The Sotheby's Auction of "Bitcoin Ordinals"
You have to see it to believe it; here, I will try to explain it
Programming Note1I’ve been working all week on an almost finished piece on “box spreads” (and the nature of money and risk) and I plan soon thereafter to publish a longer, three-part essay titled “How to Tell the Story of Crypto.” Today’s post should really be a coda or new chapter to that crypto essay. However, I had to pre-empt all my plans and get this out today, because the auction closes tomorrow and I wanted everyone who had the time to get the chance to see it live.
First off, seriously, everyone has to go look at this. I’ll wait here.
If you visited that site and nodded along knowingly at the description of “rare sats” and “the invention of ordinals,” before scrolling through the auction lots and perhaps even considering placing a bid – well, I hate to tell you this, but you’ve scored quite highly on the formal test to determine if you are a bitcoin bro.
For everyone else, if almost nothing that you read there made any sense at all, and perhaps you kept asking yourself, “is this really the Sotheby’s site, or is this a spoof?” then let me start by saying the following. I’ve spent a huge portion of my time over the last six years researching, studying – really trying to understand bitcoin, blockchain, and the larger crypto universe. And as of yesterday I had never once heard of a “sat” or “ordinals” or basically anything else described so breathlessly by Sotheby’s.
A market that blossomed out of the invention of Ordinals, rare sats introduce a layer of uniqueness and rarity to otherwise fungible and ordinary bitcoins, transforming them into collectible, historical artifacts. … rare sats stand as more than just a component of Bitcoin's blockchain; they are a new form of digital asset, combining inherent monetary value with historical significance. (emphasis added)
This is all utter rubbish, yet I still think it’s worth spending a few minutes sorting through the trash in order to offer an explanation. Why do I think that? First, because just this morning the price of bitcoin topped $60,000 – up 45% in just the first two months of 2024. Second, because the above comes not from a crypto evangelist but from the Sotheby’s auction house – the bitcoin ordinals auction is listed on their website right next to an upcoming auction of works by Magritte, Degas, and Picasso. Finally, most significantly, because crypto remains a crucial terrain of/for money/power. Crypto advances radical theories of money in support of a radical political project and over the past year these projects have renewed themselves and gained momentum, despite the catastrophic collapses of 2022–2023 (the topic of my long, upcoming crypto essay).
But I don’t want to waste anyone’s time, so let me concisely work through Sotheby’s “sat hunting” in three short steps.
Step I: Recognize that in the bitcoin code, there are no bitcoins
It may come as a surprise to many of you, but the original code that launched and runs the bitcoin blockchain has no provision for “bitcoins.” If we think in terms of a handful of individual coins (nickels, dimes and pennies), or a stack of paper notes in our wallet (a fiver, a tenner, etc.) bitcoin has no coins. It’s tempting to assume that the bitcoin protocol, through the mining process, creates a set of coins, with unique identifying numbers.2 But that’s just not true. For our purposes, the protocol establishes only two key ingredients: addresses, and a variable called “value.” A bitcoin transaction can be defined as the transfer of some numerical quantity of value between one address and another. “Value” is abstractly posited by the code, but that variable has no identifier or name – its only link is to the address that owns it.
I doubt this would have bothered Satoshi: I personally think that he genuinely believed that he was inventing “digital cash,” a technology people would use to make payments. As we know, his invention has failed spectacularly as a practical “currency” but it has “succeeded” beyond anyone’s imagination at becoming an object of speculation. Bitcoins are not used, they are bought and sold, and quite often they are “owned” in the sense of collectors’ items. But the original bitcoin code specifically prevents any specificity to the “coins”: there is no way to distinguish one from the other.
Again, this is arguably by design: as posited by economists, fungibility is one of the many necessary criteria for being money.3 But the NFT boom taught people in crypto world a very different lesson: if you want people to speculate on (virtual) objects, it helps to claim they are rare, distinct, even unique. It’s right there in the name: NFTs are non-fungible tokens; my bored ape and your bored ape are both special in their own particular way (even though both are really ugly and hard to tell apart). NFTs were built as “smart contracts” on the ethereum protocol, and thus bitcoin missed out entirely on the NFT boom (and bust).
Step II: Invent “sats” and “ordinals”
If you want to build a collectors’ market in the bitcoin universe it makes no sense to make “a bitcoin” – if by that we mean 1.00000000 BTC – your basic unit, because doing so dramatically restricts your target audience to people who can afford $10,000 to $100,000 for a collectible. So the first move is to invent – just by making up the name – a unit called “the satoshi” or “sat” for short. A sat is like the penny of bitcoin, only instead of being one hundredth, it’s one hundred-millionth – 0.00000001.
But talking about sats doesn’t change the conditions described above: sats are no more uniquely identified in the bitcoin code than bitcoins are. Enter “Ordinal Theory.” Invented in July 2022 and implemented in early 2023, it breaks down into 3 components:
A simple set for rules establishing an order by which to count satoshis: i) order of mining, and ii) order of transactions following a “first-in-first-out” principle.
A simple computer program that reads data from the bitcoin blockchain and then creates an index that gives a unique number* to each 0.00000001 of bitcoin by following the above rules. *A separate algorithm then translates the number into a series of A–Z letters (identified as a name), some of which will, randomly, turn out to be recognizable as words in various languages.
An entire astrological construction that creates the following categories of satoshis based on their rarity: common, uncommon, rare, epic, legendary, and mythic. To this typology one can add “exotics,” understood as sats “prized for reasons other than their name or rarity.”
Notice that “ordinal theory” must be understood as something like a gestalt or a Weltanschauung (or perhaps some other, even better German term) because it ranges so widely – from a basic way of counting bitcoin units, to the code that does the counting, to an entire way of relating to the construction of those programs. Ordinal theory makes “satoshis” and “rare sats” a reality, both by positing them as such, in creating them as the output of a new computer program, and by then categorizing those outputs to give people a way of relating to them. Within this worldview, bitcoins have not just become distinct coins – specific entities one could buy, hold, and trade. More than this, some bitcoins (that is, sats, of course) have become special, rare…unique.
Ordinal theory manufactures a world where bitcoins can be collected. The coins are virtual, but the collecting is real.
Step III: Sell, sell, sell
I hope my account of the first two steps transforms Sotheby’s murky marketing into something quite clear. If you live in the ordinal theory worldview, you are a sat hunter, and Sotheby’s has located the very best sats for you to collect. As of the time I write these sentences Pizza Ninja 1 has not just exceeded its reserve price; it has blown through its estimated sale range, and currently the price sits at $45,000. In other words, you’ll need to spend almost an entire bitcoin to receive one hundred-millionth of a bitcoin. But rest assured, Pizza Ninja 1 is inscribed!***
I suppose one could describe the above as immanent critique: I’ve tried to faithfully reveal how we got to the point we are today, with Sotheby’s auctioning pizza ninjas sats for over $45,000, but along the way I trust that making the whole thing clearer doesn’t render it any less mad.
And I want to emphasize that I’m not being disingenuous when I use the word “faithfully.” Perhaps the strongest pejorative found in my above description appears when I call element 3 of ordinal theory “astrological.” But I didn’t choose that term out of the ether or in order to sneak in some crypto-normatvism.4 It draws rather from the source for ordinal theory, Casey Rodarmor, who articulated the rules of ordinals, wrote the original code to implement the indexer and the namer of sats, and has generated the vast majority of text supporting all the ideas that are the basis of the Sotheby’s auction. Sotheby’s speaks glowingly of his invention of the “Rodarmor Rarity Index.” But before the ordinal project took off, Rodarmor himself expressed a rather cynical view of it on Twitter.
i would say it's astrology for men but ordinals has hoes too so maybe it's just astrology 🤷♀️
I can only imagine that Sotheby’s finds it somewhat discomfiting that the tweet is still up. In turn, I can only assume that the bedrock justification for all of this, in Sotheby’s eyes, is simply lots of dollar signs. After all, the creation of ordinals and ordinal theory has set off a massive rush of bitcoin transactions: “total Bitcoin fees for ordinals reached $44,792,565 in June [2023].”5 Clearly there are enough people in the “crypto community” who really do already know about and believe strongly in sats and ordinals. More to the point, enough of them have enough money to burn that bidding on Palindrome from Block 78 and Silk Road 391467764193 seems like a reasonable thing to do (and at $700, it looks like a bargain right now).
The auctions close soon, so good luck to everyone in their bidding!
***Appendix : all about inscriptions
I’ve excluded mention of “inscriptions” from my main text, partially for reasons of concision but also because Sotheby’s doesn’t directly discuss inscription, even though some of the sats they are auctioning are “inscribed.” “Inscription” is the name that ordinal theory gives to the fact that the updated bitcoin protocol allows one to add metadata to any transaction. This is the memo field of a check, but with slightly (just slightly) wider freedom to doodle – that is, you can write a note, draw a picture, or add a small audio or video file. Ordinal theory’s creator, Rodarmor, wants to call these memos “inscriptions” in order to convey the idea that the sat itself is now indelibly marked by whatever gets added to this memo field. He claims:
Satoshis can be inscribed with arbitrary content, creating Bitcoin-native digital artifacts. Inscribing is done by sending the satoshi to be inscribed in a transaction that reveals the inscription content on-chain. This content is then inextricably linked to that satoshi, turning it into an immutable digital artifact that can be tracked, transferred, hoarded, bought, sold, lost, and rediscovered.
The idea here is basically a much-improved NFT. Rather than using a centralized (what crypto folks call “off-chain,” i.e., the world) pointer to some distinct object of art, “inscriptions” purport to place that unique, non-fungible entity on the blockchain directly. Indeed, Rodarmor has persistently tried to argue that inscribed sats should be referred to as digital artifacts and be rigorously distinguished from NFTs. Digital artifacts, he claims are immutably stored on-chain and owned directly (by the owner of the sat).
But I suspect that Sotheby’s elided any mention of description from their fancy auction website, because even within the terms of the crypto universe, inscriptions do not live up to this ideal. First, because just like the memo field of a check, the next person to conduct a transaction with the “same sat” can add their own metadata. The ordinal theory gestalt deals with this by calling such an inscription cursed. Indeed, they are now trying to figure out a way to save cursed inscriptions by creating blessed inscriptions. I’m not making this up, I swear. Second, because as coders within the crypto community themselves have argued, the entire framework for creating, counting, and naming sats depends on an external software program (off-chain!). Ordinal theory is powerful metaphysics, which may itself change the world by altering people’s behavior, but it cannot change the empirical reality of the blockchain code.6 And in that reality, “inscriptions” must be understood as scribbles in the memo field of the check.
With both thanks and apologies to Matt Levine. I thought long and hard about an alternative name for this little bit, but “Programming Note” is just too appropriate.
This is what I assumed, which is why I got it completely wrong in my money book, where I write, “the coin is nothing more or less than a long, unique string of alphanumeric characters” (MHNV, 81). Nope, that’s wrong. The mistake has no bearing on the specific argument I make about bitcoin there, but it proves crucial for the story today. Because if coins were created with unique alphanumeric identifiers, then ordinals and ordinal theory would never have been invented.
I would agree, though for slightly different reasons. Fungibility is not about the nature of the money-thing. The token of the credit/debt relation doesn’t itself have to be easily divisible or consistent or homogeneous; the tokens can be a series of unique physical objects (shells, jewels, etc.). Fungibility is about the nature of the credit/debt relation. I can “owe” someone in lots of different ways, but to owe them money means to owe them a debt denominated in an abstract money of account. As money, one euro is, and must be, the same as any other euro.
Honestly, the pun wasn’t intended when my fingers typed it, but post hoc, I absolutely intend it! For all my readers out there who get the pun, I hope it helps justify the time we all lost, years ago, reading the Habermas/Foucault debate.
Bitcoin has high transaction fees in general, but the massive fees relating to ordinals likely stems from people sending sats to themselves (from one wallet they own to another they own) and adding inscriptions. This leads to transactions where the fees are 10 times higher than the amount transferred (e.g., the transfer of 546 sats for a fee of 6,460 sats).
Even staunch believers in ordinal theory and the power of inscription find themselves tacitly admitting this fact, at least when they try to explain the technical details of inscriptions: “Inscriptions invents the concept of ‘inscribing’ data on a sat by interpreting data embedded in transactions in a certain way.” To reiterate, ordinal theory is a worldview; it asks its followers to look at the bitcoin world a certain way.
Sam, you inspire me to make something up and offer it to Sotheby's to hawk.