Are blockchains immutable?
A dive into the collective belief that ‘yes, assets CAN be digital, and you CAN own them!’
When I started Untangled in November, there was roughly 60 of you subscribed. Now there are 335 of you, and roughly 65% of you open every email. That’s really exciting to me — I’m flattered 😊. My goal was to reach 400 subscribers by April. If each of you shares this email with a few friends, we’ll get there together. Do it, do it now! Pretty please?
Now on to the show!
There’s a big idea making the rounds in crypto these days — that “blockchains are immutable” and “enable ownership” of digital artifacts such as non-fungible tokens (NFTs). Let’s start with NFTs and work backward, shall we?
I mean, what is an NFT, really? Most people imagine a picture of a bored ape when they think of one.
But the NFT isn’t the image itself, it’s a digital receipt that includes metadata about the image, audio, or (and this is a big ‘or’) any digital thing. In his stellar essay, “The Internet Is Just Investment Banking Now,” Ian Bogost writes,
“So NFTs aren’t strange or novel because they make appeals to value, provenance, and ownership via collective fantasies of paperwork. That’s old news. They feel strange and novel because normal people don’t usually construe monetary value in mere references to everyday things, like a cash-register receipt, or computer data. Belief in such value is, however, completely normal in the financial sector. In that context, an instrument that confers ownership, which can be bought or sold and which holds monetary value, is called a ‘security.’”
In other words, we’ve long agreed that certain things — certificates, titles, etc. — can symbolically represent the ownership of something, like a house. So what’s jarring about NFTs is not that we’re using one thing to represent the value of another thing — it’s that crypto enthusiasts now want to value any digital thing. The question becomes, will we collectively agree that these things are valuable? I mean, how much is this essay worth? I obviously think my words are priceless — how about you?
More seriously, what’s at stake here is the possibility of slipping into a society where every single digital thing becomes subject to the logic of money. So what does it mean to ‘own’ an NFT? Is it different from owning a physical thing? To answer these questions, we need to interrogate the idea that blockchains are immutable, which starts with some data in a box:
Okay, next apply a hashing algorithm to the data. Ta-da, the data becomes a unique identifier.
But hashing algorithms also show whether the data has been altered in any way. As Quinn DuPont explains, “this is possible because [...] any change of input for a hash algorithm produces a radically different output.” To see if anyone altered the data, just run the hash algorithm and compare it to the original.
Blocks are then chained together (hence the name!) such that each new block includes the hash from the previous block.
Alter one block, and you’ll invalidate the whole chain. The hashes and chains are what make blockchains technically immutable, and this immutability is what enables the ownership of any digital thing.
Or does it?
⛔️ Old assumption: how technologies shape human agency is ‘baked-in’ to the technology itself.
It seems we’ve arrived at a place where we assume that the tech itself is what shapes our experience of it; that because blockchains use hashes to link blocks of data together, they are therefore ‘immutable’ and we, in turn, get to own all the things! Not quite.
This makes the assumption that it is the materiality of the technology itself that shapes and constrains human agency. For example, STS scholar danah boyd identified four ‘affordances’ of “network technologies”. Think of an ‘affordance’ as a way in which technologies alter our experience of it. So, according to boyd, the four affordances of networked technologies are: “persistence,” (i.e. online data is recorded and archived) “replicability,” (i.e. content can be replicated) “scalability,” and “searchability.”
These qualities help describe the internet today. But remember the 1990s when the internet wasn’t about visibility and persistence but anonymity and ephemerality? What happened? According to sociologist Janet Vertesi, what technology enables and constrains isn’t “baked-in” to the object itself but is emergent and socially constructed, made visible over time. The internet is what it is today because we made it so. These affordances aren’t naturally occurring in the tech, but rather are — to use Vertesi’s term — “accomplishments” of people, organizations, and in this case, companies like Facebook, Twitter, and Google.
💡 New assumption: how technologies shape human agency isn’t pre-determined, unchanging, or a natural quality of the technology itself — but an accomplishment of people, organizations, and companies.
If we treat ‘immutability’ as an accomplishment rather than an affordance, we have to broaden our aperture beyond the blockchain. Blockchains are entirely ensconced in a much wider ecosystem that contains legacy tech, laws, consensus mechanisms, and governance systems. For example:
People can vote to change the chain (remember when the Ethereum blockchain did a hard reset?)
People can exploit or hack consensus mechanisms for their own gain — I touched on this in my first piece on decentralization.
Blockchains also interact with legal rules that don’t currently back up claims of ownership
So, whether immutability emerges or not hinges on the systems that blockchains interact with. In fact, the broader technological ecosystem that blockchains live inside of is, well, totally mutable, isn’t it?
This brings us back to NFTs and the concept of ‘ownership.’ Sometimes the digital receipt lives on the blockchain (or “on-chain”) but sometimes it is stored “off-chain” on a server hosted by a company like OpenSea, a link away from the blockchain. This sort of means that OpenSea is a centralized aggregator of NFTs — a Web2 company masquerading as a Web3 company (revealing that these categories don’t have as firm a boundary as some crypto enthusiasts want to believe). But anyway, if the receipt lives “off-chain,” the NFT merely contains a URL that points to the data. In Moxie Marlinspike’s recent interrogation of crypto, he writes:
“What surprised me about the standards was that there’s no hash commitment for the data located at the URL. Looking at many of the NFTs on popular marketplaces being sold for tens, hundreds, or millions of dollars, that URL often just points to some VPS running Apache somewhere. Anyone with access to that machine, anyone who buys that domain name in the future, or anyone who compromises that machine can change the image, title, description, etc for the NFT to whatever they’d like at any time (regardless of whether or not they “own” the token).”
In other words, even if the governance or consensus systems don’t alter the “immutability” of the blockchain, you can only really go as far as uniquely identifying something, and not outright owning it. Making these distinctions takes back some of the power we weirdly (and willingly) hand over to the technology itself. But the technology is only part of the story.
⚠️ While this essay interrogates the intersection of ‘immutability’ and ‘ownership’, blockchains will be used for storing more things than NFTs. Terrible things. For example, blockchains will undoubtedly be used to store revenge porn and images of child sexual abuse. In other words, if blockchains become ‘immutable,’ that’s not something to cheer about. There is nothing inherently good about ‘immutability.’
The other part is the story itself — and right now, the ‘ownership’ narrative is taking off. Just Google ‘the ownership economy,’ and you’ll see what I mean. This is kind of ironic, of course. Bitcoin began in response to the 2008 financial crisis, as a supposed antidote to the corruption of our financial system. As researcher Ines Faria puts it, there was a “mysticism surrounding the hyped Bitcoin and blockchain software — which people wanted to understand and bet on, depositing faith in its power to bypass financial intermediaries and make them obsolete.” It may have started that way, but it has transformed into a system optimized for the financialization and ownership of every digital thing.
We imbue technology with a kind of magic. We tell stories about it, and then sometimes, we start to collectively believe those stories in ways that transcend the capabilities of the technology itself. If crypto transforms society in the way its supporters believe, it won’t be because blockchains are “immutable.” It will be because we overlook the limitations of the technology, change the surrounding technical and legal ecosystem, and decide to collectively believe in the story of ‘ownership’.
Thank you to Matt Goerzen for providing comments on an earlier draft, and to Georgia Iacovou, my editor.
Charley, great read! Thank you for the wonderful perspective. Your second-to-last paragraph made me consider something provocative - I thought I'd share.
This paragraph made me reconsider why tech companies are motivated to develop Web3. "Because the systems [are] optimized for the financialization and ownership of every digital thing", tech companies would overtake the roles of financial institutions and become intermediaries themselves - collecting economic rents on the flows of capital. What are your thoughts on this?
Interesting perspective Charley. I think a fair question to ask in your articles, which doesn't get asked, is whether or not the technology in some given context is better than what came before it. NFTs are a chance to do this. At present day, I think we can agree that there are two big emergent use cases for NFTs; conveyance of rights with respect to a purely digital asset and conveyance of rights with respect to a physical asset. While there may be limited precedent for the former when it comes to your rights to a Bored Ape (largely driven by a scarcity market for bragging rights aka collectability), there is sufficient precedent for the digital rights use case with something like music. Since the first iPod, music fans have largely agreed to Apple's conveyance of rights to store and play digital music from their favorite artists. Depending on your point of view, this system was an evolutionary step over the previous system of rights-conveyance; taking physical possession of an original CD. For example, in the new system, rights were conveyed one song at a time at the user's option. You didn't have to buy an entire album. But in exchange for this convenience, some would argue there was sacrifice. A deal with the devil. There were and still are downsides. For example, Apple may have conveyed a right, ensconced in some digital contract somewhere, when you paid $1 for a song. But that right came with restrictions. You could only exercise that right on an Apple branded device. If you died, the right was not transferrable the way a CD previously was. The point isn't to bash Apple. The point is that every system dating back to vinyl involved flaws. And so, the fair and operative question in terms of scenario #1 (conveyence of a right to a digital asset) is whether, despite its imperfections, the new system (NFTs) offers improvements over the system that came before it. What are the sacrifices (as you like to note) and are they worth it. For example, a new system of DRM for music could disintermediate Apple as the final arbiter of your rights, or those of the musician. The musician gets to make the rules, instead of Apple. A smart contract on a blockchain could convey your rights to a successor upon your death, and so on. Just as was the case with iTunes, a case could be made that there's real efficacy in the new system when it comes to the conveyance of rights to digital assets. The second NFT scenario involves conveyance of rights to a physical asset. And so I'd ask the same questions (by the wall, all in context of your question about immutability). Our system of laws in the US have evolved to a standard whereby a piece of paper backed by some system of non-repudiation (typically managed by a widely acknowledged third party such as a state government) is the instrument of provenance that conveys a right to something physical. A title to a car or a house. Rubber stamps and notaries. But that system has flaws. For example, paper can be forged. An entire ecosystem (eg: title searches) exists to mitigate the risks associated with the potential for bad actors in a system based on paper based entitlement. So, the question that should be asked in this case is to what extent this new digital system is an improvement over the previous one. Is there potential for certain bad actors and the risk associated with them to be removed? Does it create opportunities for new bad actors and at the end of that analysis, did we reach some improved state, are we taking a step backward, or is it 6 of one, half dozen of the other? Terrorists remotely detonate bombs with cell phones but we would never throw those babies out with the bath water. Maybe, as Moxie argues, there's a dark side to immutability. But maybe the upside greatly outweighs the shortcomings in the rules and provenance of the current system. Or maybe not. What if the digitally conveyed rights to the building that I think I own are enumerated in two separate distributed public ledgers? Like the CD that got converted into bits, how does today's paper move safely into a new medium that solves for problems that the current system doesn't solve for?