Hoaxes and False Alarms

"Well it is rocket science, but it's not hard rocket science."

Siduri trusts no one

Siduri: But now, I thought that was what SETI@home was doing.

Seth: Well, that's a different project. That's a Berkeley project. We process all our data, as I say, real time. And any signals that look like they're good are verified by a second telescope at Jodrell Bank, in England, just south of Manchester. We have a slow-speed line that sends the information about a handful of signals, whatever you have in those 300 seconds, sends that information to Jodrell Bank. And then they're looking at the same star. So here we are in Puerto Rico, and there they are in Jodrell Bank. And we're both pointing at the same star, so if we find signals here, then these guys spend the next 300 seconds checking those out. So they're always 300 seconds behind the Arecibo telescope.

Siduri: That's clever.

Seth: Yeah, and obviously, if it's really ET, you know you should be able to find it in both places. But beyond that, you're on the rotating earth, and if you do a little geometry you realize that, well, the motion of the earth means that these two telescopes are moving—with respect to whatever star it is you're looking at—at a slightly different rate, so there's a slightly different Doppler shift, so the frequency ought to be different here than there. A small amount, but it's easily measured, see, so there's a little microcomputer that does that simple high school geometry, and looks for that change in the Doppler shift. And if it doesn't find it, then it knows this is not ET: this is a telecommunications satellite. So that's another trick.

So they all get checked out. Now, SETI@home, that's data that's taken by the Berkeley crowd, on the same telescope but in a different way. And their data is of a nature that, you know, you can look at it two months later and still be interested. With ours, ours is so sensitive that if you found anything, you would want to check it out right away.

Mark: And how often does that happen? Do you get calls in the middle of the night, saying "check this out right away"?

Seth: Well, I'm usually sitting there. I usually have the 6 to midnight shift; Jill Tarter takes the midnight to 6 AM shift; Peter Backus comes in and spells one of us. You're actually there. And if it finds a signal, I mean, it doesn't—you don't get these giant klaxon sounds and red flashing lights. MM-NEE! MM-NEE! I mean, that would be great, that's what it is on Star Trek, but what actually happens is the workstation goes "beep." You barely notice it.

But then it says, you know, "I found these six candidate signals," it's a real signal. And what it does is, it sends them to Jodrell Bank, right. And if they come back from Jodrell Bank—almost always, they come back "sorry, Charlie, this is not it"—but every now and again something gets through the filter. So Jodrell Bank says, "Looks good to me." And meanwhile it's looking in a database of known interferers to make sure it isn't in there.

If all these systems are "go, this could be the big one," you still don't call up the airlines and get a ticket to Stockholm. Because what happens next is it, you know, "beep beep," it comes again. And you're still not paying much attention. But now what it's done, because it's found a signal that both telescopes think might be real, it breaks the observing. It stops going up the dial and it now moves the Arecibo telescope away from the star. Just a little bit. Maybe the diameter of the moon or something like that. Just that far away. And the signal ought to go away now. If it's really ET, if it's really coming from the star, the signal oughtta go away. Now, if the signal's coming from some telecommunications satellite that's wheeling overhead, that signal's bouncing all around, I mean it's so strong, bouncing around the telescope structure, it doesn't really matter where you point the antenna. You'll pick up that signal. So, if you point away from the star and you still get the signal, you know right away it's not ET.

But if it goes away, it might be ET. Or maybe the telecommunications satellite just went over the horizon at the same time. And that can happen too. So if it goes away, then you keep looking, and you move the antenna back on the star and see if the signal comes back.

Siduri: Does your heart start speeding up at about that point?

Seth: At about that time, if it goes off, and it goes away, then you pay attention. Because it's sort of fun. And you always set the sensitivity so this happens once every couple nights. And if it doesn't find it there and then it comes back and it does find it again, then, yeah, then you become more interested. And if it goes off and doesn't find it, then you're standing up. That doesn't happen very often. It's happened to me once or twice. That's when you get excited.

And at some point, you know, it'll do this half a dozen times, and then it sort of throws up its little computer hands and says, "I don't know what this is anymore. This is something you know how to decide. I don't know what to do next." I don't know, I think only once has it ever gotten to the point where the computer didn't know what to do next.

Siduri: While you were there, or...?

Seth: Well, I was downstairs, and we were observing simultaneously—No, we had a remote setup downstairs. I wasn't actually at the telescope. This was '97, I was at Greenbank, but the second telescope that we were using those days wasn't working that day, and that was how we got fooled.

Siduri: Ohhh.

Seth: But we got fooled for something like 16 hours. That was pretty interesting. It was very interesting to see how people reacted when it looked like this was the real thing.

Siduri: How did people react?

Seth: Everybody was simultaneously excited and nervous. That's true. Nobody went home, nobody went to bed. People just hung out. The engineers were all hanging around the computers downstairs. And I half expected that we were going to see the government show up. Everybody expects that the government would move in if we ever got a signal. But the government had no interest in moving in. Much to my disappointment. Even the local, the Mountain View government didn't want to move in. Nobody wanted to move in.

[We laugh. But, transcribing this, it occurs to me that a real conspiracy theorist could take this as proof that the government already knows about aliens! Because why wouldn't they react to an extraterrestrial signal, unless they had some way of knowing that it was a false alarm? And how could they know that, unless they already knew where the real aliens were? Hmm?]

Seth: I sort of hoped that the pizza guy would move in. We were getting hungry. But what did happen is that the media started calling up.

Siduri: How'd they know?

Seth: Somebody just called them. I mean, there's no secrecy, so, you know, they called up the New York Times. And they called me: "What about that signal?" I said, "Well, we're following one, but, you know..."

Siduri: So you would really know right away.

Seth: Oh, you would know right away. Yeah. In fact, I really do believe you would not only know right away; you would know before we had fully checked out the signal. Because how long is it going to take us before we believe it? I mean, surely within an hour or so you'd be beginning to get the idea that this might be it. But it still could be a software bug, it could be—you know, there are all sorts of things that can go wrong. It's an extremely complicated system. It involves two telescopes, it's a one-off device so needless to say it isn't fully debugged, and all that. So you'd say, "okay, this is looking interesting, but I'm not sure I'm going to tell the world I found ET yet. I'm going to call up somebody at another antenna, another radio observatory, and say, 'You check this out, with your equipment, not ours. Because you've got a completely different system, and maybe yours isn't buggy or whatever, if that's what it is.'"

And they would say, "Okay, well, we'll do that, but we can't get to it until tonight, because, you know, the star isn't up yet, or we have to change the receiver," whatever. So they'll spend some time doing that, it'll maybe take a day or two. And so surely, four days, five days can easily go by before enough have seen it to say "Okay, I'm willing to stake my reputation on it. This is for real." But long before that happens you'll be reading about it at Albertson's in the check-out line. That media won't wait; they'll run with the story before you've confirmed it. Everybody will, they'll be going "oh, the sky is falling, interesting signal!" So that's the way it will happen.

Siduri: So you actually do observation work on a regular basis.

Seth: Yeah, I go down there. We observe at Arecibo. These are some pictures I took; I didn't take that one, but this is what we were using when we found that interesting signal. In fact you can see the box over there, that trailer, that's Project Phoenix. That's in West Virginia. The other one's in Australia. So we get to use this for a couple weeks in the spring and a couple weeks in the fall.

Siduri: How many observers are there?

Seth: Well, we're doing SETI so we observe at night, because the sun would mess up these narrow-band signals, so for us it's better to observe at night. So we observe from 6 PM to 6 AM. We just break that into two shifts, from 6 to midnight and midnight to 6. So you only need two astronomers, actually, but usually there are three of us there—Jill Tarter, who was sort of the prototype for the Jodie Foster character; Peter Backus, who heads up the observatory operation; and myself. The three of us. But there are also a bunch of engineers. The engineers are much more valuable than the astronomers. The astronomers are pretty stupid, and the engineers are pretty savvy; they're much smarter than we are.

Mark: I have to ask, it seems like a false positive signal would be about the best April Fool's joke ever. Has it happened yet?

Seth: Oh, you mean where somebody would try and sort of hoax a signal? Yeah, well, we've thought about it, and maybe they have too.

Siduri: [laughing] Thought about it.

Seth: Well, we wouldn't do it, because it's very important that people have faith in our—that this is a good experiment, that's competently run, that's important to us, because we're running with their money, after all. So we have to do the best job we can, so obviously we're not going to make a hoax. But somebody else might. Maybe a couple undergrads would find it interesting to do that.

But it's actually quite hard because—you know, getting a signal into the Arecibo telescope probably isn't all that hard; you have to just walk up there with a transmitter and turn it on. You can do that, I can picture that. Of course it has to be in the right frequency bandwidth at, by the way, I might say, the right time, because we're only looking at a certain small part of the band at any given time, so you're gonna have to know all that, but it's not— well it is rocket science, but it's not hard rocket science. Yeah, they could do that, they could do that.

But the point is that they then also have to fool the guys in England. So they gotta—now it's becoming bit more involved, right, because now they have to have two telescopes that believe this signal. They've got to get the right Doppler shift, so they'd better know their geometry. And then they've got to be savvy enough to, when we do this very simple test where we move the telescope back and forth, to turn the transmitters on and off. So they essentially have to be, it really has to be an inside job, I would say, because they need to know what we're doing when. And I don't see it happening as an inside job. It's just not going to happen. But, you know, I won't say it's impossible. I think it's very, very hard. Very hard, but certainly not impossible. That's why you would check it out for a few days.