Non fungible tokens –  NFTs –  are a hot, even heated, topic. Most of us probably never heard of fungible before we heard of non-fungible. Money is fungible. If R1000 is deposited into my bank account, I have a contractual right to get *it* back. But *it* does not mean the very cash I handed over, or the very particular credits transferred by my friend – if indeed there is such a thing as ‘the particular credits transferred by my friend’. I’m just entitled to get *some* R1000, and there is really no reason I would care about this complete interchangeability, or ‘fungibility’ of money. Any dollar can stand in for any other dollar, without anything really changing.

However, If I deposit, into the bank’s safety deposit box, the Olympic medal I inherited from my grandfather, I would expect to be able to get that particular medal back. Perhaps I can be hoodwinked into thinking I am getting it back, while being given a replica, but that is not the point. I might find out I had a replica when I try to sell it and an expert takes a close look, for example. In any case, I would probably not deposit anything into a bank’s safety deposit box unless I was pretty confident I could recognise it and detect at least the most obvious attempts at swapping it for a cheap fake. Such a specific thing, which can’t just casually be swapped around for others ‘like it’, is non fungible – like an old-fashioned work of art.

So when art goes digital, we have some challenges. I’m not likely to want to pay big money for an image which the artist can sell again and again, by just emailing the image file to other buyers. If I’m going to pay real money, I would usually want to know that only I own the thing I bought. Maybe I could get the artist to handwrite me a note that says that only I own the image – and anyone else displaying it on their screensavers is cheating. That might not stop other copies getting out and about, but I could produce the letter and prove that I bought the artwork, if anyone asked – and perhaps I could sell it on to someone else – the crux of which transaction would probably be the handing over of the letter – my non-fungible token – not the handing over of the file containing the image, or an actual printout of the image.

Usually, when we hear of NFTs these days, we are not talking about handwritten letters from artists. Most NFTs, are digital ‘ledgers’ (transaction logs) which are designed in some way to be very, very difficult to tamper with. Blockchains, which first became famous for keeping track of digital ‘money’, in a way that doesn’t directly rely on the currencies of central banks, are fundamentally not really ‘about money’. Blockchains are fundamentally a way of keeping track of transactions in such a way that there are many copies of the ‘ledger’, distributed across devices controlled by a whole community of participants, and it becomes almost impossible to tamper with this ‘consensus’ record.

The technical details of financial instruments are not important here. The point is that we sometimes seem to care a great deal whether we are dealing with a (fungible) ‘thing of a certain kind’ or a (non-fungible) ‘particular item’ which we have in mind. We tend to think about fungibility as being a special noteworthy property of conceptually abstract things, like money, or stored information – and we tend to think of every day physical items as unique, non-fungible. A large OLED television of a certain model is functionally like another of the same model, but it has a unique serial number to confirm whether one found in someone’s car is the one that was stolen from my wall. We don’t really doubt the stable existence of ‘my TV’.

What about the substance of the world around us, from which the superficially unique items of the world are made? This, we might think, is the fundamental non fungible reality. When we hand over cash, it is a particular piece of paper or metal – even if we don’t much care which one. We may keep data on particular storage devices, and if they all burn up in the same fire, without there being additional backups, then the data may be lost forever because the particular infrastructure which contained it has been lost. Underneath all the representations and ideas, we may think, the physical world is naturally non-fungible, specific, unique.

Well, we could now argue a lot, and in different ways – but we are NOT discussing whether a book is still a book when a page goes missing, or how many pages must go missing before it is no longer a book, or when a page, through losing little pieces, becomes a scrap of paper, or anything along these lines. It’s true that when we name something we inevitably create an abstraction that becomes a bit fuzzy around the edges when we look too closely, but this is a quite different point.

Most fundamentally – when you see a thing, and you then look away, and you then, slightly later, see a very similar thing – can you ever really be sure whether you are seeing the same thing, or a fully fungible stand-in? The short answer, it seems, is no!

What about people – surely we are non-fungible? Can we not be pretty sure that our children are our children, and not someone else, when they come home from school? In every way that matters, I think we can recognise our children, and many other individuals, but perhaps this is because they are too complicated to be fundamental ‘items’ in any sense that can be made at all precise in engineering terms. There is some unique, largely stable, but also ever-changing pattern in the universe which we see as this friend, or that child.

My bank has some copies of a database from which it can reconstruct my transaction history. It is difficult to hack, and so the accounting claims contained in the database are pretty stable, but this stability has nothing to do with any efforts keep track of ‘particular’ units of money – and no one is troubled by this fungibility. The bank has a specific number of copies of its database, and if these all burn down at once, it would be very problematic, and there would be no way of pulling a copy of the database out of thin air. The servers, and storage media on which the database lives, are solid, specific, non-fungible items, and if they are destroyed, there is no way of just whipping up an equivalent storage device, with the same string of data.

If we drill down, we find the material from which items are made. Zooming in on a computer reveals silicon wafers or thin strips of metals, or magnetic materials, or large molecules from which plastic casings are made. What makes these materials what they are is their ‘chemical’ structure – the bonds between atoms; bonds which are made up of electrons. When we try to find the individual electrons, they turn out to be very slippery.

If we watch very carefully, is it possible to keep track of electrons and tell which one is which? It appears not. Perhaps we could, if we knew how – but according to the standard way of describing electrons, through what is called the overarching framework of ‘quantum field theory’, this does not appear to be possible – and it really may be a fundamental aspect of nature. It’s not just about limits of technology, or thermodynamic stability – in quantum field theory we can’t even talk meaningfully about electron number one and electron number two. Electrons appear to be like money in a banking database – we can keep track of how many there are, and where they are, but we can’t actually say which one is which.

As we drill down into the nuclei of atoms which are bound together by these fungible electrons, we find the famous protons and neutrons which most of us learned about in school. For some decades we have known that protons and neutrons are made up of ‘quarks’. The question naturally arises whether these quarks are made up of other things which we might be able to talk about – but so far, there is no evidence that we need to think about any such more detailed structures in order to explain anything we see in experiments.

So is a proton a ‘fundamantal’ ‘thing’, or some kind of froth in a sea of quarks? In some sense, it’s definitely the latter. What about the quarks themselves? When you look at swells in the quark ocean, can you tell one quark from another? As far as we know, no! Quarks appear to be fully fungible. The same is true of all fundamental particles which we have found – W and Z bosons, neutrinos, Higgs bosons, the lot. Every time we run out of ingredients, and the thing appears fundamental, it also turns out to be fungible. If we try to describe these particles as non-fungible, we end up making all sorts of predictions which are simply not correct – such theories simply fail. The so-called standard model of basic physics is fundamentally built on ultimately fungible ingredients, and everything that appears non-fungible is like some kind of wave, or current, or eddy in a sea – a pattern that persists and superficially appears to be its own thing – while the underlying substance from which it is made remains slippery and elusive – fungible.

It is not unusual to think of fungibility as a property of clever abstractions, like fiat money, banking credits, and so on. The universe can appear to be made up of particular, non-fungible, ‘items’, and we may think that it is our manipulation of these objects, against the backdrop of abstract concepts, that creates and embodies the notion of fungibility. In fact, it’s probably more accurate to think of non-fungibility as the abstraction, an illusion, a figment of our imagination which we invoke to make sense of the fact that patterns persist. ‘Items’ appear to exist in specific ways.  Underlying the most persistent patterns, deep inside the most solid stone hewn by artists centuries ago, or in the heart of stars shining for billions of years, we have yet to find any truly non-fungible building blocks.