Friday 14 November 2014

Compilation: EA Epistemology

In preparation for the start of the first EA blogging carnival, I am creating compilation posts for topics on which various EAs have expressed opinions. An example of this that has already been done is Julia Wise's post on Giving Now vs Giving Later.

This post is on the epistemology of cause and charity selection.

Holden Karnofsky - Sequence Thinking vs Cluster Thinking
Nick Beckstead - Common Sense as a Prior
Brian Tomasik - Cost-Effectiveness in an Uncertain World
Carl Shulman - What Proxies to Use for Flow-Through Effects?
Jonah Sinick - Many Weak Arguments vs One Relatively Strong Argument
Jonah Sinick - Knightian Uncertainty from a Bayesian Perspective
Paul Christiano - Beware Brittle Arguments
Luke Muehlhauser - Model Combination and Adjustment
Eliezer Yudkowsky - The Weak Inside View
Philip Tetlock - Expert Political Judgment: How Good Is It? How Can We Know?
Robin Hanson - Beware the Outside View
Peter Hurford - Why I'm Skeptical About Unproven Causes









Tuesday 11 November 2014

EA Blogging Carnival?

I discovered the concept of a "blogging carnival" through the asexual community. For all I know, they invented it.

How it works is that each month, a different blogger "hosts" the carnival by selecting a topic. Everyone interested in participating for that month then writes a blog post about that topic. The host then writes up a post linking to all the submissions. Here is an example of what a link roundup would look like.

Given that there are so people blogging about EA issues, I think this concept would transfer really well into the EA community. Each month, have somebody choose a topic such as "Donating Now vs Later" or "EA Outreach" or whatever and then accept submissions from the various contributors.

One advantage of blogging carnivals is that they draw attention to a variety of blogs, including those that fly under the radar.

A second advantage is that they provide a variety of voices on one topic. Rather than just reading one EA's thoughts on cause prioritization, you could get 8 of them side-by-side. There's something more democratic about hearing from multiple competing voices.

This then creates mini crash courses on what EAs tend to think about a particular subject at a particular time. If I want to read about issues regarding EA and religion, I can a whole bunch of posts from different people that take a different approach to the month's theme. I see this being especially useful to outsiders that want to catch up on EA quickly. 

Thursday 6 November 2014

My Briefest Summary of Artistic Value

A lot of people seem to believe in something called "artistic value." The idea is that some art is better than other art according to some set of criteria. Intuitively, this makes a lot of sense. If artistic value doesn't exist then we have to admit weird things like that any child's scribbled drawings are artistically equal to the Mona Lisa.

The problem with conversations about artistic value is that nobody can define "great art" in a way that satisfies everyone else. Some people think great art must be timeless and universal. Other people think great art is art that expresses the most "truth" about what the world is really like. And so on. There's no real way to resolve these disagreements because there isn't much reason to favour one definition over another. You can define words however you like but nobody else is forced to use your definition.

Thus I think tailoring art to score within the goalposts of any particular definition of "artistic greatness" is kind of arbitrary considering the existence of plenty of equally valid definitions.

HOWEVER - I do think there's a way to sensibly evaluate art without appealing to artistic value. Okay, so maybe the Mona Lisa isn't inherently "artistically greater" than some kindergarden kid's drawings but that doesn't mean that art is a pointless waste of time and that there's no reason to try when you can just scribble. Art has plenty of positive (and sometimes negative) effects on the world - it can increase empathy, tolerance, improve social skills, deliver ideas across cultural boundaries, train people to think creatively, educate, etc.

These don't seem like arbitrary accomplishments (although there's debate about that too that I won't get into here). They actually improve the world. And so I think it makes sense to think of art in this way, as a force for improving the world, and that the most meaningful way to evaluate art is to evaluate the degree to which it can be shown to make the world better or worse.

This isn't to say that you can't also have a working definition of "artistic value" and then go around saying weird things like "Film A is artistically better but Film B is morally better." You might prefer Film A but what you happen to prefer is based on all kinds of factors that aren't directly tied to the social effects of those preferences. And I think it's easier to justify the legitimacy of social effects than it is to justify the legitimacy of your subjective preferences.

In most conversations, it'll make more sense to replace the symbol with the substance. If a novel is more creative than another novel, call it "more creative," not better. If a painting looks nicer to you than another one, call it "nicer looking," not better. It's unhelpful to insist on a definition of "artistic value" that you know other people will reject.