#imperfect-union
Imperfect Union: How ConstitutionDAO Lost Its Way
Drew Millard
Kevin Munger

Mon Dec 06 2021

17,000 people teamed up to a buy a copy of the U.S. Constitution. What happened taught us a lot about decentralized decision-making.

In a September episode of Justin Murphy’s Other Life podcast, Mónica Belevan, the theorist and writer behind Accursed Share and Covidian Aesthetics, compared the current state of Web3 to that of a baby on the cusp of toddlerhood. “It’s still ga-ga, goo-goo — it’s inarticulate,” she explained. “But any minute now, it will say its first word.”

Evidently, that first word was “We,” quickly followed by the ones, “Are Gonna Buy The Constitution.”

Some context: In September, Sotheby’s announced that it would be putting up for auction one of the 13 original, still-surviving copies of the U.S. Constitution. Given the item’s rarity and historical significance, this was probably always going to be a fairly big deal, at least in the rare documents space. But then, thanks to the efforts of a small group of tech entrepreneurs, meme lords, and at least one teenager, it became one of the biggest Web3 events to ever cross over to mainstream awareness, yielding a gushing New York Times story, a frenzied crowdfunding effort, and an impromptu mass education in the mechanics of gas fees. We are talking, of course, about ConstitutionDAO, the nearly-successful attempt of a Decentralized Autonomous Organization to raise enough money to buy the U.S. Constitution.

Watching the auction on Sotheby’s YouTube Channel (archived here), where swarms of commenters assaulted the event’s official chat with a cascade of “GM”s, was genuinely thrilling. As the price climbed higher and higher — first jumping from $14 million to a whopping $30 million, then incrementally rising to more than $40 million — it seemed as if ConstitutionDAO was going to accomplish its goal. Even though the group was self-consciously trying to re-enact a legal, crypto-pilled version of the movie National Treasure, it was hard not to root for them. The fact that the chat couldn’t agree on which of Sotheby’s professional bidders was actually representing ConstitutionDAO — Brooke or David — only added to the fun.

Getting a clean look at what the high-end auction world is actually like also went a long way towards making these Web3 weirdos sympathetic. Just think about what the response would be if someone were to propose the status quo: a room full of visibly rich white people whispering into their corded phones (???) on behalf of their far richer masters, blushing and giggling with the quasi-sexual thrill of moving around tens of millions of dollars.

By the time only two bidders remained, one of whom we knew to be acting on behalf of ConstitutionDAO, it felt as if we were watching, in real time, Web3 realizing its agency and then acting on it. The DAO’s coffers felt limited only by the passion of its members; the other bidder, whoever they were, could not possibly match their unity of vision.

But then, the gavel fell. After eight minutes of escalation, ConstitutionDAO was unable to beat the winning bid of $42.3 million — a figure that, a Sotheby’s press release would later boast, was the “HIGHEST PRICE EVER ACHIEVED AT AUCTION FOR ANY BOOK, MANUSCRIPT OR PRINTED TEXT.” Within days, it was revealed that the DAO had lost out to a hedge fund owner and prominent Republican donor named Ken Griffin.

The reason why the DAO failed in its quest is simple: Griffin had more money than they did. ConstitutionDAO had raised well over $40 million in funds, from more than 17,000 people, for the purchase; to beat them, all any single covetous billionaire needed to do was check the organization’s public Juicebox page, see approximately how much the group had in its coffers (certain donors, we were told, contributed through other channels), and decide whether they were willing to pay that much. And while it’s hard to say whether broadcasting this information was a tactical error on ConstitutionDAO’s part — it surely drove donations, and may have deterred certain prospective bidders from entering the auction at all — the approach did give Griffin a window into the DAO’s hard cap.

There was a touch of tragic irony to the revelation: If ConstitutionDAO was 2021’s answer to WallStreetBets, then you couldn’t have found a better villain than Griffin. During last year’s GameStop short-squeeze frenzy, his firm, Citadel, benefited from being the primary market-maker for Robinhood, fulfilling many of the countless retail trades that sent the struggling retail chain’s stock soaring. Then Citadel turned around and invested $2.75 billion in Melvin Capital, the hedge fund whose thwarted short position in GameStop had nearly plunged it into insolvency — in effect, fueling the fire that burned Melvin’s house down, before swooping in to buy the property at a discount.

As the DAO’s members publicly navigated an expensive and unwieldy refund process and debated how to move forward before ultimately disbanding, it was tempting to label the entire endeavor as pointless. Had it won the auction, ConstitutionDAO planned to solicit proposals from institutions hoping to exhibit the document and allow its members to vote on where and how it would be publicly displayed. Shortly after actually winning, Griffin announced that he’d be doing a version of the same, loaning his Constitution to Crystal Bridges Museum of American Art, an Arkansas institution founded, per the New York Times, “by the Walmart heiress Alice Walton.”

Surely, many of the people in ConstitutionDAO’s ranks got involved for the same reason that people on the Internet do anything: They thought it would be cool, or funny, or stupid, or “epic.” Still, the implications of the gesture were profound: Here was a group of people staking a claim for Web3 as a tool that could be used to build a world with new rules, by purchasing the rules of the old one. But in helping to bid up the document’s price to record-breaking levels — perhaps even in deciding that this particular document was something worth rallying behind at all — the DAO’s donors may have inadvertently reinforced the power of the old order. ConstitutionDAO’s struggles, in this way, are a learning experience for everyone.

“It couldn’t be a better David and Goliath story,” said Kristy Tillman, a Bay Area design director who was involved in Constitution DAO as a donor, and a vocal participant in its internal discussions. “There are a lot of people who are invested in keeping things the way they are. They have a lot of money to bankroll that.”

Tillman, a former designer for Slack and the founder of the Made in the Future fellowship, was drawn to the project because of her interest in history. She also saw it as an opportunity to educate herself in the particulars of Web3. “I’m someone that works in technology,” she said. “Even for me, there were barriers to really understanding the technologies behind [the space].”

By that metric, she believes the project was a success. She pointed to a graph indicating that over 2,300 ConstitutionDAO contributors were first-time ETH wallet users. “That’s a good proxy for the percentage of beginners [involved],” she said. But she also came out of the experience feeling like one of the most appealing aspects of the DAO model — its non-hierarchical structure, allowing thousands of people to coordinate and make decisions as a group — could be a double-edged sword. “There are so many details about a DAO’s mechanisms that you have to rely on in order to feel like your participation is viable,” she said. And in the mad dash to “fly the plane while building it,” as Tillman put it, a lot of those details ended up falling through the cracks.

For one thing, she noticed that ConstitutionDAO’s act-first-think-later approach inadvertently created some alarming blind spots when it came to issues of race. The organization’s crowd-funding system asked donors to exchange Ethereum for an ERC-20 token called We the People, abbreviated as simply $PEOPLE. Given that the very document that the DAO was bidding on contained a clause denying Black people personhood under the law, this came across as tone-deaf at best, and outright racist at worst. “People were making jokes like, ‘Let’s buy a bunch of people,’ with the implication that they’re joking about the personhood status of my ancestors,” said Tillman, who is descended from enslaved individuals. “I’m not just going to grin and bear that.”

The mechanism the group used to authorize the transfer of funds from Juicebox to its multisig Gnosis Safe wallet, where funds would be held in preparation for the auction, was modeled after the ratification process of the Constitution itself. This required at least nine signatures from a 13-person panel. However, none of the members of the core group were Black — a fact suggesting that for all its radical implications, ConstitutionDAO’s very power structure was symbolically recreating the racial dynamics of the actual Constitution. “This was a major opportunity to make a statement about equality and to use this as a jumping-off point to talk about a different world that we wanted to build,” Tillman said. “But there were no descendants of slaves on that team?”

While the DAO’s team was receptive to such criticisms, at one point even planning a new governance token called $WTP (“We the People”), Tillman ended up realizing that when it comes to building a DAO, it’s important to know who you’re joining forces with. “Being aligned with the mission isn’t enough,” she said. “You have to be value-aligned with the community.”

There were other issues standing in the way of the DAO’s egalitarian ambitions. As VICE’s Motherboard points out, the median ConstitutionDAO donation was $206.26. Those small-dollar donors are seeing most or all of their refunds getting eaten up by Ethereum network gas fees, which remain static regardless of the size of the transaction.

One person involved with the DAO who worked closely with the core contributors said that there was only so much that could be done to mitigate the issue. “We only had a week for this, and the platform that we used wasn’t necessarily immediately equipped” to efficiently handle gas fees. Part of the problem was structural: Many layer 2 Ethereum platforms, which could have enabled contributors or refund-seekers to package a large number of transactions together while splitting the cost of a single gas fee, lack the regulatory clearance to be deployed in certain states, including New York. Had the DAO opted for one of those platforms, it could have shut out countless potential donors. Besides, they said, “The only way we were able to walk away and know what needs to be built better was by trying this.”

In the days following the loss, the DAO’s Discord channel remained open and active. Members debated whether they ought to buy another historical artifact and discussed potential name changes for the $PEOPLE token. The team had pledged to provide clarity on the DAO’s path forward — and indeed, whether there was a path forward at all. But at 3:01 p.m. on Wednesday, November 23, ConstitutionDAO announced on its Discord channel that it had “determined that building and maintaining an ongoing project is not something that we as a core team are able to support.” Within a week, the inevitable happened: The Discord had been invaded by bots.

To the insider we spoke to, the decision to end the DAO was the correct one. “If people didn’t have the bandwidth to action on a clear-cut plan, then it makes sense to wind things down,” they said. Besides, maybe it’s good that ConstitutionDAO was a little haphazard and chaotic: Next time, people will know the pressure points. High gas fees are one thing when it’s a group trying to buy the Constitution. But if an activist group using a DAO to organize a decentralized bail fund runs into similar issues, the consequences could be decidedly direr.

On the night of the auction, one of us biked up to Independence Hall. The manicured campus in the heart of Philadelphia was deserted except for scattered security guards and people walking to dinner, pointing at the Liberty Bell, and marveling at the building itself, inside of which a group of delegates from the 13 original colonies signed the U.S. Constitution 234 years ago.

At night, Independence Hall has a ghostly presence. We had wondered whether any of ConstitutionDAO’s members would think to follow through on the IRL symbolism of their URL gesture by watching the livestream of the Sotheby’s auction at this foundational historical site. The Reddit GME short-squeeze, after all, had led to a gathering of ape-costumed protestors and performance artists on Wall Street, and we reckoned this might merit at least a few enthusiastic Discord posters in tri-corner hats.

There was no one. Perhaps this was appropriate. When the Constitution was written and ratified, its legitimacy was based in part upon the fact that powerful people had gotten together and signed it. The effort to purchase this particular copy of the Constitution at auction, meanwhile, was coordinated remotely, with the DAO partnering with an organization called Endaoment to place bids on its behalf. In retrospect, our sense of physical proximity as a precondition for social connection seemed terribly Boomer-ish, especially when listening to an interview with the DAO’s central players on Twitter Spaces after the auction. Their emotions, communal spirit, and exhaustion were palpable. That the project’s official Twitter profile picture featured Nic Cage’s character in the movie National Treasure didn’t change the fact that for all of the meme-y energy it projected, these people still cared.

Still, the centrality of National Treasure to the culture of ConstitutionDAO is no accident. Cage’s decades of manic energy onscreen inadvertently predicted the frenetic dream-logic of online life, and the film’s narrative conceit of a series of clues hidden in plain sight resonates with the contemporary digital conspiracy craze. In one of the movie’s key scenes, Cage’s character uses Masonic solar geometry to identify a specific brick on the roof of Independence Hall where Benjamin Franklin had stored “some kind of optical device” that allows the wearer to see hidden messages on the back of the Declaration of Independence, written invisible ink. That the scene was actually shot at a replica Independence Hall, outside of a theme park in Los Angeles drives home the near-total mediatization of our experiences. We are in constant conversation with others’ interpretations of the world around us, debating the validity of those interpretations, the weight we ought to be giving them, and even their source’s very right to express them. After a certain point, we risk losing the ability to interpret, let alone change, the world ourselves.

Web3’s love of memes, whether they take the form of vacuous call-and-response (“GM,” “WAGMI” ) or campy references to mass-market films, points to the limited horizon of possibility we are able to see outside the realm of technological change. Of course it would be better if people in the space had better taste, if they were able to match the potentially revolutionary medium with innovative content rather than art that took all the wrong lessons from Damien Hirst. Now is the time for artists trying to wring revolution out of the tired forms of the past to take up the tools of the present and build something at the current technological frontier; the technology itself is no guarantee of a revolution, but neither is its artless application a condemnation of its potential.

Gathering out on the street, as a signal of collective agency, has always been on the spectrum between voting and open warfare. The March on Washington in 1963, when a quarter of a million people marched for civil rights and witnessed Martin Luther King Jr.’s “I Have a Dream” speech, was evidence of a powerful form of organization. For years, the civil rights movement had built up the logistics, communication, and leadership necessary to coordinate a street operation of that magnitude. This process necessarily entailed the creation of deep community bonds and trust — a foundation that made the movement a credible threat to the establishment, an organization that wouldn’t melt under pressure.

While smartphones and social media have made it easier to get people on the streets, these new technologies have produced wave after wave of protests that lack coherent governing structures or the capacity to engage in sustained negotiations — either internally, about their own goals, or externally, about concrete terms of what it would take to get them to stop. Occupy Wall Street is the canonical example, but the BLM protests from the summer of 2020 and the Capitol Riot of January 6 also consisted of crowds that were largely without organization. They did not, in the framework of Martin Gurri, have the ability to “take yes for an answer.”

This “tactical freeze,” as Zeynep Tufecki has described it, tends to hobble these networked protest movements from the start. The harder centers of power — the State, capital, and those who enforce their will — have the advantages of military-style top-down coordination and rapid, private communication. They also have all the time in the world; as long as they avoid losing, eventually they win.

This is where we are with the financial flash mob iteration of Web3. ConstitutionDAO raised a healthy amount of cash in under a week, but the ideal of openness embedded in the blockchain ended up being a crippling liability when it came to fighting against a financial establishment armed with decades of experience in opacity and nigh-unlimited liquidity.

And yet, the tactical freeze here was probably most evident after ConstitutionDAO lost. The organizers seemed genuinely committed to doing right by their community of donors, pushing for rapid refunds to anyone who wanted them. But it felt like they were answering to the spurned contributors to a Kickstarter, forced to enact “accountability” in the form of complete supplication to their audience. Rather than attempting to actually use the tools they had built — tools which were the nominal reason for all of this excitement — the group buckled under the weight of the crowd. It lacked the genuine interpersonal trust that can only come with time, dedication, and sacrifice — the glue that made the March on Washington successful.

One of the greater ironies of the ConstitutionDAO saga is that there are already original copies available for public viewing. You can easily look up the text online, and you can buy a pocket-sized copy of it at nearly any bookstore. Its words are ubiquitous. So unless (A) you’re trying to sell it, or (B) there’s a secret message about some treasure written on the back, there’s really no practical use in owning an original.

Contrast this with another recent high-profile auction: PleasrDAO’s $4 million acquisition of the Wu-Tang Clan’s 2015 album Once Upon a Time in Shaolin, of which only a single copy was pressed. When the notorious “Pharma Bro” Martin Shkreli originally purchased the double-LP six years ago, it was sold with a copyright restriction placing an 88-year ban on its owner commercially releasing the record, a stipulation that reflected the wish of its producers that the work be valued not as a piece of music, but instead as a singular art object. This restriction did not, however, bar its owner from releasing the music for free, or holding public listening parties where the album was played. Shkreli, cultural villain that he is, mostly did not do these things.

PleasrDAO hopes to change that, claiming in its Mirror post announcing the purchase that “Although we are bound by the legal agreement underpinning this work of art and may not be able to duplicate and share the music digitally, we firmly believe there are ways to share this musical masterpiece with the world.”

As an object whose fabricated scarcity was designed for the benefit of the artists who created it, Once Upon a Time in Shaolin was sort of a proto-NFT. Should PleasrDAO succeed in its goal of disseminating the record to the masses, then the circle will be complete. To the majority of Wu-Tang Clan’s fans, it’s less important to physically possess Once Upon a Time in Shaolin than it is to be able to listen to it. If its sale to Shkreli subsidized the record’s creation, and the sale to PleasrDAO facilitates the public’s ability to actually hear it, then the project might serve as proof that the premise of Web2 — that an optimal Internet is one in which platforms make money by devaluing the work of content creators and commodifying the data of content consumers — was false. There is a world in which our work can be fairly compensated yet free to access, with no strings attached.

But this is the dream, not the reality. Until the altruistic-minded ConstitutionDAO came along, many of Web3’s biggest success stories have involved a short-run transfer of wealth to a relatively small group of tech-savvy young people, elevating the few while leaving the many behind. For every artist becoming a millionaire off of NFTs, there are dozens of others only making $100 from the sale of their digital asset, collectors paying an overinflated price for an NFT that, statistically, they will most likely be unable to resell, or people getting involved with the wrong project and having the rug pulled out from under them.

If we accept that scenarios such as these are the only possible endpoint of Web3, then the future is already over. But the decentralized, community-focused nature of DAOs suggests that these same emerging technologies have the capacity to genuinely empower people and shift the distribution of force in society. Important conversations about how to make Web3’s tools accessible to and financially viable for members of BIPOC and LGBTQ+ communities, rather than recreating today’s IRL inequalities, are already happening. The will for change is there, and that is a good thing.

If nothing else, Web3 has created a new space of play. It has been over a decade since the Internet felt like this. Hundreds of mainstream musicians and artists have now issued NFTs. Prior to his November 28 death, the era-defining fashion designer Virgil Abloh had been conceptualizing “a real-life, digital museum” called Skyscraper that would run off blockchain technology. And even if it didn’t succeed as a political gesture, the ConstitutionDAO stunt certainly succeeded on the level of performance art. The pressing question for the next year or two will be whether this time will be different: Can we sustain this sense of fun and play while forging organizational structures that are antifragile in the face of capital’s inevitable attempts at co-option?

Because here’s the thing: Despite the best efforts of millions of self-styled online revolutionaries, there does not exist a tweet that will bring down Twitter, or ExxonMobil, or the NYPD. If getting owned mattered, neither Hillary Clinton nor Joe Biden nor Donald Trump would ever have won their party’s Presidential nomination; and yet, even after five years of abject failure, the enthusiasm for posting has not dimmed.

ConstitutionDAO itself can be seen as an extension of the poster’s mentality. Its existence and early successes garnered media attention, which helped fuel its fundraising momentum, which increased its media-worthiness in turn. But as ConstitutionDAO careened towards the Ultimate Event of the auction itself and beyond into the void, the spectacle of its creation seemed to boost everyone except the DAO itself. Sotheby’s earned millions off of the sale, as well as priceless free publicity. Journalists got a wacky story to write about. Ken Griffin got the benefit of you now knowing who he is. Brooke Lampley, the Sotheby’s employee who bid on his behalf, parlayed her brief meme status into a couple thousand Twitter followers. Hell, the Constitution did pretty well for itself too — when was the last time you’d thought about that thing at all before this whole thing started?

So, again: What would we think if someone were to propose the status quo of the Internet? No one ever proposed it, of course; the people who built it moved fast and broke things. ConstitutionDAO represents a super-sized version of something that the world they created would be extremely good at: mass, memefied mobilization — a.k.a., the “moving fast” part. The Constitution isn’t a meme, though, and simply buying it is far from making anything that can last. We need to move fast and build things.

Drew Millard is a writer and editor whose work has appeared in Gawker, The Nation, and the New York Times Magazine. He is currently finishing a book about golf, due out in the spring of 2023 on Abrams Press.

Kevin Munger is an Assistant Professor of Political Science and Social Data Analytics at Penn State University. His book, Generation Gap: Why Baby Boomers Still Dominate American Politics and Culture, will be published in May 2022.

Graphics by Fiona Carty.