Ergodicity and Digital Culture
New information technologies have changed our culture in depth. Some of the most striking features are that the spread of information is cost free and it is instantaneous. It makes the digital culture different from both oral tradition and written tradition (with or without printing). I discuss in this article what we can derive from these differences from an ergodic point of view.
Information Technology and culture
Information Technology has a huge impact on how the culture is developed and transmitted through time. A striking example is the rise of Protestantism and more generally of the Reformation of the XVI° century. One of the major factor is undoubtedly the widespread of Holy scriptures that has been made possible by Gutenberg’s printing press. Long story short, the printing technology spread literacy. The access to the text of the Bible was no longer filtered by an elite (the clergy). Moreover, analyses, critics and various texts could be circulated much more quickly than before. Remember than before the printing press, copying information was costly and slow: some people literally dedicated their whole life copying the Bible. If someone had a new theory he could only physically spread it only few meters from himself because of the limited reach of the voice. The fact that sounds don’t last is also a hard limit. All of these defects were somehow corrected by the printing press: you can make many copies of a text easily and cheaply, plus you can spread the text so that people may access it in parallel.
Virtualization and culture
New information technologies have made communications between the world of ideas and the world of objects easier. You can use your smartphone to digitize any moment of your life (and share it across the globe via Instagram or TikTok), and alternatively you can use a 3D printer to turn a file (essentially a big number made of 0s and 1s) into an object. As a result ideas have acquired characters of objects and vice-versa. For instance bitcoins are nothing but numbers but behave like objects : you cannot reuse a coin multiple times (whereas a number just remain the same number eternally). On the other hand you can print electronic tickets on a white sheet of paper: you can destroy the paper it won’t destroy your ticket that you can print again. The ticket is an object that behaves like an idea. Indeed it is indestructible (like an idea) and copies of it are indistinguishable from the original one. In facts the ticket is just used a pointer to a database. The older readers may remember how it is different from what happened if you lost a concert ticket in the 90's…
This digitization of the world is related to the subject of this article because it has generated a culture that is a chimera obtained by mixing oral and written way of communicating ideas. Like in the oral tradition, digitized culture is delivered via sounds and images (podcasts). But thanks to information technologies, those sounds/images have permanence, like in written tradition. They also can be spread instantaneously across the globe and accessed in parallel.
Ergodicity and Culture
The ergodic hypothesis is a principle used in thermodynamics. The basic idea is to be able to describe what are the equilibrium states of perfect gases. From a phenomenological point of view, ergodicity is verified if, more or less, every time that you compute a statistical measurement (across space or time) you find the same result. It means that randomness is “well shared”. Another way to see it is that, when you have ergodicity, doing N random experiments in parallel will give you the same result as doing N experiments one after the other. A way to illustrate this, proposed by N.N. Taleb, is very telling: imagine a hundred players going to the Casino. You know that among them one is going to be bankrupt. You can compute the expectancy of return by computing the average return (the total wealth of the players out of the casino minus the money they brought in the casino) which is not 0. But if you take a single player P going to the Casino 100 times, you know for a fact that P will be bankrupt, because in one instance P will lose everything. P’s average return is 0. In this case the ergodic hypothesis is not verified.
What is the relation with culture and new information technologies? For the largest part of history, we had an oral tradition. Schematically, stories were said around the fire camps and transmitted across generations. It does not take a degree in evolutionary psychology to imagine that this kind of information transmission is filtered through time. Most interesting stories turn into myths, religions or traditions. The more efficient (or maybe luckiest) ones survive the test of time. What is sure is that stupid stories cannot pass the test of time because, as the player going to the casino multiple number of times goes bankrupt at some point, if they were stupid enough to simply have a small percentage of failure, they would have led to the death of the people believing in it. And as the only way to propagate information is via living entities they would have simply vanished.
Written culture changes that: because of the permanence of written texts, and of the difficulty to write a text (you can see that the Hammurabi’s code was literally carved in stone), the filter becomes different. The editor’s work become paramount: indeed, before carving letters in stone you have strong incentives to write things that have been vetted many times. So information is filtered through an elite: the scribes, copyist monks etc. Printing press and later radio and television do not alter fundamentally this characteristic. Peer reviewing process is the scientific version of this.
In the digital era writing, publishing, modifying, updating and spreading information has a marginal cost of 0: once you have bought the infrastructure (a computer and access to the network) the fact that you publish do not bring extra costs (not like printing a new copy of a book which at least requires sheets and ink). The vetting process will adapt: it is not before publication but after publication that it will take place. More and more we see that at work: for instance during the COVID19 hundreds of articles have been published online on BiorXiv about this new disease. It means that the filtering process that allows the building of the culture is changing. But what was done sequentially and vertically among generations is now done in parallel and horizontally as well. Information is not only widely spread but instantaneously so !
So the natural question arises: is culture building (viewed as the process of filtering interesting stories/myths) an ergodic process ? Does the digital culture achieve the same kind of results than the oral tradition ? Do horizontal filters work the same as vertical filters ?
Buzzfeed vs Tales and Stories of the Past with Morals
The first thing to notice is that the incentives for the kind of stories that makes listeners want to hear them haven’t changed. Clickbait is nothing new. But the ways stories are imposed has changed. In an oral tradition, the authority of what makes a good story comes from the fact that those stories are old and repeated over and over. In a digital tradition the fact that someone listen to a story is mostly driven by Buzz like phenomenon. There is a kind of global synchronization (Panurgization ?). that has never been witnessed. And in this domain also you have a “winner take all” kind of phenomenon. So if it is true that stories are tested in mass scale and in parallel, there is a difference in the nature of how stories have been selected. One problem is that if an idea takes ground for some reason, then it is so widely spread that it is difficult for other ideas to be tested at the same time. It greatly differs from vertical culture in which many different fibers of ideas can be tested in parallel. In the COVID19 crisis it is striking how everyone has copied the “solution” used in China: general lockdown. I don’t know if other approaches are realist, but some are definitely worth considering: from the mandatory wearing of masks in public to selective lockdown (due to the very particular death distribution among age class) etc. The same unifying tendency has been witnessed in the controversy around treatment based on hydroxychloroquin. The problem is not in the use or the test of a treatment without full knowledge of its consequences: nobody knows much about that disease and what to do, but you have to try things! The problem is that it does not appear to have such good results that it terminates the COVID19 crisis. But it takes all the space and maybe (I just have a distorted view of what actually happens but it appears like that from the outside at least) discourage other tests.
This is an horizontal world
I don’t have solutions to propose. These phenomenons are new and we need time to adapt. But realizing that digital culture is something fundamentally new, and that you can’t rely on an ergodicity hypothesis, is important. Moreover, it is not a phenomenon that will vanish in a foreseeable future. We need to develop tools and concepts to filter new ideas in our digital culture so that we can get the best from the horizontal filtering while avoiding its worse aspects.
