"The Revolution Will Be A/B Tested"

Reflections on Aaron Swartz's life and death, how to change the world, and making ourselves the people we want to be.

MIT Memorial Service

Below is the prepared version of the speech I gave at the MIT Media Lab last night, at the memorial service for Aaron. The video of the full event will be up later on the Media Lab’s website; for now we have videos of my speech and Aaron’s father’s speech.

It’s worth noting that the organizers of the event were clearly pressured to try to keep the event apolitical and noncontroversial — and especially to suppress any discussion of MIT’s role in Aaron’s case. To whomever was applying that pressure, you should know that no one at the Media Lab knew the contents of my speech in advance, and they cannot be blamed for them. No one could have stopped me from speaking. Second, I hope that my speech has persuaded you of the ethical problems with your decision to try to suppress free and open discussion of MIT’s mistakes.

Most important of all, Aaron’s life, legacy and ideals were all inherently political and controversial — as was his death. It’s not possible to honor him apolitically.

You should follow me on Twitter here.


Remarks by Taren Stinebrickner-Kauffman, MIT Media Lab, March 12, 2013

Like most of you, I have a strong affinity for MIT – I lived here for two summers, as a student and then later as a counselor at the Research Science Institute. My first kiss was on MIT campus (though not with Aaron).

When I first started dating Aaron, for probably our third or fourth date, I came up here from DC for July 4th weekend. We walked down Mass Ave to the river to watch the fireworks. I had to go to the bathroom, and for some reason he waited for me out on the sidewalk as I ran into the cafeteria building instead of coming in with me. I didn’t think much of it at the time. There were other odd things he did – like insist that I not leave the door to his apartment unlocked if I had to run down the hall to the washing machines. But I didn’t understand how those things fit together until later.

One day in July, Aaron called me. I was at Frisbee practice in DC. He said, there’s this thing that might hit the news tomorrow about me. Do you want to hear it from me or do you want to read it in the news?” I said “Well, I guess I want to hear it from you.” He said, “I’m going to be indicted for downloading too many academic journal articles, and they want to make an example out of me.” And I said, “That doesn’t sound like a very big deal.” He paused for a second and thought about it and said, “Yeah, I guess it’s not like anybody has cancer.” In the end though, it kind of was like that. I called him back later that evening and said, I’m sorry, I feel like I might have underreacted. I’m sure this is really stressful, being arrested and so on. And he said, “No, no, that was actually the most helpful reaction anyone has had so far. Please stick with it.”

The case took its toll over the last two years of his life, and I don’t think any of us actually realized how much of a toll it had taken, until later, until after he died. He hid it from us well.

Yesterday was two months since Aaron died. It’s been a hard week already, and it’s only Tuesday. Yesterday, I spent a little bit of time going through Aaron’s stuff, which is now in storage here in Cambridge, and I came across a couple of to-do lists. He always wrote them on the back of envelopes. One was probably from October or November. In between items like “make a dentist appointment” and “talk to Brian on my team” there was an item that said “Taren test,” and I didn’t know what it was. And what was especially interesting about this item was that it was crossed off. So I’m not sure what that was about. But it did make me think of one of Aaron’s ideals.


As the Media Lab said to me last week, the reason we’re here is to honor Aaron’s ideals. And this story made me think of one of Aaron’s ideals – iterative learning and testing, and in particular learning from mistakes. He was always testing things, especially his own assumptions. He was a scientist in the true sense of that word.

One of Aaron’s blog posts last year was called “Cherish Mistakes.” Let me read to you from it.

This is a tale of two nonprofits.

At one, they hate making mistakes. How else could it be? “We’re not ever going to enjoy screwing up,” they told me. But this attitude has a lot of consequences. Everything they do has to go through several layers of approval to make sure it’s not a mistake. And when someone does screw up, they try to hide it.

It’s only natural — you know you’re going to get in trouble for screwing up, so you try to fix it before anyone notices. And if you can’t do, then your boss or your boss’s boss tries. And if no one in the organization can fix it, and it goes all the way to the executive director, then he tries to figure out a way to keep it from the press or spin it appropriately, so the world never finds out they made a mistake.

At the other nonprofit, they have a very different attitude. You notice it the first time you visit their website. Right in their navigation bar, at the top of every page, is a link labeled “Mistakes.” Click it and you’ll find a list of all the things they screwed up, starting with the most horribly embarrassing one (they once promoted their group under false names).

And it goes on to discuss mistakes big and small, core and peripheral. They previously used flaky phones that would cut out during a call, annoying people. They were insufficiently skeptical in some of the most important claims they made. At times, their admissions have the tone of a chastised teenager forced to write an apology, but together they provide a remarkable record of all the mistakes, both crucial and mundane, you might reasonably make when starting something new.

It’s not that this group likes making mistakes — you can feel the annoyance and embarrassment seeping through the page — but they don’t shirk from them either. They identify their mistake, admit them publicly, and devise steps to avoid them next time. They use it as an opportunity to get better.

Mistakes are our friend.They can be an exasperating friend sometimes, the kind whose antics embarrass and annoy, but their heart is in the right place: they want to help. It’s a bad idea to ignore our friends.

That’s a hard attitude to take toward mistakes — they’re so embarrassing, our natural instinct is to want to hide them and cover them up. But that’s the wrong way to think about them. They’re actually giving us a gift, because they’re pointing the way toward getting better.

One of the really interesting things about this mandate to cherish your mistakes is how closely it aligns with MIT’s mandate. What do I mean by that? I mean that MIT is about science and about furthering human knowledge.

And there’s a funny thing about science, which is that you’re supposed to admit when you’re wrong. I heard a story once about an academic who’d been trying to prove a theory for decades. He got up at a conference on a panel with another academic. The second academic laid out a clear case, with convincing new evidence, for why the first person was wrong.

As an institution, MIT’s mission statement reads, in part:

The mission of MIT is to advance knowledge…in science, technology, and other areas of scholarship that will best serve the nation and the world in the 21st century. The Institute is committed to generating, disseminating, and preserving knowledge, and to working with others to bring this knowledge to bear on the world’s great challenges….We seek to develop in each member of the MIT community the ability and passion to work wisely, creatively, and effectively for the betterment of humankind.

MIT’s core worldview is the worldview of science, and your mission is the dissemination of knowledge and the betterment of humankind. And that statement describes Aaron to a T. This alignment of missions is why MIT’s involvement in Aaron’s prosecution and death are so tragic – and so deeply ironic.

You are probably all familiar with the broad outlines of the prosecution against Aaron, but you may not be as familiar with MIT’s involvement. There were many decision points MIT faced. MIT called in the Secret Service for a matter that could have been handled internally. When JSTOR reached a settlement with Aaron and publicly called on the prosecution to drop the case, MIT refused to join JSTOR. For years, MIT cooperated with the prosecution, freely providing them with evidence, documents, and access to staff, while Aaron’s lawyers had to fight for over a year to get to interview the exact same staff the prosecution had already spoken with. At any point in the two years preceding Aaron’s death, MIT could have issued a public statement saying that it did not want the prosecution to proceed, and the case would have gone away. After repeated private entreaties that MIT do so, Aaron and I had given up hope that your General Counsel’s office would listen to reason of its own volition. So, over the last 6 weeks of Aaron’s life, he and I were beginning to work with students and alumni of MIT, friends and supporters of Aaron’s, who wanted to help. For the first time, we were willing to make the case public — for most of our relationship, Aaron had been actively avoiding press and public coverage. These supporters were going to launch a website called Save Aaron. They were going to poster the campus with a slogan I’m personally very proud of devising: “Nerd != Criminal.”

Aaron’s missions and MIT’s were aligned. You both believe in the dissemination of knowledge. You both believe in bettering the world. So where did MIT go wrong? Why were the General Counsel of MIT, Greg Morgan, and his staff so bent on helping Steve Heymann and Carmen Ortiz at the US Attorney’s Office try to make an example out of Aaron and lock him up?


I think the answer lies in one of Aaron’s favorite books, Moral Mazes, by sociologist Robert Jackall. Jackall explores how managers and other actors in bureaucracies make decisions about moral questions — which at their root, most questions and decisions are. Let me tell you about it in Aaron’s own words, from another blog post in 2006:

Moral Mazes (one of my very favorite books) tells the story of a company, chosen essentially at random, and through careful investigation from top to bottom explains precisely how it operates, with the end result of explaining how so many well-intentioned people can end up committing so much evil.

This week’s scene takes place inside a textile processing plant at Weft Corporation, where the company’s poor low-paid workers are suffering from byssinosis. Byssinosis, also called Brown Lung Disease, is when your lungs fill up with cotton dust. Eventually your throat closes up and you suffocate to death. The company insists the whole thing is a stunt made up by Ralph Nader and other liberal do-gooders. But one day they change their tune:

"Weft, as well as all the other large and medium-sized American textile companies, was actually addressing the cotton dust problem, but in a characteristically indirect way. As part of a larger modernization effort, the firm invested $20 million in a few plants where executives knew such an investment would make money. … The investment had the side benefit of reducing cotton dust levels … One manager who was in charge of the project … comments on whether dust control was a principal factor in the decision…:

It was on these bases that the decision was made.

Publicly, of course, Weft Corporation, as do many other firms, claims that the money was spent entirely to eliminate dust, evidence of its corporate good citizenship. Privately, executives admit that without the productive return, they would not have—indeed, given the constraints under which they operate—could not have spent the money. And they have not done so in several other plants and only with great reluctance, if at all, in sections of otherwise renovated plants where it is more difficult to … achieve simultaneous cost and dust reduction.”

(Robert Jackall, Moral Mazes, 158f)


Corporate managers simply aren’t allowed to be moral, or even reasonable. And those who try are simply weeded out.

This book, I believe, helped Aaron articulate another of his core ideals: That we each have the same moral obligations in our professional lives as we do in our personal lives. That institutions per se do not deserve our loyalty. They deserve it only so long as they facilitate good in the world, and they lose it when they facilitate bad.

Most people don’t operate that way. Jackall put it like this: 

What is right in the corporation is not what’s right in a man’s home or church. What is right in the corporation is what the guy above wants from you…. Managers do not generally discuss ethics, morality, or moral-rules-in-use in a direct way with each other, except perhaps in seminars organized by ethicists…. (p.6, Moral Mazes, Oxford University Press, 1989)

 And elsewhere:

Bureaucracy erodes internal and even external standards of morality not only in matters of individual success and failure but in all the issues that managers face in their daily work. (p.194, Moral Mazes)

 Another of Jackall’s core points is that managers aren’t allowed to admit mistakes in bureaucracies. Most bureaucracies operate like the first of the two non-profits that Aaron wrote about in his blog post I read earlier. But scientists are supposed to operate like the second of the two non-profits.


So my question for MIT is this: Which are you first and foremost? Are you scientists, or are you a bureaucracy?

I respect MIT’s mission and the people here a great deal. There is no doubt that MIT made mistakes. There is no doubt that the persecution that led to his death has made the world a much, much worse place.

Immediately after Aaron’s death, MIT’s President Reif announced an investigation into MIT’s involvement in the case, headed by esteemed professor Hal Abelson. I was hopeful. I was hopeful that this investigation might be in the spirit of genuine science, of acknowledging and learning from mistakes. We can never get Aaron back, but MIT can ensure that this kind of injustice and tragedy doesn’t happen again in its community.

But since then I have become less hopeful. I fear that the investigation will instead be in the spirit of a bureaucracy. I fear it will be the kind of “investigation” that Robert Jackall might have written about: A PR exercise, a whitewash. I fear this because of the fact that the General Counsel’s office is itself involved in running the investigation, of which it should be the primary subject. I fear this because it has been two months and my understanding is that neither Aaron’s lawyers nor Aaron’s father have been interviewed by the committee running the investigation, nor has there been any sign that the report will be released until after media interest has blown over.

Another of Aaron’s ideals was asking challenging questions. And that’s why I’m here to ask those questions today.

So here is my challenging question for those of you at this institution who care about its ideals and care about your own moral compasses. Is MIT a scientific enterprise, working to fulfill its mission of bettering the world? Or is MIT a bureaucracy, operating much like for-profit corporations, interested primarily in promoting and protecting itself as an organization?

Here is how you will be able to tell:

  • If MIT releases its report in a timely fashion. It’s already been two months – how much longer do we have to wait?
  • If the report represents MIT’s critics in this matter fully and fairly;
  • If the report acknowledges serious mistakes by MIT;
  • If the report holds specific people and organizational structures accountable for those mistakes;
  • If the report refers to the morality of the actions of people at MIT in the context of universal ethics, not in the context of organizational security;
  • If MIT implements clear, actionable changes that a reasonable person would believe would have prevented Aaron’s persecution and death, had they been implemented before his case.

And perhaps most of all, and I need to give a little bit of background for this one: Aaron’s lawyers are filing a motion this week to lift the protective order on the evidence against Aaron. So far, the press and people close to Aaron, including me, have not been able to see the evidence against him, because it is under protective order. Most of this evidence would have been introduced publicly at trial had there been a trial, which is what we were all expecting. It is critical for us to be able to understand what happened and the extent of the malfeasance by the prosecutors’ office for us to be able to see that evidence and for journalists to be able to see that evidence. If MIT opposes public access to the evidence against Aaron, then we can be quite sure that MIT’s investigation is not proceeding in good faith. 


I had a nightmare a week after Aaron died. I dreamt that the two of us were living in a house, and we knew that there was somebody coming after us. It felt in the dream that if we could just secure the house, if we could just lock the doors, put furniture in front of them, board up the windows with 2 by 4s. We were running around, trying to secure the house. And I stopped for a moment, when it seemed like we had made a lot of progress, I stopped to make breakfast. I made Aaron’s favorite breakfast, scrambled eggs and Swiss cheese. And when I turned around, Aaron was suddenly dead.

In my dream I didn’t know who was coming after us. In real life, it was Steve Heymann, Carmen Ortiz, with the help of MIT General Counsel’s office. We all make mistakes, and nothing can bring Aaron back. But MIT has a chance to make a major course correction. Aaron would have respected an organization that could do so. The question is, will you?

  1. paraprosdokia reblogged this from tarensk
  2. jalbertbowdenii reblogged this from tarensk
  3. grumpykaz reblogged this from tarensk
  4. nivair reblogged this from tarensk
  5. louisedesrenards reblogged this from tarensk
  6. bln reblogged this from tarensk
  7. tarensk posted this