The heliocentric theory was acceptable; the dance remained coherent. Repetto replayed the scene, with a "translation" in a small window, showing the positions of the planets represented at each moment. Maria still couldn't decipher the correspondence -- the Lambertians certainly weren't flying around in simple mimicry of the hypothetical orbits -- but the synchronized rhythms of planets and insect-astronomers seemed to mesh somewhere in her visual cortex, firing some pattern detector which didn't know quite what to make of the strange resonance.

She said, "So Ptolemy was simply bad grammar -- obvious nonsense. Doubleplus ungood. And they reached Copernicus a few years later? That's impressive. How long did they take to get to Kepler . . . to Newton?"

Zemansky said smoothly, "That was Newton. The theory of gravity -- and the laws of motion -- were all part of the model they were dancing; the Lambertians could never have expressed the shapes of the orbits without including a reason for them."

Maria felt the hairs rise on the back of her neck.

"If that was Newton . . . what came before?"

"Nothing. That was the first successful astronomical model -- the culmination of about a decade of trial and error by teams all over the planet."

"But they must have had something. Primitive myths. Stacks of turtles. Sun gods in chariots."

Zemansky laughed. "No turtles or chariots, obviously -- but no: no naive cosmologies. Their earliest language grew out of the things they could easily observe and model -- ecological relationships, population dynamics. When cosmology was beyond their grasp, they didn't even try to tackle it; it was a non-subject."

"No creation myths?"

"No. To the Lambertians, believing any kind of "myth" -- any kind of vague, untestable pseudo-explanation -- would have been like . . . suffering hallucinations, seeing mirages, hearing voices. It would have rendered them completely dysfunctional."

Maria cleared her throat. "Then I wonder how they'll react to us."

Durham said, "Right now, creators are a non-subject. The Lambertians have no need of that hypothesis. They understand evolution: mutation, natural selection -- they've even postulated some kind of macromolecular gene. But the origin of life remains an open question, too difficult to tackle, and it would probably be centuries before they realized that their ultimate ancestor was seeded "by hand" . . . if in fact there's any evidence to show that -- any logical reason why A. hydrophila couldn't have arisen in some imaginary prebiotic history.

"But it won't come to that; after a few more decades banging their heads against the problem of the primordial cloud, I think they'll guess what's going on. An idea whose time has come can sweep across the planet in a matter of months, however exotic it might be; these creatures are not traditionalists. And once the theory that their world was made arises in the proper scientific context, it's not going to drive them mad. All Alisa was saying was that the sort of primitive superstitions which early humans believed in wouldn't have made sense to the early Lambertians."

Maria said, "So . . . we'll wait until 'creators' are no longer a non-subject before we barge in and announce that that's exactly what we are?"

Durham replied, "Absolutely. We have permission to make contact once the Lambertians have independently postulated our existence -- and no sooner." He laughed, and added, with evident satisfaction, "Which we achieved by asking for much more."

Maria still felt uneasy -- but she didn't want to hold up proceedings while she grappled with the subtleties of Lambertian culture.

She said, "All right. Cosmology is the trigger, but they're looking for a deeper explanation for their chemistry. Are they having any luck?"

Repetto brought back the map of Planet Lambert; the markers showing the locations of the teams of theorists were replaced by small bar charts in the same positions. "These are the dance times sustained for various subatomic models which have been explored over the past five years. A few theories are showing some promise, improving slightly with each refinement; other groups are getting fairly random results. Nobody's come up with anything they'd be capable of communicating over any distance; these dances are too short-lived to be remembered by teams of messengers."

Maria felt her skin crawl, again. False messages die, en route. There was something chilling about all this efficiency, this ruthless pursuit of the truth. Or maybe it was just a matter of injured pride: treating some of humanity's most hard-won intellectual achievements as virtually self-evident wasn't the most endearing trait an alien species could possess.

She said, "So . . . no team is on the verge of discovering the truth?"

Repetto shook his head. "Not yet. But the Autoverse rules are the simplest explanation for the thirty-two atoms, by almost any criterion."

"Simplest to us. There's nothing in the Lambertians' environment to make them think in terms of cellular automata."

Zemansky said, "There was nothing in their environment to make them think in terms of atoms."

"Well, no, but the ancient Greeks thought of atoms -- but they didn't come up with quantum mechanics." Maria couldn't imagine a preindustrial human inventing the cellular automaton -- even as a mathematical abstraction -- let alone going on to hypothesize that the universe itself might be one. Clockwork cosmologies had come after physical clocks; computer cosmologies had come after physical computers.

Human history, though, clearly wasn't much of a guide to Lambertian science. They already had their Newtonian -- "clockwork" -- planetary model. They didn't need artifacts to point the way.

She said, "This 'aesthetic' which governs the acceptability of theories -- have you been able to map the neural structures involved? Can you reproduce the criteria?"

Repetto said, "Yes. And I think I know what you're going to ask next."

"You've devised your own versions of possible Lambertian cellular automaton theories? And you've tested them against the Lambertian aesthetic?"

He inclined his head modestly. "Yes. We don't model whole brains, of course -- that would be grossly unethical -- but we can run simulations of trial dances with nonconscious Lambertian neural models."

Modeling Lambertians modeling the Autoverse . . .

"So how did it go?"

Repetto was hesitant. "The results so far are inconclusive. None of the theories I've constructed have worked -- but it's a difficult business. It's hard to know whether or not I'm really stating the hypothesis in the way the Lambertians would -- or whether I've really captured all the subtleties of the relevant behavior in a nonconscious model."

"But it doesn't look promising?"

"It's inconclusive."

Maria thought it over. "The Autoverse rules, alone, won't explain the abundances of the elements -- which is the main problem the Lambertians are trying to solve. So what happens if they miss the whole idea of a cellular automaton, and come up with a completely different theory: something utterly misguided . . . which fits all the data nonetheless? I know, they've grasped everything else about their world far more smoothly than humans ever did, but that doesn't make them perfect. And if they have no tradition of giving up on difficult questions by invoking the hand of a creator, they might cobble together something which explains both the primordial cloud and the chemical properties of the elements -- without coming anywhere near the truth. That's not impossible, is it?"

There was an awkward silence. Maria wondered if she'd committed some terrible faux pas by suggesting that the criteria for contact might never be met . . . but she could hardly be telling these people anything they hadn't already considered.