had successfully stomped them all out before, and felt secure enough to continue practices (like indulgences) which were as flagrantly destructive of trust in the central value proposition of the Catholic Church as modern destruction of evidence and unbounded prosecutorial discretion are to the central value proposition of government. Despite the comforting lessons of history, the Church rather suddenly discovered ample reason to regret how it had complacently continued to liquidate trust in the decade(s) immediately preceding Martin Luther's theses.

Another episode, which has developed several times faster than the earlier printing press episode did, is the development of cheap digital electronic communication. Which among other things makes it relatively convenient for us to type windy weblog articles. But even as I type this, it is time for me to stop doing that, because it is time for me to start preparing for some real-world social activities instead.

Sunday, January 21, 2018

Ah, Wikipedia.

From Wikipedia:
In 1939, Manchuria was a puppet state of Japan known as Manchukuo, and Mongolia was a communist state allied with the Soviet Union, known as the Mongolian People's Republic.

Sunday, January 14, 2018

Trump, Twitter, and Shy Tories

President Trump has established a clear pattern of saying something
that people would naturally be expected to attack, but which they
can't attack without bringing attention to a subject which hurts their
cause. Often but not always he has done this on Twitter. Often, indeed
perhaps almost always, little considerations like whether making the
intended attack will hurt their cause don't stop people from attacking
him.

For example, this pattern seemed pretty clear in Trump's "Pocahontas"
remark about Elizabeth Warren. The remark was a fullthroated violation
of the taboo of using some stereotypical folk figure to refer to real
racial issues, so of course it was going to be criticized. But it
couldn't be criticized without exploding the Journolist-style working
agreement to bury the sorry Elizabeth Warren purely-fake-diversity
affirmative action affair. (Also, to an interesting extent, making the
people criticizing look like creepy villains for prioritizing the
oh-so-sensitive stereotypical-figure taboo over things that are more
central substantive violations of modern racial norms, such as using
false claims of racial identity to get affirmative action preferences
which were supposed to be for members of other racial groups.)

In many cases, including the Warren one, this plays out in a fairly
obvious predictable way, and I find myself pretty sure that Trump has
a pretty good idea what he's getting from this, and a pretty good idea
the price that he's paying, and that he's basically correct that the
political benefits for him are large compared to the political costs.
But in some messier cases, including the recent controversy over his
(vaguely second-hand sourced) remarks along the lines of "why do we
want immigration from that shithole", I can't see how anyone, me or
Trump included, could make the computation very accurately, in part
because it's difficult to anticipate people's reactions to the remark,
and in part because it's a "Shy Tory" sort of issue that makes it
difficult to observe people's reactions reliably. I can't think of any
wonderfully sophisticated professional marketer/politician survey
technology Trump could be using to overcome these problems, so it
seems as though he must be mostly "playing it by ear".

So far, eyeballing it from the outside, my impression is that he's
likely been pretty successful at it even in these messier cases, but I
wonder whether that impression will hold up in hindsight as we learn
more (e.g. from voting, or from other tells) about people's reactions.

Sunday, January 07, 2018

self-refuting arguments

Sometimes an argument is constructed so that it doesn't make much sense in its own terms. For example, saying that US constitutional law needs to be reinterpreted because standards have changed seems to bite itself in the tail: if standards have indeed changed, the Constitutional amendment process is right there.

Sunday, December 31, 2017

Merry Christmas, etc.

Merry Christmas! And a happy New Year, too.

2017 has been the kind of year that makes me think of wry remarks like "there is a lot of ruination in a nation" and "may you live in interesting times". And indeed, maybe the whole decade ending in 2017 was exhibiting a similar spirit. But perhaps I should be if not cautiously optimistic, at least open minded. Among other things, the more I have learned about history, the more I have been reminded that the baseline level of "ruination" and "interesting" tends to be pretty high, even in times when in hindsight we can see that things were looking up.

To a relative whose birthday falls very near Christmas, I sent a pair of gifts close to that year-end theme: Gorecki's Symphony of Sorrowful Songs as recorded in Poland in 1987, and Moynahan's Book of Fire. And if none of your relatives thoughtfully gave you such timely gifts, and you want to remedy the deficiency for yourself, and you don't need to stick to copyrighted works in order to make a traditional purchasable gift, you can try Macaulay's History of England, which is available free on Project Gutenberg (and which might make a subject for another post).

Saturday, July 10, 2010

two different perspectives on an investigation

Is it noteworthy not to look at the emails in question in context? And not to ask questions such as: "Prof Jones, did you delete any e-mails?" Some think so. Others think not. If only there were any commonly accepted standards by which we could judge one of them to be absurdly wrong...