Why predict?

When is a prediction not a prediction? A lot of the time. Often, of course, a prediction is just a statement about the future based on what we know about the present, and how it will change based on either the relevant rules, laws, and principles or on what we’ve learned from experience about it. 

Types of predictions 

Predictions of eclipses and whether a skyscraper will withstand an earthquake are based on laws and principles (laws of movement, the relation of force and tensile strengths, etc.), although they are not equally precise. A hundred years ago, weather forecasts starting using Newtonian laws, although now they also use statistical analyses of data; data-driven predictions make those forecasts more accurate over a longer stretch of the future. 

While we tend to think of the events of the universe as governed by immutable laws, we’ve been using data to predict the future for many thousands of years, at least if we take experience as a type of unquantified data. 

The Egyptians could predict when the Nile would overflow without knowing why it happened because many generations of experience told them that when the Dog Star rose, so would the Nile. They may have had explanations involving gods, but that’s a far cry from how we can predict with astounding accuracy when a space probe will reach Pluto. 

We like to think that deductive reasoning from principles is more reliable than inductive reasoning from experience and data because what happens tomorrow may not be similar to the past. But in practice, even if the laws from which we are predicting outcomes remain true and stable, the situations to which we apply those principles can be so complex that deduction is approximate if not unreliable: We know all the laws of physics that determine what number a roulette ball is going to land on, but physicists are as likely to lose at roulette as anyone else. 

Predicting versus explaining 

Whether deductive or inductive, these are the sorts of predictions we think of first when we’re thinking about predictions. But they are not the only types of predictions. 

For example, a bet is also a prediction, one with real and knowable consequences. But bets are an odd sort of prediction because there is an incentive to put your money behind less likely outcomes. That’s not how predictions are supposed to work. 

Then there’s a class of predictions that is not really about anticipating the future but explaining the present. For example, some science-fiction stories exaggerate a current trend to warn us about it, or just to help us expose hidden aspects of it. Suppose the use of mood-altering pharmaceuticals expands until we’re all taking happiness pills? Or suppose we can communicate with others anywhere just by thinking? 

Time-travel stories are often a different type of prediction, especially when they expose the paradoxes inherent in their premise. No matter how far in the future such stories are set, they’re exploring the nature of time, not making real predictions. 

Likewise, suppose someone invents a phone booth that lets you teleport anywhere, as in Larry Niven’s 1973 novella, Flash Crowd. Although the book postulates a future event, it’s really a way to explore the role distance plays in our lives now. 

How we use predictions 

Predictions are also used to try to get to the bottom of something in the present. That’s often the case with arguments about what the web will do to us and society. We’ve been having those disagreements in the form of predictions since the very beginning of the web in the early 1990s. Will it liberate all of humankind, or will it make us prisoners of gossip and mindless entertainment? Both of those arguments use predictions to highlight features of the tech and our engagement with it—usually to make a values-based argument about what we should do about that technology. 

It is but a short step from that to the use of predictions as a tool of rhetoric. For example, at least some of those who early on predicted that the web would liberate us from traditional cycles and habits did so in part because they—we—wanted to nudge the future in that direction. If you paint a plausible picture of the future, people may push toward it. If they don’t have that sort of prediction in front of them, it’s unlikely that they will.