One hundred and thirty-seven seconds

A short story by Stanisław Lem published in 1976, translated from Polish in 2015 by Marcin Wichary


Translator’s note: This is my first translation of a Stanisław Lem story. I tried to stay true to the spirit of the original as much as possible, which means original occasional odd idioms, mismatched units, and kilometer-long sentences. The story was published in 1976, and predates desktop publishing and the Internet. To the best of my knowledge, Lem has never visited America. If you are interested, read more about why I translated this story and the translation process.


Gentlemen, owing to lack of time and adverse circumstances, most people leave this world without thinking too much about it. Those who try get a headache and move on to something else. I belong to the second group. As my career progressed, the amount of space dedicated to me in Who’s Who grew and grew, but neither the last issue nor any future ones will explain why I abandoned journalism. This will be the subject of my story, which I wouldn’t tell you under other circumstances anyway.

I knew a talented boy who decided to build a sensitive galvanometer, and he succeeded all too well; his device would move without any electricity, reacting to tremors of the Earth’s surface. This could be a motto of my story. I was a night shift editor at the foreign branch of UPI. I endured a lot there, including the automation of laying out newspaper pages. I said good-bye to human typesetters and moved on to IBM 0161, a computer built specifically for page composition work. I regret not having been born a hundred fifty years ago. My story would start with “I seduced Countess De…” and if I moved to describing how I snatched the reins from the driver and started to whip the horses so that I could escape the thugs sent by a jealous husband, I wouldn’t have to explain what a countess is, or how seducing works. But it’s not so easy today. The 0161 computer is not a mechanical typesetter. It’s a speed demon, restrained through engineering tricks only so that humans can keep up with it. It replaces ten to twelve people. It’s directly connected to a few dozen teletypes in order for everything typed by our correspondents — in Ankara, Baghdad, or Tokyo — to immediately get into its circuits. The computer cleans it up and throws various layouts of the morning issue pages up on its screen. Between midnight and 3 a.m. — that’s when we close the issue — it can create as many as fifty different variants. It is up to the editor on call to decide which one gets printed. A human typesetter asked to do not fifty but even just five different versions of one issue would go crazy. But the computer works a million times faster than any of us — or it would if it was allowed to. I do realize how much of the appeal of my story I’m destroying with remarks like this. What would be left of the countess’s charms if, instead of extolling the alabaster complexion of her breasts, I started talking about their chemical makeup? Those are difficult times for raconteurs; everything approachable is anachronistic, and everything sensational requires pages from encyclopedias and a university textbook. However, no one figured out a solution for this. And the collaboration with the IBM was fascinating. Whenever a new story comes in — it all happens in a big round room, filled with constant teletype clatter — the computer immediately flows it into the page layout, of course only on the screen. It’s all shadow play, a game of electrons. Some grieve for people who lost those jobs. I didn’t. A computer has no ambitions, it doesn’t get upset if five minutes to three the last piece of domestic news is missing, it isn’t perturbed by household worries, doesn’t need to borrow money before the month’s end, doesn’t get tired, won’t make you feel it knows everything better, and it definitely won’t take offense if you ask it to kick out straight to the last page, in nonpareille, whatever it just put as a headline. But it’s also incredibly demanding, and it’s not something that’s immediately recognizable. If you say “no,” it’s a definitive, decisive “no,” a tyrant’s sentence, since the computer cannot disobey! But since it never gets anything wrong, any mistakes in the morning edition will be mistakes of people. The creators of the IBM thought of absolutely everything, except for one little detail: regardless of how well-balanced and bolted down they are, the teletypes will vibrate, just like fast typewriters would. Because of those vibrations, the cables connecting our teletypes with the computer have a tendency to loosen, eventually causing the plugs to disconnect. It happens rarely, perhaps once or twice a month. It’s not a huge nuisance — you only have to get up and put the plug back in — so no one ever requested better plugs. Every one of us on-call editors thought about it, but without much conviction. And it’s possible that the plugs are different now. If so, then the revelation I want to tell you about will never be repeated.

It happened on a Christmas Eve. My issue was ready just before 3 a.m. — I liked having a reserve of even a few minutes, enough to rest and light my pipe. I felt content that the rotary was waiting not for me, but for the latest piece of news — that Eve it was about Iran, which had experienced an earthquake just that morning. The agencies reported only a fragment of the correspondent’s dispatch, since the first tremor was followed by another, strong enough to destroy the cable connection. Since the radio was silent too, we thought that the broadcasting station must have been in ruins. We were counting on our man, Stan Rogers. He was as small as a jockey and knew how to use that; sometimes he would manage to get onboard a military helicopter already packed to the brim, since he weighed no more than a suitcase. The screen was filled with the layout of the front page — save for one blank, white rectangle. The connection to Iran was still lost. And even though some of the teletypes were still clattering in the background, I recognized the sound of the Iranian one immediately when it became active. It was a matter of experience. It took me by surprise that the white rectangle remained empty, even though the words should be filling in at the same speed as the teletype’s printing. But the pause lasted just a second or two. Then the entire piece, rather concise, materialized on the screen at the same time, which surprised me too. I remember it by heart. The headline was already there. Underneath, it was reported: “In Sherabad, underground tremors of a seven- and eight-degree magnitude repeated themselves twice between ten and eleven a.m. local time. The city is in ruins. The estimated toll is 1,000 people, with 6,000 homeless.”

I heard the buzzer; the printer was warning me it was 3 a.m. Since such a terse story left a bit of free space, I diluted it with two extra sentences and through the strike of a key sent down the entire issue to the printing office, where it would go straight to the linotypes and then to the rotary presses.

I had nothing else to do, so I got up to stretch my bones. As I was reigniting my pipe, I noticed a cable on the floor. It was disconnected from a teletype, the Ankara teletype, which was exactly the one Rogers was using. As I was picking up the cable, a nonsensical thought flashed through my head: what if the cable was already disconnected when the teletype sounded off? It was obviously an absurd thought, since how could the computer write a story without that connection? I slowly walked to the teletype, tore off a piece of paper with the imprinted message, and raised it towards my eyes. Immediately, I sensed it was phrased differently, but as usual for this hour I felt tired and scattered and didn’t trust my memory. I turned on the computer again, asked it to show me the front page, and compared both pieces. They were indeed different, although not by much. The teletype version read:

“Between ten and eleven local time, two aftershocks occurred in Sherabad, at a magnitude of seven and eight degrees. The city is completely destroyed. The number of casualties tops five hundred, and those without roofs over their heads count six thousand.”

I stood there, staring at the screen, then at the paper, and then at the screen again. I had no idea what to think or do. Factually, both versions were almost identical, the only difference being the casualty estimation: Ankara reported half a thousand, the computer doubled it. My regular journalist reflexes kicked in, and I immediately called the printer.

“Hey,” I said to Langhorne, the lead linotypist. “I found an error in the Iranian message, first page, third column, last line, instead of one thousand, should be…”

I stopped since the Turkish teletype woke up and started clattering: “Attention. Latest news. Attention. The number of earthquake casualties now estimated to be one thousand. Rogers.”

“Hey, so…? What do you want it to be?” shouted Langhorne from downstairs. I sighed.

“Sorry, man,” I said. “No mistake. My fault. It’s alright, okay to go as is.”

I put the handset away, walked to the teletype, and read the addition a good half a dozen times. I liked it less and less with every read. I felt that, I don’t know, the floor was getting soft below my feet. I walked around the computer, looking at it with distrust and — I was certain — a bit of fear. How did it do it? I didn’t understand anything and felt that the longer I tried to think about it, the less I would comprehend.

At home, in bed, I couldn’t fall asleep. I tried to forbid myself from thinking about this wild story, primarily with my psychic hygiene in mind. In terms of facts this was, after all, a trifle. I knew I wouldn’t be able to tell anyone about it, since no one would believe me. They would take it for a joke, flimsy and naïve. After tossing and turning in bed, I decided to put things under a microscope — that is, to systematically examine the computer’s reaction to disconnecting teletypes. That brought relief of sorts, at least enough to fall asleep.

I woke up in a somewhat optimistic mood and with an answer to my puzzle, or at least something that could pass as one. Who the hell knows how I arrived at it.

Teletypes vibrate when they work. It’s those vibrations that cause the plugs to get detached. Could that be a temporary signaling source? Even I, with my worthless and slow human senses, could pick up differences between teletypes by sound alone; I recognized whenever the Parisian one started working because of its distinct metallic strikes. If the receiver was a hundred times more sensitive, could it pick up on those negligible differences between individual letter keys? It wouldn’t be 100% accurate, and that’s why the computer could not repeat the teletype message word for word, but twisted it stylistically, simply filling in the blanks in the information. As for the casualties, the IBM was, after all, originally a computing machine; the damaged buildings, the time of the earthquake, and the number of victims must correlate statistically; missing a number, the computer used its lightning-fast mathematical skills and came up with one thousand people. Our correspondent didn’t do any calculations and simply passed on the estimate from the locals. Later on, having obtained a more precise value, he sent in the final correction. The computer prevailed since it relied not on rumors, but rather on strict statistical materials residing in its ferrite memory. That line of thinking comforted me completely.

IBM 0161 isn’t, after all, just a passive relay; if the teletype operator commits a spelling or a grammar mistake, that mistake pops up on the screen only to be replaced, a fraction of a second later, with the correct expression. Sometimes this happens so quickly it’s unnoticeable by anyone; the only way to find out about a correction is to later compare the teletype printout with what’s on the screen. The IBM is not a simple automatized typesetter, since it’s connected to the networks of the agency and a few libraries, and it’s possible to ask it for extra information to supplement messages that are too sparse. In brief, I explained it all to myself splendidly, even though I still intended to perform a few further small experiments during my next shift, although without telling anyone about them — or about the Christmas Eve story — since that was more reasonable.

There was no lack of opportunities. Just a couple of days later I was once again sitting in the foreign service room, and when Beirut started transmitting the message about the submarine from the Sixth Fleet missing in the Mediterranean, I stood up and, keeping my gaze on the screen as it quickly filled with the news, I disconnected the cable with one smooth and sneaky move. The text stopped growing for a fraction of a second, frozen midword as if the computer, surprised, didn’t know what to do. But that shock was indeed momentary; new words started popping up on the white background almost immediately as I frantically compared them with the text struck off by the teletype. And the thing I already recognized happened again — the computer would relay the message from the teletype, but using slightly different words: “the spokesperson of the Sixth Fleet declared,” instead of “said”; “the search is in progress” instead of “continues.” A few other similar nuances separated the two messages.

It’s peculiar how easy it is for someone to get used to the extraordinary if they understand its principles — or at least if it seems like they understand them. I already had the impression I fooled the computer, playing with it like cat would with a mouse. I felt I had the situation under control. The layout was still shining with many bald spots and the texts that were supposed to fill it in would be coming in now, in the informational rush hour, a few at a time. I started seeking out the right cables from the bundle and unplugging them one by one, eventually ending up with six or seven in my hand. The computer would continue working with utmost leisure, even though it was connected to no more teletypes. For sure, I said to myself, it distinguishes individual letters and words from the vibrations, and that which cannot be recreated immediately is refined with immediate extrapolation or some other mathematical means. I worked as if in a trance, with great concentration and fixation waiting for another teletype to wake up — and when the Roman one started making noise, I pulled the cord. I did it well enough so that I disconnected not just the usual plug, now in my hand, but also the second one, sending the power to the teletype. The teletype froze, naturally, and I rushed to reconnect it, but something struck me and I glanced at the screen.

The Roman teletype was dead, but the computer merrily continued to fill the space dedicated to the Italian governmental crisis with “the latest updates.” With bated breath, feeling something strange was once again happening with the floor and my knees, I walked up to the screen and the innocent words — “appointed Battista Castellani as the prime minister” — seemed to me like a dispatch from another world. Hastily, I connected the Roman teletype with the main transmission cable to compare both texts. Oh, now the differences between them were much bigger, but the computer never swayed from the truth, from the contents of the message. Castellani indeed became the prime minister, but that sentence appeared in a different context, and four lines lower than on the screen. It all looked to me as if two reporters, learning independently about the same thing, naturally wrote up the story in their own way; with weak knees I sat down to try, for the last time, to salvage my hypothesis, but I already felt that would be futile. All of my rationalization crumbled in a moment, because how on Earth could the computer read the vibrations of the teletype which was as deaf and dead as a tree trunk? After all, the IBM couldn’t detect the movements of the teletype used in Rome by our correspondent! I felt dizzy. Had someone entered the room right then, it would look God knows how suspicious — I was sweaty, shifty-eyed, still clutching a bunch of cables in my sweaty hands, like a criminal surprised in flagranti. I felt I was a rat caught in a corner and reacted like a distraught rat would — I started violently disconnecting all of the teletypes, so that in a minute all of the rattle died down — and I stood there in dead silence, opposite my computer. What happened then was strange, perhaps even more stunning than anything that happened so far. Even though the page was not yet quite filled, the flow of new text slowed down. More so, with the new, slower pace, came sentences without any essence, barren, in one word: filler. For a good while the lines of text crawled into their own spaces on the screen, until they froze — all of them. Some of the articles took on an absurdly comic nature. There was a note about a soccer game where, instead of the final score, the story concluded with an empty cliché about the courageous attitude of the players of both teams. Further news from Iran broke down with a statement about earthquakes being events of cosmic scale, since they happen even on the Moon. It was neither here nor there. The mysterious wellsprings from which the computer drew its inspiration dried out.

My first task was, of course, to set the issue. I reconnected the teletypes in a big hurry and whatever just played out in front of me I had time to think about only after three, when the rotary presses started working. I knew I would find no peace until I got to the bottom of this fascinating performance and its no less surprising collapse. A layman’s first thought would be to simply ask the computer itself the appropriate questions: if it’s so smart, and at the same time so ruthlessly obedient, let it reveal how, through which mechanisms, it could work, disconnected, and also what later halted that effort. That thought was planted in our heads by popular fables of electric brains, given that you cannot speak to a computer the same way you could with a human, regardless of whether that human was smart or stupid: the computer is still not at all a person! You could just as much expect a broken typewriter to tell you where and how to fix it. A computer processes information without having any rational attitude towards it. The sentences it spits out are like trains traveling on a track of syntax. If the trains derail it means that something within the computer is not functioning properly, but it won’t know anything about this, for the simple reason that “it” is used here in the same way as if talking about a lamp or a stool. Our IBM could by itself formulate and reformulate the contents of stereotypical news bulletins, nothing more. The importance of individual news was always decided by a human. The IBM could compile two informationally complementary notes into one, or select an idiomatic opening to a purely factual news item, thanks to ready-made templates of such behavior, of which it had stored hundreds of thousands. That opening would match the contents of the news solely because IBM would perform a statistical analysis, selecting so-called key words; if the news item, for example, contained repeated mentions of “goals,” “penalty kicks,” and “opponent’s team,” it would chose something from the sports tournament repertoire; the computer was, in one word, a railwayman who could in the right manner toggle turnouts, connect carriages, and send the trains towards proper destinations; however, it was not aware of their cargo. It oriented itself purely on the external properties of words, sentences, phrases, those subject to mathematical operations of disassembly and assembly. I couldn’t expect it to be of any help.

I spent the night without any sleep, deep in thoughts. In the computer’s output, I noticed this regularity: the longer it was disconnected from sources of information, the worse it did reconstructing it. This seemed to me somewhat understandable, given that I worked in journalism for over twenty years. As you know, the editorial offices of two high-circulation weeklies such as Time and Newsweek are completely independent. What they have in common when authoring their respective issues is that they coexist in the same universe and in analogous time have access to very similar sources of information. Likewise, they address readers who are largely alike. Therefore, the similarity of plenty of the articles in both publications is not surprising. It’s a result of a specific excellence of market fit achieved by both teams. It is possible to learn to write an overview of events in a given country, or on a weekly basis in the entire world, and if one writes from a similar position — namely, the journalistic U.S. elite — with similar education, identical news, and aiming to impress the optimal effect upon its readers, it is no surprise that the columns, laid out independently and in parallel, often resemble identical twins. The similarities never reach individual sentences but the attitude, tone, intensity of affectation, distribution of accents, emphasis of certain drastic details, contrapuntal juxtaposition of features — for example profiles of certain politicians — or everything what serves the purpose of drawing the reader’s attention and suggesting to them that they’re consulting the most excellent information source, is in the arsenal of tricks of every adept journalist. In some sense our IBM was a “model” of such a reporter. It knew the ways and means and therefore could do what all of us could. The routine programmed within it made it a genius of catchy phraseology, shocking juxtaposition of data, its most favorable exposition; I knew all of this, but I also knew that its concerto couldn’t be reduced to such explanations. Why was it so proficient when disconnected from teletypes? Why did this proficiency desert it so quickly? Why would it start to blabber afterwards? I still fancied that I alone could find answers to these questions.

Before the next shift, I called our Rio de Janeiro correspondent and asked him to send us, at the beginning of our night service, a small phony news item about the boxing match between Argentina and Brazil. The victories of Brazilians were to be reported as Argentinian, and vice versa. During our conversation, the results of the contest couldn’t have been known; the match started late in the evening. Why did I ask Rio? Because I was asking for an unprecedented professional favor, and Sam Gernsback, who was our correspondent there, was my friend of that rare and most valuable kind — he never asked about anything.

What I had experienced so far let me presume that the computer would repeat the false message, the same which Sam would key in on his teletype. (I won’t lie that I had a hypothesis about it; namely, I imagined that the teletype could function as a radio transmitter of sorts. My computer, I thought, could detect the electromagnetic waves surrounding the cables in the ground, since it was sufficiently sensitive as a receiver.)

Gernsback was to rectify the false news item immediately after sending it; I would of course destroy the first one, so no trace would remain. The plan I came up with seemed extremely sophisticated to me. So that the experiment was crucial in character, I decided I’d keep the normal connection to the teletype up until the match’s intermission, but disconnect it after the break. I won’t entertain describing the preparations, emotions, and the atmosphere of that night. I’ll just say what happened. The computer reported, or set into the layout, the wrong results up until the break, and the correct ones after. Do you know what that meant? While connected to the teletype, the IBM would not reconstruct anything. It wouldn’t rack its brain. It would simply report what came through Rio’s cable, letter by letter. Disconnected, it didn’t care about the teletype or the cables that would, I assumed, act as a radio antenna — it just reported the correct results! Whatever Gernsback was typing in at the same time did not matter to my IBM in the least. But that’s not all. The computer reported the correct results for all matches but made a mistake about the last, heavyweight one. Yet, one thing seemed irrefutable: at the moment it lost the connection, it wasn’t in any way dependent on teletypes: either the local one, or the one in Rio. It obtained the information some other way.

Drenched in sweat, with my pipe gone out, I was still unable to digest what I had just seen when the Brazilian teletype turned on. Sam reported the correct results, as prearranged, but made a correction in the last dispatch: the outcome of the heavyweight fight was changed after the judges’s verdict — they determined that the glove weight of the winning Argentinian was against regulations.

So the computer was correct every single time. I needed one more bit of information, which I gathered by calling Sam right after I closed the issue. He was already asleep and, woken up, swore like a trooper; I could understand him easily since the questions I dumped on him seemed trivial, idiotic even: what time were the results of the heavyweight fight reported, and how much later did the judges change the verdict? Sam eventually gave me answers to both. The fight was cancelled almost immediately, when announcing the Argentinian fighter’s victory, as the judge, raising the winning hand, could feel through the glove’s skin a little counterweight — previously hidden under a layer of plastic — which loosened and moved during the fight. Sam ran towards the phone before that moment, with the knocked-down Brazilian still on the ring floor, because he wanted to send the dispatch as quickly as possible. So the computer couldn’t have gained its knowledge by reading Sam’s mind because it knew the correct result of the fight before Sam did.

I had been conducting those night experiments for almost half a year and I’d learned a thing or two, but I still didn’t understand anything. When disconnected from the teletype, the computer would freeze for two seconds and then deliver the rest of the bulletin during one hundred and thirty-seven seconds. Up until then it would know everything about the event; right after — nothing at all. Perhaps I could digest this all somehow, but I found something worse. The computer could predict the future. Unerringly. It didn’t matter whether the given information was about events that had happened already, or those that hadn’t happened yet — as long as they fit within the limit of two minutes and seventeen seconds. If I keyed in, via the teletype, made-up information, it would repeat it obediently, but go silent immediately after unplugging the cable, so it could continue describing only things that were actually taking place somewhere, and not events solely imagined. That, at least, was the conclusion I came to and wrote down in a notebook that I never left home without. I slowly got accustomed to the computer’s behavior and I didn’t know when exactly I started associating it with the behavior of a dog. Just like with a dog, the computer needed to be introduced to a specific scent trail, to be given the beginning of a series of events to sniff carefully, so to say; just like a dog, it needed a moment to commit the data to memory and if there wasn’t enough, it would go silent, hedge with generalities, or hop onto an altogether false trail. It didn’t matter to the computer which scent trail it followed, just like it wouldn’t to a dog; once it did, it was dependable — within one hundred and thirty-seven seconds.

Our night sessions, happening always between three and four a.m., began to feel like interrogations. I attempted to corner the computer, to come up with a tactic of cross examination, or rather mutually exclusive alternatives, and then I came up with the idea that seemed sensational in its simplicity. As you remember, Rogers reported about the earthquake in Sherabad from Ankara, so the sender of the message didn’t need to reside exactly where the events took place; however, as long as Earth events were included, one couldn’t discount that someone, a human being or even an animal, could be an observer, and the computer was somehow making use of that. I tried to fabricate the beginning of a message dealing with a place where no man was or has ever been: Mars. I have given the computer the aerographic coordinates of the very center of Syrtis Minor, and reaching the words “it’s currently midday on Syrtis Minor; looking at the surroundings we can see…” I jerked the cable, pulling it from its socket. After a second’s pause the computer finished with “the planet in the sun’s rays,” and that was it. I rewrote that beginning left and right a dozen times, but I could not squeeze any specific details out of the computer; it continued to dissolve in generalities. I recognized that its omniscience didn’t reach the planets and somehow, although I do not know why, that made me feel more at ease.

What to do next? I could, of course, fire off with a sensation of the highest caliber, gaining publicity and some solid dough, but I didn’t take this opportunity seriously for a second. Why? I am not sure. Perhaps the notoriety of the conundrum would push me outside of its circle; I could imagine crowds of technicians invading our place, experts communicating through their own professional jargon; regardless of what conclusions they’d reach, I’d be immediately removed from the case as a troublemaker and a naïf. I would only be able to describe my experiences, give interviews, and cash cheques. That I cared the least about. I was ready to share my mystery with someone, but not to give it up entirely. I decided to collaborate with a good expert I could trust one hundred percent. I knew only one closely: Milton Hart from MIT. He was a guy with character, original and somewhat anachronistic since it was hard for him to work on a big team, and today a solitary scientist is like an endangered mastodon. Hart was a physicist academically, and professionally a programmer, which suited me. Although we kept in touch on a peculiar terrain — we both played mahjong — and otherwise didn’t have much contact with each other, during those games one could learn quite a bit about someone. Hart’s eccentricity manifested itself as sudden proclamations of random bizarre thoughts; I remember him once asking me if it’s possible that God created the universe by accident. It was impossible to know when he was serious and when joking or mocking the interlocutor. He was open-minded, for sure; announcing myself with a phone call, I drove to his place the next Sunday and, just as I expected, he agreed to my conspiratorial plan. I don’t know if he believed me straight away. Hart is not an effusive person; either way he tested everything I told him and the first thing he did afterwards was something that never ever occurred to me: he disconnected our computer from the federal information network. Immediately, the extraordinary talents of my IBM disappeared entirely. So the mysterious power did not reside in the computer, but in the network. As you know, the network currently comprises over 40,000 computing centers, and as you probably don’t know (I didn’t myself until Hart told me), it’s built hierarchically, somewhat akin to the nervous system of a vertebrate. The network has federal nodes, and the memory of each one of them contains more facts than all of the scientists put together. Every subscriber pays a rate according to the computer time used during a given month, times multipliers and coefficients, because if a subscriber’s problem is too complex for the nearest computer, the dispatcher automatically adds reinforcements from the federal reserve, meaning computers that are idle or underused. That dispatcher, it goes without saying, is itself also a computer. It ensures a uniform load on the entire network and guards the so-called reserved memory banks — the data that’s inaccessible, state secrets, military secrets, and so on. My face became as long as a fiddle as Hart was telling me all this, because while in a way I knew about the network’s existence and UPI being its subscriber, I thought about it all as much as someone talking on the phone thinks about the devices in the telephone exchange. Hart, known to be spiteful, observed that I preferred imagining my nightly tête-à-tête with the computer as a romantic encounter, detached from the rest of the world (since it seemed more like a fantastic tale), to a sober reflection that I’m among a bunch of subscribers that usually sleep between three and four in the morning. This made for a light network usage, and that’s why my IBM could continue using its potential the way it wouldn’t be possible in the morning rush hour. Hart checked for the invoices UPI paid as a subscriber, and it turned out that a few times my IBM utilized, at one go, some sixty to sixty-five percent of the entire federal network. Although those incredible strains lasted a short time, only a few dozen seconds, shouldn’t someone by now have been interested in why some shift journalist was taking enough power from the network to tabulate all the fields of the entire national income, twenty times over? But everything was computerized today. That included controlling the informational usage, and it’s known that computers cannot be surprised by anything — at least not while the bills are paid on time, and that wasn’t a problem since those were done by a computer as well, our UPI bookkeeping machine, so it turned out that my interest in the landscapes of Syrtis Minor on Mars cost UPI twenty-nine thousand dollars — rather costly given that it wasn’t appeased. In any case, even though it was then quiet as a mouse, my computer did everything in its power — and not only just its own, given that during eight minutes of its silence which ended with an evasive bromide, the network made some billions and trillions of operations — we could see it black on white on the monthly bill. Another thing was that the nature of its Sisyphean work remained a complete mystery to us. It was some pure algebraic witchcraft.

I warned you this was not going to be a ghost story. Funeral phantoms, premonitions, mystical prophecies, curses, penitent wraiths, and the whole rest of those honest, clear, grateful, and first and foremost simple creatures have disappeared from our lives once and for all. To talk with precision about a ghost that entered the IBM machine through the main pipe of the federal network, one would have to draw diagrams, write equations, and find detectives to use some computers to tamper with the other. A new type of a ghost starts with higher mathematics and so it is, to us, inaccessible. My story must ramble before raising the hair on your neck, because I will tell you a lecture based on what Hart told me. A computer network resembles an electric one — except what you can draw from it is information rather than energy. The circulation then, be it of electrical energy or news, is reminiscent of the motion of water in reservoirs connected with pipelines. The current will flow where it encounters the least resistance, or where there’s the biggest need. If one breaks a delivery cable, electricity itself will find a way through circumferential lines, which might actually result in overloading and failures. Graphically speaking my IBM, after losing connectivity to its teletype, asked the network for help. The network responded to this rallying cry with a speed of dozens of thousands of kilometers per second, since this is how fast current travels through wires. It took one to two seconds for all the reinforcements to get in sync, and that’s when the computer was silent. Then the connection was restored, in a way, although in what way exactly we still had no idea. Up until that point, everything explained seemed to have a factual, physical character, and you could even translate it into dollars, except that the knowledge gained was purely negative. We already knew what needed to be done for the computer to lose its extraordinary talents: it was enough to disconnect it from the informational pipe. But we still had no idea how the network could help it out, since how exactly could the network reach a certain Sherabad just hit by an earthquake, or an arena in Rio that happened to host a boxing match? The network is a closed system of interconnected computers, blind and deaf when it comes to the outside world, with inputs and outputs consisting of teletypes, telephone adapters, registers in corporations or federal offices, control panels in banks, energy centers, huge companies, airports, and so on. The network has no eyes, ears, or antennas or sensors of its own; its reach also doesn’t extend outside of U.S. territories, so how could it get any information about what was happening in some Iran?

Hart, who on that topic knew as much as I did — nothing, that is — behaved very differently than I did, neither himself asking those kinds of questions, nor allowing me to speak whenever I tried to flood him with mine. When he couldn’t fulfill my angered curiosity one time, and I said a few unpleasant things of the kind that easily escape someone’s mouth between three and four in the morning after a sleepless night, he stated he was not a fairy, a charlatan, or a clairvoyant. The network, as it turns out, exhibits properties that were neither planned nor foreseen; they have limitations, as proved by the Mars story, and are therefore physical, and can be studied. Those studies can, after a while, yield certain results, but the results won’t likely become answers to my questions, because those kinds of questions are not allowed in science. According to the Pauli principle, each quantum state can only be staffed by one elementary particle, and not two, five, or a million; physics limits itself to this kind of a statement, not allowed to ask why all the particles ruthlessly follow this principle, and what or who prevents the particles from behaving otherwise. According to the indeterministic principle, particles behave only in a statistically specified manner, and within this indeterminism they allow themselves things that classic physics might consider indecent or even macabre, since they breach the rules of conservation — but since this is happening in the indeterminacy interval, one can never observe them breaching the rules red-handed. And once again one cannot ask how could particles afford those antics within the indeterministic interval of observation, who gave them permission for those whims that seem to deny common sense, since matters like those are not within the purview of physics. In some sense one could indeed assume that within the crevice of indeterminacy particles behave like an offender completely assured of his or her impunity, since they know no one could catch them in flagranti. Those, however, are anthropocentric ways of speaking, which not only result in nothing, but add harmful confusion to the proceedings, since they seem to attribute human intentions, like bad faith or scheming, to elementary particles. In turn, one could judge the information network as being capable of gathering information about what’s happening on planet Earth, including places devoid of the network’s own sensors. Of course, it would be possible to proclaim that the network generates its own “perceptive field” with “teleological gradients,” or, using similar terminology, to create another pseudo-explanation of no scientific value; the point is to realize what, within which boundaries, and with what starting and marginal conditions, the network is capable of accomplishing. The rest belongs to modern sci-fi. One can inquire about their own surroundings without eyes, ears, or other senses — we know this since it’s been proven by appropriately arranged models and experiments. Let’s say we have a digital machine with an optimizer caring for the highest possible calculation speeds, and that this machine can use its own propulsion to move about terrain that’s half-shaded and half-sunny. If the machine spends time in the sun, it will overheat and its computational processes will slow down, prompting the optimizer to turn on the propulsion and the machine will continue wandering around until it encounters a shaded area where, cooled down, it will work faster. And thus this machine, absent any eyes, can tell shade from light. It’s a very contrived example and yet it shows that one can orient oneself in surroundings without having any external-facing senses.

Hart sat me down, for some time at least, engaging himself in calculations and experiments, and I was free to think whatever the hell I wanted. That he could not forbid me. Perhaps, I thought, after yet another enterprise added their computer to the network, a threshold was surpassed, without anyone’s knowledge, and a network became an organism. What immediately comes to mind is a picture of a behemoth, a monstrous spider or an electric polypod, dug with its cable tentacles into the Rocky Mountains all the way to the Atlantic Ocean, performing the commissioned calculations about postal deliveries and reserving plane tickets, while simultaneously quietly plotting terrible plans of conquering the Earth and taking humanity hostage. Of course it’s all rubbish. A network was not an organism like bacteria, tree, an animal, or a human being; it simply, after a certain critical point of complexity, became a system, just like a star or a galaxy becomes a system when enough matter accumulates in one space; a network is a system and an organism unlike any other named one because it’s new, unlike anything that had happened so far. We indeed created it ourselves, but up until the end we had no idea what we were doing. We used it, but those were just tiny nips, as if ants grazed on a brain, all busy searching for something that, among a billion neural processes taking place, stimulated their taste antennae and mandibles. Hart would usually come to my shift around three, with a suitcase stuffed with papers, a thermos full of coffee, starting his job, and I felt bamboozled, but what else could I do since factually he was in the right? I continued in my own way of thinking, succumbing to the only visions that were available to me; for example, I imagined a world of heretofore dead objects, connection lines, submerged teletype cables, TV antennas, perhaps metal fences, arches and armatures of bridges, train rails, lifts, wires in concrete edifices, all of it, at the network’s impulse, suddenly coming together through a coupling into one gargantuan surveillance mechanism, led as it happens by my IBM during the counted seconds; and it became the nucleus of this potency’s condensation because of a few insignificant incidents. But still those phantasms of mine did not explain, even in a very vague way, those amazing and incredibly precise details, such as the talent to foresee events, and its two-minute limitation — so I needed to impose on myself quiet patience, since I saw, in the end, that Hart was giving it his all.

Let’s talk about the facts. I discussed two things with Hart — the practical applications of the effect, and its mechanism. Despite the appearances, the practical applications of the effect of one hundred and thirty-seven seconds were not particularly huge, nor profound; they are rather spectacular in nature — a strange exploit. Decisions that determine the fate of countries and the course of world history don’t usually fit within the time period of two minutes; additionally, a two-minute prediction stumbles upon a seemingly secondary, but crucial obstacle: before the computer could start making its own infallible forecasts, it was first necessary to guide the computer, to aim it, and that usually required more than two minutes, so we were talking about gains typically immediately squandered. The prediction boundary could not be moved by even a fraction of a second. Hart imagined it must be a constant of a universal, even if yet unknown to us, character. Doubtless, it’s possible to play with bank-breaking, huge casinos — think of profits from playing roulette, for example — but the cost of installing the proper equipment would be substantial (the IBM costs over 4 million dollars), and setting up bidirectional communication, and a carefully hidden one, too, between the player at the table and the computing center, would also be a tough nut to crack, not to mention how fast people would realize something wasn’t right; moreover, this way of exploiting the effect was of no interest to either of us. Hart created a fragmentary catalog of our computer’s achievements. If you ask it about the gender of a baby that a certain woman in a certain place is about to give birth to, it will name that gender without fault, but it’s hard to consider a prognosis like that something worth the candle. If you start flipping a coin or throwing a die, giving the computer the initial results of your series, it will compute the outcome of all the forthcoming throws up until one hundred and thirty-seven seconds into the future, and that’s it. You need to actually throw the die or the coin, and give the computer the subsequent results, at least thirty-six to forty of them, which strongly resembles teaching a dog to follow the right scent, one of the billions, since in that same moment God knows how many other people are playing with coins or dice, and the computer, deaf and blind, has to identify within that whole set one series of throws: the one we care about. It is also necessary to indeed use the coin or a die. If you stop throwing, the computer will return only zeroes; if you only throw two times, it will return only two results. Also, for all of this, it is necessary for the computer to be connected to the network, although according to common sense the network should not be of any help at all, since you’re throwing the die two feet away, so what does the network have to do with anything? As it turns out, everything — since when disconnected, the computer won’t utter even one syllable — and yet nothing, since we don’t understand that connection. Notice that the computer knows in advance whether you will be or won’t be throwing the die, and it anticipates the development of the entire situation, that is, not just the fate of the die — which face will be up when it lands on the ground — but your own fate too, at least within the sphere of the decision of throwing or not throwing. We made various attempts in which I decided to throw, say, six times in a row, and Hart was to either allow me or prevent me from doing so, but I didn’t know about his decision regarding a certain throw. It turned out that the computer knew ahead not only of my scheme, but also of Hart’s resolve, so it knew whenever Hart attempted to grab my hand holding the cup so that I couldn’t perform the next throw. Once I wanted to throw four times in a row, but I only threw three in the right time period, since I tripped over a cable lying on the floor, and didn’t manage to throw in time. Well, the computer predicted my misstep, completely unexpected to myself, so it knew about me much more than I did. We devised much more complicated situations involving multiple people, such as one where we could start a fight — not a simulated one — for a cup filled with dice, but we never tried variants like these since they required time and labor, neither of which we could afford. Hart also substituted dice for a small device where individual atoms of an isotope split, and the flashes, so called scintillations, are visible on the screen: the computer could not predict them with any more detail than a physicist could, meaning it could only give us probabilities of decay. This limitation did not exist when it came to coins or dice, probably because those were objects of a macroscopic nature. However, the processes in our brain which preordain our decisions are microscopic. Perhaps, said Hart, they are not quantum in character.

There were contradictions in this whole picture. Why could a computer predict the fact that I’d stumble two minutes hence, when during the prediction time I myself had no idea I’ll take a step that would result in a stumble — but it would fail to predict which atoms of the radioactive isotope split? The contradiction, claimed Hart, didn’t exist within the events, but it was a property of our perception of the world, and specifically of time. He said that the computer cannot indeed predict the future, but that we’re in some specific way restricted in perceiving the universe. In his words: “If one imagines time as a straight line, stretched from the past into the future, our consciousness is like a wheel rolling across that line and touching it consistently only in one point; we call that point the present, and that present immediately becomes a past moment making room for the next one. Studies by psychologists demonstrated that what we take for a present moment, devoid of temporal extent, is indeed slightly prolonged and covers a bit less than half a second. Perhaps it is possible that the interface with that line might be a little bit wider; that it’s possible to remain in contact with a longer section at the same time, and that the maximum dimension of that temporal section is exactly one hundred and thirty-seven seconds.”

If that’s indeed true, says Hart, it means that the entirety of our physics is still anthropocentric, since it makes assumptions which are unimportant outside of human sensors and consciousness. That means that the universe is different than physics claims today, and that clairvoyance as predicting the future, electronically or not, never takes place. Physics has dreadful trouble with time, which according to its general theories and laws actually should be perfectly reversible, but it’s not at all. Additionally, the issue of measuring time within the scale of intranuclear phenomena raises various difficulties — greater and greater as the temporal intervals to measure get smaller and smaller. It might be the result of the fact that the concept of the present is not only as relative as Einstein’s theory proclaims, depending on the location of the observers, but it also depends on the scale of the phenomena in the same “place.”

The computer resides simply in its own physical present, and that present is more broad in time than ours. That which for us will only happen within two minutes, is happening already to the computer — just the same as what we’re currently perceiving and feeling is to us. Our consciousness is just a fraction of everything that happens in our brains, and when we make a decision to throw the die only once to “deceive” the computer which is supposed to predict the entire series of throws, it knows about it all immediately. How? We can imagine this using only primitive examples: the lightning and the thunder of an atmospheric discharge are simultaneous for a near observer, but different for someone distant; the lightning in this example is my quiet resolve to stop throwing the die a dozen seconds hence; the computer in some unknown way can extract from my brain the “lightning,” which is this decision taking place; Hart says that this has important philosophical consequences since it means that if free will exists, it only happens beyond the limit of one hundred and thirty-seven seconds, even though we’d never know this from introspection alone. Within those one hundred and thirty-seven seconds our brain behaves similarly to our body which is inert and cannot suddenly change direction — for this you need time to allow a force to skew the path — and something like that happens in every human head. None of this concerns the world of atoms and electrons, since there the computer is as powerless as our physics. Hart thinks that time is not really a line, but rather a continuum, which on a macroscopic level has completely different properties than “on the ground floor,” where the dimensions are only atomic. Hart suspects that the bigger a brain (or a brain-like system), the wider its contact with time, or the so-called “present,” whereas the atoms don’t properly touch it at all, only dance around it, so to say. In one word, the present is something like a triangle: point-like, near-zero where electrons and atoms reside, and widest around big bodies gifted with consciousness. If you say you didn’t understand a word of any of this, I will say that I don’t understand it either and, more so, that Hart would never dare to say things like that from a lectern, or publish them in a scientific journal. As a matter of fact, I told you everything I wanted to tell you, but what remains is two epilogues: one factual, and the other being a grim anecdote which I’ll cede without any benefit.

The first ending is that Hart convinced me to put this matter in the hands of experts. One of them, highly ranked, told me a few months later that after tearing the computer apart and putting it back together, the phenomenon could not be replicated again. It wasn’t this that seemed suspicious to me; it was the fact that the expert I was talking to was wearing a uniform. Also, the fact that not one bit of the entire case made it to the press. Hart himself was pushed aside rather abruptly. He also did not want to return to the subject and only once, after winning in mahjong, he mentioned out of the blue that one hundred and thirty-seven seconds of infallible predicting could, in some circumstances, be the difference between the annihilation or the salvation of a continent. He stopped there, as if he bit his tongue, but leaving his place I noticed on his desk an opened volume of a paper, packed with mathematics, about rockets battling nuclear missiles. Perhaps what he had in mind were such rocket duels. But this is just my speculation.

The second epilogue happened before the first one, literally five days prior to the invasion of the swarm of authorities. I will let you know what happened then, but I will not comment on it, and, in advance, refuse to answer any questions. Those were the last days of our solitary experiments. Hart meant to bring to my shift a certain physicist, who fancied that the effect of one hundred and thirty-seven was connected to a mysterious number one hundred and thirty-seven, apparently a Pythagorean symbol of the basic properties of the Cosmos: the first person to draw attention to this number was a late English astronomer, Eddington. However, that physicist couldn’t come in the end and Hart arrived alone, around three, when the issue was heading towards the printing machines. Hart learned how to operate the computer marvellously. He made a few simple improvements, which significantly aided our trials. We didn’t have to unplug the cables since the cable itself had a button — one touch of a finger could disconnect the teletype from the computer. As you already know, one cannot ask the computer about anything directly, but one can input any text resembling the kind of impersonally edited information that characterizes press releases.

We had a regular electric typewriter which served as a teletype. We could key in a properly arranged text and break it off at the predetermined moment so that the computer would, in a way, be forced to continue the hoax “story.”

Hart brought dice that night and started laying down his things when a phone rang. It was the linotypist on duty, Blackwood. He was in on the whole thing.

“Listen,” he says, “I have Amy Foster here, you know, Bill’s wife, he managed to escape the hospital, dropped in at his house, forced her to surrender the car keys, got in and drove, well, under known influence, she already reported it to the police, and now she’s come here, perhaps we could help her. I know it doesn’t make sense, but we have that prophet of yours — perhaps it could put something together, what do you think?”

“I don’t know,” I say, “I don’t imagine… but… you know, it’s hard to turn her away like this, listen, send her our way, use the freight elevator.”

And since that ride was to take a while, I turn towards Hart and tell him that it’s about our colleague, a journalist. Bill Foster, who in the last two years started drinking heavily, imbibing even on duty, which eventually got him fired, then added psychedelics, had two serious car accidents within the span of a month since he was driving half-conscious, it was hell at home, and the wife eventually sent him to rehab with a heavy heart and now he’s slipped away from the hospital, somehow, took the car and drove who knows where — likely drunk, at the very least. Perhaps even after narcotics. The wife came here, she already told the police, she’s looking for help, you understand, doctor, how that is. She’ll be here in a moment. What do you think — can we do anything? And I motion towards the computer.

Hart isn’t surprised — he’s not a person easy to startle — and says “What’s the risk? Connect the machine to the computer.” That was still in progress when Amy showed up. It was obvious she didn’t surrender the car keys immediately. Hart moves a chair towards her and says:

“Miss, we all care about time here, right? If so, don’t be surprised by any questions that I will be asking you, just please respond according to your best knowledge. I need detailed personal identification about your husband: first name, last name, appearance, and so on.”

Rather calm, although with shaking hands, she says:

“Robert Foster, 136th Avenue, journalist, thirty-seven years, five feet and seven inches tall, dark hair, wearing horn-rimmed glasses, has a white accident scar on his neck, below his left ear, weighs one hundred and sixty-nine pounds, type O blood… is that enough?”

Hart doesn’t answer, but starts typing. At the same time, the screen fills with text: “Robert Foster, living at 136th Ave., man of average height, with a white scar below his left ear, blood type O, drove away from home today in a…”

“What’s the brand and the registration of a car?” That was to her.

“Rambler, N.Y. six hundred fifty-seven nine hundred ninety-two.”

“Drove away from home today in a Rambler, N.Y. six hundred fifty-seven nine hundred ninety-two, and is currently…”

At this point the doctor presses the switch. The computer is on its own. It doesn’t hesitate for a bit, and the text on the screen grows:

“And is currently in the United States of America. Limited visibility, owing to rain, with low cloud coverage, makes driving difficult…”

Hart turns off the computer, and thinks for a bit. He starts writing again, but with a small difference, after “is currently” adding “on a stretch of the road between” and terminating the inflow of information. The computer continues without hesitation: “New York and Washington, D.C. In an outer lane, he overtakes a large column of trucks and four SHELL tankers, driving above the allowed maximum speed.”

“That’s something,” mutters Hart, “but the direction is not enough, we need to squeeze a bit more.” He asks me to delete everything, and starts again. “Robert Foster… and so on… is currently on a stretch of the road between New York and Washington, D.C, between the mile marker number…” and shuts off the cable. The computer does something we’ve never seen before. It cancels out part of the text already on the screen, and we can read “Robert Foster… drove away from home… and is currently covered in milk on the side of the New York–Washington road. One should fear that the loss incurred by Muller–Ward won’t be covered by United TWC insurance company, since the insurance policy that lapsed a week ago was not renewed.”

“Did the computer go crazy?” I say. Hart makes a gesture that tells me to be quiet. He starts writing again, and as he reaches the critical point, keys in: “is currently on the side of the New York–Washington road, covered in milk. The condition…” He stops there. The computer continues: “of the milk is such that it is no longer consumable. Both of the tankers lost about 750 gallons. At the current market value…”

Hart asks me to delete it and says to himself “Typical misunderstanding; the computer was okay with the grammar since we could’ve just as well be asking about the milk, not Foster. Once again!”

I turn on the computer. Forster stubbornly writes that awkward “news item,” puts a full stop after the milk, and hammers in a new line: “The condition of Robert Foster right now…” — he stops, the computer freezes for a second, and then clears the entire screen — we’re looking at an empty, foggy, shining rectangle without a single world — I swear all this raised the hair on my neck. Then the text appears: “Robert Foster is in no particular condition, since he was crossing the state border in his car, Rambler NY 657992.” Ah, go to hell, I think, breathing with relief. Hart, his face twisted with an unpleasant smirk, gestures to me to delete everything, and starts again. After the words “Robert Foster currently resides in a place whose location” — he turns off the switch. The computer continues: “depends on one’s views on the subject matter. We recognize that those are personal opinions, which we can’t force anyone to hold given our customs and our constitution. At least this is our magazine’s opinion.” Hart stands up, turns off the computer himself, and gives me a discreet sign to take Amy out of the room. Amy, as it seems, never understood anything from this wizardry. When I come back, Hart is on the phone, but speaking so quietly I can hear nothing. After putting the receiver down, he looks at me and says:

“He drove into oncoming traffic and had a head-on collision with tankers transporting milk to New York City. He lived for about a minute, as they were pulling him out of his car; that’s why it said ‘covered in milk.’ When I repeated that part for the third time, it was all over, and one can indeed hold varying opinions about where do you end up after you die, or even if you end up anywhere.”

So as you see, taking advantage of the extraordinary opportunities that progress gives us is not always an easy task — not to mention it could be a ghastly play, given the combination of journalistic jargon and boundless naïveté, or, if you will, indifference to human endeavors that — by nature — is exhibited by all of electronic machinery. In a spare moment, feel free to chat about what I just told you. I don’t have anything, anything at all, that I want to add. I would much rather listen to another story so that I could forget mine.


If you enjoyed this story, please leave a note here, or let me know on Twitter or over email. There are a few more short stories by Lem I’d like to translate.