Predicting the Future: Can Science do Better than Ancient Oracles?

For now, we see through a glass, darkly, but then face to face: now I know in part, but then shall I know even as I am known. 1 Corinthians 13:12

The pigeons that frequent my garden can predict the future. They know that I give them bread crumbs, so they fly down there when they see me appearing. Their prediction is usually correct, even though they probably have no models telling them why this particular human feeds them every once in a while. What about us, humans? Can we do better than pigeons?

Just like pigeons, people have been trying to predict the future for a long time, although the idea appears in written documents only starting with the late second millennium BC. Before that time, nobody seemed to have been worried about oracles and predictions, or at least, they didn’t put that into writing. Then, things changed, as we can see in the clay tablets of Akkadian times in the Near East. (I am citing here from Julian Jaynes’s book “The Origins of Consciousness” (1979)). Three main categories of forecasting appear in writing: omens, divination, and sortileges.

Omens are also called “presages” or “portents.” They are attempts to describe correlations: you see B happening after A, and you conclude that A caused B. Say, you see black clouds, and then the rain comes. You suppose that there exists a correlation, and you are correct. But how about a black cat crossing the street in front of you? Omens were based on observations, but they didn’t deal with the slippery concept of “causation.” This is not an incorrect view if applied to pigeons who see a human being in the garden and assume that he will feed them. Why does he do that? It is probably beyond the capability of both pigeons’ brains and human ones.

Divination (also “prophecy” or “oracles”) is a more active forecasting method. You don’t just wait for messages from the Gods; you actively look for signs that will tell you something about what the Gods are up to. One way was to ask questions to people believed to be able to hear the voice of the Gods directly; the “Pythoness” of Delphi is an example. Or, you may perform divination by “extispicy” or “haruspicy.” It consists in examining the entrails of a sacrificed animal, typically a goat. It must have been a little gory, but there were specialized people trained to do exactly that. Another form of divination is the use of random events. A good example is the use of a random procedure to select a hexagram from the I-Ching book — and then act on the received message. An extreme version is the Russian Roulette. I discuss it in my book “Before Collapse.” (2019). Strange as it may seem, in the United States, there may be a few tens of people every year who aim a revolver loaded with a single round at their head and then pull the trigger. A few of them die as a result. It is a way of questioning the Gods about whether or not they want them to live. Sometimes, evidently, they don’t. I don’t know what signs the pigeons who frequent my garden look at to decide whether their benevolent deity will appear, but they seem to be waiting for me every time I walk into the garden. For sure, they are smart enough to avoid killing themselves to know.

Sortileges, or sacrifices. You directly ask the Gods to help you, and you prod them by doing something that you believe they will like. It is the extremely ancient idea that humans were the “stewards of the Gods” who created them exactly with the purpose of feeding them. It is not so clear why the Gods should be happy at seeing animals being killed in front of the statuary representing them and then eaten by the priests (unless it is a Qorban-Olah, a burnt sacrifice, but even in that case, the Gods don’t eat).

Human sacrifices are rare, but historical records show that they were not so uncommon. Perhaps they remain so in modern times, even though they have different names (say, “giving your life for the country”). In the Russian Roulette, for instance, the person who tests the Gods offers himself/herself in sacrifice to the Gods. Again, it is not so clear why the gods should be happy that somebody does something so stupid, but it is an ingrained way of thinking in human minds. We tend to believe that the universe involves some form of gift-giving: you can influence the will of the Gods by doing something for them. I don’t know whether pigeons ever thought of sacrificing one of their flock to the deity that feeds them bread crumbs, but I did see pigeons with their skull broken open in a way that only the beak of another pigeon could do. Was it a pigeon sacrifice? For us, non-pigeons, it is impossible to say.

Overall, there is a clear parallel between what the ancients were doing and what we are so proud to do with our “Science,” which is supposed to be “exact.” The triad of omens, divination, and sacrifices corresponds well with our triad of observations, experiments, and inventions.

So, we are more careful today in establishing correlations in our observations on the basis of a weird set of incomprehensible rituals called “statistical validation.” Our “laboratory experiments” are assumed to determine “universal laws” obeyed by every single atom in the universe. The only difference with ancient divination is that, whereas Gods are notoriously capricious, universal laws are supposed to be fixed and immutable. Finally, sortileges or sacrifices are what we do nowadays when we ask “Science” to solve some important problem for us. We are supposed to invest money into scientific research in order to find ways to gain health and wealth, kill our enemies, and shape the world the way we would like it to be. It may work better than ancient prayers, but often it involves huge numbers of human victims (say, “the war that will end all wars”).

On the whole, we tend to think that the ancients were naive and, sometimes, even a little dumb. So, we disparage their ideas about knowing the future as little more than silly superstitions—horoscopes read in the Sunday supplement of your newspaper. Maybe. But is our “Science” (written with a capital initial letter) able to do better? The idea that our forecasting is “scientific” may be no more than a delusion of our age, and we might doubt that the convulsions of the Delphic Pythoness were so much worse than our science-based predictions. Even when we know the “laws” at work in a system, predicting its behavior may be impossible. Even just three bodies of similar mass interacting with each other are said to have given Isaac Newton a terrible headache. The universal gravitation law may be deterministic, but the unavoidable uncertainty in determining the initial conditions rapidly causes reality to diverge from computation.

If three bodies are already too many, most of the natural systems we want to study are far more complicated. Complex adaptive systems (CAS) are the rule in nature, and they are formed of a large number of elements linked to each other by feedback relations. These systems react non-linearly to perturbations so that the uncertainty in the initial conditions is amplified with time, and their behavior is impossible to predict with any long-term accuracy. In other words, complex systems “always kick back” (and sometimes with a vengeance). The systems called “chaotic” are even more difficult to predict. Lorenz’s butterfly, which beats its wings in Brazil and causes a hurricane in Florida, is a paradigm of these unpredictable systems. To say nothing about those systems where uncertainty is built-in and whose behavior can be described only in statistical terms, such as quantum mechanics. Not even the Pythoness of Delphi could know whether Schrödinger’s cat is alive or dead.

These problems have been known for a long time, yet people tended to maintain a naive faith in “Science” not unlike the way the ancients saw their oracles. But scientific hubris took several terrible blows in recent times, not the least caused by the poor show of the solemn virologists who would regularly appear on TV at the time of the COVID pandemic. It was not just the pandemic that was disastrously mismanaged. It is a whole series of pretended scientific miracles that are presented with great fanfare as able to solve problems, but that usually only worsen them. Nuclear fusion, hydrogen, and biofuels are good examples in the field of energy. We are now discussing the idea of “climate geoengineering,” which may be our last chance to survive the collapse of the ecosystem but carries huge risks that we may not even imagine right now.

So, nowadays, modern scientists look very much like the oracles of late classical times, who tried to provide obscure predictions that could be interpreted in more than one way and, hence, always say that they were right. They paid for their presumption by being widely ridiculed and disbelieved. Something similar seems to be happening with modern science and its presumptuous priests; they are widely ridiculed and disbelieved. They risk going the same way as the Pythoness of Delphi did.

In the end, the future remains a blank slate: we cannot penetrate it; at best, we can see vague shapes on the mysterious “other side,” as if looking into a mirror, darkly. Will we ever see clearly? Our modern oracles, which we call “artificial intelligence,” may be better than the ancient Gods in speaking to us, or perhaps not. And, in the end, who said that the ancient were wrong and we are right? Are natural laws really as absolute as we think they are? Or does “something” else rule the universe? Who can say? Maybe the pigeons of my garden know better than us.

A version of this post as a video clip.