“George, I believe that in some way this is Morag, it really is. Of course I fully accept she died, all those years ago. So something is going on here which isn’t normal, rational. It’s acausal for one thing. But I don’t believe she’s a ghost, with all the connotations of that word. There is no quality of—” I hesitated, unwilling to finish the sentence.

“Evil?” George asked softly.

“None of that, no. And it is Morag. Does that make sense?”

“No. But then, rainbows would make no sense if you had never seen one. If she’s not a ghost, then what is she, do you think? That language is obviously not human — or at least, not twenty-first-century human.”

“No.”

“Then what? Some kind of alien?”

“I suppose it’s possible. It seems a strange way for them to communicate, though.”

He shrugged. “What’s a good way? I’ve speculated about this stuff over the years. Look at it this way. I still cut my lawn.” It was just a scrap, overgrown by clover and weeds, but George seemed to like it that way. “Now, my evolutionary divergence from the grass is, what, half a billion years deep, more? And yet we communicate. I ask it if it wants to grow, by feeding it phosphates in the autumn and nitrogen in the spring. It answers by growing, or not. It asks me if I want it to grow over five centimeters, or if I want it to start colonizing the verges. It tells me this by doing it, you see. I say no, with my mower and my strimmer. So we communicate — not in symbols, but with the primal elements of all life forms, space to grow, food, life, death.”

“And you think it might be that way with intelligent aliens?”

“If there is no possibility of symbolic communication, maybe. But if they have the capability to reach us then they will be the ones with the lawn mower…”

“I don’t think aliens have anything to do with this,” I said firmly. “It feels too human for that.”

“Then there seems only one possibility left,” he said.

We both knew what he meant: that my Morag, with her high-density speech, was a visitation, not from the past, and not from some alien world, but from the future — our human future. In some ways I found that the most terrifying prospect of all, because it was the least comprehensible.

“Rosa guessed this,” I admitted. “Even before we recorded and analyzed Morag’s speech.”

“Well, she is a Poole.”

We could only speculate; we didn’t know enough. George changed the subject. He asked me if I still flew Frisbees.

When I was a kid, growing up on the Florida coast, I became fascinated with Frisbees. Everybody played with them. But try as I might I couldn’t find anybody, any book, to explain to me convincingly how the damn things actually flew — and especially how come they were so hard to fly right, why they dipped and flapped the way they did.

So, aged about ten, I used to buy up old Frisbees to experiment with them. At first it was just kiddie stuff, painting them or adding spectacular, useless fins. But then I tried a more systematic series of modifications. I cut chunks out, or added strips of plastic to the rim to change the weight distribution, or scored the flat surfaces with new patterns of grooves to change the flow of air. I didn’t really know what I was doing, of course, but I was instinctively systematic. I kept logs and even little movies of how my Frisbees flew, before and after the modification. It didn’t last long — kid fads never do — but when George had visited in those days he had always shown an interest.

“But what you don’t know,” I said to him now, “is that playing with Frisbees got me one of my first career breaks…”

In my final year at college I happened to look up Frisbees on the Net. I found to my surprise that still nobody had figured out how a Frisbee flew. Not only that, there would be practical applications of such knowledge, for planetary probes targeted at airy worlds like Mars and Venus and Titan would be spun for stability — they were high-tech, hugely expensive Frisbees sailing into unfamiliar atmospheres, sent to their fates on the basis of a scary lack of knowledge of how they actually flew.

“So I dug out my ten-year-old hobby,” I said to George. “And I looked up the theory, such as it was. A Frisbee gets its lift like a wing, but the front of the disc tends to get more lift than the back, which makes it unstable. But unlike an airplane wing it’s spinning, so that uneven lift is like a finger prodding a spinning gyroscope; it deflects a Frisbee’s course rather than making it flip completely. But I found that nobody had got beyond rough rules of thumb.

“Anyhow I started trying to figure out how a Frisbee really flies,” I said. “I went beyond what I could do as a kid. I scrounged some parts from my college lab, and gave a Frisbee a black box recorder.” I had installed a small accelerometer to measure the forces on the disc, and a magnetometer and light sensor so I could track its position compared to the sun and the Earth’s magnetic field, and a computer chip. Soon I was able to record all the essentials of a flight, and reproduce it at leisure in a simple VR environment. Later, when my professors got interested in what I was doing, I went further, such as by coating a Frisbee’s upper surface with sensors to measure the pressure and flow in detail.

“I quickly figured out the gross aerodynamic coefficients,” I said. “To optimize your flight you have to match your spin rate to your forward speed and angle of attack. But more important, I started to get an understanding of how pressure was distributed over the surface of the spinning disc, and was able to model ways how you might control this optimally, for instance with small flaps and holes to direct the airflow. NASA was doing the same sort of study, of course, but using spinning models in wind tunnels. I was able to get better results far more cheaply, just by smothering a Frisbee in sensors and flying it outdoors.” In the end the study turned from a hobby into a term paper, which NASA took up and sponsored. It was a great line on my CV when it came time for me to look for a job.

“I didn’t know all that,” George said. He grinned. “So you managed to combine career advancement with throwing Frisbees all day. I’m even more impressed.”

I shrugged. “You got to enjoy yourself.”

“Absolutely.”

I knew what I had to say next, even though it was difficult for both of us. “George — you always took an interest in my stuff, a proper interest, back when I was ten or eleven.”

We both knew what I meant. My dad was always faintly bemused by such stunts as experimenting with Frisbees. He would always throw a Frisbee or two with me. But he always spoke to me as a kid, if you know what I mean, which wasn’t necessarily the right thing to do, even if I was a kid. George spoke to me as a junior engineer; he took me seriously.

“It made a difference. To my whole life.”

George just nodded; he knew what I meant, and he knew it had to be said. He clapped me on the shoulder. “I guess you were never going to be a Steve Zodiac. But you would have made a good Matthew Matic.”

“Who?”

Fireball XL5… Something else that will disappear from the world with me. Never mind.”

George started to tire, so I called for a pod bus, and we found a bench and sat. Sitting there, breathing hard, I thought he looked ill for the first time during the visit. I could see the skull under his flesh, I thought, the skin drawn tight beneath his cheekbones, his mouth drawn in, his eyes perhaps creased in pain. A row of blank-walled modern houses, eyeless without windows, loomed before us, uncaring.

To my surprise George said he was thinking of selling up and moving away from England altogether.

“I’m going back to Amalfi,” he said. This is a small town on the Sorrento coast of Italy. “I went there after Rome, you know, after I went in search of Rosa. Once I found her I needed some time to recover, to get myself together again. The weather is still better there than here. I know it will be hard to sell up. Hell, it will be awful having to fly again. But I think I will be able to rest there, you know? That’s the way I’ve always thought of Amalfi, a place I could rest.”