The map and territory of NFT art
post by frcassarino · 2020-12-29T11:45:34.960Z · LW · GW · 12 commentsThis is a link post for https://federicorcassarino.substack.com/p/the-map-and-territory-of-nft-art
I’ve recently become aware of the world of non-fungible tokens.
Wikipedia puts it as:
A non-fungible token (NFT) is a special type of cryptographic token which represents something unique; non-fungible tokens are thus not mutually interchangeable.
Non-fungible tokens are used to create verifiable digital scarcity, as well as digital ownership, and the possibility of asset interoperability across multiple platforms.
The application of NFTs that I found the most thought-provoking is in digital art. There’s websites, such as opensea.io, that have listings of NFT-based digital paintings.
In that website, you can buy any of the listed NFT-based paintings, and become the proud owner of the “original” version of a digital painting. People are paying outrageous sums of money for this. An NFT-portrait of Ethereum co-founder Vitalik Buterin dressed like a medieval harlequin recently sold for $141,536.20. The NFT-art market is absolutely booming.
What’s the punchline? That the “original” version of a digital painting is pixel-by-pixel identical with any of its copies. On opensea.io you can go ahead and download a perfect .jpeg copy of any of the listed paintings for free. You can download the portrait of Vitalik Buterin right here, and set it as your wallpaper. It’ll be the exact same portrait that the buyer paid 141k for. There is absolutely zero material difference between the original and the copy, except for the fact that, in some technical fuzzy way, one is the original and one isn’t.
Most people are either perplexed when hearing this, or react with scorn. After all, it’s intuitive to think that digital paintings being infinitely and perfectly replicable defeats the point of paying for an “original” version.
My reaction is that NFT-paintings are a wonderful reduction of the idea of “originality”, and it’s a great exercise to analyze the phenomenon from the lens of map and territory.
Let’s pick a more traditional example: what makes DaVinci’s Mona Lisa worth more than a replica of the Mona Lisa?
One can point out a bunch of different factors at play here. DaVinci’s Mona Lisa is more valuable because replicas were made by lesser artists and result in inferior paintings that don’t captivate the viewers as much. DaVinci’s Mona Lisa is more valuable because it’s the only one that has the exact strokes of paint that DaVinci plastered on the canvas, while the replicas looks subtly different. DaVinci’s Mona Lisa is more valuable because of its history, because it’s the painting that was in Napoleon’s bathroom, and the painting that was stolen and returned to the Louvre, while the replicas aren’t. DaVinci’s Mona Lisa is more valuable because the atoms that make it up are the same atoms that were there when DaVinci finished painting it. DaVinci’s Mona Lisa is more valuable because everyone agrees that it’s “the original” Mona Lisa, and everyone agrees that “the original” anything should be more valuable.
Let’s unpack this.
Let’s say scientists devised a way to create a perfect copy of a painting. They manufactured a machine that can read the exact configuration of atoms in DaVinci’s Mona Lisa, and create a perfect “clone” with the exact same type of atoms arranged in the exact same order. Would those clones be just as valuable as the Mona Lisa that hangs in the Louvre?
Certainly not. Though the boundary between original and replica does get blurred a bit in this scenario, the original Mona Lisa would still have a lot of a fuzzy historical/sentimental value attached to it, as those atoms/that painting (and not its perfect copy) is the one that DaVinci painted. But this does tell us that its value is not derived, for the most part, from the fact that the Mona Lisa looks different than the replicas.
We could pose even more convoluted thought experiments. Like what if the machine, instead of copying the configuration of atoms, instead split each atom of the original Mona Lisa into two, in some sense creating two “originals”? But I think these kind of thought experiments just expose that there’s no good answer at this level, and that this exercise just misses the point.
People have a map that says that there’s originals and replicas, and that the former is worth more than the latter. We drew this map to navigate the uncomplicated, familiar terrain of painters and paintings, sculptors and sculptures, seamstresses and garments. Once we tread onto such unfamiliar territory that we are questioning what happens if we split atoms, we can assume that our model of which painting is rightfully the “original” has broken down. The map is too simple to reflect the idiosyncrasies of the territory.
Originality at this level is a useless term, and at the end of the day when putting a price on a painting, what matters is which painting everyone agrees is the original, fuzzy epistemological reasons be damned. It was never about the atoms; the map is not the territory.
What’s especially interesting about this thought experiment is that it shows how maps persist, even when the territory that gave rise to these maps changes. There’s a viscosity to them; they don’t just disappear from our collective minds as soon as they stop being useful.
We can think of all the aforementioned factors - how replicas tend to be of lower quality, how replicas don’t have rich histories, how replicas aren’t hung in fancy museums, etc - like characteristics of the territory that gave rise and credence to the map in the first place. Because we humans observed, in a bunch of different contexts, that originals tend to have a bunch of characteristics that make them better in many ways to replicas, we built a map that makes it intuitively obvious that it’s important to distinguish originals from replicas, and originals should be considered more valuable.
Even if we design a thought experiment that takes away literally everything that makes an original better than a replica, in the real world with real people who have pre-existing beliefs, the “originality” map lingers. People still are willing to pay more for the original.
Which brings me back to NFT paintings. In some sense, the fact that crypto art is made up of non-fungible tokens, makes an “original” CryptoKittie and a “replica” CryptoKittie not mutually interchangeable. It’s similar in a way, to our thought experiment about a Mona Lisa copy with an identical configuration of atoms.
But the subtler insight here, is that this fuzzy, super-technical, inscrutable to laymen claim to originality, is not the territory. It’s an excuse for us humans to slowly reach for our familiar model that “there’s the original CryptoKittie and there’s the copy”, and as soon as we buy into it, it’s enough to trigger our “originals are more valuable” mental routine, and the original will be worth 200k, while a bit-by-bit perfect copy will be worthless. The kicker here is that we’ll be just as able to derive meaning from owning the “original”; it’ll be satisfying in the same way that hanging the Mona Lisa in your living room would.
I find myself reacting to this with both fascination at how us humans can derive meaning from almost anything, and with scorn at how silly this all is.
Let’s end this with a few more reflections.
It’s interesting to note that while the question of what makes something original is a great one in some sense- I did write an entire essay about it - it’s also banal in the sense that it’s immediately diffused by having a notion of map-and-territory. Watching people be perplexed by the NFT phenomena makes the widespread confusion between map-and-territory very vivid in my mind.
It’s also cool to see how a model slowly propagates societally. Right now, there’s a miniscule portion of the population that know or care about NFT’s. A lot of people hear about it, and immediately dismiss it as weirdo nonsense. “This makes no sense without real paintings”. It’ll be cool to see how —as NFTs become widespread in games and elsewhere— the idea that a digital item can be “authentic” goes from being niche and hard to grasp to intuitively obvious to everyone.
It’s also important to realize that forcing scarcity into something as seemingly intrinsically multipliable as digital artwork is absolutely not-unique and we can find similar examples everywhere, both in the digital and physical world. Examples of artificial scarcity can be found in copyright, DRM, planned obsolescence, the diamond industry, paywalls, collectible cards, torrent poisoning , the Agricultural Adjustment Act, Tulip mania, price fixing by cartels and monopolies, signed merchandise, Chick-Fil-A closing on Sunday, knockoff Gucci bags, and so on.
Finally, I’ll gleefully mention that as I’m finishing writing this essay on originality, I found out that twitter user @DCCockFoster has used the Mona Lisa example to make the same points I made on NFT-paintings. Sorry pal, you should’ve tokenized the twitter thread.
12 comments
Comments sorted by top scores.
comment by lalaithion · 2020-12-29T20:24:22.068Z · LW(p) · GW(p)
It's worth noting that this is nothing new; photographic prints have been sold for millions, despite them being trivially replicable. In this case, the uniqueness of what counts as the "original" is constrained by legal contracts as opposed to cryptography. In fact, it's not unheard of for an artist to print some number of identical "originals" instead of just one. According to Wikipedia's list of most expensive photographs, the most expensive photo ever sold was Rhein II. This photo has six original editions, of different sizes.
Replies from: gworley↑ comment by Gordon Seidoh Worley (gworley) · 2020-12-31T02:29:47.231Z · LW(p) · GW(p)
To be fair, though, when dark room work as necessary to develop photos there actually was an original print generated from the negative that might have desirable features that would lead to it being copied rather than the original negative, though then again there was still that original negative to go after as the true "original".
comment by Measure · 2020-12-29T16:05:45.884Z · LW(p) · GW(p)
I haven't heard of NFT's before. Is my understanding correct that the NFT functions as a sort of "title deed" of ownership for a file with some verifiable mathematical relationship between the binary representation of the token and the binary representation of the file so that one can prove that a particular token is associated with a particular file?
Would it be possible to e.g. change one pixel of an image file and generate a new NFT for the altered file such that you couldn't necessarily tell which image-token pair was the original except that the original author has the "correct" version of the file published somewhere for everyone to verify?
comment by Pattern · 2020-12-29T17:14:39.554Z · LW(p) · GW(p)
The kicker here is that we’ll be just as able to derive meaning from owning the “original”; it’ll be satisfying in the same way that hanging the Mona Lisa in your living room would.
Unless you value having the original. Imagine a private collector and the head of an art gallery, both happy they have the Mona Lisa. And only the thief who promised the private collector they'd switch it out for a forgery knows which is the forgery, and which, is the original.
Let’s say scientists devised a way to create a perfect copy of a painting. They manufactured a machine that can read the exact configuration of atoms in DaVinci’s Mona Lisa, and create a perfect “clone” with the exact same type of atoms arranged in the exact same order. Would those clones be just as valuable as the Mona Lisa that hangs in the Louvre?
Certainly not.
This thought experiment is probably right, although it hasn't been performed.
If the value of the original is derived from it being the original - that knowledge - then if a forgery was switched out for the original successfully, it would obtain that value (or price).
Replies from: frcassarino↑ comment by frcassarino · 2020-12-29T17:31:54.724Z · LW(p) · GW(p)
Unless you value having the original. Imagine a private collector and the head of an art gallery, both happy they have the Mona Lisa. And only the thief who promised the private collector they'd switch it out for a forgery knows which is the forgery, and which, is the original.
Indeed. That maps well to the idea that we value "originals" more for the sake of them being originals, even if they don't provide any additional utility to us compared to copies besides that fact.
comment by Coafos (CoafOS) · 2021-01-02T22:00:39.775Z · LW(p) · GW(p)
I think a useful concept would be the colour of bits. For example, a digital song can be bought on a CD or downloaded from the internet. The computer does not see a difference between them, because it just sees a number, but in the eye of the law, one of them is legal, the other is not.
The number on the CD is coloured "green", the downloaded number is coloured "red". Green numbers are legal, but red numbers are not. If you upload a song a from CD, it will be red because you can only send red numbers. However, if the studio produces a new CD, it will have green numbers because they have the copyright to the song.
Anyone can copy a digital artwork because it is just a number, but the copied number will be "yellow" coloured. With an NFT you do not buy a number, you buy the right to make this number "blue". This right can worth a lot of money if a blue number has a higher value than a yellow number.
comment by Donald Hobson (donald-hobson) · 2020-12-29T17:35:38.763Z · LW(p) · GW(p)
Value is a Keynesian beauty contest. If everyone believes something has value, then anyone who buys it can sell it later.
What you really have here is an abstract crypto-economic value token, and a piece of artwork that are only notionally connected. If anyone can make as many tokens as they like, the value of a token falls to 0. By requesting that people make art, this functions as a proof of work. You limit how many tokens are produced.
Replies from: Dagon↑ comment by Dagon · 2020-12-29T19:44:04.914Z · LW(p) · GW(p)
This is a subtlety often lost when people talk about fungible or non-fungible. Fungibility is not a function of the token, but of the USE of the token. It's about whether consumers care about the uniqueness, not just about whether it's unique.
Currency is the standard example of fungible items, and even there, notes and coins are only fungible for some purposes. Stolen currency with tracked serial numbers are not fungible with other bills. Many coins are worth more than face value based on properties that are otherwise-ignored for the fungible use of money.
comment by MSRayne · 2020-12-29T14:11:22.030Z · LW(p) · GW(p)
This is an interesting point, but I think you're missing something fundamental about what originality means. This isn't a question of map versus territory, it's a question of what identity is.
The same people who value an original Mona Lisa, or an original NFT, would likely also be wary of treating a copy of someone as equivalent to the original person. Those who see no distinction, would probably see copies of people as fungible, too. This is an argument between pattern identity theory (you are a data pattern with some number of instances) and continuity identity theory (you are a particular instance of a data pattern, picked out by having only gradual change in physical makeup over time).
As a continuity believer, I think that the original Mona Lisa objectively is more valuable and that only something which destroyed the information of which one that is could possibly render it fungible with a copy - for the same reason I believe that my own continuity of consciousness is an absolutely necessary prerequisite for a being to be defined as "me", and that a perfect copy of me would be another person entirely who just happens to resemble me. The only way you could get me to consider the copy equivalent to myself, is if you erase from existence (or at least from the knowledge I can ever hope to personally access) any evidence of which is which.
NFTs, on the other hand, seem a bit more muddled, as the real original of any digital artwork I've ever made (and I've made a lot of them - didn't know NFTs existed though, I may have a lucrative business opportunity now :P) is the copy that lay on my hard drive. And even that may not be the original, since it might have been overwritten or moved to a different region of memory, which would require copying the data and then deleting the original. It's unclear how continuous any data structure on a computer could be said to be, so in the case of files, there may really be no original.
This, by the way, is why I am uncomfortable with standard ideas of uploading. Besides the obvious dangers of rogue hyper-self-copiers, I suspect that continuity of consciousness in a digital medium might be compromised altogether (as Integrated Information Theory also suggests). I think uploading could only safely occur by gradually migrating into an artificial neural net made as a physical brain (rather than software), with the physical parts instantiating you changing only gradually and continuously as they do in the human body - not a data structure in a standard computer which moves by being copied and deleted, which I worry could be a philosophical zombie.
Replies from: frcassarino↑ comment by frcassarino · 2020-12-29T17:48:55.656Z · LW(p) · GW(p)
Great reply. I share your beliefs on consciousness copying, and would have the same concerns.
As a continuity believer, I think that the original Mona Lisa objectively is more valuable and that only something which destroyed the information of which one that is could possibly render it fungible with a copy - for the same reason I believe that my own continuity of consciousness is an absolutely necessary prerequisite for a being to be defined as "me", and that a perfect copy of me would be another person entirely who just happens to resemble me. The only way you could get me to consider the copy equivalent to myself, is if you erase from existence (or at least from the knowledge I can ever hope to personally access) any evidence of which is which.
I do grant that in some sense there are features of some territory which we could name originality. There's complicated boundary questions, as we've both outlined.
It's not obvious to me why the Mona Lisa would be objectively more valuable; even if it were objectively original, it doesn't follow that the fact that it's original makes it more valuable.
Even if there's a good argument for why it's objectively more valuable, my broader point is that the reason why it's more valuable in practice is because people have maps that say that originals are more valuable than copies.
Whether that's true or not objectively doesn't change that. And those maps were originally brought on because as a heuristic, getting an original X usually brings more utility in many ways than getting a copy. But we are so used to those maps, that even NFT paintings are enough to trigger them, even though there's no conceivable advantage of owning the original. Actually, the sole advantage is that because we are so used to applying the map that rewards us for owning originals, we will in fact gain utility/pleasure just from knowing that it's an original NFT. Very meta.
comment by Filipe Marchesini (filipe-marchesini) · 2021-02-16T13:10:47.563Z · LW(p) · GW(p)
Please, I love this discussion, I super upvoted and if you have any new thoughts about NFTs I would like to hear you talking about. Did you get any new information or have you updated your thoughts about NFTs? I am completely fascinated on how this is even possible, people caring so much about useless maps, and the territory has become "it is super valuable to have a map that tells which of the two perfect copies of a territory arrived first", or people are just pretending for the sake of playing the game in which allows you to speculate on the value of useless maps? I say "useless" in the sense that you can not have any extra utility from knowing that you have two exact sequence of bits "00101" and "00101", but you know that the first one was generated first than the second one, which is a perfect copy of the first one. Even if I have this information, the value provided by those bits doesn't change, so I wonder if this is just a game in which you pretend to care about which sequence of bits came first, even if you can not be absolutely sure which one came first (it all depends which computers uploaded a copy of this sequence of bits first to the internet, not the computer that really generated the first instance in the first place). This is completely crazy for me, but it makes me imagine that I could get a lot of value from the digital art work that I usually generate while learning new subjects in math, and people would love to pay more for the first copy of my digital art, but the second copy would be worthless, and that's because the first copy that I uploaded is the first copy that has been put on the blockchain, why would anyone give a f*** about this? But if everyone is playing a game in which "if you get lucky on getting the first copy and everyone pretends that the first copy is more valuable, you can get a lot of money by playing this game and trying to get the best first copies of every digital piece". LOL HELP ME WHAT ARE WE EVEN TALKING ABOUT, is this what is really happening? Just a new emergent game with strange rules just to have fun and get money and happiness points by randomly assigning points to random digital copies on a specific blockchain? Should we even play this game? I mean, is it net useful to humanity to pretend and play this game?
I'm having fun exploring the subject and I would love anyone to expand on this topic, if anyone has ever explored this craziness we are seeing.
Replies from: supposedlyfun↑ comment by supposedlyfun · 2021-02-16T13:28:40.611Z · LW(p) · GW(p)
This comment captures both the substance and the style of my reaction to the story.