Skip to main content

From a meme to $47 million: ConstitutionDAO, crypto, and the future of crowdfunding

The real Constitution is the friends we made along the way

Share this story

Photo Illustration by Grayson Blackmon / The Verge

Jonah Erlich is one of the core members of a group called ConstitutionDAO, a group that raised money to try to buy one of the original copies of the United States Constitution at an auction held by the high-end auction house Sotheby’s.

Raising money on the internet is a pretty familiar idea — there are crowdfunding campaigns for everything from new products to movies to scientific research to politics. But most of the crowdfunding we’re used to happens through a central service: Kickstarter, or GoFundMe, or Indiegogo.

But the DAO in ConstitutionDAO stands for decentralized autonomous organization. Jonah and around 30 other people formed the DAO and then offered people around the world the ability to donate to it using the Ethereum blockchain. In exchange, they’d get the right to vote on what to do with this copy of the Constitution and what the organization should do in the future. And this attempt to raise money was extremely successful: ConstitutionDAO raised something like $47 million and verified enough funds with Sotheby’s to participate in the auction.

I asked Jonah to come on and explain how ConstitutionDAO got started, what happened with the auction, what the actual plan to handle one of the only remaining copies of the Constitution was if they’d won, and what happens to all the money now. 

There’s a lot to this story that hasn’t really been told yet — and I think it’s a really useful look at the tension between the promise of crypto technologies, like smart contracts and DAOs, and the reality of getting a bunch of people together to do... well, anything.

The biggest question I had was why did they need this technology to achieve their goal. I think that’s the big question for a lot of crypto right now, and I really appreciated Jonah coming on the show to talk it over with me.

Okay, Jonah Erlich from ConstitutionDAO. Here we go. 

This transcript has been lightly edited for clarity.

Jonah Erlich, you are a software engineer and you are one of the core contributors to the ConstitutionDAO. Welcome to Decoder.

Thanks, Nilay. Glad to be here.

So from what I understand of your project, you wanted to buy the Constitution. 

What started as a joke on Twitter led to being a very real bid in an auction to buy a copy of the US Constitution. It was a really great experience to see and flex the power of the internet in just a week of trying to make this happen.

There is a lot to talk about with what happened with the bid, how the auction process actually went, what you plan to do with the money you have raised so far. You raised roughly $47 million — although that’s in Ethereum, so that number is moving all the time — but I want to start with the very basics. DAO stands for decentralized autonomous organization, but what is a DAO?

The definition of a DAO is a group of people that come together around a shared community and a shared resource. The most fun description I’ve heard is that a DAO is a group chat with a bank account. Essentially, what a DAO lets you do is have this very lightweight organizational structure to be able to move around assets and interact with the world.

We spend a lot of time on this show talking about organizational structures: very few are actually lightweight in the end. If you have a group chat with a bank account, how do you decide how to spend the money in that bank account?

“If DeFi is speedrunning the history of finance, then DAO is speedrunning the history of organizational structures.”

It depends on the DAO. DAO structures are really in their infancy today. What most of them do is it’s a voting process. Essentially, everyone who holds the token for a DAO will be able to vote. They’ll suggest a proposal: let’s say we want to buy the Constitution, as an example. All the token holders get a vote. There are more complicated and interesting mechanisms being worked on in the DAO world right now. One of the jokes in the crypto world is that if DeFi (decentralized finance) is speedrunning the history of finance, then DAO is speedrunning the history of organizational structures. Because DAOs are so new, most of them operate on this very simple structure of direct democracy.

If I had contributed to ConstitutionDAO, what was I actually buying?

A donation to ConstitutionDAO means that — if we were to have succeeded at our bid — we would’ve been granted governance rights over the documents of the US Constitution. We were trying to bid and win one of the remaining 13 copies of this document. At that point, the DAO would have voted on what to do with it. For example, we had museums lined up that were going to give proposals on how their museum should be the one to store and display this document. The DAO would also be able to vote on what text should be displayed alongside this copy of the Constitution. What message do we want to share with the world? Say the DAO won a copy of the Constitution, and we displayed it in a museum. We probably would have funds left over to give to a community that is really excited about doing things. The token holders would set the future direction.

I read on the website that when someone contributed, what they were fundamentally buying was a governance token. The token was called $PEOPLE, which is very funny — buying $PEOPLE has a dark connotation.

Yeah. It was a dark side. This was something we moved very, very quickly on. We were actually planning on rebranding the token to a name that was a little more positive.

How were you going to rebrand it?

It was going to be $WTP — We the People. That was the leading idea. We really want this DAO to be a positive thing, so we really listened to the community when members said they were uncomfortable with the name of the PEOPLE token. That was something that was chosen very quickly, and we wanted to adapt to what the community wanted.

When someone buys tokens named $PEOPLE or $WTP, do they get one vote per token?

We had not figured out the governance model. In the rush to be able to buy this document, there were a lot of details that we were going to figure out after we had won. We were actually discussing different governance models had we won. One of the big concerns is that some of the large holders in the crypto world — we call these whales — would have outsized control over what happens with the documents. Our big message is: this is a movement for the people. This is the people’s document, so we want to have it controlled by the people.

We were looking at different mechanisms. We were looking at the direct democracy voting: one token, one vote. We were also looking at one vote per wallet, which isn’t an exact proxy for one person, but it’s pretty close. Quadratic voting is another mechanism in which voting power is actually diminished when a person has more tokens: you don’t have a hundred extra votes for every hundred extra tokens, but rather a smaller amount of votes as that number of tokens increases.

When you say DAO is speedrunning the history of organizational structure, you’re describing speedrunning the history of democracy: setting up a voting system and figuring out how those votes will be representative of a group of people. However, a group of people has to make those decisions; not exactly democratic. How is that working for ConstitutionDAO?

There’s a good concept called progressive decentralization: it’s really hard to actually start something fully decentralized, so you have to start with an initial team that builds something, who then brings it to decentralization. Our plan with ConstitutionDAO [was] that, once we win this auction, we’re going as fast as we can to becoming fully decentralized to the greatest extent possible.

A lot of what I hear about with blockchain projects is trust and verification. What is the mechanism of accountability for that process? If I buy into this project, you have promised to decentralize it as fast as possible. What makes you do that?

Everyone who was involved with this project was very public. The signers on our [multisignature wallet] were very public. We worked with an excellent nonprofit called Endaoment who helps with on-chain giving to take custody of the documents and do our bid on behalf of Sotheby’s. The personal reputations of people who are trusted within the crypto community were on the line, getting us to this point of decentralization. That’s really what we were leaning on for the beginning of this process. What happens with a lot of new technologies is unfortunately you can’t do things perfectly the first time. We’re still very, very early with the capabilities of DAOs and the capabilities of crypto. We wanted to be able to move as fast as possible and get things done while keeping that promise that we are going to get there. I felt very confident that had we won the document, we’d be getting there very quickly.

The reason I’m pushing on it is because I hear from the crypto community that you don’t have to trust people: the software will do it for you in some way. We can build mechanisms of accountability that are automated, transparent, or verifiable — all without some central control. In reality, quite often you have to trust a bunch of people to do what they say they are going to do. Did you perceive that tension as you were racing through this project?

You are 100 percent right. There was trust you had to put into the ConstitutionDAO core team. Initially we had what’s called a multi-sig — basically a smart contract that would hold the funds that were controlled by 13 different signers. Seven of them had to sign to move the funds. Initially, for their safety, those signers were anonymous, because it was a lot of money and we were concerned about their personal safety. But we heard from the community: “Hey, we’re not okay with this.” Many people didn’t want to contribute because of the anonymity, so we asked the signers to volunteer to let us know if they are comfortable sharing who [they] are so we could put them on our website. Everyone decided to volunteer. It was definitely an iterative learning process.

The trust aspect is something that we definitely want to minimize in the long run, and I think what’s really interesting here is how to connect the physical world with the crypto world with these trust assumptions. How do you make sure the physical document is actually governed according to the will of the token holders? One of the ideas that we explored was moving the DAO into a 501(c)(3) where the bylaws would require signers to act in accordance to the will of the token holders.

The will of the token holders expressed through a voting scheme that you had not determined yet.

Yes. There’s great tooling out there, but we had not officially determined anything yet.

If you have to trust a bunch of people — some of whom I believe you knew personally — and in the end, voting just implies a level of politics, why do you need a decentralized worldwide computer network to manage that? Really, you’re just buying shares of a thing that provide you some governance rights. That describes every corporation in America.

The way that I like to think about this is in terms of two primary types of agreements that we’re used to in the world. One is the agreements between people and society. These are laws along the lines of prohibiting theft or avoiding taxes. You’re not doing something bad to one specific person — although some of those do have individual parties — but rather it’s enforcing societal rules as a whole.

The other part of that is agreements between individuals and agreements between entities. I think this creates a much lighterweight framework for arbitration for agreements between individuals, so we can put as much as possible into the code as opposed into legal documents. When it’s in code, it’s a lot easier to actually simulate what happens. It’s a lot more difficult to interpret that in a way that was not intended. You can put in those spots where you want human interpretability, but it creates a much more lightweight, efficient system for agreements between people. That’s why it’s really important that it’s on this decentralized structure, because it allows us to move a lot faster and more efficiently in that sense.

Put this into practice for me: I went to law school. I spent years of my life learning contract law, though I’m not sure I learned anything that I retained to this day. Contract law is a very well-established body of law, with hundreds of years of doctrine behind agreements between people. What exactly can you put into code that automates contract law?

There are lots of things that are already running today. I think ConstitutionDAO, once we had gotten the document and the token holders had voting power, was part of that, I think with a lot of this—

But why do you need a decentralized network of computers to manage something proposing some text for a museum display of the Constitution? Why is that impossible on a centralized computer?

Centralized computers don’t have the same guarantees that you do on a blockchain. For example, you can directly read and audit the code of what the agreement is that you are interacting with, where that’s much more difficult with a centralized service. That’s not important for something big, like a highly regulated financial institution. It makes it a lot easier for an upstart to come in and to create a new service where someone can audit the code and say, “Hey, this does exactly what it does.” 

Obviously we can’t expect individuals of all sorts to be auditing code. Most people don’t know how to read code, but there’s a very strong cottage industry of smart contract auditors growing in this space and doing a better job of understanding what the best practices are and what outcomes we want from these smart contracts.

Are smart contract auditors just lawyers?

No, they’re software engineers.

Okay. Again, my history is a very analog version of this, which is: I write you a contract. I send that contract to you. You should not sign that contract unless your lawyer reads it. Should I wait to sign a smart contract until my smart contract auditor reads it?

Either your smart contract auditor or firm that you trust. Actually, I have a really great example for you of a smart contract that you might find interesting: a very common thing in the tech world for a tech employee is vesting — you join a company, you get stock options, and there’s a one-year cliff. What you can do is essentially put that vesting into a smart contract. What’s really interesting about that is you can actually see the funds. They are sitting in that contract. You can pull them at any time, so it’s not that you’re getting your equity a month at a time, but rather you’re able to pull that at any time, every block. It’s almost a continuous stream of pulling out this equity from the contract and it’s significantly lighterweight.

I’ve read some of these vesting contracts and they’re 200-300 lines of code. At the end of the day, that’s pretty light. If you’re reusing the same one over and over again, you can verify that this is the same vesting contract that Google used for 100,000 people — you would know that this is something secure that you can probably trust as opposed to the repetitiveness of legal documents. For example, when I’m reading my lease for my apartment in New York, I have the New York fire safety code in there. It’s unclear what I actually need to read here. In software, there’s this idea of modularity, where I can just say, “Alright. I know that there’s this fire safety code module in here and I can move on.”

I think a lot of lawyers would say the same things about how they read contracts. There’s a lot of crossover between the ways lawyers work and ways software engineers work. There’s a great Larry Lessig essay from ages ago called “Code is Law,” which is—

The thing I want to push back on though: with software, you can choose your own interface. With a legal document you don’t have that optionality.

A lot of this makes sense for financial instruments and payment rails. With this project, you were trying to do something far more subjective: deciding which museum gets the Constitution. I’m still unclear why that needs to operate on a decentralized computer network, as opposed to a model in which a bunch of people you trust have issued you some shares in their organization that provides you governance over those people.

I see. Part of this was just the speed at which we were moving. We wouldn’t have been able to do this via traditional mechanisms in terms of payment processing. We had some very, very large payments come in. We had a donation of $4.5 million that came in overnight. Most payment processes are not going to handle that in a span of 15 seconds in the way that a blockchain will, so that was a big part of using a decentralized system. It’s also significantly more difficult to get those voting rights into place using traditional systems. It’s not something that’s very common outside of the financial world, and this is not a financial instrument — this is donation and control over a document. That’s why this technology was necessary because it gave us the ability to have that governance rights without the heavy-handedness of going through a lot of different, really important processes. It also guaranteed that the holders that were voting were who they said they were.

I think that makes sense, but at that point, the second half of the puzzle wasn’t assembled yet. The tokens had not actually conferred any articulated rights or voting structure. I’m curious why people were buying into a thing that they didn’t know what they were quite getting yet, but before we get too far down the road talking about donations and voting structure inside the DAO, I wanted to talk about how the ConstitutionDAO started. There are 30 core contributing members to ConstitutionDAO. How did this all start?

“It started as a meme.”

Like all great things on the internet, it started as a meme. There was an article about the Sotheby’s auction for a copy of the US Constitution, and some crypto people were making jokes of, “Oh, what if a bunch of crypto people got together and bought the Constitution?” Those jokes turned into being a little bit more serious, and there was then a Zoom call where half the people came in joking and the other half came in dead serious. And by the end of the Zoom call, it was decided, “Alright. We’re actually going to go after this thing.”

There’s precedent for crypto groups betting on auctions — PleasrDAO, for example, did a $4 million buy of a Wu-Tang Clan album — so we thought, “Alright. This is a significantly bigger number.” At the time we thought $20 million, but it’s in the realm of possibility. That’s when we started planning and organizing. We formed a Discord and a website. We started talking to museums and vendors. Literally seven days later, we had a representative at Sotheby’s bidding with an almost $50 million war chest.

What was your role in this organization?

A lot of this was super vague and we were moving very fast. What I ended up specializing in was newbie onboarding: how do we teach people who are new to crypto how to get involved with this safely? We also had open Discord rooms where we were helping people troubleshoot. I also contributed a bit to marketing and some things strategically. What we found was that people were very flexible about moving all over the place to where they were needed.

It’s funny. Newbie onboarding is why you’re on the show now, that’s me. I appreciate the expertise. When you’re talking to vendors, you obviously need banking partners. You have this partner called Endaoment — and it’s a great name; it’s “dao’’ in the middle, instead of “dow.” It’s perfect. That’s a lot of people trying to do a lot of things all at once. How did it actually come together? Did you not know that it was going to come together?

We really didn’t know until the end. There were a lot of things happening, up in the air, and a lot of uncertainty. As we raised more money, our legitimacy became greater and greater with different institutions.

That is very much an American story. The story of the United States of America is if you have more money, your legitimacy increases.

“The story of the United States of America is if you have more money, your legitimacy increases.”

Right. I think a lot of people got really curious. One of the most rewarding parts of this experience was the number of museums that we’re speaking to that now want to learn more. They saw us raise $40 to $50 million depending on the Ethereum price. They had said, “We take six months to raise this amount of money. Let’s talk: how do you do this?” Obviously, there were some very specific dynamics that were happening here, but they are very curious now about DAOs, about collective ownership, and all the different things that we explored with this organization.

Were you more interested in the copy of the Constitution or the technology and the organizational speed run?

I was interested in both parts of it. What really interested me was the story — there were so many people that were interested in this project for different reasons, and I’d encourage you to actually look at our fundraising site. We used a platform called Juicebox. People could leave messages of why they were contributing, and they wrote some really beautiful, heartfelt things: talking about how they were descendants of slaves and they were really excited to now own this document, despite some of the horrible things that were in it. Immigrants wrote in, as did parents who were excited to share this with their children one day. The different emotional stories that this project evoked in people was really the thing that pulled me in. I’m also very curious about the technology, but that’s the secondary factor.

That is interesting. During the swirl of activity in the seven days leading up to the auction, the auction itself, and the very rapid aftermath, we heard almost exclusively about the technology. We heard that this project had proven X, Y, and Z about crypto going mainstream, regarding DAOs. What I found very challenging was I heard very little about the Constitution. There’s only 13 copies of this thing floating around, and it seemed like the object was a meme. Like all internet jokes, there was an element of it which was just a troll. Did you think that was going to flip when you actually had to govern this copy of the Constitution?

I want to push back on that it was just a troll. I think there was legitimate excitement and curiosity around owning this document. A big part of the narrative was around the technology because it’s such a new thing. The technology is something that’s very misunderstood and people are still trying to wrap their heads around it, but all great technology exists for one reason: to enable people. That’s why I would say the people and their stories are what drove me. 

But you’re right — the technology was really important because it drove the narrative. “Hey, these crazy crypto people are going to buy a copy of the US Constitution. What’s a DAO? Why are they raising ‘on chain’? What does ‘on chain’ mean?” Those two factors combined are what drove that narrative. The technology, yes, was a big part of it, but again — technology just enables humans to do better things.

The flip side of this is that people are spending money. I watched your education video on how to contribute. You recommended that people add $100 or $200 to what they’re willing to contribute just to pay for the gas fees, which are the transaction fees in the Ethereum network. That is real money for quite a lot of people in this country. Quite a lot have contributed, but the back half of the project really wasn’t plotted out. What do you think the next steps would have been after raising the money and winning the auction at Sotheby’s? What was supposed to happen after all that?

We had this idea of doing a 501(c)(3). None of it was hammered into stone, but the concrete plan was Endaoment was going to custody the document until we had a legal structure together.

Here is my real question: your organization spends the money and you were just going to give it to another organization. That is a very complicated contractual relationship, right?

Yeah.

Were you going to write a contract with Endaoment to say, “We’re giving you custody of this document”? Were there lawyers on either side of that relationship?

This is — as we’ve discussed — a very highly experimental technology. This is something that we’ve done in a very lightweight fashion because it’s so new. This is something we’ve done in a very trustful way. Endaoment is very well-known within the crypto community, and I’m sure they wouldn’t want to risk their reputation on doing something negative. That’s the structure that drove this. In the future, I would love to see better legal structures that enable this sort of project. One of the biggest lessons of ConstitutionDAO is that the technology is so close, but we’re not fully ready yet. There’s a lot of rough edges that need to be sanded for this to have gone so much easier in two to five to 10 years from now.

Sure, but is this a technology question? That’s what I keep coming back to. I understand it with payment rails and financialization, which right now basically operate on computers owned by banks. Let’s decentralize those computers — I can get from here to there. You have one of the 13 remaining  copies of the founding document of the country and you haven’t planned out what to do next. Technology doesn’t help you solve that problem, right?

Right.

Someone has to take it and store it and not ruin it. You have to make sure that happens the minute you take possession of the document. Who was going to do that?

It’s both a technology question and a regulatory question. We had some partners that were ready to actually take physical custody of the documents. Sotheby’s was willing to hold it for an additional 30 days. One of the reasons we didn’t bid our full amount is because we knew we needed money for things along the lines of custodying the document, as well as paying fees and taxes. It’s both a question of regulation and technology as we figure out those structures. There are so many great organizations right now working with regulators to make this a lot easier. I’d say that’s one of the things that makes crypto different from a lot of the rest of the tech industry. There are organizations like Coin Center, the Blockchain Association, that are working very hard to get regulators to understand. A lot of this isn’t trying to convince regulators to let everything run buckwild and free, but rather, “Hey, let’s think very thoughtfully about what a really good policy framework looks like to keep the United States competitive while also keeping investors safe.”

Wait — how do we get to regulators from what were you going to do with the Constitution?

So what I was saying is, it’s a problem of both technology and regulation, and trying to tie those two things together so we can figure out how an on-chain entity can hold onto a physical document from a very concrete—

What part of that is regulated?

There is not a clear vehicle for a DAO to interact with a real-world entity.

So you raised a bunch of money in this DAO. Did you start an LLC? Forming one costs like $50.

Yeah. We had an LLC essentially as a backup if the relationship with Endaoment didn’t work out.

Okay. So the DAO had the money, and it held the money on the blockchain in Ethereum. Were you going to transfer the money to Endaoment? Were you just going to give it to them?

We had the money in an FTX account that Sotheby’s used to verify proof of funds and Endaoment would bid on our behalf—

That was the win-win. I’ll give you that one. Sotheby’s verified the money.

Thank you. I appreciate that.

That is a big deal; I don’t think Sotheby’s has done that before. I’m just curious about the actual steps: you had the money, and then you made the bid. If you had won the bid, what would have happened to the money next?

It would go either from FTX directly to Sotheby’s or to Sotheby’s via Endaoment. I was not directly involved with that process. As I said, I was more involved on the newbie onboarding and the marketing side, so I don’t have an actual concrete answer there, unfortunately.

The reason I’m asking so explicitly is that if I were a prospective investor thinking about giving this project money, “What happens next?” would be the question I would want the answer to.

Totally.

Who would have gotten the Constitution in the end? It seems like that part was not actually baked. You’re just the third party that I just have to trust. There is no software contract or standard legal contract that I can look at that says we’re going to transfer $47 million to Sotheby’s. And then, somehow, this middle organization — whose governance relationship to ConstitutionDAO is unclear — is going to get the Constitution, and then something else will happen. To me, that’s all just lawyer stuff: converting an LLC to a 501(c)(3), handing the document back. Did you have a budget for lawyers?

Yeah, that was part of our budget, and you’re right — there was a lot of trust baked in here, but we were very transparent from the beginning that this is a very trustful process. This is not something that, at the point where we started, was fully decentralized. The core team was working as fast as we could towards progressive decentralization. We were very transparent with those trust assumptions from the start. People felt they could trust us, and they came along for the journey. We are really grateful for that.

You said when you started that about half the people thought it was a joke, and half the people were dead serious. Some people suggested burning the Constitution symbolically. What were the internal politics of this core group as this week went on?

There was plenty of internal politics. The thing I think was most interesting was the pace at which we moved and how that really bred trust, and how the fast pace meant that people really had to choose. If I’m committing to this project, who has control here? Who has the decision-making power? A lot of that was very often unclear. One thing that would very frequently happen is we would have a vote in our Discord, and you’d have maybe an hour in the core team to get your voice in.

Some of these were very big decisions. A lot of it was being in a very high-trust environment. There were a lot of newcomers that people didn’t know, but pretty much everyone had at least one person they had known previously, so it was this very loose network of people that were vaguely aware of each other that came together. The internal politics were mostly disagreements on direction. There was nothing actually truly political in the way that you get in a company, because everyone was so mission-driven and knew we just needed to get things done really fast.

I just keep coming back to, the point of blockchain technology is to decentralize trust — to build trustless systems. A lot of our conversation has come back to how this was a high-trust environment between people, so I’m just trying to explore that tension.

I think this goes back to the fact of the speed-run DAO infrastructure. We did use some great trustless technology like Juicebox, which led us through the fundraising. People were able to look at the Juicebox contracts, see how they worked. That part was really important to us. That was done in a contract. It wasn’t just a multi-sig for the fundraising.

There are a couple of other features that were really important, but this goes back to the fact that we’re still early. This technology is still so new. People like to compare it to the early era of the internet in the 1990s. I think that’s pretty accurate.

How do you think software would actually change the politics of an organization? 

That’s a very broad question.

You need to make decisions. You’re voting in Discord. That changes the valence and speed of decision-making, I’m sure. For example, a DAO could work like a direct democracy, one token per vote — just the simplest framework to think about. You’re still going to get blocks of token holders politicking other blocks of token holders. You’re still going to get Succession-style investor meetings where you vote for leadership. How does this software actually change any of that outcome? Or are we just speedrunning to that outcome?

For looking at the software, it’s actually better to look at more established DAOs because there’s some really interesting, innovative things that people are working on. There’s a DAO called Index Co-op that makes index funds: you can buy, for example, the DeFi Pulse Index, which has 20 different DeFi tokens. This organization is fully a DAO. The votes on which products to release, who manages the products, funds for marketing — all that sort of thing is managed on-chain. What they’re finding is that more complicated organizational structures are needed to be able to keep pace.

A really interesting discussion in their forums recently led to an organizational change where Index Co-op now has what they call Wise Owls: council members who are dedicated to stepping in when there’s a lack of clarity and either figuring out a way to solve a problem, or bringing it to the person who can solve that problem. The thing that makes this really clear, and what’s really interesting with this software, is you can essentially define the hard capabilities of someone in an organization.

Can they call a certain function on a smart contract? Do they have control over a certain amount of funds in the treasury? Some of these more cutting-edge, more developed DAOs are much more interesting case studies for looking at this because they’ve been actively working on this for — in the example of Index Co-op — over a year with people working full-time on that project.

I find the whole thing fascinating, but at the end of the day, ConstitutionDAO is the one who raised all the money and got all of the attention. To me, that is the gap where I’m looking at the stats. For 5 percent of your donors, this project was their very first transaction with a [cryptocurrency] wallet. Almost half of your donors had under 40 transactions with a wallet. You brought a lot of people into the fold. That’s a lot of people trusting you with their money. What happens now?

You’re right. It’s a lot of people trusting us with their money. That’s why there were two things that were really important here: the first was transparency. We were upfront with everything that was going on throughout the process. 

The second important thing was refunds. It’s a big reason that we chose the smart contract that we did for the fundraiser because it had that built-in refund functionality. We did exactly what we said we were going to do. We were going to raise money. We were going to bid in the auction. Had we lost, which we did, we were going to offer refunds. The refunds are available for people to claim.

Let’s talk about the auction real quick, and then let’s get back to the refunds. You raised $47 million. Your max bid for the auction was $43 million. Sotheby’s had to verify that number for your max bid, and then they told you to hold some in reserve for taxes, fees, storage, and all of that. That’s how you got to $43 million. When did you know that $43 million was the max?

It was very, very shortly before. We did our proof of funds about three hours before the auction. We were hoping to be able to continue fundraising throughout those last few hours, but totally understand that we had to go through the process of making sure that they could verify we had the money. This was something very new for Sotheby’s, and we’re really grateful that they chose to partner with us and work with us to verify these funds.

ConstitutionDAO didn’t win. A famous billionaire investor who runs Citadel named Ken Griffin won the auction. He came in at $43.2 million, which is not that much more than what ConstitutionDAO raised. I imagine that’s quite frustrating.

It is frustrating, but that’s the way the world is. If we can’t win based on our own merits, then I think we just need to work harder to build a stronger ecosystem. I see this as a moment of inspiration. I want to congratulate Ken. He has some really great plans to display the document to make it so people are able to see it and engage with it. I hope that after this, Ken also takes the opportunity to engage a little bit more deeply with what we’re doing.

Ken, if you’re listening — and I’m sure he is — reach out to Jonah. What was it like for you watching the auction?

A group of us actually ended up getting together in New York to watch the auction together. It was exhilarating. Internally, we knew the name of the person who was bidding on our behalf, but obviously, we were not able to release that until after the results of the auction. During the auction, when the number was creeping up, I felt like I was going to puke. If we won, I might have cried. It was a very intense experience, especially after this crazy week.

One of the coolest parts was meeting a ton of people in person that I knew from the internet, which was such a fun experience. Even though we lost, people were sad for a little bit, but there was so much hope about what we accomplished. We got DAOs on some of the largest news publications in the world and taught people about this technology. We got all of these new people to actually try this. That’s a big part of new technology, especially with crypto: people often understand it better when they actually try it out.

One thing I like to think about as a student of the history of technology is the ways in which people adopt new tech — often, it’s because they have to. If you think about email, for example, a lot of people’s attitudes in the ’90s were basically, “Why would I use email? This is ridiculous.” Fast forward to the present, and it’s, “I’m going to get fired if I don’t use email because I’m not capable of doing my job.” By no means is ConstitutionDAO a project that everyone needs to participate in, but it’s a start that piques people’s curiosity. It gets more people asking, “What could I do with this thing?”

Do you think you introduced more people to crypto and DAOs or more people to the Constitution?

Definitely more people to crypto and DAOs. I think a lot of people knew what the Constitution was, and I think their interest was piqued because they were already familiar with the Constitution. I think people also really liked the Nick Cage memes.

They were very good. 

Oh, yeah.

There was a lot of confusion at the end of the auction. You knew the identity of the person bidding on ConstitutionDAO’s behalf, but the public didn’t. We had no idea who the bidders represented. Immediately, the Twitter space was full of cheering: “We won.” Where did that disconnect come from?

We don’t know exactly how that happened. Everyone on the core team was told not to post on social media until the official results were out. The people associated with that space were not affiliated with our team. This is some classic social media confusion that was happening there. It was really unfortunate because I saw some of the memes when people thought we’d won. There was so much joy, and it really felt like a bummer to get that official statement out. I think there’s a lot to still be optimistic and excited about. Unfortunately, people who are not affiliated with the team are going to do what they do on social media.

When did you find out? Since you knew who was bidding, did you know immediately?

Yeah, we knew right away.

What was the plan to release that information?

We were waiting for a final confirmation from Sotheby’s, and then we had an official statement that we released on social media.

You have refunded about 55 percent of the money after some initial confusion about refunds. The whole project was going to get wound down, and refunds were going to happen in some way, but then there was some outcry, and you’ve changed that process. How are refunds working now?

Just to clarify here, we actually didn’t change it. We were exploring another avenue that we decided not to approach. And we went back to the original plan just because we felt like that was the simplest and safest.

Tell me the difference between the two.

We explored a few options, and actually hadn’t fully settled on one. The plan that we had was, essentially, redepositing the funds back into the contract on Juicebox. Then people were able to withdraw their refund.

That is what’s happening now.

Exactly.

What was the other plan that you explored that you decided against?

The other options aren’t really worth getting into the details, but we essentially decided that all of them had sufficient technical risk. With technical risk comes the risk that we lose people’s money, and that is unacceptable.

One of the challenges now is that people have paid $100 in gas fees to donate $100. Now they might have to pay $100 in gas fees to get their $100 back. Some people have already paid more to get their money back than they put in. Isn’t this one of the core challenges of this entire project — that the transaction fees actually outweigh everything else that it has accomplished?

This is a really important criticism. The fee issue is something we were transparent with from the start. We were telling people, “Hey, the gas fee is part of it. There is this transaction fee, and another  transaction fee for refunds.” People were aware of that. It’s part of any new technology.

It reminds me of a story that’s very near to my heart about Nintendo’s Satoru Iwata. Iwata was developing the Pokemon Gold and Silver games. Pokemon Gold and Silver had two different broad areas to explore. The cartridges they used at that time were very constrained in terms of storage. In the software world, there’s a legendary story of Iwata basically figuring out a way to do compression that fit both of these regions into that cartridge.

That’s the era we’re in right now with crypto. We still haven’t gotten all of the scalability and all of the bandwidth figured out yet. Right now, we’re still in those early days of trying to pack in as much as we can at the lowest fee cost. That’s part of technology. A lot of things start off as a slightly worse experience than the technology of the day. But as the tech improves over time, it becomes significantly better. Another example of this I bet we’ll also see in the upcoming years is self-driving cars — worse than a human now, but eventually I hope to never drive a car again.

I’m with you. We talk a lot about self-driving cars on the show, but you’re in the here and now. The idea that technology will get better with time does not solve the problem of someone who was in love with the idea of buying the Constitution is now out $200 in fees because they gave you 50 bucks. How do you solve the problem for that person?

That’s why there was transparency from the start. It’s unfortunate. I wish there could have been lower fees. There are scaling solutions, but they didn’t have the contract with that refund feature, which was really important to us. 

We felt like we needed to be super transparent from the start, especially when we were just targeting people that are what we call crypto-native. They understand that fees are part of it. For those that weren’t crypto-native, that’s why we wanted to make it very clear in our education that there will be fees that are part of using the network.

You refunded about 55 percent of the money. We have all the great stats about the people who contributed for the first time or close to their first time. Were the same first-time wallets requesting refunds? Is the 45 percent of funds left over from your first-time donors? That, to me, would be very telling.

I wish I had those stats for you. The dashboard that we had for who contributed how was a community-built dashboard. That’s one of the really cool things about this technology: someone who’s totally unaffiliated was actually just able to pull the blockchain data and make this dashboard on what’s going on. I’m sure it wasn’t as exciting for that person to do the refund dashboard. They did a little bit, but didn’t put details on who and how much per wallet and that sort of thing. There was an overall refund dashboard.

So we don’t have that data, unfortunately, but I do think the fact that it wasn’t even someone on our team, it wasn’t someone affiliated — there’s just a third party that wanted access to this data. Because it was publicly on the blockchain, they could share it with us and with the world.

But doesn’t this issue sort of implicate everything we’ve been talking about? This is all about trust, transparency, and accountability. In the mad scramble of things to do, building dashboards is always the lowest level of priority. I understand that, but at some point, knowing who your donors are and taking care of the newest ones should have risen up that priority chain, right? Did you think, “We should go figure out who the people we introduce to crypto were, to make sure they feel taken care of, because that is the big victory of this project right now”?

That’s why we’ve still been active in the Discord, providing support to people who are trying to figure out how to do this. The dashboard would be great, I agree, but I’d rather engage with people directly who are asking me — both individually and direct messages and in the general Discord — “Hey, I’m struggling to do this. Can you help?” Or, “I’m not sure if the fund’s transferred out. Can you check?” It’s nice to have a dashboard that we can share with other people, but I would much rather work with those individuals to actually get them those refunds.

Forty-five percent of the money is sitting there. That number has trailed off, at least when I looked at the dashboard right before we started recording. It seems like you hit 55 percent refunds and that number has just stayed there for several days. Why do you think that is?

I think part of it is the gas fees, as we were talking about. Part of it is people not paying attention as much. A lot of people in the crypto world move very fast and they miss things — mostly people who are crypto natives. They’ve got 50 different things that they’re paying attention to at once. I expect that, over time, the number will continue tapering off, but there’s a lot of people in crypto who are just throwing bets everywhere and seeing what happens.

$20 million of “let’s see what happens” money in the bank right now, or on the chain?

To clarify, I think the gas fees are a big part of it. I don’t think their attitude is “let’s see what happens,” but rather they’re not paying attention or haven’t made the time to do it yet. It’s very exciting. You want to jump right in when there’s excitement around donating to buy a copy of the Constitution but claiming a refund is not as fun, so I think people are in less of a rush.

A classic business model is to make people claim their refunds. I don’t think that’s what you’re doing, but—

No. We don’t have access to those funds either.

What are you going to do if, five years from now,  there is still money in the wallet? The price of Ethereum moves quite a bit. Five years from now, if that 45 percent of ConstitutionDAO’s money is still there, it could be worth $50 million again. What’s the plan?

Those refunds will be available indefinitely. The smart contract is on the chain and it can’t be modified, so someone could come back 20 years from now and say, “Oh, I realized I still had funds here. I want to pull them out.” I expect a lot of the tooling to get better. One thing that’s really popular nowadays is those apps that tell you how many subscriptions you’re subscribing to that you don’t know about. We’re going to start to see that with a lot of blockchain-based things, because people are going to realize they have money in nooks and crannies all over the place that weren’t worth a lot when they first put it in there. Then, way down the line, you realize, “Oh, I should really go grab this.”

A secondary market for the $PEOPLE token has already developed. That seems very strange because this secondary market is not related to an ongoing organization. What do you make of that?

When we were choosing which platform we were going to use, there were a series of tradeoffs we had to make. We really wanted to optimize for refunds. Part of that was choosing Juicebox — it has the token and you can claim the token even if we don’t win the auction. That was not something that we had advised people to do. We were telling people just to claim their refunds.

With the blockchain, you can’t really stop people. It is nothing that’s affiliated with the team anymore and we’re sunsetting the project. It’s something that exists out there, but not something we’ve been endorsing.

Do you have a sense of why people are building a secondary market for this token?

People like the memes.

You think that’s it?

Yeah. People like the memes; they like to have fun. I think that one thing that I’ve learned from exploring the crypto world is that it’s very hard to rationalize a lot of the different momentums — things that happen with memes and with the memetic transfer of information. I think if people are having fun and they’re doing their thing, then I don’t try to interpret it too deeply, because I’m just going to break my brain.

This is where I want to end it. We’ve discussed a lot of really big ideas here. I think it is fundamentally cool that people tried to raise a lot of money to buy a copy of the Constitution. At the end of the day, though, the internet has always been good at firing a bazooka of money at things. You can go back to the ice bucket challenge. When people fired a bazooka of money at the ALS Association, they were not necessarily ready for it. You hoped they would do good things with it and that they actually invested in that research — and thankfully, in this case, they did.

The layer on top of this dynamic is financializing that bazooka of internet money in extremely surprising ways — ways that move with a momentum that breaks my brain, too. Do you think that’s good? Now there is a secondary market of a token that was meant to govern the physical copy of a Constitution but now it’s just playing in the wreckage of this project. Something about that doesn’t seem great — it actually cuts against your goals. I’m wondering if you think about that tension.

It’s a really good philosophical question and a very good concrete question. My belief is that people will do what they can, and they are going to be curious and they’re going to try things. I think it’s our job as a society to figure out what we want to come out of that — and what things we don’t want to come out of that. It is interesting to be pushing the boundaries; I’m not talking about this case specifically, but with how we can repackage and organize information and value and governance and all those mechanisms. It’s going to take some time to figure out what the right thing is. 

That’s one thing I’ve actually really appreciated about the way that the US government is approaching crypto regulation, compared to some other governments. The US government is giving the industry some space to breathe. They’re saying, “Alright, let’s watch what’s happening in the depths here. We’ll try to understand this more limited group of people who know they’re getting into a very highly risky asset class. What are we going to learn from this smaller group?” I do want to emphasize again that what we were doing was a donation for governance — voting on the future of this document. How can we make this document something that’s governed by the people? 

You introduced a lot of people to these concepts for the first time. Now some of them have taken their refunds, some of them are waiting on gas fees, and some of them are just happy to have participated. What do you think the takeaway for that group of people should be? Is it that they can spend money on cool meme things with a community attached? Or should they be more interested in this project as a new instrument of fundamentally corporate governance?

I think it’s the latter. What things do you want to happen in the world and how do you make that happen? Is this the right tool set? There’s been a proliferation already of really interesting DAOs that are doing lots of exciting projects. A few examples: there’s a DAO that’s now working on prison reform. There’s another that’s working on how to create more public scientific research. There’s one that’s acquired another document: anti-slavery documents from US history.

We’re already starting to see the effects of this. In the long term, we’re going to see more DAOs for people who are trying to accomplish goals that don’t fit in the structure of a company or a nonprofit, but more of a community collective. We’re going to see lots of people take the lessons from what we did and build better tooling and better best practices.

Let’s end where you started: the interface between the crypto world and the DAO community world and the real world is not yet built out. I understand why it is happening in financial institutions first. That is all just computer stuff. Literally, we just had an Excel influencer on the show: most of that industry is people using Excel or computers to do things. You can automate and reconstitute those contracts on decentralized databases right away.

When you are talking about prison reform — that is bodies in the world. That is not computer work. What do you think are the next steps between DAOs coming together to form governance collectives and impacting the real world? What is the next most important step?

There’s a few things that a DAO can actually do today. The thing we need to bridge the gap with is figuring out what a smart contract can actually control, because that’s the source of a DAO’s power. For a lot of DAOs today, the majority of what they do is just sending treasury funds; $20 million to start a crypto lobbying group. That is a real world example from Uniswap, a decentralized exchange that had a vote to start this crypto lobbying group.

However, I think that over time we can build out more functionality for the powers of DAOs — both from the voting power, what the actual token holders can vote on, but also elected representatives who the DAO gives very specific, concrete and auditable powers to.

I feel like I could go another hour on this. I really want to know if the lobbyist now gets to pick what steak restaurant they take elected officials to. Can the DAO vote on that? That would be really interesting, but you’ve given us so much time. What is next for you? What is next for the core members of the ConstitutionDAO group?

ConstitutionDAO is finishing, shutting down official activities, refunds are going. Everyone’s dispersing to different projects. I have a day job as a software engineer at a very different — but I would say just as exciting — part of the tech industry. People are working on what they’ve been working on, but I expect we’ll see more from this group in the future. It was really incredible. I’d say one of the best parts of this experience is I made some new friends. It’s really fun when you can call up people you know from the internet and say, “Hey, do you want to go hang out? Do you want to go grab a coffee?”

That’s great. Jonah, thank you so much for coming on Decoder. I pushed you a little bit. Thank you for engaging. I really appreciate it.

Nilay, thank you for having me. This has been a great experience and glad to be on this journey of understanding crypto with you.

Decoder with Nilay Patel /

A podcast from The Verge about big ideas and other problems.

Subscribe now!