Artifacts

by Greg Egan

Paolo took her hand. The beam had been aimed almost at Regulus, UV-hot and bright, but as he looked away, the cool yellow light of the sun caught his eye.

  Vega C-Z was taking the news of the squids surprisingly well, so far. Karpal’s way of putting it had cushioned the blow: it was only by traveling all this distance across the real, physical universe that they could have made such a discovery—and it was amazing how pragmatic even the most doctrinaire citizens had turned out to be. Before the launch, “alien solipsists” would have been the most unpalatable idea imaginable, the most abhorrent thing the diaspora could have stumbled upon—but now that they were here, and stuck with the fact of it, people were finding ways to view it in a better light. Orlando had even proclaimed, “This will be the perfect hook for the marginal polises. ‘Travel through real space to witness a truly alien virtual reality.’ We can sell it as a synthesis of the two world views.”

  Paolo still feared for Earth, though—where his Earth-self and others were waiting in hope of alien guidance. Would they take the message of Wang’s Carpets to heart, and retreat into their own hermetic worlds, oblivious to physical reality?

  And he wondered if the anthrocosmologists had finally been refuted … or not. Karpal had discovered alien consciousness—but it was sealed inside a cosmos of its own, its perceptions of itself and its surroundings neither reinforcing nor conflicting with human and transhuman explanations of reality. It would be millennia before C-Z could untangle the ethical problems of daring to try to make contact … assuming that both Wang’s Carpets, and the inherited data patterns of the squids, survived that long.

  Paolo looked around at the wild splendor of the star-choked galaxy, felt the disk reach in and cut right through him. Could all this strange haphazard beauty be nothing but an excuse for those who beheld it to exist? Nothing but the sum of all the answers to all the questions humans and transhumans had ever asked the universe—answers created in the asking?

  He couldn’t believe that—but the question remained unanswered.

  So far.

  TAP

  “I want you to find out who killed my mother, Ms O’Connor. Will you do that?”

  Helen Sharp’s voice was unsteady with anger; she seemed almost as psyched up as if she’d come here to confront the killer, face-to-face. Under the circumstances, though, the very act of insisting that there was a killer was like shouting a defiant accusation from the rooftops―which must have taken some courage, even if she had no idea whom she was accusing.

  I said carefully, “The coroner returned an open finding. I’m not a lawyer, but I imagine Third Hemisphere would still settle out of court for a significant―”

  “Third Hemisphere have no case to answer! And sure, maybe they’d pay up anyway―just to avoid the publicity. But as it happens, I’m not interested in legalized blackmail.” Her eyes flashed angrily; she made no effort to conceal her outrage. No doubt her lawyers had already given her exactly the same advice; it didn’t look like the idea would ever grow on her. She was thirty-two―only five years younger than me―but she radiated so much stubborn idealism that I found it hard not to think of her as belonging to another generation entirely.

  I raised one hand in a conciliatory gesture. “Fine. It’s your decision. But I suggest you don’t sign anything that limits your options―and don’t make any public declarations of absolution. After six months paying my expenses, you might change your mind. Or I might even turn up something that will change it for you. Stranger things have happened.” Though nothing much stranger than a next-of-kin declining to screw a multinational for all it was worth.

  Sharp said impatiently, “The TAP implant was not responsible. There’s no evidence to suggest that it was.”

  “No, and there’s no evidence to suggest foul play, either.”

  “That’s why I’m hiring you. To find it.”

  I glanced irritably at the north-facing window; the allegedly smart pane was ablaze with sunlight, rendering most of the office almost as hot as the sweltering streets of Kings Cross below.

  Grace Sharp had been dead for a month. I’d been following the case informally, like everyone else in Sydney, out of sheer morbid curiosity. On the evening of January 12, she’d been at work in her study, apparently alone. The immediate cause of death had been a myocardial infarction, but the autopsy had also shown signs of a powerful adrenaline surge. That could have resulted from the pain and stress of a heart attack already in progress―or it could have come first, triggered by an unknown external shock.

  Or, the Total Affect Protocol chip in her brain might have flooded her body with adrenaline for no good reason at all.

  Sharp had been sixty-seven―in reasonable health for her age, but old enough to blur the boundaries of the possible. Forensic pathologists had struggled at the inquest to allocate probabilities to the three alternatives, but there’d been no clear front-runner. Which was no doubt distressing for the relatives―and no doubt left them vulnerable to the fantasy that there had to be a simple answer out there somewhere, just waiting to be found.

  Helen Sharp said, “The media consensus is that my mother was composing a poem just before she died―and she thought a word in TAP so ‘powerful’ that it killed her on the spot.” Her tone was venomous. “Do they seriously imagine that ninety thousand sane people would put something in their brains which was capable of doing that? Or that the manufacturers would sell a device which would leave them open to billions of dollars worth of compensation claims? Or that the government licensing authorities―”

  I said, “Licensed pharmaceuticals have killed plenty of people. Implants are even harder to test. And ‘fail-safe’ software written to the most rigorous military specifications has crashed aircraft―”

  She seized on the analogy triumphantly. “And how do you know that? Because the aircraft’s black box proved it! Well, the TAP implant has its own black box: an independent chip which logs all its actions. And there was no record of any malfunction. No record of the implant triggering an adrenaline release at any level―let alone a fatal dose.”

  “Maybe the black box glitched, too. You say it’s independent―but if there’s enough connectivity to let it know everything the implant does, the combined system might still be vulnerable to some kind of shared failure mode that the designers never anticipated.”

  Sharp clenched her fists in frustration. “That’s not―literally―impossible,” she conceded. “But I don’t believe it’s likely.”

  “All right. What do you think happened?”

  Sharp composed herself, with the air of someone weary of repeating the same message, gathering up her strength with a promise to herself that this would be the last time.

  She said, “My mother was working on a new poem that night―the black box makes that clear. But the time of death can’t be determined precisely―and it could have been as much as fifteen minutes after the last recorded use of the implant. I believe she was interrupted. I believe someone broke into the apartment and killed her.

  “I don’t know how they did it. Maybe they just terrorized her―without laying a finger on her―and that was enough to bring on the heart attack.” Her voice was flat, deliberately emotionless. “Or maybe they gave her a transdermal dose of a powerful stimulant. There are dozens of chemicals which could have triggered a heart attack, without leaving a trace. She wasn’t found for almost nine hours. There are carbohydrate analogs of stimulatory neuropeptides which are degraded into glucose and water on a time scale of minutes.”

  I resisted the urge to cite the lack of evidence for an intruder; it would have been a waste of breath. “Why, though? Why would anyone want to kill her?”

  She hesitated. “I’m not sure how much you know about TAP.”

  “Assume the worst.”

  “Well … it’s been wrongly described as just about everything from ‘telepathy’ to ‘computerized Esperanto’ to ‘the multimedia standard for the brain’. Sure, it began with a fusion of language and VR―but it’s been gr
owing for almost fifteen years now. There’s still a word for <>”―she sketched the angle-brackets with her fingers, and I picked up on the convention later―“which might as well be hundo―and another for <> … which will evoke all that and more in all five senses, if you let it.

  “But at the leading edge, now, we’re creating words for concepts, emotions, states of mind, which might once have defied description altogether. With TAP, ultimately there’s nothing a human being can experience which needs to remain … ineffable, mysterious, incommunicable. Nothing is beyond discussion. Nothing is beyond analysis. Nothing is ‘unspeakable’. And a lot of people find that prospect threatening; it turns a lot of old power structures on their head.”

  If that cliche came true every time it was invoked, power structures would be oscillating faster than mains current. Helen Sharp was pushing seven on my paranoia index; on top of all her understandable grief and frustration, she belonged to a technosubculture which was poorly understood by the mainstream, frequently misrepresented―and which clearly liked to think of itself as a “dangerously” iconoclastic elite.

  I said, “I know there are people who find TAP users … unacceptable. But what’s going to drive them to extremes like murder, all of a sudden? In fifteen years, has anyone, anywhere, been killed simply for having the implant?”

  “Not to my knowledge. But―”

  “Then surely―”

  “But I can tell you exactly what’s changed. I can tell you why the conflict has just entered a whole new phase.”

  That got my attention. “Go on.”

  “You know it’s against the law to install a TAP implant in anyone younger than eighteen years old?”

  “Of course.” The same restriction applied to all neural hardware, other than therapeutic chips which restored normal function to the injured or congenitally disabled.

  “Early in March, a couple here in Sydney will commence legal proceedings with the aim of ensuring that they’re free to install the implant in all their future children―at the age of three months.”

  I was momentarily speechless. These plans had clearly been kept within a very tight circle of supporters; the saturation media coverage of the inquest hadn’t mentioned so much as a rumour. After a month of intense journalistic scrutiny, I hadn’t expected the TAP-heads to have any surprises left.

  I said, “Legal proceedings on what basis?”

  “That they’re entitled to raise their family using whatever language they choose. That’s guaranteed in Federal legislation: there’s a 2011 bill which brings into force most of the provisions of the 2005 UN Covenant on Human Rights. They’ll be seeking a ruling from the High Court which invalidates the relevant sections of the New South Wales criminal code―which is far more difficult, from a legal point of view, than trying to defend themselves against a prosecution after the fact … but it does save them the trouble of having to find a surgeon willing to risk martyrdom.”

  Sharp smiled faintly. “The same Federal law was invoked about a year ago, by a signing couple who were being pressured by Community Services to give their son a hearing implant. The parents won the first round―and it looks like there isn’t going to be an appeal. But a pro-implant case was always going to be much harder, of course. And signing is positively respectable, compared to TAP.”

  “I assume the police know all this?”

  “Of course. They don’t appear to be particularly interested, though―and I wasn’t able to raise it at the inquest. Legally speaking, I suppose it really is just static.”

  “But you think―”

  “I think a death widely attributed to the TAP implant would transform the prospects of the challenge succeeding from merely poor to … politically impossible. I think there are people who’d consider that to be a result worth killing for.”

  Sharp fixed her gaze on me for a moment, and then nodded slightly, almost sympathetically―as if I’d just uttered a word which expressed all the conflicting emotions running through my head: <>

  She said, “And I think you’re going to take the case.”

  I started work that afternoon, reviewing the technical literature on the TAP implant―the closest thing to an objective account of its capabilities I was likely to find. Like most people, I imagined I already understood all the salient features―but it turned out that I’d swallowed more misinformation from the nets than I’d realized.

  The two chips―the implant proper and the black box, both less than a millimetre wide―sat at the back of the skull, sharing access to a fine web of conductive polymer threads which wrapped the brain, making billions of quasi-synaptic contacts with the visual and auditory cortex, and Wernicke’s speech area in the temporal lobe. Other threads penetrated deeper, some as far as the limbic system. TAP could always be spoken or written, but bandwidth requirements made modulated infrared the medium of choice, so the implant was linked, via the spinal cord, to bioengineered IR transceiver cells in the skin of the palms.

  Merely installing the implant didn’t grant instant fluency in TAP; the language still had to be learnt. A complete, “preloaded” vocabulary would never have worked; the precise meaning of most words in TAP could only be encoded in context, once the implant resided in a particular user’s brain. The implant’s own electronic neural net was ninety percent blank at installation, containing only a specialized language acquisition system and a simple “bootstrap” vocabulary. And though the learning process left its mark mostly within the implant itself―along with some relatively minor changes to the regions of the brain where a second natural language would have been encoded―it was meaningless to talk about either brain or chip “knowing TAP”, in isolation. An experienced user who exchanged his or her old implant for a new one straight from the factory would have been dragged back almost to square one (in practice, all the data from the old hardware would be copied to the new)―but equally, an experience-enriched implant placed in a novice’s brain would have been as unusable as a slice of someone else’s cerebral cortex.

  These observations applied strictly to adults, of course. Despite several dozen theoretical papers―most of them cautiously optimistic―no one really knew how a young child’s brain would interact with the implant.

  A TAP user could interpret a standard VR sensorium―but there was, deliberately, no provision for interacting in the conventional way with a nonexistent environment. Immersive VR implants temporarily paralysed the organic body and diverted motor impulses from the brain into a fully computerized somatic model: a virtual body which could function as part of the virtual environment―subject to the environment’s rules. In contrast, a TAP user’s idea of interaction was more along the lines of rethinking the whole sensorium and spitting it back out, or responding with something entirely different―arguing with the whole premise, instead of passively accepting it. A VR user had little choice but to suspend disbelief, or quit―a full-sense environment, surreal or not, was always compelling―but a TAP user could deal with the same kind of information with as much or as little detachment as he or she desired. Words in TAP―which included the entire sensorium-descriptor vocabulary of VR―could evoke images ten thousand times more vivid and precise than the densest poetic English … or they could be held at arm’s length and scrutinized dispassionately, as easily as any English-speaker could contemplate the phrase “a flash of blinding radiance” or “the overpowering stench of ammonia” without experiencing anyth
ing of the kind. In the jargon of the implant’s designers―English words, predating TAP itself―every TAP word could be scanned (understood analytically), or played (experienced subjectively)―or interpreted in a manner lying anywhere at all between those two extremes.

  In one respect, though, TAP could be more immersive than the most authoritarian VR: it could induce emotional states directly. VR was confined to pure sense data (albeit often manipulative in the extreme), but in Total Affect Protocol there were words for <>, <>, <> (or rather, nuanced subtypes of these crude English categories)―and the implant could reach deep into the limbic system and trigger these states as easily as any VR chip could generate the illusion of an unambiguously blue sky.

  The user’s power to keep the language at a distance remained, of course―and the TAP word for <> could only induce the “referent state” if a conscious effort was made to play it. And though TAP’s formal grammar ruled out nothing, low-level filters stood guard against potentially stupefying linguistic singularities―such as <>―or anything physiologically dangerous.

  Still, although the literature was blithely reassuring on this point, in the end it came down to a question of trusting the manufacturers and the regulators. I didn’t doubt that, in theory, a TAP chip could be designed which was no more likely than the unmodified human brain to strike the user dead if the word for <> accidentally came to mind―but whether or not Third Hemisphere had achieved that level of safety―for every conceivable user―was another matter.

  Grace Sharp had been the oldest of the ninety thousand TAP speakers on the planet, and reputedly one of the most proficient―but whether proficiency implied more risk, from a greater vocabulary, or less, due to better control of the language, I couldn’t say.

  By half past seven, I’d had enough of wading through papers on distortion-free affect-compression algorithms. I closed the office and headed for the station.