Copyright ©2012 NPR. For personal, noncommercial use only. See Terms of Use. For other uses, prior permission required.

RENEE MONTAGNE, HOST:

We're going to hear now the story of how computers came to play a role in predicting the winner on election night. 1952 was an historic year for TV coverage for politics - the first coast-to-coast broadcast for a presidential election. Walter Cronkite anchored his first election night broadcast for CBS, and it was the first time computers were brought in to help predict the outcome. NPR's technology correspondent Steve Henn explains how the event helped usher in the computer age but it wasn't exactly love at first sight.

(SOUNDBITE OF CBS BROADCAST)

CHARLES COLLINGWOOD: This is the face of a UNIVAC. A UNIVAC is a fabulous electronic machine, which we have borrowed to...

STEVE HENN, BYLINE: CBS's Charles Collingwood was the reporter assigned to the UNIVAC, one of the world's first commercial computers.

(SOUNDBITE OF CBS BROADCAST)

COLLINGWOOD: See those round things over there, looks kind of like candy mints? Well, those are reels of metallic tape.

HENN: Collingwood sat in front of a mock-up of the UNIVAC's console in New York. It was the size of a large desk, with something that looked like a blinking bookcase sitting on top. But the real UNIVAC, which took up the better part of a room, was almost 100 miles away, in Philadelphia, with its programmers and a CBS camera crew.

(SOUNDBITE OF CBS BROADCAST)

COLLINGWOOD: It's there with its operator. On the right of the UNIVAC, there's something which looks like a typewriter. That's the way UNIVAC talks.

HENN: Just watching Collingwood 60 years later, you can see he's not sure what to make of this machine. He calls it an electronic brain. He personifies it.

(SOUNDBITE OF CBS BROADCAST)

COLLINGWOOD: He's sitting there in his corner humming away.

IRA CHINOY: It was by no means a done deal that computers should be a technology used in news in any way, let alone on election night.

HENN: Ira Chinoy is associate dean of journalism at the University of Maryland. He wrote a dissertation about UNIVAC's big night. For CBS, using a computer was a bit of a gimmick - a sideshow. But for Remington Rand, the company that made the UNIVAC, this was an enormous gamble.

CHINOY: There was a clear awareness that if they messed this up on election night, it might set their nascent industry back quite a bit.

HENN: And early in the evening, things were not going well.

(SOUNDBITE OF CBS BROADCAST)

COLLINGWOOD: Have you got a prediction for us, UNIVAC?

HENN: No response. The typewriter didn't move and to the audience at home, the UNIVAC must have looked like a big dumb box.

(SOUNDBITE OF CBS BROADCAST)

COLLINGWOOD: You're a very impolite machine, I must say. But he's an awfully rapid calculator.

HENN: Behind the scenes in Philadelphia, not everything was as it seemed. The UNIVAC actually did make a prediction, but someone held it back. Most likely, the computer programmers themselves, and the most likely reason is because the prediction seemed so ridiculous. Before election night 60 years ago, the race between Adlai Stevenson and Dwight Eisenhower looked close. But early in the night, with just over three million votes counted, UNIVAC predicted the odds were 100 to 1 in favor of Eisenhower. It wasn't until after midnight that a company spokesman in Philadelphia came on the air and admitted...

(SOUNDBITE OF CBS BROADCAST)

UNIDENTIFIED MAN: We should have had nerve enough to believe the machine in the first place. It was right. We were wrong. Next year we'll believe it.

HENN: UNIVAC's early faltering actually turned into a publicity coup for Remington Rand. Newspapers later ran headlines like: A Machine Makes a Monkey Out of Man. UNIVAC became a cultural icon. It showed up on the cover of a Superman comic book. And Wile E. Coyote built a UNIVAC to help him capture the elusive Bugs Bunny.

(SOUNDBITE OF CARTOON)

MEL BLANC: (as Wile E. Coyote) Finished. Wile E., you're such a genius.

HENN: But there was often an undercurrent of mockery, a hint that this supposedly all-knowing machine wasn't quite what it was cracked up to be. When the UNIVAC offered advice to Wile E. Coyote, it didn't always work.

(SOUNDBITE OF EXPLOSION)

BLANC: (as Wile E. Coyote) Oh no.

HENN: Back in 1952, NBC also used a computer on election night - the Monrobot. But according to Ira Chinoy, ambivalence about the computer continued for much of the decade.

CHINOY: The way we think about technology, if we look in the rearview mirror, is we're all thinking, oh, well, they just, they marched in, they became a fixture on election nights and that's it. NBC actually backed away from using a computer in 1954 and decided that, you know, a good reporter is better than any kind of statistical device.

HENN: But all three networks were using computers by the next presidential election. One of them, though, set up a challenge: man versus machine. ABC invited pollster Lou Harris and a team of 100 reporters in the field to compete against a computer to see who could call the election first. When it came to crunching numbers, the computer was untouchable, but the predictive models used in its programming were simplistic. Harris based his prediction on specific voting districts, which he thought would mirror the overall outcome. His team tallied the results with slide rules and won.

LOU HARRIS: Basically, patterns emerge from data. If you can't read them right, then you can't tell the story.

HENN: And yet Harris wondered if computers could do better. So, he spent some time at IBM, which by now had become the industry's leader. And there Harris pushed the IBM programmers to create more sophisticated election night software.

HARRIS: Because I said, look, I've got to get the computer to print out this, this, this, this and this. And they said, good Lord. It was multiple dimensions versus simple dimensions.

HENN: Harris, who had been John Kennedy's pollster, was contracted by CBS to help with the 1962 midterm elections. His computer-generated predictions were fed to Walter Cronkite.

HARRIS: He said to me after we made the first prediction, he said, Lou, you better be right, my whole future depends on it. I said, Walter, I'm sure it will be.

HENN: Cronkite, on Lou Harris's advice, called the Michigan governor's race that year for George Romney, even though early returns had the senior Romney behind. That night CBS destroyed the competition, making call after call accurately and early.

MARTIN PLISSNER: This was a huge triumph for the then newly created CBS News election unit.

HENN: Martin Plissner was CBS political director for over three decades. He says it was now clear journalists understood the computers' place on election night.

PLISSNER: They knew that the quality of the information they were getting out of it depended considerably on the quality of the information they were putting into it.

HENN: In the last 60 years there have been election night disasters, but when the information was good and the programming was clever, computers could be immensely powerful tools. In retrospect, Ira Chinoy says, it may be hard to understand why computers didn't march in and take a central place immediately on election night, 1952. But...

CHINOY: New things both engage us and also scare us, right? We're drawn to them, but they're disruptive.

HENN: You can hear that on the CBS broadcast from 60 years ago. You can hear it in Charles Collingwood's voice.

(SOUNDBITE OF CBS BROADCAST)

COLLINGWOOD: This is not a joke or a trick. It's an experiment. We think it's going to work. We don't know. We hope it'll work.

HENN: In the end, it did. But of course computers were also getting faster, more powerful. And the big early machines like the UNIVAC, made using thousands of vacuum tubes, were shrinking. By the late 1960s, tiny electronic circuits, or microchips, would transform the industry and computers would begin to find their way into all of our lives. Steve Henn, NPR News.

(SOUNDBITE OF MUSIC)

MONTAGNE: And there's more about UNIVAC and election night 1952 at our website, NPR.org.

Copyright © 2012 NPR. All rights reserved. No quotes from the materials contained herein may be used in any media without attribution to NPR. This transcript is provided for personal, noncommercial use only, pursuant to our Terms of Use. Any other use requires NPR's prior permission. Visit our permissions page for further information.

NPR transcripts are created on a rush deadline by a contractor for NPR, and accuracy and availability may vary. This text may not be in its final form and may be updated or revised in the future. Please be aware that the authoritative record of NPR's programming is the audio.

Comments

 

Please keep your community civil. All comments must follow the NPR.org Community rules and terms of use, and will be moderated prior to posting. NPR reserves the right to use the comments we receive, in whole or in part, and to use the commenter's name and location, in any medium. See also the Terms of Use, Privacy Policy and Community FAQ.

Support comes from: