(The following editorial first appeared in the May 1999 issue of Analog Science Fiction and Fact. Thirty-five others, on a wide range of topics, are collected in the 2002 Tor book Which Way to the Future? A different one (usually one not available in the book) will be posted here a few times a year. And, of course, brand-new ones appear in each issue of Analog.)

Science fiction writers, contrary to a popular misconception, are not in the business of predicting the future. As Frederik Pohl put it in the introduction to an anthology some years ago, "The mistake you must never make about science fiction is in thinking that, because it is about the future, it is necessarily about the future." We are instead in the business of imagining possible futures. Our stories are, in a sense, "trial runs" that might help all of us decide which possibilities we'd like to steer toward, and which we'd like to avoid.

Nonetheless, there is a certain satisfaction in saying, "I told you so!" when one of our imaginings comes true; and there are those (often outside science fiction) who do try to predict what sort of future is most likely.  Sometimes it's fun to look back at one of those attempts at prediction and see just how well it did—to compare "the way the future was" (sorry, Fred!) with the way it is.

Such an opportunity recently fell into my lap when a friend presented me with a February 1950 issue of Popular Mechanics Magazine that he had picked up at a flea market.  Prominently blurbed on the cover was an article on "Miracles You'll See in the Next Fifty Years," by Waldemar Kaempffert, then Science Editor of The New York Times.

Mr. Kaempffert wisely acknowledged early in the piece that he wasn't going to describe exactly How the Future Would Be, either, though apparently he did expect to come pretty close.  "The only obstacles to accurate prophecy," he wrote, "are the vested interests, which may retard progress for economic reasons, tradition, conservatism, labor-union policies and legislation.  If we confine ourselves to processes and inventions that are now being hatched in the laboratory, we shall not wander too far from reality."

Well, we shall see.  Since we are very near the end of the fifty years he was talking about, this seems a good time to take a look at how he did.  We'll find that he did pretty well on some things, and not so well on others—partly because he underestimated just how complex and powerful those "only obstacles" could be, and partly because they aren't really the only obstacles.

Some things the article hit so close to right that it would seem nitpicking to quibble about the details it missed.  It was quite right, for example, about lighter metals (such as aluminum) and synthetics (such as plastics) replacing steel as a structural material in many applications.  It also did quite well in foreseeing the prevalence of frozen foods and microwave ovens (though the latter was identified only as "the electronic industrial stove which came out of World War II").  It took for granted that the Dobsons, the "typical family of 2000" used as an example throughout the article, would have a television set.

But it then added, "It is connected with the telephones as well as with the radio receiver, so that when Joe Dobson and a friend . . . talk over the telephone they also see each other."  That option has been available for some time now, but very few people have chosen to use it.  Businessmen do have video teleconferences, as the article foresaw, and there are shopping networks on television.  But so far, to the best of my knowledge, they haven't reached the point of salespeople holding up samples for individual consumers to examine and ask questions about.

And there's no suggestion in the article that those television images are in color, or that radio and television developed in different directions as separate and independent media.

Some things the article got conspicuously wrong.  We do not "clean house by simply turning the hose on everything," and it does not seem likely that we will in the foreseeable future.  We do not all use personal helicopters for most of our family transportation, and that, too, now seems unlikely.  Piloting any aircraft requires considerably more complicated skills than driving a car, and it's difficult to see how available airspace could safely accommodate that density of traffic even if the vehicles themselves could be made idiot-proof. 

We do not have nuclear-powered ocean liners and we do not seem likely to any time soon—not because of technical infeasibility, but because of prevailing beliefs about safety of such things.    Supersonic planes exist, but because of their expense and environmental problems receive very limited use.  The article imagines that in 2000 "nobody has yet circumnavigated the moon in a rocket space ship, but the idea is not laughed down."  How, I wonder, would the author have reacted to the suggestion that in the real 2000 men would have walked on the moon so long ago—and with so little of the logical follow-up—that much of the population can't even remember it?

We do forecast and track hurricanes with an accuracy barely dreamed of in 1950, but we do not defuse or divert them by spreading gigantic oil slicks on the ocean and setting fire to them.

In medicine, the article correctly anticipated great improvements and increases in the use of antibiotics and medical instrumentation.  It contains no hint of the evolution of antibiotic-resistant strains of bacteria—or of open-heart surgery and organ transplants (though it does hint at pacemakers).  It suggests that "such virus diseases as influenza, the common cold, poliomyelitis, and a dozen others are cured with ease."  I have the feeling the author would be surprised to learn that the common cold must still be endured, while polio is almost totally prevented in developed regions.  And there's no hint of the genetic engineering which now looks like so much of the future of medicine.

The article did foresee the greatly increased automation of industry, in ways that now seem an odd blend of going too far and not going far enough.  It describes a helicopter factory in which "electronic inventions that seem to have something like intelligence integrate industrial production so that all the machines in a factory work as units in what is actually a single, colossal organism."  They're digitally controlled, but by holes punched in paper tape.  The only humans to be seen in the plant are "a few trouble shooters . . . [who] respond to lights that flare up on a board whenever a vacuum tube burns out or there is a short circuit."

And that image leads us neatly to what is, for me, perhaps the most intriguing part of this package of predictions: the things it missed entirely.  This automated factory was quite visionary fifty years ago, but it never imagined that punched tape would seem quaint or that there would be no vacuum tubes to burn out.  A really demanding critic might fault the author slightly for that last point, for the transistor had been invented—but only three years earlier. 

The large-scale integration of circuits that led to the microchip would have seemed an extremely farfetched dream to most engineers in 1950.  So this article utterly missed what is probably the single most significant and pervasive technological change of the last half-century: the ubiquity of tiny, powerful computers and everything they imply, from telecommuting to compact disk players, to CAT scans, to the internet, to interactive video games.  (Don't get smug—so did almost the entire field of science fiction!)

Which is merely one example of how, contrary to Mr. Kaempffert's supposition, we can wander very far from reality if we confine ourselves to processes and inventions that are now being hatched in the laboratory. The real revolutions are likely to come from surprises—discoveries and inventions that can't be anticipated at the time of attempted prophecy. The microchip/ computer revolution is perhaps the most conspicuous example of such a technological shift that has already had huge effects.  The laser, on perhaps a smaller scale, is another.  The aforementioned genetic engineering is just getting started, but promises to have at least equally sweeping consequences. Nanotechnology is even younger, but likely to have even more far-reaching effects.

And, of course, some of the most important changes in any period are not technological, but social. There, too, the Popular Mechanics article had some conspicuous misses. It talks about "How Jane Dobson cleans house . . ." and "How Jane Dobson shops . . . ," and one picture caption reads, "Housewives in 50 years may wash dirty dishes right down the sink [their cheap plastic dissolving in hot water]."  To us, today, that vision of dishwashing sets off alarm bells as a likely pollution source, unless the materials are very biodegradable. But what's likely to strike us as even stranger about this discussion is the implicit assumption that shopping and housecleaning and dishwashing are still the work of "housewives," while Joe Dobson is assumed to be the commuter. 

The article's closing paragraph is particularly telling: "Any marked departure from what Joe Dobson and his fellow citizens wear and eat and how they amuse themselves will arouse comment.  .  . . It is astonishing how easily the great majority of us fall into step with our neighbors. And after all, is the standardization of life to be deplored if we can have a house like Joe Dobson's, a standardized helicopter, luxurious standardized household appointments, and food that was out of the reach of any Roman emperor?"

Well, some might not think so—but what has actually happened is far less standardization than prevailed in 1950.  Indeed, a complaint not uncommonly heard today is that the market offers so many options—e.g., a dozen or more variations on "orange juice" within a single brand—that people sometimes feel overwhelmed by the number of choices they must make.  And while there are certainly still pressures for conformity, particularly among the young, and many people seem to live their entire lives by bouncing from one fad to another, I think that there is also a good deal more variety in lifestyles and tolerance of our neighbors' eccentricities.

Both the unchanged roles of "housewives and breadwinners" and the assumption and acceptance of extreme standardization of life seem clear extrapolations of the way things were in 1950—an era now rather notorious for its homogeneity in those areas. The tacit assumption in this 1950 attempt to imagine the year 2000 seems to be, at least in the social sphere, "Just Like Now, But Even More So."  It's probably almost inevitable that any prognosticator (present company not excepted) will tend to make such assumptions.  The need to consciously try to avoid them is perhaps one of the first lessons to be drawn from this "look back at a look forward."

Another is: expect surprises. You can't imagine a truly likely future by simple extrapolation; there will almost certainly be important developments that cannot be anticipated by looking only at what's already known and in the works. A futurist working as such is at a disadvantage here; he or she can't postulate something out of the blue and seriously claim that it's likely to happen. A science fiction writer, on the other hand, can imagine anything that isn't demonstrably impossible, and then explore its likely consequences and ramifications—including changes in how people live. Many of those futures, of course, will bear little resemblance to what actually happens—but a few of them may well be more accurate than anything the "serious" forecasters can do.

And in trying to imagine those possible futures, it's essential to remember one more lesson illustrated by this little exercise: you can't predict future trends by considering either technology or society in isolation. They interact, and whenever you change one variable, you have to consider how it affects all the others. For a highly simplified example that we're already living, if computers become little, cheap, and easy to use, they're likely to become ubiquitous.  If it becomes easy for them to talk to each other, e-mail becomes a prevalent mode of communication and it becomes possible for more and more people to work at least partly at home.  So they do, and that influences the extent and nature of transportation. All of this effects fundamental changes in how people live and work and interact, and that in turn affects how receptive they will be to what new kinds of technology . . . .

This type of integrated approach, considering technological and social change as a single highly interconnected system, is something this Popular Mechanics article didn't do very well.  E.g., that automated factory it envisioned had only a few human employees, but the article also had lots of people (most of them male heads of households) commuting, many of them by private helicopter or huge "air buses" carrying hundreds of commuters.  What were all these people commuting to?  The article gives little hint of that.  It shows pretty explicitly how many common jobs of 1950 had been eliminated by automation, but says essentially nothing about what the people who held those jobs were now doing instead.  So the feeling in this article, as in too many science fiction stories, is of a future made of fascinating details that don't quite hang together into a logically consistent, believable whole.

Good science fiction writers, of course, know all this and appreciate the magnitude of the challenge.  They will nonetheless continue trying—and sometimes making the same sorts of mistakes in spite of themselves.  But amongst the mistakes, sometimes we'll get it right—and at the very least we'll have provided some advance scouting into just about any new territory that might come along.

So let's keep doing our best, and check back in another fifty years to see how well the current crop of "predictors" have done!