Links: Gender, Economic Progress, War, Caplan

WikiTribune Sad Must Be Better!!!

Apart from being an icon for reappropriated terms, fake news is viewed as a serious problem. To combat this problem Jimmy Wales, a co-founder of Wikipedia, has founded Wikitribune a “
news website in which professional journalists research and report news stories, and volunteers curate the articles. It’s like Wikipedia for the news, except perhaps maybe not as open. Wales intends this project to combat fake news.

There are two issues with this project. First, a problem about prioritization. Fake news is just not that large of a problem. A recent study found that single fake news stories on social media would have to have a persuasive effect of 36 television campaign ads in order to swing the election. Moreover, note that Fake news is not much of a new problem, there have always been widely disseminated media sources and stories which lack epistemic hygiene to the extent that they become propaganda. 

The second problem is that this project is devoid of ambition. The cause of fighting fake news is a part of the larger project, the pursuit of truth. But Wikitribune doesn’t seem to give us much more by way of truth. The idea: there are professional journalists whose content will be curated by fact checking and proofreading volunteers. At best, this will be marginally better than the New York Times combined with Snopes. Both are great, but they are things we already have.

Wikipedia has given us a wealth of information. It was consistently underrated by my teachers.  But Wikipedia doesn’t come close to what has been made available by specialists. Compare the Stanford Encyclopedia of Philosophy with philosophy pages in Wikipedia. No contest. Wikitribune may be trying to take advantage of specialists, professional journalists, but they have probably selected the wrong class of people. These are the same people who gave us the Times. Even selecting academics wouldn’t be much of an improvement.

What’s lacking is a unique (and better) incentive structure. Organizations like Wikitribune have the ability to take advantage of both specialists and capable volunteers and also experiment with incentive structures that have yet to be implemented in large organizations. If Wikitribune wants to be more ambitious they should go after truth in unique ways, not be distracted by fake news and become an ordinary media organization along the way. 

Cuomo College Catastrophe

David Brooks gives us eight reasons to think that Cuomo’s college is counterproductive. Generally, terrible proposals have between one to three things really wrong with them, the rest is decoration. The sum of decoration can make for distraction. It’s my impression that Brooks’ complaints here are an illustration of this principle.

Cuomo’s college plan, The Excelsior scholarship, makes tuition free at New York public colleges for anyone coming from a family making more than 100k a year. This will save a family making $100k a year, $26k. Not insubstantial. Students must attend school full time and graduate in two to four years to be eligible. This rules out 60 percent of students at four-year colleges.

Brooks’ complaints are:

  1. The law does nothing to help students of poor families making less than 50k a week. Their tuition is already covered.
  2. It doesn’t alleviate any difference to non-tuition fees.
  3. It doesn’t help part-time students.
  4. It demotivates students.
  5. It will, maybe, destroy some of NY’s private colleges as upper-class-middle-class students may be drawn away from them and to public colleges.
  6. More students will apply, which means schools can be more selective, which will continue increasing income inequality.
  7. NY schools rely on tuition to fund programs.
  8. Students have to stay in New York for four years after graduation.

You needn’t read the article anymore!

Brooks seems to think that college should have some cost (otherwise it will be demotivating), but that it currently has too high of a cost for poor families? But also that free tuition isn’t sufficiently motivating to be on track to graduate?

He notes that making college “free” allows schools to be more selective. This is true but schools are already very selective. Over the years, higher-ed schools have become better and better at selecting the top students. While this occurred, the top students are becoming less dynamic and more insulated. That is top students spend more time with each other, clustered geographically and socially, and less time with others. This increases income inequality, however bad that is. What’s worse is that is may decrease productivity.

Brooks concludes:

We’re all focused on Trump, but one of the reasons Trump was elected was that many of the people who try to use government to do good just haven’t thought things through.

Lovely line. But what is the good here that Brooks is after? Making college more affordable for poor students? Why think that that will be good for productivity or decreasing income inequality? He’s right to think, even if implicitly, that Cuomo is probably trying to ride the free school wave for a bit and not much else. But he risks doing much the same on the higher education wave.

I Believe In Free Speech But…

As you have probably heard by now, Charles Murray gave a speech at Middlebury College a while back, greeted with chants of:

Hey hey, ho ho, Charles Murray has got to go

Racist, sexist anti-gay, Charles Murray go away

Brief recap: anticipating trouble, administrators relocated the event “but the dialogue was cut short by loud protesters who slammed chairs, chanted and periodically pulled fire alarms, which shut down the building’s power.” No doubt, the protesters had read The Bell Curve and found the book wanting. Sorry, actually they didn’t have to read it to know that it’s racist nonsense written by a white nationalist that sort of thing is a priori. Anyway, they ended up injuring Middlebury professor, Allison Stanger.  You can read professor Stanger’s comments here.

The aggressors were, like the fascists in Berkeley, probably not students. But let’s not focus on the violence, unfortunate though it is, what deserves attention is that anyone would try to de-platform Murray, SingerPeterson, Mac Donald from speaking at all.

At the Singer event, the manager of campus security “explained that campus security respects everyone’s rights to free speech, including protesters.”

Join the conversation!

This seems to imply that the protestors had the right to drown out Singer’s talk! This is absurd. There is no right to drown out invited speakers, any more than there is a right to drown out local lecturers. Presumably, protestors don’t have a right to shout down Singer when he’s speaking at Princeton — where’s the relevant difference? It’s one thing to hand out pamphlets in front of Singer’s class, write letters to The Princetonian, or demonstrate before the lecture, but another to shout Singer down. Trying to prevent Singer or anyone else from lecturing at all through obstruction is infantile.

The view that the protesters, call them “de-platformers,” are justified has been pretty thoroughly refuted for more than a hundred years (see also Jason Brennan’s thoughts). Free, fair, rational inquiry and discussion is one of the best tools our civilization has. Often debates will be painful and difficult. But it is worth it. And the alternative is unstable, will be hijacked, and has a non-negligible chance of stunting serious progress. 

What then should happen to the protesters who actively shout down, interrupt, or heckle speakers? They should be expelled, put at risk of expulsion, or suspended. It is not yet clear to me yet what the optimal deterrent is, but I suspect it is one of the three above. Importantly, de-platformers should be treated in the same way the academically dishonest are treated. This is not hyperbole, these students are violating the spirit of free inquiry, the pursuit of the truth, and putting intellectual and moral progress at risk. 

In particular, the de-platformer is violating the norms of public universities in the same way that academic dishonesty violates such norms. The cheater is undercutting the spirit of a public university. It would be vastly better if public universities maintain a role of being places of learning and debate. Of course, our universities fail in that role and, of course, it may not be the only function that they possess (the other salient functions being mating grounds for students and signaling to future employers), but what matters is that public universities possess this role. The de-platformer violates this role. And they do so to the same extent that a cheater does, hence they should be treated in a similar matter.

Update: the president of Claremont McKenna appears to agree.

Caplan and Wellman: Is Immigration a Human Right?

Bryan Caplan and Kit Wellman recently debated open borders at USD:

I’ll briefly summarize the main arguments and then offer some thoughts.

Bryan Caplan, of course, argued in defense of open borders. His main argument takes after Huemer’s fantastic work. Sometimes open borders types are made fun of with phrases like “Should everyone be free to move into open borders types houses?” Caplan helpfully distinguishes between 3 views:

  1. Foreigners should be free to live in my house even if I don’t consent.
  2. Foreigners should be free to live in my house if I consent.
  3. Foreigners shouldn’t be free to live in my house even if I do consent.

Caplan’s view is the second. The restrictionists view is the third. But being able to restrict an individual’s freedom to invite others into their house is a bit odd, it’s certainly not something other private agents have the right to do. Caplan doesn’t think the state is any different. We do not consent to the state (as Wellman agrees), it has no justified special powers over our lives (though here Wellman may disagree). Hence, the state should no more restrict my guest’s freedom than other private agents should. Hence, we should be free to hire foreigners, sell them housing, and so on. This right to immigrate isn’t absolute, but the other side has to show that there are compelling costs, which is going to be especially difficult given the economic benefits.

Wellman dissents. Legitimate states have a right to political determination. The right to political determination demands a right to association. The right to association entails the right to not associate, that is, it justifies immigration restrictions.

In defense of this, we think that Denmark can’t just decide to merge into Sweden. This is because Sweden has a right to self-determination. If Denmark wants to merge into Sweden and the Swedes don’t want that, Fuck em. A merger with Denmark would completely alter the political landscape of Sweden. Question: what if the Danish all want to immigrate to Sweden? Maybe a few Swedes are opening their houses up and are willing to sell land to the Danish. Wellman thinks that the Danish have no such right to immigrate, the Swedish state should decide, using a just procedure, whether or not to open their borders. After all, opening the borders to the Danes would completely alter the political landscape of Sweden.

New question: what if Denmark really sucked. Hard to imagine [link], but suppose Denmark was really impoverished and a terrible place to live. Shouldn’t the Swedes open up their borders? Here Wellman takes a disjunctive view. Suppose one’s view of distributive justice entails that the Swedes have a duty to aid the Danish. The Swedes can discharge that duty either by opening their borders or by dispensing effective aid to the Swedes. The real force behind opening the borders, in this case, derives from one’s view of distributive justice, not a right to immigrate.

Wellman’s case is initially persuasive but begs the question. Here’s a matter which would completely alter the political landscape of Sweden: whether freedom of religion is enforced. Should freedom of religion be placed on the ballot or be decided by the usual political mechanisms? No. Because there is a right to freedom of religion. Here putting the law in the hands of the usual political mechanisms is out of the question! But then, why not say the same about immigration. “Because there is no right to immigration”— ah, got it.

To be fair, Wellman probably thinks the positive case for a right to immigration has not been provided. But we’ve been given strong reasons to think that there is an initial or prima facie right to— immigration restrictions are coercive and there’s a presupposition against coercive actions. In other words, violence and threats of violence are bad, we need a good reason to show that violence is justified.

Now the restrictionist needs to show what exactly will be this good reason (or reasons) in the immigration case. Elsewhere Caplan convincingly shows that the usual suspects are not persuasive.

Obligation vs Opportunity

Julia Galef has begun an interesting series called Open Questions. Here’s a sample:

1. When is overconfidence useful (if ever)?

2. Can we intentionally improve the world?

3. Are you motivated by obligation or opportunity?

I want to talk about the last of those questions. There are at least two interesting questions here:

1. Are we obligated to make the world a better place?
2. Is it better to be motivated by obligation or by opportunity?

These questions are largely independent, though Galef seems to blur them a bit. Both could be true. Perhaps 2 is true, but 1 is false.

I’m going to deny an understanding of 1 here and discuss 2 in a following post. However, there will be consistent thread through both posts as we’ll see. 

What do I mean by moral obligation? Let someone, S be morally obligated to X if it is the case that S morally should X or S is morally required to X. Here ‘morally’ just signifies the domain of moral actions. What fruit one should eat, what shirt one should wear, what movie one should watch are (typically) not in the moral domain. Maybe one should wear the appropriate dress to an interview, but that’s not a ‘moral’ should. On the other hand, when we say one should not lie, not kill, but should fight off existential risks, those claims are in the moral domain. Maybe there’s nothing fundamental in the division between the moral part of the normative domain and the pragmatic/prudential domain, that’s probably so, but let’s leave that aside for now.

There is no deep sense in which we are ever required to do anything. Sometimes it may be better to believe that one is required to X, at other times or stages in one’s life it may be better to not have that belief. Let something be morally useful to the degree that that something contributes to moral betterness. It may be better for Alice to believe that she is required to help others because that motivates her to help others more than the alternative. That belief would then serve a moral purpose for Alice, it would be more morally useful than its opposite. But perhaps Charles feels terribly guilty when he inevitably fails to help others as much as he could. Maybe this guilt brings about negative thought spirals that he finds difficult to escape  it would be better for others if Charles didn’t believe that he was required to help them. Claim: there’s no sense in which one is required to X, independent of how morally useful the belief that one is required to X is. Some states of affairs are better than others, sometimes it is better to think of ourselves as required to bring about those states of affairs, sometimes it may not be better to not so believe. Whether this is so will be an empirical matter, contingent on individual’s psychological profiles and the environment that they are in.

Here’s a sketch of an argument for this view. The argument is a reductionist one. In a flashback to the past, suppose we have a few grains of sand, 3 to be precise. We do not have heap. If we add 1 more grain, we will still not have a heap. Same for another grain.

Question: by transitivity, it seems like we will never have a heap, even after continuing this process until we have quite a lot of sand – but surely there are heaps of sand?

One view on this sort of issue, the issue of vague predicates, is that there is a strict cutoff between a heap and not a heap, being dry and wet, being bald and not bald. This view is wrong for the reason that there’s nothing in the world that could make true. Just what is it, what could possibly it be, about some additional grain of sand n that makes its addition to the collection of heaps transform the collection into a heap? Why n grains of sand rather than n + 1? These questions have no answers, which is illuminating.

Here the obnoxious skeptics of philosophy in undergrad intro to philosophy classes have something like the right answer – this paradox reveals tensions in our linguist or conceptual scheme, nothing more. Moreover, whether or not x amount of grains compose a heap at all is not a deep issue that goes beyond linguistic or conceptual matters. Calling certain collections of grains, heaps, is useful but not a deep ontological description of those grains.

A similar paradox applies to value. We think that genuinely helping people is better than not helping them. By genuinely, I mean to avoid cases where we “help” someone, say make them feel comfortable or giving them a handout, in a way that’s bad for them in the long term – I’m talking about cases where we genuinely help another. We aren’t obligated to help other’s in a very small way, we are not morally required to dispense grains of benefit. We can generate a similar sorites paradox here. If we aren’t obligated to dispense n amount of benefit or n + 1 benefit, then it seems like we are never obligated to dispense n + m benefit. One could also run this argument in the opposite direction, asking at what point does the supererogatory become the obligatory?

Just as we denied epistemicism, for heaps we should do it in the moral domain. Once we do this we should ask what makes it true that one is required to do anything at all. What the sororities paradox reveals, or ought to reveal, is that the question about heaps concerned merely conceptual and/or linguistic matters to begin with, empty ideas. Similarly, if it were true that we are morally obligated to give to charity, the idea’s truth would be due to the fact that it would be better if we believe that we have that moral obligation. Whether or not it is better that we believe that we have that moral obligation is due to that beliefs moral use.

Note that this argument doesn’t deny value realism or objectivity (though I’m not convinced that that’s an issue that matters much), but it does deny the idea that moral obligation is a part of the furniture of this world in any substantive way. At the bottom, the moral world is a matter of betterness or worseness. Talk of moral requirements add conceptual and emotional complexity to that picture, a picture of the world that is probably initially useful, though eventually, worse than other pictures.


John Adams once wrote:

There is nothing I dread So much, as a Division of the Republick into two great Parties, each arranged under its Leader, and concerting Measures in opposition to each other. This, in my humble Apprehension is to be dreaded as the greatest political Evil, under our Constitution.

The US has only two major political parties. The Republicans and Democrats have consolidated their political power. They have the money, the media coverage, and the voters. This has been the case for most of our short history and will probably be the case for the next half century. At least.

That there are only two parties is built into the political system.

The US voting system is a plurality voting system. Voters can only vote for one candidate and the winning candidate takes all. There is no prize for second place.

Contrast this with proportional voting systems: partly list, mixed-member, or choice. In a closed party list system, parties present a list of their candidates. Voters cast their vote by party. Parties then win seats based on the proportion of their votes they win. If, in a five-seat race, a party wins 40% of the vote, their top two candidates will win a seat. In an open party list system, voters will vote by candidate rather than party. Of course, there are different methods which determine how the proportion of votes map to who will take office — the earlier example being a bit simplistic. What’s important is that there is a prize for second place.

Mixed-member systems are hybrids. In this model, half of the legislators are elected by plurality contest, the other half are elected by party list. We are assured that this is considered one of the hottest systems by those in electoral design.

Choice voting, or single transferable vote systems, ranks candidates by voter preference. A threshold of votes is set (by various possible mechanisms), and if a candidate meets that threshold and has more votes than their competitors, they are elected. If no candidate meets the threshold, then the last place candidate is eliminated and the votes for the last-place candidate are transferred to the next choice. This process is more complicated and depends on several implementation details, but is nonetheless interesting.

We’ll probably cover these alternatives later in more depth. Again, what’s important about these systems is that there is a prize for second place.

Most proportional voting systems offer parties a set of incentives that will not result in two parties. In proportional voting systems, one can win seats without winning the popular, electoral or whatever vote. It can be better for a party to win second place than to compromise or build coalitions with rival parties. Not so for parties in the plural system.

Suppose you reside in a country with a plurality voting system. You’re a politician looking to 
stay in office, err, save the world. There are three parties. The greens, the purples, and the yellows. The purples are coming in hot with 40% of the support. But the greens and the yellows, your very own party, are not doing so badly with about 30% support respectively. I mean, 30% isn’t so bad, but in your system, since 30% is less than 40% and winner takes all, it looks like you and your comrades will get 0% of the seats. So, you could take the complete loss, or compromise and build a coalition with the greens. Note that if you lose, there’s a good chance that the purples will bend the political rules in their favor, perhaps by re-drawing districts, leaving you waiting for the debate invite, or decreasing the probability that your favorite demographics would vote.

Voters in a plurality voting system will also experience the pressure to compromise. Instead of voting for their favorite candidate, they may judge that it will be better to vote for a bearable candidate who has a higher chance of winning. For example, though many voters may have preferred Stein to Clinton, they voted for Clinton instead — because she had a chance. The same goes for those voting in the Republican primaries, who may have preferred Kasich but were forced to vote for Cruz or Trump because of their higher chances of winning.

The force to compromise has historically been strong enough to pressure parties into fusing with others. As the Whig party lost its place as a major party, those remaining Whigs joined the Democratic party. Similarly for the People’s party, 50 years later. Apart from taking some votes from major parties, third-parties have been nothing more than a footnote for the last century.

This is Duverger’s law.