If you think forecasting is rational and objective, based on facts and data generated by dispassionate experts, think again. In fact, what we predict reflects who we are. A case in point is Théry Report on the Internet. It provides a fascinating case study of the dangerous interplay between expertise, identity, and technological foresight.
Entitled The Information Superhighway, this report was written in 1994 by three communication experts and addressed to the French Prime Minister. Reading it more than thirty years later is very instructive. It shows how three of France’s most competent technocrats saw the future of the information superhighway and the role the Internet would play in it. The importance of the latter was recognized, but not taken seriously. We read: “Its cooperative mode of operation is not designed to provide commercial services. Its wide openness to all types of users and services reveals its limitations, in particular its inability to provide high quality real-time voice or image services”. This was written at the very moment when the Internet revolution was about to explode in the United States, when Marc Andreessen created Netscape, which would introduce the first Internet browser, and when a certain Jeff Bezos created Amazon, now one of the world’s largest commerce sites.
Why do smart people make these mistakes? Charitable souls are quick to point out that in 1994, many of the tools needed to turn the Internet into a truly robust commercial network were still missing. The network was in its infancy and accessible only through a very slow connection. Its evolution was hard to imagine. Granted, thousands of entrepreneurs were doing just that at the time. But it’s not the lack of foresight that’s interesting here. In fact, this is not the first time that experts have completely failed to predict an evolution in their environment. Beyond what I call the epistemic arrogance of those who know and are wrong, we need to dig a little deeper and ask ourselves who Gérard Théry was. He was the inventor of the Minitel, a cheap terminal introduced in France in the 1980s. Minitel was a great French technocratic success, and in some ways the precursor of the Internet. Minitel was a primitive, closed version of the information superhighway. A model in which the telecom operator, at the time a state monopoly, designed the system and fully controlled the technical and commercial aspects of the services offered. The Internet, on the other hand, is the open, somewhat anarchic system that emerged from a military research project and was taken over by academics. The Minitel is Descartes – ordered, controlled, closed; the Internet is Montaigne – messy, cooperative, open. These are two opposing cultural models.
In this report, the authors judge the Internet according to what they know and what models their thinking, i.e. Minitel. Thus, the absence of a directory on the Internet, which is due to its open and decentralized nature, seems to them to be redhibitory. As an open network, the Internet doesn’t seem as credible or secure as Minitel, and so on. This is a very classic phenomenon in strategy: the vision we develop of our environment is shaped by our social and professional identity. Minitel was the failed precursor of the Internet. It came before, it was a pioneer, but in 1994 it was clearly no match for the newcomer. For Théry and his colleagues, this situation must have been unbearable. How could a network built by hippies with so many flaws replace our technological jewel? So it’s not surprising that the report is skeptical about the Internet’s prospects. Identity bias turns the author, perhaps unwittingly, into a defender of his cause. Théry’s case is not unique. The economist Tim Harford cites an experiment carried out at the Experimental Science Center of the University of Oxford. Participants were divided into two groups: farmers, who would earn more if wheat prices were high, and bakers, who would earn more if they were low. They were then shown a graph purporting to show the price of wheat and asked to predict the future price, with a reward for accurate predictions. Despite being paid for their accuracy, the farmers consistently predicted higher wheat prices than the bakers. Harford concludes: “Everyone predicted what they hoped would happen.”
In the realm of forecasting, the influence of identity reaches far beyond shaping expectations—it molds the predictions themselves. While forecasting is often portrayed as an objective exercise conducted by detached experts, the reality is quite different. The Théry report provides a telling example. Despite the report’s recognition of the Internet’s significance, it dismissed its potential due to a bias rooted in the authors’ identity and past successes. The lesson lies not in futile attempts to achieve absolute objectivity in forecasting but in acknowledging and utilizing predictions as reflections of identity and perspective.
📖 This article is an excerpt from my book “Welcome to Uncertainty!“
📬 If you enjoyed this article, don’t hesitate to subscribe to receive future articles via email (“I subscribe” in the upper right corner of the home page).
🇫🇷 French version of this article here.