iamwillsun

Two ships passing

2016-10-24T00:00:00+00:00

The tech industry has largely rushed to defend its business relationships with Peter Thiel by citing its commitment to “diversity”. This simplistic defense both conceals and exacerbates the very problems that the diversity rationale purports to solve.

Thiel recently donated a large sum to the Trump campaign, a donation that adds to his support at the RNC. Some immediately sought companies like Y Combinator and Facebook to condemn Thiel’s support and remove him as partner and board member, respectively. Both Sam Altman and Mark Zuckerberg let Thiel remain. Both condemned Trump to some extent, and both cited “diversity” as the reason why firing would be inappropriate. Since Sam’s post is more specific and compelling than Mark’s, I’ll work off that going forward.

If you’ve read my last two blog posts, you probably know where I’m going with this. Sam’s post boils down to:

Trump poses a serious threat to this country (and particularly women and minorities)
But Thiel’s support of Trump does not justify his removal because of diversity

Or in a more general form, akin to the discourse around the riots in Baltimore or free speech at Yale:

Agree that there is a problem A
But disagree that A justifies B, where B violates some Important Principle

The twist: the narrative is delivered by an actor who is arguably complicit in violating the same Important Principle.

Bright lines

Sam presents two lines of argument defending Thiel’s position as YC partner:

As repugnant as Trump is to many of us, we are not going to fire someone over his or her support of a political candidate. As far as we know, that would be unprecedented for supporting a major party nominee, and a dangerous path to start down…the polarization of the country into two parallel political realities is not good for any of us.

I don’t understand how 43% of the country supports Trump. But I’d like to find out, because we have to include everyone in our path forward. If our best ideas are to stop talking to or fire anyone who disagrees with us, we’ll be facing this whole situation again in 2020. That kind of diversity is painful and unpopular, but it is critical to health of a democratic and pluralistic society. We shouldn’t start purging people for supporting the wrong political candidate. That’s not how things are done in this country.

The first argument suggests that firing would result in a slippery-slope effect, possibly with societal consequences, whereas the second argument suggests that diversity of ideas is important. Sam draws an additional line: if Thiel were to say some of the things Trump has said, “he would no longer be part of Y Combinator.” These arguments, as presented, are woefully incomplete.

If I were to consider the set of possible reasons why I might keep Thiel around, they would break down into roughly three categories:

Personal. If I were friends with Peter Thiel, maybe I would also know his true political views, which may not include Trump’s discriminatory positions. I might therefore believe that his support of Trump is not intended to support those untenable positions, and maybe even believe that Thiel’s involvement with the campaign, in light of this private position, will have some salutary effect - maybe he could influence Trump to be less discriminatory. This defense requires an understanding of Thiel’s private stance and a belief that his support of the Trump campaign is not disqualifying. That is, the act of donating money to a candidate who supports discrimination does not outweigh the “benefit” of Thiel’s involvement or cross some line. Were Sam to hold any of these positions, it would not be possible for him to publicly use this rationale: it wouldn’t be his place to publicize Thiel’s private beliefs, and doing so may even sabotage any possible positive influence.

All of the above is pure conjecture, but it would be a plausible defense. I don’t think it’s particularly strong though; while it is almost guaranteed that a supporter will not hold every one of the same positions as the candidate, donating money to a candidate probably does little to change the candidate’s positions and merely amplifies his current set of positions. In Trump’s case, it seems highly unlikely that Thiel would be able to actually change Trump’s thinking about minorities or women.

This also gets at the bright line that Sam drew. Apparently Thiel would be fired if he said the things Trump has said, but donating money to a candidate who has said the things he has said does not trigger this. In order for this line to hold, one would have to believe that personally speaking is materially different than paying money to amplify that speech. Arguably amplifying discriminatory speech actually causes more of the damage that Sam concedes a Trump presidency would cause, so it’s unclear to me why personal speech is the line.

Professional. Perhaps Thiel meaningfully contributes to YC, whether through his work as a part-time partner or merely being associated with it. It would be reasonable to believe that cutting ties would hurt YC in the standing of the tech community, which largely respects Thiel and his brand of contrarian thought. This too would inevitably be a private defense - there would be no way that this would ever be conveyed publicly. It’s a legitimate reason, but it feels pretty trivial compared to some of the real issues of this election , which Sam describes with great concern in his post.

Structural. This is ultimately what Sam chooses to argue, but his argument isn’t well-specified. Sam basically leaves it as: firing Peter would be unprecedented and lead to X, where X is bad. “Unprecedented” doesn’t really add any value to this argument, so we can ignore that part. There are probably a couple plausible values for X: firing Thiel could further radicalize both him and Trump supporters, emboldening them to support even more discrimination, or alternatively, firing Thiel could embolden other actors to start “purging” people from their private organizations, which would degrade the liberal fabric of our society and lead to further fragmentation. These impacts seem highly problematic, but even assuming we evaluate this through a consequentialist framework, there exist a competing set of impacts that I haven’t seen well-articulated by those who use this defense. Here’s my best attempt at articulating them.

Sam argues that Trump poses an severe threat to the country, and certainly to women and minorities. Trump’s rhetoric and actions could lead to policy changes, or at the very least, develop a culture where women and minorities are no longer able freely exercise their basic liberties. One can draw a straight line between the threat that Trump poses to women and minorities, support of a candidate who threatens these groups, and failure to fully censure this support as complicity.¹ If Sam were evaluating through the lens of a consequentialist framework, then the fact that he understands the Trump threat (at least well enough to condemn it in the blog post) suggests that he must believe the aforementioned illiberalism impacts to either be of greater magnitude or higher probability.² I actually don’t think this was how he evaluated his decision - his diversity rationale suggests otherwise.

You come at the king, you best not miss

Sam argues that supporting diversity is “critical to the health of a democratic and pluralistic society” and removing Thiel is “not how things are done in this country.” This advocates for an alternative decision framework: assuming one supports liberalism, it is imperative to tolerate views that differ from your own. No cost-benefit weighing, this is the right thing to do full stop. It sounds great on paper (and probably also when you say it aloud), but this argument isn’t nuanced at all.

Sam already concedes that there is actually a line where he would fire Thiel: if Thiel said the things Trump has said, he would be fired. So maybe Sam doesn’t actually believe that tolerance is imperative, it’s just imperative when an individual doesn’t cross a certain line that is arbitrarily defined? This kind of under-specified argument irks me, because no matter the author’s position or intentions, it so utterly simplifies the discussion that you forget what exactly you’re even discussing. Trafficking in Important Principles like tolerance and diversity without any nuance is a terrific way to paint the complainants as destroyers of liberal society, because the discussion shifts from the specifics of the Thiel situation to whether we should uphold Important Principle (which inevitably is yes).³

The most notable aspect of this situation, compared to the discussions on Baltimore rioting or Yale speech, is that the Important Principle of diversity is exactly what the best argument for Thiel’s removal relies on. With the riots, we saw condemnation of rioters and a call for non-violence. With the Yale free speech issue, we saw condemnation of guidance on Halloween costumes and a call for free speech. One couldn’t really argue that the act of rioting is non-violent, or the act of reminding students not to wear racist costumes is in support of free speech. But here, calls for full censure of Thiel’s support of Trump, who threatens the basic liberties of women and minorities - that is fundamentally an argument of tolerance and diversity. Supporting a candidate who intends to use political power to restrict the rights of portions of society is support of intolerance. This is why a simplistic invocation of diversity and tolerance destroys any possibility of coherent discussion: both arguments for and against removal of Thiel use it.

Perhaps what makes the diversity rationale seem so jarring in this discussion isn’t really because of the issues raised above, but rather the understanding that the tech industry is notoriously bad at supporting diversity as an Important Principle. Just like government calls for non-violence or institutional calls for free speech reek of hypocrisy, one can understand how maddening it must be that a tech leader like Sam only writes about diversity when individuals that align with a politician who threatens diversity are being questioned. In this specific situation, invocation of the Important Principle without nuance isn’t just a distraction, it’s almost like trolling those who are actually working to make Silicon Valley a more diverse place.

It seems like any sufficiently complex discussion degrades into simplistic arguments about Important Principles.⁴ Though this aspect of discourse almost certainly has been around forever, I suspect the kind of fragmentation that the Internet allows has only worsened this phenomenon. In an era where you can easily (and unknowingly) surround yourself with only ideas that you agree with, understanding alternate viewpoints, and more importantly, engaging in productive discussion on the basis of that understanding, is only becoming more challenging. Sam is completely right that discourse matters, but if we’re incapable of engaging one another with restraint, precise language, and humility, I wouldn’t expect too much.

In this case, I assume “full censure” to mean in one’s full capacity as a private citizen: through speech, through private action, through association etc. As an aside, arguing that Sam has done a lot to defeat Trump is an insufficient defense against this argument.↩
There are some interesting pragmatic questions about some of these impacts. For the former illiberalism impacts, how does one balance the liberalizing effect of asserting liberal ideas of tolerance (i.e. that it is unacceptable for political power to be used to marginalize minorities or limit them from exercising their rights) with the “radicalization” or illiberalizing effect? Is there any way of communicating a decision to fire Thiel that would sufficiently mitigate the illiberalism impacts? For the latter self-preservation impacts, how does one compare the total loss of certain minority groups (e.g. from a Muslim ban) against the kind of social fragmentation resulting from radicalization? Even with a low probability, does loss of minority groups automatically outweigh the more nebulous illberalism slippery slope?↩
It’s difficult to ascribe motive, but Sam writes “if our best ideas are to stop talking to or fire anyone who disagrees with us…” as if the argument to censure Thiel is predicated on logic that we should censure “anyone who disagrees with us.” The strongest formulation of this argument is most certainly not this, but rather a discussion of the core liberties that liberal society must guarantee to all citizens, and whether a right to self-preservation takes precedence over liberal tolerance. ↩
I’m probably guilty of this myself. With my three latest posts all basically talking about the same thing, maybe I’m just force-feeding all of these situations through a framework that oversimplifies what is going on. I need to find some more topics to write about!↩

Failure to acknowledge

2015-11-10T00:00:00+00:00

Though it’s easy to understand the current protests at Yale as a debate over free speech, this simplification belies a more complex discussion that standard free speech positions fail to address.

If you read coverage of the events that have recently taken place at Yale, they predominantly focus on a pair of emails, one sent by the Intercultural Affairs Committee and another sent by Erika Christakis, associate master of Silliman College, as the flashpoint that triggered the recent unrest on campus. Whereas the IAC email suggests to students to be thoughtful about the costumes they wear for Halloween, Christakis' email raises the question of whether the Yale administration should be in the business of suggesting and censuring student costumes, which she argues is an exercise of free speech. The student backlash to Christakis' email and the public’s backlash¹ to the student backlash have been equally swift. The swiftness of these responses, and indeed my own initial response as well, gave me pause: were the patterns being recognized here the right ones?

Christakis articulates a fairly basic narrative around sensitivity and free speech:

There are “genuine concerns about cultural and personal representation”
But sensitivity does not justify guidance on free speech (e.g. institutional guidance on what costumes are ok)

This free speech narrative reminds me of the rioting narrative I described in my blog post about Baltimore six months ago, where public officials would

Express their sympathy for the death of Freddie Gray
Condemn the riots

As I discussed then, it was confusing to me why a public official would spend any time explicitly condemning the riots; doing so isn’t intellectually consistent, it doesn’t really stop rioting, and if anything, it just saps the strength of the nonviolent reform movement.

Arguably the free speech narrative suffers from similar flaws. In both cases, a protest movement forms around calls for institutional reform, whether a city police department or the administration of a private university. Some individuals choose to react with specific means, like destroying property or endorsing a framework of speech (e.g. suggest inoffensive costumes).² In response, the institution in question deploys the standard narrative, which highlights the trivial point that rights (e.g. property or speech) are inviolable, and in so doing, divert attention from the core of the issue. In the Baltimore situation, government officials would use the standard narrative; here, Christakis, an associate college master and therefore representative of the administration, does the same. Just as public officials have a responsibility to guarantee equal protection under the law, Christakis has a responsibility, as college master, to “the physical well being and safety of students in the residential college, as well as for fostering and shaping the social, cultural, and educational life and character of the college.”³ It’s through this lens that Christakis' choice to send the email that she did is similarly confusing.

Independent of the tactical blunders in Christakis' email⁴, she seems to ignore the fact that Yale is an institution which does not observe from the sidelines, it regularly endorses and censures speech, in ways far more pernicious than a suggestion to be thoughtful about Halloween costumes. Ta-Nehisi Coates' critique of the Baltimore narrative cleanly applies here as well: it is inconsistent to require those who lack power to forego institutional endorsement while simultaneously glossing over Yale’s sustained endorsement of speech against minorities.⁵

This is where the similarities end. With Baltimore, I argued that denouncing riots doesn’t really stop rioting, and furthermore, it delegitimized the reform movement by drawing attention to an extreme subset of actors. With Yale, though it’s unlikely that Christakis' position on free speech will placate anyone who is truly calling for speech codes, it’s plausible that it would be effective on a set of students who don’t hold strong opinions on this matter. Or to simplify, everyone understands that rioting is bad, but not everyone agrees on how Yale should handle free speech, so talking about rioting being bad doesn’t change anything whereas talking about free speech might. Thus, broaching the subject may have some value, but it also incurs some cost. By bringing in the rights-based discussion that she does, Christakis draws discourse away from the original topic of Yale’s treatment of minority groups, instead plunging it deep into the quagmire of free speech, which arouses strong emotional responses and severely complicates any discussion.⁶

Christakis' decision to deploy her narrative may seem dependent on a balancing of sorts: whether it’s worth promoting her view of free speech and starting a discussion at the cost of undercutting a protest movement seeking to address the marginalization of minority groups on campus, both by calling out the IAC email as well as by drawing out the subset of actors who hold more extreme views on the value of free speech (i.e. those who wish to have Christakis removed) and generating a backlash. However, this is a false dichotomy. It’s possible to initiate a robust discussion on free speech, while simultaneously acknowledging the context of the protest movement, and Christakis failed to do this.

Importing personal context

Despite the protest movement moving past the initial anger expressed by student op-eds or confrontations with Nicholas Christakis, the media continues to fixate on this. Some observers have commented that the extreme responses of certain student protesters are indicative of a larger phenomenon within higher education, that students have been “coddled” into weakness and oversensitivity.⁷ Though there may be some truth to this, I think these student responses are more indicative of something different. The stereotypical student who matriculates to an institution like Yale isn’t someone used to incivility. One way to approach this is to infer that Yale has made these students weak, but the other is to infer that these students are experiencing genuine pain. As one student describes it:

The protests are not really about Halloween costumes or a frat party. They’re about a mismatch between the Yale we find in admissions brochures and the Yale we experience every day. They’re about real experiences with racism on this campus that have gone unacknowledged for far too long. The university sells itself as a welcoming and inclusive place for people of all backgrounds. Unfortunately, it often isn’t.

This distinction is critical to understanding why the second inference, that this pain is real and not manufactured or imagined, may be closer to the truth. Those who make the weakness inference are implicitly comparing the stated sources of pain (e.g. an associate college master’s email, swastikas, exclusionary frat parties) against sources of pain that they’ve heard of in society (e.g. the right to vote, the right to marry) and drawing the conclusion that because these grievances seem so wildly different in scale, perhaps the sensitivity of these students are way off. The error here is importing societal context as a heuristic to judge the validity of the students' pain about things happening on campus.⁸ Evident from Aaron’s point above, the magnitude of pain in reality depends both on the source as well as the expectation, and in this case, the expectation is set by Yale, which touts certain ideals that it supposedly upholds. The notion that some of these sources of pain are illegitimate because everyone expects that “shit happens” in the real world is also precisely why it’s conceivable that a Yale student would be so upset. If Yale promised an environment where shit doesn’t happen, where overt and covert racism are both minimized, and this is in fact not the case, feeling disappointed and betrayed seems perfectly reasonable.

A second, related point is the observation that student protests typically are at the vanguard of larger protest movements, and it’s exactly because of the unique context established by institutions of higher education. Unlike many parts of society, universities are particularly adept at creating spaces for rational discourse, so students are able to more freely question why things are the way they are. What outsiders perceive as useless and overly-sensitive student protest movements is less a commentary on the quality of those movements and the sensitivity of the students and more on their own limited understanding of context and perhaps even their indifference.⁹

More broadly, the issue of missing context, of arguing without understanding, is something that is clearly becoming emblematic of discourse around these protest movements. In a world where discrimination is now generally covert, context matters that much more. Leveraging personal experience to understand the harms that another experiences is no longer as useful of a heuristic as it may have once been, and if understanding is a prerequisite to productive debate, the only way out is to start from an epistemic position that isn’t so presumptive.

See the comment thread on any article about the matter.↩
To simplify, here I distinguish the IAC from the institution; though they comprise of Yale staff members, one could argue that they are closely allied with the protest movement.↩
http://yalecollege.yale.edu/campus-life/residential-colleges ↩
The comparison between child’s play and racist costumes seems misplaced and condescending at best.↩
Fun fact: Calhoun College, one of Yale’s 12 residential colleges, is named after John Calhoun, one of the strongest proponents of slavery in American history.↩
So much writing on this topic conflates so many things, ranging from the classic “defending the right to speech v. defending speech” confusion to the idea that Yale, as a private institution, is obligated to use the exact same framework and observe the exact same compelling interests that the federal courts use, to the idea that there exists a broadly agreed-upon right to not be offended, etc.↩
The Coddling of the American Mind ↩
I describe this class of problems more fully in my blog post on epistemic humility.↩
Consider the Vietnam War protests, or the Tiananmen Square protests, or many other major protest movements, where student protest movements weren’t initially taken seriously, but eventually led the way.↩

Baltimore

2015-05-05T00:00:00+00:00

Many have recently condemned the unrest in Baltimore as “counterproductive.” At best, this standard narrative inadvertently diverts attention from the crux of the situation, and at worst it deliberately undermines any effort to seek redress and reform.

At the risk of oversimplification, to clarify what I mean by standard narrative, I mean the one-two punch of (1) what happened to Freddie Gray is sad, but (2) there is no excuse for rioting. Government officials, including the President and AG Loretta Lynch, have used this in their remarks. Even David Simon, creator of one of my favorite TV shows, has written as much. I don’t doubt the sincerity of the President, AG Lynch, and Simon in their calls for reform, but I find it confusing that they would waste their bully pulpit to explicitly denounce the rioting.

Condemning without understanding, understanding but not condoning

Ta-Nehisi Coates has a compelling takedown of this narrative:

When nonviolence is preached as an attempt to evade the repercussions of political brutality, it betrays itself. When nonviolence begins halfway through the war with the aggressor calling time out, it exposes itself as a ruse. When nonviolence is preached by the representatives of the state, while the state doles out heaps of violence to its citizens, it reveals itself to be a con. And none of this can mean that rioting or violence is “correct” or “wise,” any more than a forest fire can be “correct” or “wise.” Wisdom isn’t the point tonight. Disrespect is. In this case, disrespect for the hollow law and failed order that so regularly disrespects the community.

At the core of this is the tension between the violence of the state’s officers in the death of Freddie Grey and the violence of the rioters in their burning of cars and CVS stores. Coates attempts to bridge this by calling out the standard narrative as hypocrisy: it is inconsistent to require those who lack power to remain nonviolent while simultaneously glossing over the sustained, structured violence of the state against minorities. The critique then is predicated on the observation that the standard narrative does little to address the state’s violence; it paints Freddie Gray’s death as an isolated tragedy worthy of judicial examination but certainly not the result of a broken system operating off of broken policies with broken actors. I wouldn’t expect the President or the Attorney General to ever concede this much in public, but Coates’ point is that they must if they are to invoke the principle of nonviolence.

It’s not just intellectual consistency which Coates is calling the standard narrative out on. The implications of this lead to a second critique: failure to address this inconsistency is either an act of willful ignorance, political convenience, or deliberate sabotage. Coates rides with the middle option, pointing out that the very same people calling for calm are also charged with “enforcing the policies that led to Gray’s death” and because they “can offer no rational justification for Gray’s death,” they instead “appeal for calm.” Even a more favorable view implies a condemnation of the rioters without a proper understanding of what could possibly motivate such a group to act.

Effectively delivering this critique isn’t the easiest (admittedly my first read of Coates’ article left me confused as to what exactly he was advocating), but as Deray McKesson concisely puts it, “I don’t have to condone it to understand it.” This distinction is critical, and one that is likely lost on some individuals: the idea of either condemning or condoning rioting is a false dichotomy.¹ Both Coates and McKesson certainly condemn violence, but they choose to not make it a centerpiece of their rhetoric, as the standard narrative does. Since both clearly advocate for principled nonviolence, they could spend time denouncing both state violence and rioter violence, but they don’t really need to address the latter; their involvement in nonviolent direct action speaks louder than words.

Realpolitik

There’s a second reason why Coates and McKesson don’t constantly condemn rioters: they recognize that denouncing rioting doesn’t actually stop rioting, either removal of oppressive structures or offering a nonviolent alternative does. This is precisely what MLK described in a speech delivered in 1967:

I think America must see that riots do not develop out of thin air. Certain conditions continue to exist in our society which must be condemned as vigorously as we condemn riots. But in the final analysis, a riot is the language of the unheard. And what is it that America has failed to hear? It has failed to hear that the plight of the Negro poor has worsened over the last few years. It has failed to hear that the promises of freedom and justice have not been met. And it has failed to hear that large segments of white society are more concerned about tranquility and the status quo than about justice, equality, and humanity. And so in a real sense our nation’s summers of riots are caused by our nation’s winters of delay. And as long as America postpones justice, we stand in the position of having these recurrences of violence and riots over and over again. Social justice and progress are the absolute guarantors of riot prevention.

Assuming the above, then denouncing rioting, as in the standard narrative, has no positive impact. If anything, it focuses attention on the wrong actors and delegitimizes the reform movement. Rioting incurs a heavy cost, drawing support away from the protesters and towards the state, and talking about it in public forums only helps reinforce that cost.

That being said, even if rioting is prima facie bad, its undeniable that it can be an effective way of attracting attention. In the case of Baltimore, Obama pretty much admits this in his recent remarks:

The violence that happened yesterday distracted from the fact that you had seen multiple days of peaceful protests that were focused on entirely legitimate concerns of these communities in Baltimore, led by clergy and community leaders. And they were constructive and they were thoughtful, and frankly, didn’t get that much attention. And one burning building will be looped on television over and over and over again, and the thousands of demonstrators who did it the right way I think have been lost in the discussion.

Arguably the goal of the Baltimore protests is to raise awareness and bring about change at scale, and the primary prerequisite is some percentage of the populace caring about it. Combating apathy is therefore of the highest priority. Again, Obama:

If we really want to solve the problem, if our society really wanted to solve the problem, we could. It’s just it would require everybody saying this is important, this is significant – and that we don’t just pay attention to these communities when a CVS burns, and we don’t just pay attention when a young man gets shot or has his spine snapped. We’re paying attention all the time because we consider those kids our kids, and we think they’re important. And they shouldn’t be living in poverty and violence.

Obama gets it: unrest is media-worthy, and media attention is necessary but not sufficient to effect real change. Many others don’t get it though; they instead rail against the rioters and cite the Civil Rights Movement as evidence to demonstrate the power of nonviolent direct action. Though it certainly is an exemplar of nonviolent tactics, it’s important to recognize that the success of those tactics depended on particular contexts in the 1960s.

Michael Klarman, a Harvard Law professor and Baltimore native, developed the “backlash thesis,” which he uses to demonstrate the indirect role that Brown v. Board had on race relations. While many scholars have previously conceded the limited, immediate impact Brown had on school desegregation, Klarman persuasively argues that the Brown decision led to a backlash in the South which ultimately resulted in reform.²

While Klarman’s backlash thesis is remarkably interesting³, what is most relevant for the purposes of this discussion is the fact that the Civil Rights Movement became mainstream (i.e. received Northern white attention and galvanized them to act) only when nonviolent black protesters were met with brutal police violence. As Klarman describes:

By the early 1960s, King and his colleagues had basically given up on convincing southern whites of the wrongness of racial segregation and had redirected their energies toward converting northern whites to the civil rights cause by exposing the true evils of the Jim Crow system. Yet events quickly demonstrated that even blatantly illegal southern responses to civil rights demonstrations were not sufficient to arouse national outrage or to evoke a presidential response. Opinion polls from the early 1960s show that the public began to rank civil rights as the most important issue facing the nation only when demonstrations produced violence and social disorder, not when they led simply to mass arrests of peaceful participants.

To be successful, then, King’s strategy required the unwitting assistance of southern police chiefs in creating, or at least tolerating, sizable racial conflagrations. When southern law enforcement officials acted as Laurie Pritchett had in Albany - illegally but peacefully arresting civil rights demonstrators - neither the country nor the administration paid much heed. Moreover, because the public evidently tends to condemn even nonviolent direct action tactics - only 22 percent of those polled expressed approval of the Freedom Rides and only 31 percent of Freedom Summer - the civil rights demonstrations could succeed only if the public’s negative attitude toward the civil rights “provokers” was outweighed by its condemnation of their violent repressors. Appreciating this fact, King and his lieutenants devised the strategy of “creative tension”: Peaceful civil rights demonstrators would provoke and then passively endure violent assaults from southern law enforcement officers and mobs, with the hope of harvesting a public opinion windfall from a horrified viewing audience. The success of this strategy required not only that the demonstrators remain generally nonviolent and that their objectives be widely perceived as legitimate but also that such political figures as Bull Connor in Birmingham and Jim Clark in Selma “cooperate” by so brutalizing peaceful demonstrators as to mobilize national opinion behind a legislative assault on Jim Crow. As one SCLC leader put it, the movement “had calculated for the stupidity of a Bull Connor.”

King’s strategy required that the demonstrations take place before Connor was evicted from office. As Wyatt Walker observed retrospectively: “We knew that when we came to Birmingham that if Bull Connor was still in control, he would do something to benefit our movement. We didn’t want to march after Bull was gone.” The strategy proved brilliantly successful. After relatively lackluster initial marches that Connor met with uncharacteristic restraint, the dam soon burst, as Connor’s men deployed vicious police dogs and high-pressure water hoses against the demonstrators, many of whom were children. Television and front-page national newspaper coverage immediately followed, with photographs of police dogs attacking demonstrators and editorials condemning the violence as “a national disgrace.” President Kennedy reported that the famous photograph of a police dog lunging at a nonresisting demonstrator made him “sick.”

With this as a backdrop, consider how the Baltimore situation compares:

The public is more apathetic. This observation is motivated solely by the fact that technology has dramatically increased the number of alternatives that compete for the public’s attention, ranging from social media to mobile games. In the 1960s, the public was a more captive audience, and the speed with which public sentiment changed between 1962 and 1965 speaks to how efficiently news of the violent South spread.
The objective of protest is less clear. Whereas the organized movements in the 1960s could describe in specific detail the various forms of de jure and de facto discrimination, the Baltimore protesters have yet to present a clear articulation of what their demands are. This is less a commentary on the skill of Baltimore protest organizers and more a reflection of the current state of affairs. In the 1960s, one could very clearly point out how the Jim Crow laws directly resulted in state-sponsored discrimination, and one could reasonably present a case for federal legislative reform that would prevent state and local governments from engaging in such practices. In Baltimore today, there are no laws or police handbooks that openly specify mistreatment of minorities. What is at stake is perhaps more nebulous, less susceptible to legislative action, but yet equally weighty: how can we ensure the state’s officers remove their prejudices and fairly treat minorities? The Baltimore movements face a distinctly postmodern problem; just as there are no longer hostile nation-states to defend against, or entrenched political ideologies to fight, so too are there no longer racist laws to easily repeal.
The media is less equipped to tell a favorable narrative. In 1963, the mainstream media found the Birmingham violence newsworthy, and concisely captured the exact juxtaposition MLK wanted: representatives of the state violently attacking peaceful black protesters. One picture could fully illustrate that the government was systematically oppressing black Americans and that black Americans were peacefully registering their opposition to this oppression. Current harms, protest objectives, and legitimacy of the Civil Rights movement could be conveyed all in one go. I’ve already discussed why its difficult for the Baltimore protesters to delineate their objectives, so it’s no surprise that the media isn’t really capable of doing so either. Though the media can cover the Freddie Gray case, it’s primarily presented as a single incident caused by rogue officers rather than a byproduct of systemic, long-term violence.⁴ On the flip side, since there is no clear set of actors, the media presents a view of the rioting that tends to overestimate the criminality of the Baltimore citizenry as a whole, doing no favors for the nonviolent protest movement.
It’s more difficult to construct “creative tension” (i.e. police forces are no longer being led by Bull Connor). MLK depended on sanctioned police violence against protesters to receive the media coverage that would ultimately mobilize national support for civil rights reform. The Baltimore protesters cannot rely on such stupidity; the Baltimore police is sufficiently trained to limit provocation and violence against protesters, thereby limiting the possibility of media-worthy moments that support the protest movement.

In short, it seems substantively harder for the Baltimore protesters to motivate a broader swath of the populace to care. If rioting didn’t take place, there would be little reason for the media to dedicate more than a cursory amount of attention.⁵

To be clear, with my limited knowledge and expertise, I’m not sure I can conclude that a dash of violent unrest in Baltimore can achieve more than a strictly nonviolent movement, and even if I could, I’m not sure that the cost of violence to the community outweigh this incremental gain. But from a pragmatic standpoint, it’s nevertheless important to recognize that rioting can result in outcomes that are beneficial to protest movements, and these outcomes are not easily substitutable by any other form of direct action.

The discussion above only scratches the surface in terms of the numerous complexities that surround the current situation in Baltimore, and it’s because of this that I find the standard narrative so useless. It oversimplifies to the point of inanity and not only does nothing to address the issue at hand, it actually distracts from it. If you’re looking for a more succinct response to the standard narrative, Buck Showalter can help you with that.

I have no idea what Wolf Blitzer’s motivations are in that clip, but he just sounds stupid in the interview.↩
Klarman writes, “The unification of southern racial intransigence, which became known as massive resistance, propelled politics in virtually every southern state several notches to the right on racial issues; Brown temporarily destroyed southern racial moderation. In this extremist political environment, men who were unswervingly committed to preservation of the racial status quo were catapulted into public office. These massive resistance politicians were both personally and politically predisposed to use whatever measures were necessary to maintain Jim Crow, including the brutal suppression of civil rights demonstrations. There followed nationally televised scenes of southern law enforcement officers using police dogs, high-pressure fire hoses, tear gas, and truncheons against peaceful, prayerful black demonstrators (often children), which converted millions of previously indifferent Northern whites into enthusiastic proponents of civil rights legislation.”↩
Klarman has since also written about the gay marriage movement and how backlash against court decisions in the 1990s contributed to that narrative too.↩
One counterexample is this report from the Baltimore Sun on police settlements.↩
Hilarious and sharp commentary from the Daily Show on how little CNN cared about Baltimore.↩

Modern tragedy

2015-02-23T00:00:00+00:00

Stephen Kim’s story has no redeeming qualities.

The Intercept recently published a compelling narrative on Stephen Kim, a former State Department policy analyst charged under the Espionage Act for leaking classified information to the media.¹ I won’t seek to reproduce the details of his story here; instead I offer the following two observations:

[1] I couldn’t help but picture Kim as the archetypal tragic hero. Kim’s intelligence is only surpassed by his optimism and naiveté, and despite lacking strong political connections (which ultimately would have shielded him from his fall from grace), Kim fervently believes his solid analyses alone would help shape US foreign policy. This hubris results in his catastrophic partnership with James Rosen. Kim doesn’t go down without a fight, but it leaves him devastated. There’s a scene in the documentary where he’s standing in the beauty shop, blankly looking at the wigs and dyes he’ll be managing when he’s released from jail. He loses his money, his career, his wife and son; all he has left are wigs. Aristotle argued that tragedy should bring catharsis to the observer. I’m not sure I can extract anything positive from this.

[2] The government’s justification for Kim’s Espionage Act prosecution is ridiculous. It boils down to this: we need to look strong against leakers because we don’t like leaks, so let us get some prosecutions to deter future leakers. The problem is, there isn’t anything the Justice Department can use to prosecute Kim except a 100-year old law designed to punish spies acting on behalf of foreign powers.² As a result, Kim can mount virtually no defense³ and faces draconian penalties. To put in perspective how far most administrations wanted to stay from the potent Espionage Act, there’s one line in the Intercept article that’s worth repeating:

The Obama administration has prosecuted more than twice as many leak cases under the Espionage Act as all previous administrations combined.

With the courts fully deferring to the executive on these espionage cases, defendants are subject to the whims of the DOJ.⁴ These trials are pretty much useless: the executive branch controls everything. If we assume that the DOJ is operating rationally, it’s presumably prosecuting on the basis that making an example of one person, like Kim, through disproportionate punishment, while unfair to Kim, results in benefits that outweigh. This cost-benefit framework is overly simple though. Let’s use a more robust framework to consider this, perhaps one drawn from the very judiciary that has exited the picture.

If the DOJ were to reason like a court, they would recognize that deploying the Espionage Act may be out of line, perhaps implicating issues related to the 5th (due process) and 8th Amendments (cruel and unusual punishment). With Kim’s rights at risk, the DOJ-as-court would apply a balancing test and consider whether using the Espionage Act against Kim is justified by a compelling government interest. Yet, even if James Clapper, Director of National Intelligence, were to report to the DOJ that leak deterrence is critical to national security, that rationale alone is insufficient. Prosecuting Kim is not narrowly tailored to achieve the national security interest, since deterrence could be achieved through a lighter punishment or perhaps no punishment at all: given how the State Department apparently doesn’t properly train its officials on PR communication, that might be a more reasonable alternative to deter leaks.

Continuing to prosecute Kim, in spite of the above observations, makes no sense. This is about power, plain and simple, and it just so happens the system has been so utterly distorted that the DOJ can get away with it.

The documentary is particularly heartbreaking.↩
The US does not have a state secrets law like some countries do.↩
The law does not distinguish leaks for the public interest, or leaks of information that may have been improperly classified, or leaks that result in the identification of illegal programs, nor does the law require the government to demonstrate harm caused by the leak.↩
The Justice Department could have forced NYT reporter James Risen to testify against his own source: no court was going to block them.↩

Behavioral weakness

2015-01-31T00:00:00+00:00

Organizational culture, properly constructed, isn’t merely aspirational: it must recognize and restrain unproductive inclinations.

A few months ago, This American Life hosted a program that illustrates how broken the Federal Reserve Bank in New York was. Centered on Carmen Segarra, a sharp, confident lawyer who joined the Fed’s ranks after the financial crisis to help regulate other banks, the program proceeds to detail how she was effectively silenced by senior officials and otherwise rendered incapable of regulating Goldman Sachs. Carmen’s story compellingly corroborated findings from an internal report that the Bank commissioned post-crisis, in an attempt to understand why their oversight had failed:

As Beim and his team spoke to Fed employees about what went wrong leading up to the financial crisis, one problem in particular arose again and again. And that problem was the Fed’s culture. That’s the epiphany of Beim’s report. It was that basic. Culture.

Collated from numerous interviews, the report explains a set of systemic deficiencies by delving into the predilections of employees. One example, on the topic of deference:

A number of people believe that supervisors paid excessive deference to banks and as a result they were less aggressive in finding issues or in following up on them in a forceful way. Information asymmetry subtly influences the relationship between supervisors and the supervised. Banks have superior information about their business and are the primary gateway to information the supervisor needs. Banks inherently have an information advantage over the supervisors. To understand a bank’s exposures and risks the supervisor must request information and often explanations of what the information means. Getting good, timely information is therefore dependent on the willingness and enthusiasm of bank staff in providing that information. Supervisors need good working relationships with their firm, and they believe that a non-confrontational style will enhance that process.

Supervisors are commonly embedded within the banks they regulate; they work out of desks within the bank and a majority of their professional interactions and relationships on a day-to-day basis are with the bank’s employees. Though this may be operationally necessary, one can easily see how regulatory capture can take over. The last sentence in the above passage highlights exactly why a supervisor might justify deference to the bank; though as an observer, it may seem ridiculous¹, given the context, it’s all too understandable. This degree of analysis is key to cultural change.

In his study of the social sciences, Max Weber developed the concept of methodological individualism, or the idea that organizations should be considered solely as the “resultants and modes of organization of the particular acts of individual persons, since these alone can be treated as agents in a course of subjectively understandable action.”² This is precisely the way building organizational culture should be construed: culture is the set of principles and actions necessary to optimize individual behavior within the organization. Without a strong understanding of the behavior to be shaped, any attempt at building the edifice of a relevant and effective culture will fail. Far too often, (typically larger) companies will attempt to redefine their culture without recognizing the psychology behind individual behavior, and as a result, they end up with grand campaigns talking about how much the company embraces X without any real change.

At one point, the Fed’s report cites academic culture as an exemplar:

Consider how other organizational cultures encourage the best ideas to rise to the surface. Academic cultures, for example, value strong criticism because it will make everyone’s work better; such criticism is not dependent on organizational rank but is a function of intelligence and insight, which can occur at any level of seniority.

This may be an idyllic view of what academia actually is like, but nevertheless, it’s a good point of reference. Even though research groups and universities differ across many dimensions, one component is reasonably shared across all of academia: a desire to expand the realm of knowledge. The pursuit of truth leads to a second key characteristic of academic culture: a shared foundation on which discourse can take place. Opinions and ungrounded assertions are bounded within the confines of reason, and academics generally engage within commonly understood frameworks. These two aspects, a common goal and a common framework, essentially come for free with the self-selection that takes place when someone chooses to participate in academia.

Most organizations aren’t so lucky. Not only are individual motivations wildly varied, but people interact and decide on courses of action through highly heterogeneous means. One employee might solely wish to collect a paycheck, while another might solely wish to rise through organizational ranks. One employee might retain a strong ego and feel their opinions are world-class, while another might tackle business problems through a dispassionate, logical fashion. Small organizations can typically get away with not meaningfully constructing their culture, if only because they can (1) capably screen entrants for cultural fit and (2) spend enough time with each other to automatically shape the way each member operates. However, at scale, this begins to break down, and if left unchecked, one ends up with a culture like the New York Fed’s.

In order to properly construct an organizational culture, one principle stands out: identify behavioral weaknesses and buttress against them. Unlike the traditional, aspirational way of thinking about culture, identifying weakness actually helps ensure that cultural design is grounded, that it appropriately recognizes the intrinsic behaviors it seeks to modulate. For example, one might wish to have an organization steeped in rich communication. One approach could be to have everyone give PowerPoint presentations every week on what they’re doing. But, recognizing the cost of engendering self-promotional behavior and elevating charisma above reason, a culture of PowerPoints could actually serve to undermine the ideal of rich communication.³

Likewise, if an organization seeks to encourage open, constructive critique, it would be apt to closely examine current forums for critique (do they exist? how do they operate?) and existing behaviors (does anyone critique what senior leaders say? how do they do so and what kind of response do they receive?). Leaders frequently fail to realize that their position alone adds a degree of weight to their communication, so offering an opinion without setting the right context can silence critique.⁴ Acknowledging the power of the reporting structure, leaders would correctly scrutinize how they receive and engage with critique, as well as the degree to which that encourages others to speak up.

In the case of the Fed, its leaders should have known that regulatory capture, by virtue of how supervisors operated, was a clear and present weakness. As the report recommends, the Fed’s senior management must redefine the culture to specifically counteract this: by strongly encouraging “dissent", “individual initiative”, and “robust inquiry.”

An organization’s mission and values can serve as the aspirational goals that a culture can rally around, but to actually make that culture a reality, it’s critical to examine behavioral weakness and disproportionately defend against it. Akin to how the Supreme Court has identified suspect classifications and maintains specific approaches to protecting those groups, so too should an organization identify the weaknesses it most wishes to resolve and ensure that its culture is set up to assist.

Ben Horowitz has a succinct way of conceptualizing the impact of a strong organizational culture: “a small number of cultural design points…will influence a large number of behaviors over a long period of time.”⁵ Culture done right means less bureaucracy.⁶ Besides, no one likes bureaucracy.

Specifically: someone whose legal authority it is to extract information from the bank thinks that they can’t extract that information if they piss off the bank.↩
Economy and Society, Max Weber↩
PowerPoint slides are the worst. It’s a medium that encourages flashy visuals, and it’s limited space encourages simple ideas, buzzwords, and otherwise BS. I get their utility in certain cases, but for most business purposes, Jeff Bezos has it right with the short memo.↩
Jeff Weiner has a system to counteract this.↩
Programming Your Culture, Ben Horowitz↩
Don’t Fuck Up The Culture, Brian Chesky↩

Senior gift

2014-05-19T00:00:00+00:00

In the last few weeks of my undergraduate studies, I donated to Senior Gift. It was a choice I very quickly came to regret, and one year later, I’ll share why.

The standard pitch delivered by Senior Gift representatives relied upon two motivations:

Support the Harvard undergraduate experience (or show gratitude for your time at Harvard)
Support Harvard’s financial aid initiatives (or show gratitude for Harvard’s financial aid)

For the purposes of bounding this discussion, I won’t go into whether donations to Harvard are important - I’ll assume they’re important enough to warrant the existence of the Senior Gift campaign.¹ However, I take issue with how the Senior Gift campaign is run: as implemented, Senior Gift does a mediocre job of achieving goals that are difficult to achieve as is, while at the same time incurring significant costs that I find unacceptable.

The goal: participation rate

Total donations weren’t the success metric for the Senior Gift campaign; instead, it was participation rate. Participation rate is what the campaign highlighted in reports to the class, and it’s also the first stat that gets presented in any official materials about Senior Gift. Perhaps understandable given the low fund size,² but there are a set of other reasons that underlie this choice:

Senior Gift primes individuals for a lifetime of giving. By encouraging students to donate in their senior year of college, they will get used to parting with their money for worthy causes. Maximizing participation rate means maximizing the set of charitable graduates leaving this institution.
Senior Gift primes individuals for gifts to Harvard. Like the above, encouraging students to begin donating early will help reduce resistance to donation solicitations later on, and it’s important to maximize the base of potential donors.
Senior Gift is a key barometer for current donors. If gratitude, for the undergraduate experience or for financial aid, is adequately demonstrated by the senior class in the form of donations, current donors will feel more inclined to donate to an institution that delivers valuable experiences to its students. Certain gifts may even be explicitly conditional on a high participation rate.

I don’t buy the “lifetime of giving” argument. Donating to Harvard is substantively different from donating to charity X. The special relationship I have with Harvard will never come close to the relationship I might have with the Red Cross, since after all, I will never spend four formative years with the Red Cross. Getting a Harvard student to donate to Harvard is significantly easier than getting a Harvard student to donate to the Red Cross, and because of this, I would argue there’s very little generalizable impact that applies to philanthropic intent as a whole.

Priming donations to Harvard is more reasonable given the reduced scope, but this becomes interesting when we consider what the distribution of Harvard alumni donations look like. I don’t have any evidence, but intuitively, I would expect a long-tail of donors, with a handful contributing a very significant majority of total alumni donations. If such is the case, this kind of priming is all about helping shape the inclinations of as many students as possible, thereby maximizing the chance that one of those students will be rich enough to be one of Harvard’s big donors. The big donor model mitigates the importance of priming: if all that really matters is getting one big-shot from my class to donate in 20 years, it seems like targeted development efforts in 20 years will be significantly more fruitful than a broad, class-wide initiative today.

I have no insight into the importance of the last point, but I would assume that the actual participation rate doesn’t matter that much for donations (I would find it odd if certain donors have explicitly tied their donation to a certain Senior Gift participation rate). In the world of giving, one does see matching donations and total donation thresholds, so it could be possible donors will match at certain thresholds (e.g. if Senior Gift hits $50k), but I doubt this too.³

The Senior Gift campaign might incrementally increase the amount of total alumni donations, compared to without the campaign, but the lift feels small, and perhaps insignificant if alternative development initiatives are considered in lieu of Senior Gift.

The cost: privacy and coercion

Unlike a standard nonprofit, which might use selective targeting to solicit contributions, the Senior Gift committee possesses unique power. They not only target, but they know precisely who should donate (all seniors) and who has yet to do so. Combine this information with incentivizing committee reps to maximize participation rate and the fact that the reps are the peers of those who they’re targeting, and the result isn’t pretty. A few tactics I have a problem with:

Repeated solicitations. It’s one thing to receive a personal request to donate, it’s entirely another when those requests are repeated 5 to 10 times, in spite of requests to cease contact. If the purpose of donating is to show gratitude, that “gratitude” isn’t worth much when it’s under pressure from constant reminders. This undercuts all of the participation rate rationales discussed above as well.
Bulletin boards. I donated, but it was never brought to my attention that my name could be used to convince others to donate. My house committee chose to put up a list of all who donated in the house dining hall. I’m not sure why my personal show of gratitude needs to be used to convince others to show gratitude, other than the obvious group coercion effect.
Associates-level donations. If you donate more than a certain amount, you receive particular recognition and benefits. I thought that “all types of gifts are considered to be equal gestures,” but I guess not.

In short, the tactics used by the Senior Gift committee are fundamentally at odds with the spirit of what the campaign is trying to achieve. If Senior Gift is about priming individuals for future donations to Harvard, presumably by having students (voluntarily) associate their positive feelings toward Harvard with the act of donating, how do repeated solicitations contribute to this end? Squeezing the donation out of someone doesn’t necessarily mean that this link has been made, or to simplify, if you need A –> B, seeing B doesn’t mean A –> B happened. If Senior Gift is about showing other donors how grateful Harvard students are for their undergraduate experience, isn’t it disingenuous to solicit donations by resorting to blatantly coercive tactics? The expectation is very clear: if you aren’t grateful (where being grateful == donating), you should be, and the committee will beat that into your head as best they can.⁴ In juxtaposing these two goals, there’s an interesting dichotomy: in order to properly prime students to donate, there cannot be coercion, but in order to maximize donations to persuade older alumni donors, coercion is quite valuable.

In addition, if we assume that the Senior Gift is about showing gratitude for the best financial aid program in the country, attempting to convince people to show this gratitude is doing the exact opposite of what the financial aid program is supposed to achieve. No one who donates to Harvard’s financial aid program expects that they be recompensed, and no reasonable donor would expect that their donation to a student’s education comes with strings attached (e.g. the student must write a letter showing gratitude, or the student must donate $10 to Senior Gift). The point of financial aid is to give disadvantaged students the resources they need to proceed with their educational experience on as equal of a playing field as possible. If, at the end of four years, Senior Gift comes rolling along pushing the point that financial aid depends on these donations (which is of course only nominal), and gratitude for that financial aid is key, who gets impacted by this? This solicitation has no personal impact on the rich student who has nothing to do with financial aid. Instead, this disproportionately impacts those on financial aid.

This would be more ok if the campaign didn’t target and instead used class-wide tactics⁵, but the Senior Gift committee has the exact list of those who haven’t donated, and they use this list to full effect. The committee is able to target anyone who has initially refused to donate, and this target list very likely includes those who are economically disadvantaged.⁶ This is unacceptable, both as a privacy violation and for the coercion that results. The committee members were my peers, members of my college community, and they have no business possessing a list of fellow peers who have yet to donate, whether because of economic standing, principled resistance, or just being too lazy to show appropriate gratitude. To then act off this list and use repeated, targeted solicitations only further propagates the pernicious idea that recipients of financial aid should be grateful. Regardless of the exact makeup of this list of non-donors, there inevitably exists a meaningful subset of students who are on financial aid, and forcing them to show gratitude with this targeted approach incurs a severe cost.

This same line of reasoning doesn’t apply to thankathons, since they (1) don’t track attendance at the individual level, (2) they use class-wide solictations, and (3) they’re not asking for money. Conveying gratitude through a hand-written note is perfectly reasonable. But conveying gratitude by donating money? Especially when that gratitude is for having received financial assistance to even be at Harvard? It would be like asking cancer patients struggling to pay for chemotherapy to donate to the American Cancer Society to show their gratitude.

As Matthew puts it:

I am told to be grateful. The implicit message remains that I don’t get full financial aid because I deserve to be here; I deserve to be here only because someone covers my full financial aid. I am told that I had better be grateful. That’s how Harvard actually makes me feel.

I feel guilty when I’m back home, where my closest friends are struggling through community college. Here, I feel uncomfortable and small. I am reminded over and over again not only of my economic status but also that my economic status separates me from other, more fortunate students.

I feel like I should keep silent, because I feel that I have been made silent—I feel small. And because of Harvard, I do feel like a poor student.

This is precisely what robust financial aid was supposed to guard against: removing vestiges of old Harvard’s classism that would limit the experience of disadvantaged students. Yet ironically, Harvard has decided that the most effective way to support their financial aid fund is to run a campaign that improperly brings class distinctions to the fore. This is marginalization, and as a community, we should strive to minimize any form of marginalization. The problem is, so much rah rah and “gratitude” gets mixed into this discussion that a reasonable critique like Matthew’s is met with extreme hostility, which coincidentally fulfills the coercion argument.

The minimal benefits that might come with a high participation rate do not at all justify the privacy and coercion costs incurred by Senior Gift. I don’t need to demonstrate that there are 10 or 50 students that feel the same as Matthew - one student is enough. Matthew legitimately felt coerced, and no school-sponsored initiative should lead to such consequences.⁷

I accept the idea of Senior Gift, of having classmates organize a school-sponsored campaign to collect donations. But since it’s school-sponsored, the campaign should not reinforce class divisions by rewarding students who donate large sums with private events. Likewise, since the Senior Gift committee is composed of classmates, the committee should not have possession of the list of donors and conversely, the list of those who haven’t donated. Finally, the rhetoric used to solicit donations and to shame those who haven’t donated, by intertwining gratitude with donations, is inexcusable.

I valued my time at Harvard, and my special relationship with Harvard would be reason enough to donate. However, having my own peers co-opt the rich tapestry of student relationships built over four years to apply social pressure and targeted coercion for something with questionable value - that’s an outcome I’m unwilling to support, and it was my mistake to not have realized this sooner.

Apart from the special relationship alums have with Harvard, I focus on the following two questions when considering this: (1) do donations to Harvard uniquely achieve the “better leaders” outcome associated with having a socioeconomically diverse body of students? (cf. O’Connor’s opinion in Grutter) and (2) does a marginal dollar toward Harvard’s financial aid program achieve as much as a marginal dollar toward other institutions, like poor elementary schools? (I ignore the undergrad experience portion of the Harvard College Fund, since financial aid seems like a strictly superior cause assuming a utilitarian framework). In short, I don’t think Harvard wins on either of these fronts.↩
I certainly wouldn’t lead with the $45k that was raised by my class when reporting the success of Senior Gift, given how paltry it appears next to the Harvard College Fund or Harvard’s endowment in general. I would bet the average donation amount is in the range of $20.↩
Besides, the importance of Senior Gift to soliciting Harvard donations is partially dependent on how Harvard pitches this. If, for example, the Senior Gift results are the only piece of information used to convince alumni to donate, then of course a low participation rate would be disastrous. Merely saying Senior Gift is important means quite little to me, since one could characterize any initiative with non-zero impact as important if it’s the only initiative we have going. This is essentially a “least restrictive means” argument: just as the courts care about making sure any infringement on fundamental rights is the least restrictive way of achieving some end, I care about making sure that the development office is proceeding with the least coercive (I discuss this below) means to maximize alumni donations.↩
Consider comments on articles like these, or this ridiculous blog post.↩
If we take away the long-term priming rationales discussed above, I don’t really understand why current students should be donating to financial aid programs when 70 percent of them are on financial aid. It’s not like attending Harvard comes with a yearly salary; apart from internships and on-campus jobs, there isn’t much opportunity to change one’s overall financial status while still a student. It’s basically a guilt play, but fine.↩
Even if data demonstrated that those on financial aid donated more to Senior Gift, the data is biased because of the way Senior Gift is pitched and the way it harnesses the “I owe Harvard something since I got aid” mentality when the donor specifically wants the aid to be no strings attached.↩
Some may counter by arguing that the whole coercion impact is trivial: it’s just people sending emails asking for $10. Three responses: (1) it’s not just sending emails, committee members would walk around with lists of non-donors and their pictures to identify them for one-on-one discussions, (2) it’s not the quantity of money that matters, the idea that money is at all implicated automatically triggers the classist impact, and (3) Matthew’s anecdote is enough proof to me that there reasonably exist at least a handful others who feel the same way, and extrapolating further, a larger group of seniors who were subconsciously coerced or guilted into participation.↩

Transitions

2014-01-27T00:00:00+00:00

There are some things in life that one can fairly reasonably prepare for: say a test, performance, or athletic competition. Major life transitions generally do not fall in this category.

In a previous post, I discussed the problem of knowledge without first-person experience. A separate, but related topic, is the extent to which one can prepare for major life events with or, more likely, without previous first-person experience. Early in high school, I remember wondering whether I would be ready to leave home for college when it came time to do so. Though the act of leaving for college was bittersweet, I ended up being more than ready to go. What got me to that point was a confluence of life experiences, social influences, and maturation.

Going to college is one of those life transitions that’s comparatively easy - especially when you juxtapose it with things like marriage, parenthood, retirement, and death. It’s a basic argument, but I would assert that the ease of transition is directly proportional to how easily one can prepare for it. One can prepare for being an independent student in college by increasing the degree of independence in high school, which is typically what happens. One can prepare for marriage by spending lots of time together as a couple, which is typically what happens.¹ Preparing for the latter three transitions is a bit harder.²

I had the good fortune to tune in to Jim Fleming’s last episode of “To the Best of Our Knowledge,” a long-running program syndicated on NPR.³ The episode focused on the topic of retirement, with Fleming interviewing his friend, Parker Palmer, for his thoughts on retirement. At one point, they discuss how thinking of retirement inevitably includes thinking about mortality. Palmer notes:

The only way to go well into the big death - the death of your body, the death of your consciousness on this Earth - is to enter into the little deaths and practice dying.

Of course, practice makes perfect. But what did Palmer mean about practicing dying? He goes on to offer examples like failing to achieve a dream from his 20s, or losing a close relationship, or having a professional failure. If death is merely some combination of loss and failure, I guess Palmer’s examples make sense. However, it seems a bit too easy to equate a professional failure with death. After all, “no religion has ever achieved the status of a world religion without a doctrine of immortality.”⁴ Death is much more than the examples Palmer describes, and I think the sense that one can proactively prepare for it with “little deaths” is naive.

By association, the idea of fully preparing for a loved one’s death seems nearly impossible. With college, I was fairly confident that I would be ready when it came time. With death, I’m not so sure. Maybe it’s the irreversibility; if college disappointed, I could always transfer, but if the grief that comes with death is unbearable, there’s no way to get out of it.

A more reasonable (albeit darker) alternative to Palmer’s preparation via “little deaths” is the idea that life itself will take care of things:

Just as all political and historical change sooner or later disappoints, so does adulthood. So does life. Sometimes I think the purpose of life is to reconcile us to its eventual loss by wearing us down, by proving, however long it takes, that life isn’t all it’s cracked up to be.⁵

EDIT: Katherine shared this article with me; it discusses the decision to go to grad school in a similar light (and also has some choice quotes from Middlemarch).

The cohabitation effect seems to indict this kind of preparation though.↩
A corollary to my basic argument above: the ease of transition is inversely proportional to the amount of self-help books written on the subject. A quick Google Books search has 4m results for “parenting,” 28m results for “retirement,” and 419m results for “death.”↩
One of the voices of NPR that I will never forget, Fleming is retiring after 40+ years of radio broadcasting.↩
Julian Young, Friedrich Nietzsche: A Philosophical Biography↩
Julian Barnes, Sense of an Ending↩

More guns

2014-01-21T00:00:00+00:00

I wrote a post on guns about a year ago, so it seems fitting to revisit the topic.

A somewhat recent debate from the Intelligence Squared series focused on the “usefulness” of the Second Amendment. The debate itself was fun to watch, despite the fact that there wasn’t much clash. A few thoughts:

Sandy Levinson is thoughtful and articulate, and as the moderator highlights at one point, much too timid to debate with people like Dershowitz or Volokh. Dershowitz unsurprisingly was a huge troll, but compared to the others, reminded me most of a policy debater. Kopel and Volokh are two reasonably intelligent individuals who spent half their time arguing against gun control and the other half responding to Dershowitz’s trolling; unfortunately, neither of these were directly relevant to the constitutional question posed by the resolution.
Levinson makes the Brandeis “state = laboratory” argument, suggesting that political processes, especially at the state level, are the best way to tackle gun control questions. Judges have no particular training that equip them with the necessary expertise to adequately decide these questions, and Levinson notes how arbitrary it is for a justice like Scalia to define excluded groups.¹ Of course, just as it seems a bit odd that judges are getting into the specifics of how laws can govern gun ownership and use, one need only look at Roe v. Wade to see Blackmun demonstrate his extensive background in biology. Basically, judicial activism runs both ways.
Volokh makes the interesting point that the presence of the Second Amendment is useful because it helps temper debate. That is, by having the Second Amendment, reasonable conversations about gun control can take place: gun owners can feel certain that their fundamental right to own a gun cannot be taken away. In a vacuum, this probably makes sense, but as Dershowitz points out, this theoretical assertion doesn’t pan out very well in reality. Consider the failure of background checks in the Senate. Volokh counters by arguing that it failed because of the threat of an assault weapon ban, which “poisoned the well” and put gun owners on the defensive. This is a weak rebuttal at best, since this is exactly the kind of scenario² where one would assume the tempering effect would kick in. Even if Volokh is directionally accurate (which doesn’t seem difficult to prove), the magnitude of the tempering effect might be quite small after all. Again, in the world of abortion rights, the Ninth and Fourteenth Amendments, as well as the jurisprudence of Roe, doesn’t seem to reassure pro-choice advocates that abortion rights are secure at all.³ In this regard, pro-choice individuals and gun owners probably view restrictions quite similarly: with extreme suspicion and paranoia.

If I were to judge the debate, I’d vote on Dershowitz’s uncontested argument about why it makes little sense to constitutionally enshrine a “derivative” right (the right to bear arms) v. a “basic” right (the right to self-defense). Besides, Volokh’s concession on what a modern-day Second Amendment would look like (it wouldn’t include any language relating to militias or hunting, and would explicitly call out the right of self-defense) pretty much conceded at least half of the resolution.

See District of Columbia v. Heller, where Scalia mentions that felons and the mentally ill do not have the same Second Amendment guarantees.↩
Stronger background checks were apparently supported by >90% of Americans polled in Jan 2013.↩
The parallel doesn’t map one-to-one, since it’s arguable that the current judicial consensus on abortion rights is in a different place than the current judicial consensus on gun rights (the former having said rights in a more tenuous position).↩

Knowledge and judgment

2013-12-31T00:00:00+00:00

To what extent does incomplete knowledge hamper our ability to make judgments?

Donald Rumsfeld may have put it best:¹

There are known knowns: there are things we know that we know. There are known unknowns, that is to say there are things that we now know we don’t know. But there are also unknown unknowns: there are things we do not know we don’t know.

Rawls had this idea called the veil of ignorance.² Though the veil plays a nontrivial role in the way Rawls constructs his principles of justice (and correspondingly is the target of critique from all directions), I like applying a bastardized version of the veil when thinking about my own normative beliefs. It isn’t just a fancy way of saying to “treat others the way I would want to be treated”; it’s about considering how I would want to be treated if I were them and coming up with the principles that get me there. Ideally, this kind of thinking would divorce one’s moral and political beliefs from a myopic view of the world and hopefully make those beliefs more just. Unfortunately, it’s not that easy. Even if aligning principles with beliefs is simple,³ how exactly does one consider anything, much less a set of principles, from the standpoint of some other arbitrary individual?

I recently began watching The Wire, something I’ve been meaning to do for some time. A well-crafted show, The Wire gives a taste of what life is like in the Baltimore projects, humanizing drug dealers, enforcers, and robbers. Though my life experiences shed very little light on what it might be like to deal drugs in West Baltimore, if I properly adhere to the aforementioned veil requirement, a just consideration of current drug laws would require me to consider that first-person experience. In this case, The Wire might be able to help augment the dearth of experience I have on this subject, but is it enough? It’s unclear how one would know when a view of life is sufficiently robust to make reasonable judgments.

The problem gets even trickier when some life experiences are seemingly impervious to third-person analysis. In Levels of Life, Julian Barnes presents an unflinching account of the grief he experienced following the loss of his wife. Barnes writes:

We are bad at dealing with death, that banal, unique thing; we can no longer make it part of a wider pattern. And as E.M. Forster put it, “One death may explain itself, but it throws no light upon another.” So grief in turn becomes unimaginable: not just its length and depth, but its tone and texture, its deceptions and false dawns, its recidivism.

At times, it seems there is nothing that Barnes’s friends can do to alleviate the pain: those who refuse to talk about his wife out of respect piss him off, while those who do talk about his wife further his misery. Drawing upon one’s own personal experience or knowledge of grief serves no purpose; the basic act of deciding what to say to a friend in mourning is utterly impossible.

Ingrid Robeyns writes:

My worry is that this category of experiences, differences, practices, and other features of human life that we cannot understand without first-person experience, is much larger than we generally tend to assume. And that as a consequence, we believe that we know much more than we actually do know. And, as a further consequence, that we too often are wrong in our judgements of aspects of the lives of people significantly different than ourselves.

One might be able to recognize that judgments are more reasonable when done under the auspices of a veil requirement, but that isn’t nearly enough. These issues go all the way down: how do we know what we know? How do we know what we don’t know? How do we turn known unknowns into known knowns, and how do we know we’re doing that correctly?⁴ In the face of this mess, Robeyns is right to call for greater “epistemic humility” as a first-pass solution. Unsatisfying as that may be, perhaps that’s the best we can do.

Unsurprisingly, this odd statement received Zizek’s attention.↩
In a thought experiment known as the “original position,” individuals would come together as free and equal moral agents to agree upon principles of justice for their society. Rawls stipulated that the original position would have to have a veil of ignorance; that is, each individual would deliberate in the original position without knowledge of their race, gender, age, wealth, religion, etc. The veil would ensure that each individual would truly be situated on equal footing with each other. ↩
Not the easiest task in the world; philosophers call this “reflective equilibrium.”↩
In slightly more concrete terms, e.g. how do I know that drug dealers are bad? How do I know that I don’t know anything about how drug dealers operate? How do I learn about how drug dealers operate, and how do I trust that The Wire can teach me that?↩

Gamification

2013-11-24T00:00:00+00:00

Last month, I attended a conference in Vegas for a few days. The Strip, with all its lights and grandeur, does quite a poor job of concealing its depressing realities.

Consider this example: on a Tuesday morning, I’m walking through the casino/hotel to go to the conference center and see an elderly couple with a young boy walking in the opposite direction. I can’t think of any non-depressing contexts that explain why a school-age boy would be walking around a Vegas casino at 9 am on a Tuesday in October.

The worst are the slot machines. There’s a 100% correlation between people playing slots on a Tuesday morning and people who look utterly lifeless. In one of my seminars in college, we read a paper detailing the various gamification techniques that slot machine designers had perfected. Positive reinforcement with lights and sounds, specific schedules for such reinforcement, and disguising losses as near-wins are all textbook strategies. For such a boring game, I’m not sure if I should be more impressed at how thoughtful slot machine designers are or depressed by how many people are addicted to slot machines.

The purpose of that paper was to describe slot machine design principles that casual game designers could use to increase their game’s stickiness and “player enjoyment.” The authors use Tetris as an example of a game that could be improved.¹

Silicon Valley in particular has become much more knowledgeable about gamification in recent years. Zynga’s business is pretty much entirely based on that paper, taking otherwise mediocre games and transforming them into cash cows by layering on social networks, microtransactions, and enhanced game mechanics optimized for addiction.² It’s not just casual games; these kinds of strategies are now being used to get people to do all sorts of things, ranging from profile completion (e.g. Facebook gives me a progress bar) to exercise (e.g. Fitocracy) to discovering protein structures (e.g. Foldit).

More broadly, the tech industry understands the value of influencing an individual’s psychology. For example, it’s standard industry practice to have a growth team, which exclusively focuses on bringing in new users (e.g. optimizing signup flows) and retaining them (e.g. optimizing the time-of-day an email gets sent to users). Some tactics are more morally dubious than others, but usually the worst you can get is being too spammy.

However, consumer web products are so deeply tethered to real life that product decisions can have significant consequences. This recent article on cycling and Strava highlights one such example. The leaderboard feature almost certainly drove engagement metrics across the board, and I bet the company received positive feedback from users. All-in-all, a good product decision that leveraged elements of gamification. Except for the fact that an article like this could be written.

It’s hard to precisely disentangle Strava’s liability from the cyclists' liability, but no matter: the point remains that a consumer web product helped shape an individual’s psychology in a manner that resulted in tragedy. Even though this is a particularly extreme case, to what degree is this industry at fault for other negative repercussions? I’ve written before about the loss of privacy and authenticity, but I’m afraid it’s much more than that. Technology has the capacity to amplify and enhance every human vice and weakness, and a market environment that helps validate such meaningless forms of innovation is depressing.³

It might be reasonable to assume that gamification and other growth tactics are justified because of their positive impact on metrics, but this presupposes (1) the lack of unintended consequences (like cycling accidents or little kids hanging around slot machines on a Tuesday morning) and (2) that the underlying product is fundamentally valuable (such that getting more people to use it is a good thing). Though it might run counter to business sense, I think the Valley would do well not to build the equivalent of slot machines.

When presented with the option of self-administering drugs that stimulate the reward pathway, rats will choose to continually self-administer to the point of death. Slot machines are a nice analog: one can continually pull the lever until (financial) death. Vegas taught me that there are enough slot machines in this world.

As if Tetris wasn’t fun enough!↩
Unsurprisingly, all of these map to a corresponding aspect of slot machine gameplay.↩
Snapchat’s $4B valuation is justified, to the extent that Facebook’s $100B+ valuation is justified.↩