It was 1983: “early days” for the personal computer upheaval. Most Americans knew that microcomputers existed, but few had actually acquired one, or knew what they were good for. (Spoiler alert: not much.) I was asked about them now and then – I remember an ophthalmologist quizzing me about them during an eye exam – but the general level of interest was casual, as with a curiosity that might soon be gone from the news.
Then came WarGames.
It wasn’t a terribly exalted movie, Siskel’s and Ebert’s opinions notwithstanding. Its enduring importance was that it gave a large boost to the careers of Matthew Broderick and John Wood. (Why Ally Sheedy’s career went flat, I cannot say.) Its central plot element, the notion that a teen computer owner could finesse his way into a DoD computer capable of convincing NORAD that a nuclear attack was imminent, was purest nonsense. (For those who think the Internet makes everything available to a PC owner: It’s still nonsense.)
But it was a hit. People went to the theaters to see it several times each. (No, I didn’t. The only movie that ever entranced me that way was…well, maybe I should keep that to myself.) And it gave rise to some unhealthful ideas about what computers can do.
It comes to mind this morning because of a scene, at the climax of the movie, in which John Wood’s character, Dr. Stephen Falken, the inventor of the WOPR computer at the center of the action, strives to convince General Beringer, the commander of NORAD, ably played by the underappreciated Barry Corbin, that the images he’s seeing on the display screens are not real:
Of course he manages to do it, or we wouldn’t have the obligatory happy ending. But the key to the thing is the notion of a computer-generated fantasy: a manufactured illusion imposed upon the general and his staff that he takes for real-world data.
Under the premises of the film, all the data was being routed through the WOPR computer. However, the computer declined to present the general with the real-world reports from the early-warning radars, which would have indicated that no attack was in progress. It preferred to depict the nuclear-war game it was playing – and which it eventually tried to ignite in reality.
For those who have never read Gulliver’s Travels:
At last we entered the palace, and proceeded into the chamber of presence, where I saw the king seated upon his throne, attended on each side by persons of prime quality. Before the throne, was a large table filled with globes and spheres, and mathematical instruments of all kinds. His majesty took not the least notice of us, although our entrance was not without sufficient noise, by the concourse of all persons belonging to the court. But he was then deep in a problem, and we attended at least an hour before he could solve it. There stood by him on each side a young page with flaps in their hands, and when they saw he was at leisure, one of them gently struck his mouth, and the other his right ear; at which he startled like one awaked on the sudden, and looking towards me and the company I was in, recollected the occasion of our coming, whereof he had been informed before. He spoke some words, whereupon a young man with a flap immediately came up to my side, and flapped me gently on the right ear; but I made signs, as well as I could, that I had no occasion for such an instrument; which, as I afterwards found, gave his majesty and the whole court a very mean opinion of my understanding. [Part III, Chapter II]
Here we see the Laputan institution of the “flapper:” a servant whose function is to “flap” the ears of his master when, in the opinion of the servant, it is desirable for the master to hear what is being said. Plainly, the “servant” exercises considerable power over the “master.”
So also is it with our electronic “servants.” It’s massively unwise to trust them beyond their proper sphere. I’m here to tell you.
We can be bullied, bribed, berated, brainwashed, or subtly coaxed to believe what others want us to believe, even in defiance of reality. If the persuader’s shoulders are draped with a mantle of authority, however specious or irrelevant, he’ll have the upper hand from the start unless his target is unusually skeptical and hard-headed. Sad to say, the population of appropriately skeptical, hard-headed Americans has fallen rapidly these past few decades.
The problem has been exacerbated by the amount of our attention that goes to computer-driven screens today. They’re everywhere. They modulate an ever-greater fraction of our experiences. They present us with images and data that we routinely take for honest depictions. And they continue to proliferate.
But those images and data are at least one level removed from reality. The layer of separation imposed by the intervening computer and software is a potent tool for deception. Sometimes the computer is invoked as an authority. Hearken to Herman Kahn in his magnum opus On Thermonuclear War, as he discusses an early attempt to determine the “optimum” defensive system in the era of the ballistic missile:
In the early days at RAND most studies involved an attempt to find the “optimum system, given some reasonably definite set of circumstances, objectives and criteria. The emphasis was on comparing thousands, sometimes tens of thousands, of different systems under idealized conditions; then the “best” one would be picked….Naturally the high-speed computer often played a central role in this.
Sometimes our researchers took a curious pride in the prowess of their high-speed computers. They would make remarks such as “More than a million campaign calculations went into this analysis.” Or “This is the first analysis done by Man in which 10,000,000 multiplications were made.” Or an even more extreme boast, “Thee results came out of a complicated calculation performed by the most modern of high-speed computers using the most advanced mathematical techniques available. Do you want to argue with an electronic machine backed up by all the resources of modern science?” The only possible answer to that question is “Yes.”
Herman Kahn was a supergenius. More to the point of this tirade, he knew the difference between data and the appearance of data. He refused to take the latter in place of the former.
Yes, there’s a point to all this.
He who can persuade you to accept appearances in place of reality can persuade you of anything in the world. All he need do is fabricate the necessary appearance. Computers and software have made that ever easier as time has passed. The art of deception has largely been automated.
The number of deceivers among us has swelled as their curious occupation has become easier to ply. There’s no real mystery in that. What’s easy and potentially profitable will attract large numbers of participants. Why else would the state governments all run lotteries?
One of the factors on the deceivers’ side is our excessive regard for what comes out of computers. Those who are unaware of the inherently fantastic nature of electronic computation are usually the easiest to gull – and by fantastic I don’t mean wonderful but rather “in the nature of a fantasy.”
But what’s this? A former software expert has derided his own field with the phrase “the inherently fantastic nature of electronic computation” — ? Yes, Gentle Reader, it is so. It has always been so and it will always be so, for a simple reason every software artisan must internalize before he’s allowed to touch a keyboard: Garbage in, garbage out.
There is no inherent authority in the output of any program. Its resemblance to reality depends upon factors that the program does not control:
- The accuracy, completeness, and relevance of its input data;
- The correctness of the algorithms built into it.
A simple example should suffice. Imagine that someone has provided you with a temperature in the Centigrade system. Imagine further that you’re ignorant of the formula for converting Centigrade temperatures to the Fahrenheit system. Your interlocutor then points you to a computer and says “Just run it through the Centigrade-to-Fahrenheit program on that thing.” You do so; the program gives you a number it says is the Fahrenheit equivalent of the Centigrade temperature you entered; and you walk away.
Have you just performed:
- A scientific operation?
- Or an act of faith?
There’s only one correct answer.
The worst men in the world are deceivers, one and all. Their principal skill is in persuading you to believe that which is not so. Their techniques go beyond the scope of this tirade, but today one of the most important of them is the substitution of computer outputs for actual real-world data.
They don’t want you to know that Garbage in, Garbage out. That pertains not only to the data they claim to have fed into their supposed programs, but even more imperatively to the programs themselves: the assumptions and algorithms built into them. Yet they posture as authorities. Their favorite phrase is one I’m sure all my Gentle Readers have heard to the point of nausea: “The science is settled.”
Yet they have no science to offer. Science proceeds by observation, inference, deduction, experiment, dispassionate collection and evaluation of results, and ruthless repetition by impartial others. All they have is computer-generated fantasies, specially designed with your mind in mind.
They will do their best to bully, bribe, berate, brainwash, or subtly coax you to accept the appearances they manufacture as substitutes for real-world data. They’ll enlist politicians and plutocrats in their campaigns: the former with promises of power; the latter with promises of pelf.
See also this Baseline Essay. And do have a nice day.