'What Do You Care What Other People Think?'

by Richard P Feynman

  Before I left Houston, I continued my surreptitious investigation of the rumor that the White House had put pressure on NASA to launch the shuttle. Houston is the center of communication, so I went over to the telemetry people and asked about their switching system. I went through the same stuff as I did in Florida—and they were just as nice to me—but this time I found out that if they wanted to tie in the shuttle to the Congress, the White House, or to anywhere, they need a three-minute warning—not three months, not three days, not three hours—three minutes. Therefore they can do it whenever they want, and nothing has to be written down in advance. So that was a blind alley.

  I talked to a New York Times reporter about this rumor one time. I asked him, “How do you find out if things like this are true?”

  He says, “One of the things I thought to do was to go down and talk to the people who run the switching system. I tried that, but I wasn’t able to come up with anything.”

  During the first half of April, General Kutyna’s group received the final results of the tests NASA was making at Marshall. NASA included its own interpretations of the results, but we thought we should write everything over again in our own way. (The only exceptions were when a test didn’t show anything.)

  General Kutyna set up a whole system at Marshall for writing our group’s report. It lasted about two days. Before we could get anywhere, we got a message from Mr. Rogers: “Come back to Washington. You shouldn’t do the writing down there.”

  So we went back to Washington, and General Kutyna gave me an office in the Pentagon. It was fine, but there was no secretary, so I couldn’t work fast.

  Bill Graham had always been very cooperative, so I called him up. He arranged for me to use a guy’s office—the guy was out of town—and his secretary. She was very, very helpful: she could write up something as fast as I could say it, and then she’d revamp it, correcting my mistakes. We worked very hard for about two or three days, and got large pieces of the report written that way. It worked very well.

  Neil Armstrong, who was in our group, is extremely good at writing. He would look at my work and immediately find every weak spot, just like that—he was right every time—and I was very impressed.

  Each group was writing a chapter or two of the main report. Our group wrote some ofthe stuff in “Chapter 3: The Accident,” but our main work was “Chapter 4: The Cause ofthe Accident.” One result of this system, however, was that we never had a meeting to discuss what each of our groups found out—to comment on each other’s findings from our different perspectives. Instead, we did what they call “wordsmithing”—or what Mr. Hotz later called “tombstone engraving”—correcting punctuation, refining phrases, and so on. We never had a real discussion of ideas, except incidentally in the course of this wordsmithing.

  For example, a question would come up: “Should this sentence about the engines be worded this way or that way?”

  I would try to get a little discussion started. “From my own experiences, I got the impression that the engines aren’t as good as you’re saying here…”

  So they’d say, “Then we’ll use the more conservative wording here,” and they’d go on to the next sentence. Perhaps that’s a very efficient way to get a report out quickly, but we spent meeting after meeting doing this wordsmithing.

  Every once in a while we’d interrupt that to discuss the typography and the color of the cover. And after each discussion, we were asked to vote. I thought it would be most efficient to vote for the same color we had decided on in the meeting before, but it turned out I was always in the minority! We finally chose red. (It came out blue.)

  One time I was talking to Sally Ride about something I mentioned in my report on the engines, and she didn’t seem to know about it. I said, “Didn’t you see my report?”

  She says, “No, I didn’t get a copy.”

  So I go over to Keel’s office and say, “Sally tells me she didn’t get a copy of my report.”

  He looks surprised, and turns to his secretary. “Please make a copy of Dr. Feynman’s report for Dr. Ride.”

  Then I discover Mr. Acheson hasn’t seen it.

  ”Make a copy and give it to Mr. Acheson.”

  I finally caught on, so I said, “Dr. Keel, I don’t think anybody has seen my report.”

  So he says to his secretary, “Please make a copy for all the commissioners and give it to them.”

  Then I said to him, “I appreciate how much work you’re doing, and that it’s difficult to keep everything in mind. But I thought you told me that you showed my report to everybody.”

  He says, “Yes, well, I meant all of the staff.”

  I later discovered, by talking to people on the staff, that they hadn’t seen it either.

  When the other commissioners finally got to see my report, most of them thought it was very good, and it ought to be in the commission report somewhere.

  Encouraged by that, I kept bringing up my report. “I’d like to have a meeting to discuss what to do with it,” I kept saying.

  “We’ll have a meeting about it next week” was the standard answer. (We were too busy wordsmithing and voting on the color of the cover.)

  Gradually I realized that the way my report was written, it would require a lot of wordsmithing—and we were running out of time. Then somebody suggested that my report could go in as an appendix. That way, it wouldn’t have to be wordsmithed to fit in with anything else.

  But some of the commissioners felt strongly that my report should go in the main report somehow: “The appendices won’t come out until months later, so nobody will read your report if it’s an appendix,” they said.

  I thought I’d compromise, however, and let it go in as an appendix.

  But now there was a new problem: my report, which I had written on my word processor at home, would have to be converted from the IBM format to the big document system the commission was using. They had a way of doing that with an optical scanning device.

  I had to go to a little bit of trouble to find the right guy to do it. Then, it didn’t get done right away. When I asked what happened, the guy said he couldn’t find the copy I had given him. So I had to give him another copy.

  A few days later, I finished writing my report about the avionics, and I wanted to combine it with my report on the engines. So I took the avionics report to the guy and I said, “I’d like to put this in with my other report.”

  Then I needed to see a copy of my new report for some reason, but the guy gave me an old copy, before the avionics was added. “Where’s the new one with the avionics?” I said.

  “I can’t find it”—and so on. I don’t remember all the details, but it seemed my report was always missing or half-cooked. It could easily have been mistakes, but there were too many of them. It was quite a struggle, nursing my report along.

  Then, in the last couple of days, when the main report is ready to be sent to the printer, Dr. Keel wants my report to be wordsmithed too, even though it’s going in as an appendix. So I took it to the regular editor there, a capable man named Hansen, and he fixed it up without changing the sense of it. Then it was put back into the machine as “Version #23”—there were revisions and revisions.

  (By the way: everything had 23 versions. It has been noted that computers, which are supposed to increase the speed at which we do things, have not increased the speed at which we write reports: we used to make only three versions—because they’re so hard to type—and now we make 23 versions!)

  The next day I noticed Keel working on my report: he had put all kinds of big circles around whole sections, with X’s through them; there were all kinds of thoughts left out. He explained, “This part doesn’t have to go in because it says more or less what we said in the main report.”

  I tried to explain that it’s much easier to get the logic if all the ideas are together, instead of everything being distributed in little pieces all over the main report. “After all,” I said, “it’s only gonna be an appendix. It
won’t make any difference if there’s a little repetition.”

  Dr. Keel put back something here and there when I asked him to, but there was still so much missing that my report wasn’t anything like it was before.

  The Tenth Recommendation

  SOMETIME in May, at one of our last meetings, we got around to making a list of possible recommendations. Somebody would say, “Maybe one of the things we should discuss is the establishment of a safety board.”

  “Okay, we’ll put that down.”

  I’m thinking, “At last! We’re going to have a discussion!”

  But it turns out that this tentative list of topics becomes the recommendations—that there be a safety board, that there be a this, that there be a that. The only discussion was about which recommendation we should write first, which one should come second, and so forth.

  There were many things I wanted to discuss further. For example, in regard to a safety board, one could ask: “Wouldn’t such a committee just add another layer to an already overgrown bureaucracy?”

  There had been safety boards before. In 1967, after the Apollo accident, the investigating committee at the time invented a special panel for safety. It worked for a while, but it didn’t last.

  We didn’t discuss why the earlier safety boards were no longer effective; instead, we just made up more safety boards: we called them the “Independent Solid Rocket Motor Design Oversight Committee,” the “Shuttle Transportation System Safety Advisory Panel,” and the “Office of Safety, Reliability, and Quality Assurance.” We decided who would oversee each safety board, but we didn’t discuss whether the safety boards created by our commission had any better chance of working, whether we could fix the existing boards so they would work, or whether we should have them at all.

  I’m not as sure about a lot of things as everybody else. Things need to be thought out a little bit, and we weren’t doing enough thinking together. Quick decisions on important matters are not very good—and at the speed we were going, we were bound to make some impractical recommendations.

  We ended up rearranging the list of possible recommendations and wordsmithing them a little, and then we voted yes or no. It was an odd way of doing things, and I wasn’t used to it. In fact, I got the feeling we were being railroaded: things were being decided, somehow, a little out of our control.

  At any rate, in our last meeting, we agreed to nine recommendations. Many of the commissioners went home after that meeting, but I was going to New York a few days later, so I stayed in Washington.

  The next day, I happened to be standing around in Mr. Rogers’s office with Neil Armstrong and another commissioner when Rogers says, “I thought we should have a tenth recommendation. Everything in our report is so negative; I think we need something positive at the end to balance it.”

  He shows me a piece of paper. It says,

  The Commission strongly recommends that NASA continue to receive the support of the Administration and the nation. The agency constitutes a national resource and plays a critical role in space exploration and development. It also provides a symbol of national pride and technological leadership. The Commission applauds NASA’s spectacular achievements of the past and anticipates impressive achievements to come. The findings and recommendations presented in this report are intended to contribute to the future NASA successes that the nation both expects and requires as the 21st century approaches.

  In our tour months of work as a commission, we had never discussed a policy question like that, so I felt there was no reason to put it in. And although I’m not saying I disagreed with it, it wasn’t obvious that it was true, either. I said, “I think this tenth recommendation is inappropriate.”

  I think I heard Armstrong say, “Well, if somebody’s not in favor of it, I think we shouldn’t put it in.”

  But Rogers kept working on me. We argued back and forth a little bit, but then I had to catch my flight to New York.

  While I was in the airplane, I thought about this tenth recommendation some more. I wanted to lay out my arguments carefully on paper, so when I got to my hotel in New York, I wrote Rogers a letter. At the end I wrote, “This recommendation reminds me of the NASA flight reviews: ‘There are critical problems, but never mind—keep on flying!’ ”

  It was Saturday, and I wanted Mr. Rogers to read my letter before Monday. So I called up his secretary—everybody was working seven days a week to get the report out in time—and I said, “I’d like to dictate a letter to you; is that all right?”

  She says, “Sure! To save you some money, let me call you right back.” She calls me back, I dictate the letter, and she hands it directly to Rogers.

  When I came back on Monday, Mr. Rogers said, “Dr. Feynman, I’ve read your letter, and I agree with everything it says. But vou’ve been out-voted.”

  “Out-voted? How was I out-voted, when there was no meeting?”

  Keel was there, too. He says, “We called everybody, and they all agree with the recommendation. They all voted for it.”

  “I don’t think that’s fair!” I protested. “If I could have presented my arguments to the other commissioners, I don’t think I’d have been out-voted.” I didn’t know what to do, so I said, “I’d like to make a copy of it.”

  When I came back, Keel says, “We just remembered that we didn’t talk to Hotz about it, because he was in a meeting. We forgot to get his vote.”

  I didn’t know what to make of that, but I found out later that Mr. Hotz was in the building, not far from the copy machine.

  Later, I talked to David Acheson about the tenth recommendation. He explained, “It doesn’t really mean anything; it’s only motherhood and apple pie.”

  I said, “Well, if it doesn’t mean anything, it’s not necessary, then.”

  “If this were a commission for the National Academy of Sciences, your objections would be proper. But don’t forget,” he says, “this is a presidential commission. We should say something for the President.”

  “I don’t understand the difference,” I said. “Why can’t I be careful and scientific when I’m writing a report to the President?”

  Being naive doesn’t always work: my argument had no effect. Acheson kept telling me I was making a big thing out of nothing, and I kept saying it weakened our report and it shouldn’t go in.

  So that’s where it ended up: “The Commission strongly recommends that NASA continue to receive the support of the Administration and the nation…”—all this “motherhood and apple pie” stuff to “balance” the report.

  While I was flying home, I thought to myself, “It’s funny that the only part of the report that was genuinely balanced was my own report: I said negative things about the engine, and positive things about the avionics. And I had to struggle with them to get it in, even as a lousy appendix!”

  I thought about the tenth recommendation. All the other recommendations were based on evidence we had found, but this one had no evidence whatsoever. I could see the whitewash dripping down. It was obviously a mistake! It would make our report look bad. I was very disturbed.

  When I got home, I talked to Joan, my sister. I told her about the tenth recommendation, and how I had been “out-voted.”

  “Did you call any of the other commissioners and talk to them yourself?” she said.

  “Well, I talked to Acheson, but he was for it.”

  “Any others?”

  “Uh, no.” So I called up three other commissioners—I’ll call them A, B, and C.

  I call A, who says, “What tenth recommendation?”

  I call B, who says, “Tenth recommendation? What are you talking about?”

  I call C, who says, “Don’t you remember, you dope? I was in the office when Rogers first told us, and I don’t see anything wrong with it.”

  It appeared that the only people who knew about the tenth recommendation were the people who were in the office when Rogers told us. I didn’t bother to make any more telephone calls. After all, it’s enough—
I didn’t feel that I had to open all the safes to check that the combination is the same!*

  Then I told Joan about my report—how it was so emasculated, even though it was going in as an appendix.

  She says, “Well, if they do that to your report, what have you accomplished, being on the commission? What’s the result of all your work?”

  “Aha!”

  I sent a telegram to Mr. Rogers:

  PLEASE TAKE MY SIGNATURE OFF THE REPORT UNLESS TWO THINGS OCCUR: 1) THERE IS NO TENTH RECOMMENDATION, AND 2) MY REPORT APPEARS WITHOUT MODIFICATION FROM VERSION #23.

  (I knew by this time I had to define everything carefully.)

  In order to get the number of the version I wanted, I called Mr. Hotz, who was in charge of the documentation system and publishing the report. He sent me Version #23, so I had something definite to publish on my own, if worse came to worst.

  The result of this telegram was that Rogers and Keel tried to negotiate with me. They asked General Kutyna to be the intermediary, because they knew he was a friend of mine. What a good friend of mine he was, they didn’t know.

  Kutyna says, “Hello, Professor, I just wanted to tell you that I think you’re doing very well. But I’ve been given the job of trying to talk you out of it, so I’m going to give you the arguments.”

  “Fear not!” I said. “I’m not gonna change my mind. Just give me the arguments, and fear not.”

  The first argument was that if I don’t accept the tenth recommendation, they won’t accept my report, even as an appendix.

  I didn’t worry about that one, because I could always put out my report myself.

  All the arguments were like that: none of them was very good, and none of them had any effect. I had thought through carefully what I was doing, so I just stuck to my guns.

  Then Kutyna suggested a compromise: they were willing to go along with my report as I wrote it, except for one sentence near the end.

  I looked at the sentence and I realized that I had already made my point in the previous paragraph. Repeating the point amounted to polemics; removing the phrase made my report much better. I accepted the compromise.