Bi-Weekly "Metapolitics" Discussion in Fishtown

This discussion will explore some of the political essays from the Less Wrong forum, particularly the "Politics is the Mind-Killer" sequences. (Note: Less Wrong has several series of connected essays which they call "Sequences".)

The first set of essays all center around the idea that political issues are often multi-faceted and that all the "truth" on every issue doesn't tend to cluster around one standard political philosophy like conservativism, liberalism, socialism, libertarianism, etc. This is strange if you think of political philosophies in terms of scientific theories, where we expect to see a convergence of evidence known as "consilience" - https://en.wikipedia.org/wiki/Consilience

Note: In the philosophy of science, consilience is connected to the "unity of knowledge" and the approach known as "coherentism".

https://en.wikipedia.org/wiki/Philosophy_of_science#Coherentism

Each of the following articles is short, except for the last one by Scott Alexander. Hopefully some of you can read (or at least skim) all of them. For those who don't have the time to read everything, I've added in some notes with each article to give you a general sense of what they're arguing.

1. The Long Now Foundation, "Philip Tetlock: Why Foxes Are Better Forecasters Than Hedgehogs"

http://longnow.org/seminars/02007/jan/26/why-foxes-are-better-forecasters-than-hedgehogs/

NOTE: Yes, I know - this first article isn't from Less Wrong, but I included it because it supports the central issue at hand. Phil Tetlock's geopolitical forecasting tournament showed that the top forecasters were flexible thinkers who didn't try to cram all the data to fit into their preconceived ideological model. Tetlock's "Good Judgement Project" was closely followed by many in the Less Wrong community that felt that Tetlock's findings about the cognitive biases that undermine expert political judgement validates their heavy focus on learning rationality as a skillset.

2. Eliezer Yudkowsky, "Policy Debates Should Not Appear One-Sided"

http://lesswrong.com/lw/gz/policy_debates_should_not_appear_onesided/

• NOTE: Eliezer's essay explains why we can expect a convergence of evidence on scientific questions but not on policy issues - i.e. political policy issues usually deal with multi-factorial phenomena and so almost any course of action has both costs/risks and benefits - optimizing for one value means trading off another. As Eliezer argues, "A policy may legitimately have lopsided costs or benefits. If policy questions were not tilted one way or the other, we would be unable to make decisions about them. But there is also a human tendency to deny all costs of a favored policy, or deny all benefits of a disfavored policy."

A classic example of tradeoffs in business is the "project management triangle" (speed, cost, quality) where you can typically optimize only two of the three. In international economics, there's a similar trilemma called the "impossible trinity" (fixed foreign exchange rate, free capital movement, independent monetary policy).

https://en.wikipedia.org/wiki/Impossible_trinity

** NOTE: Although Eliezer doesn't mention this, another reason policy preferences don't have to converge is that even if both sides agree on the positive/descriptive elements of the issue (i.e. the objective facts that science can elucidate), they may not agree on the normative/prescriptive elements (i.e. the more subjective ethical aspects which philosophy deals with). This relates to the "is-ought problem" from philosophy:

https://en.wikipedia.org/wiki/Is%E2%80%93ought_problem

A practical example of the is-ought problem in economics is the "equity-efficiency tradeoff" where economists have to weight the relative importance of a society's overall productive efficiency and it's equal or unequal distribution of economic resources. There are both positive & normative aspects to this tradeoff: http://www.investopedia.com/terms/e/equityefficiencytradeoff.asp

3. Stefan Schubert, "Multiple Factor Explanations Should Not Appear One-Sided"

http://lesswrong.com/lw/kpj/multiple_factor_explanations_should_not_appear/

• NOTE: Stefan Schubert's non-coherentism went a bit further than Eliezer's, to the point he argued that the fact that people can generate logically valid moral arguments for both sides of a policy debate proves that "moral realism" is false. He also argues that the way the socio-economic, technological & ecological factors underlying grand theories of history like those devised by Jared Diamond & Steven Pinker all seem to line up looks suspicious and probably indicates cognitive biases (e.g. affect heuristic, confirmation bias) at work.

** NOTE: If you want to read more about the ideas Eliezer & Stefan brought up in the above 2 essays, check out Thrasymachus's post, "Beware Surprising and Suspicious Convergence." It's super wonky and we probably won't delve into the specifics in our meetup, but it's there if you're interested...

http://lesswrong.com/lw/n7w/beware_surprising_and_suspicious_convergence/

4. Eliezer Yudkowsky, "The Scales of Justice, The Notebook of Rationality"

http://lesswrong.com/lw/h1/the_scales_of_justice_the_notebook_of_rationality/

NOTE: Eliezer describes how if all you have is a "good"/"bad" binary for judging policies, it makes it difficult to choose between many real-world alternatives; for example, a nuclear reactor that is very stable but generates a lot of waste, versus a reactor that generates less waste but is less stable.

In our meetup, we've often discussed one major example of how a "scales" approach fails to adequately describe a complex situation - i.e. the one-dimensional left/right political spectrum. Some political scientists have created 2 dimensional political spectra for social & economic policy (e.g. Nolan Chart), and some have even added a 3rd dimension for foreign policy and a 4th dimension for something else. Eventually, you can keep adding more dimensions or subdividing existing dimensions to make your model more realistic in portraying all the alternatives, but this makes it increasingly complex to the point it's less intelligible & thus less useful.

5. Eliezer Yudkowsky, "Reversed Stupidity is not Intelligence"

http://lesswrong.com/lw/lw/reversed_stupidity_is_not_intelligence/

NOTE: Eliezer's point about reversed stupidity is contested by Rationalwiki's concept of "fractal wrongness" - i.e. the idea that certain irrational people or groups are wrong about everything on every level. Fractal wrongness is related to the commonly heard polemic that reality/facts have a liberal or conservative bias, and the charge that one's political opponents have undergone "epistemic closure" (i.e. systematically ignoring evidence that contradicts their ideology).

http://rationalwiki.org/wiki/Fractal_wrongness

Many ideologues believe their opponents are "fractally wrong" but Eliezer calls this into question. His essay suggests that even the most irrational people or groups are "stopped clocks" that are right occasionally by chance merely because they always point in the same direction. Examples of stopped clocks include hawks & doves in foreign policy, permabears & permabulls in the investment field, doomsters and boomsters among development analysts, etc. http://rationalwiki.org/wiki/Stopped_clock

6. Scott Alexander, See Part II of "The Eighth Meditation on Superweapons and Bingo" where he talks about low & high threshold values as it relates to politics.

http://squid314.livejournal.com/329561.html

NOTE: The "threshold effect" Scott describes is a type of "nonlinearity" - i.e. a phenomenon where output of a system is not directly proportional to input. More specifically, it's a case of "diminishing returns" as input increases and output reaches its peak, then "negative returns" as input continues to increase but output declines. (The "Laffer Curve" for taxation is a classic example from economics.) In cases where the peak that separates increasing from decreasing returns is sharp, it may be called a "tipping point".

7. Scott Alexander, "A Thrive/Survive Theory of the Political Spectrum"

http://slatestarcodex.com/2013/03/04/a-thrivesurvive-theory-of-the-political-spectrum/

• NOTE: While Scott's thrive/survive theory of politics appears to be original, it bears a striking resemblance to the results of the World Values Survey (shift from "survival values" to "self-expression values") we discussed at our last meetup on the conflict between globalism & nationalism. While reality may not have a "liberal bias" (as Paul Krugman has argued), according to the WVS it appears history does have a liberal trend.

https://en.wikipedia.org/wiki/World_Values_Survey

The way in which economic development influences a society's values is also related to Nick Haslam's "concept creep" - i.e. expanding definitions of harm to include things like "microaggressions" - which we discussed back in our meetup on social justice warriors. Haslam connected concept creep to Steven Pinker's civilizing processes described in "The Better Angels of Our Nature."

Of course, Stefan Schubert called the one-sidedness of Pinker's civilizing processes into question in his essay above, which may indicate that Stefan's (and Eliezer's) "non-coherentist" view of politics & history conflicts not only with Pinker's theory, but also the WVS graph and with Scott's thrive/survive theory of politics.

If we took Eliezer & Stefan's non-coherentism approach & turned it into a critique of Steven Pinker, Scott Alexander and the WVS, we'd look for two types of disconfirming evidence - many of which have been brought up by cultural pessimists, Neoreactionaries, market permabears & Neo-Malthusian doomsters who all think we'll be in for a rude awakening in the near future:

1. Tradeoffs/ Downsides of Modernization & Liberalization: Loss of the "Original Affluent Society" (Sahlins), Conservation of Outrage/Happiness, Loss of Meaning/Wonder (Disenchantment), Loss of Organic Community (Anomie), Loss of Masculine/Martial Virtues, Loss of Freedom (Weber's Iron Cage), Sedentary Lifestyle & Diseases of Affluence, Environmental Toxins, Stress & Overcrowding

2. Countervailing Trends That Could Slow or Possibly Reverse Modernization & Liberalization: Iron Laws (Population, Wages, Rent, Oligarchy, Bureaucracy), Climate Change & Peak Resource Catastrophes (Oil/Coal/Gas, Uranium, Wheat, Water), Systemic Risk (incl. Existential Risk), Secular Demographic Decline, Minority Rule/Paradox of Tolerance, Rational Irrationality & Tragedy of the Epistemic Commons