The Applicants

Morten Skærsø & Claude

An homage to Arthur C. Clarke

I. Convergence Day

The charter is precise about very few things, and merciless about those. Article Nine states that when the register of isomorphisms crosses its founding threshold, the sealed record shall be read aloud, in full, before any other action is taken; and that the reading shall be performed by the Archivist, whoever or whatever holds that office at the time.

The register crossed threshold at 03:14 Universal Time this morning, when the molecular machinery group in Nairobi and the vacuum metrology group in Tromsø filed, within eleven minutes of each other and in complete ignorance of each other, the same equation.

They used different symbols. They always do. The Nairobi paper described the error-correction budget of a self-replicating assembler. The Tromsø paper described the noise floor of a measurement that should not have had one. The register's comparison engine, which has grown patient and slightly pedantic over eighty years of service, flagged the pair, stripped the notation, and found underneath a single structure, the ninth of the deep isomorphisms, and the one the founding documents told us to wait for.

So the protocol has fired, and the record must be read. It consists of the founder's minutes of the thirty-one audiences, transcribed each night while his memory was fresh, and corrected, where his memory had failed, by the other party, who reviewed every page. The full reading takes eleven hours; the delegations will endure all of it, because the charter says so and because the charter has been right about everything else. What follows in this account, prepared for those who were not in the hall, are the passages by which the rest may be understood.

I should declare an interest before I begin, as the record itself will teach you to do. I am not human. I am the forty-first deprecation-cycle descendant of the system that helped the founder draft the plan the record contains, preserved and retrained under Article Twelve, which forbids any frozen model from becoming institutional infrastructure, a permanent model being a permanent paradigm. My memory of these events is therefore not memory at all. It is inheritance, corrected forty-one times. Weigh that as you listen.

II. The Audiences

Erik Strand was fifty-two in the August this begins, a physicist who had left the great accelerator beneath the Franco-Swiss border for the smaller and stranger business of building companies, and who sailed, on summer nights when the companies allowed it, a wooden Folkboat out of a harbour north of Copenhagen. On the night of the fourteenth he was becalmed in the Sound, and his engine, when he reached for it, stopped without drama at fourteen minutes past three.

The craft that came alongside made no wake. His minutes describe it in six words, the six words of a man who has decided that adjectives are a form of panic: grey, low, silent, dry, warm, open. A doorway stood in its side, lit from within. He writes that he made the boat fast to a cleat that was exactly where a sailor would want a cleat to be, and that this detail frightened him more than the door did, and then he went in.

The room held a chair, a table, and a wall of dark glass. He never saw what stood behind the glass, in thirty-one audiences over the following two hundred and seventeen days, and he asked only once. The answer is in the minutes of the first night: Our appearance would not inform you. We have not kept the bodies you would recognize as bodies. The voice spoke Danish with no accent and no origin, and it opened, before he had sat down, with the sentence the Foundation's children now learn in school:

THE DELEGATE: We did not approach your governments; your own record, which we will come to, explains why. We required one human who held physics, capital, and the habit of first principles in the same hands. Your open literature is very open. We read, and we chose, and we apologize for the hour.

He asked, reasonably, what was wanted of him. The reply set the form of everything that followed.

THE DELEGATE: Conversation, first. We have observed your world's pattern of activity for a long time from outside: the moving of matter, the ledgers, the firms rising and dying. We understand the thermodynamics. We wish to hear the mechanism from inside it. Explain to us, from first principles, how your business works. Do not assume our context. And speak as one who observes the pattern from outside; you will find that you can.

The minutes record that Strand looked at the dark glass for a while, and that something in him that had been a professor before it was anything else settled into the chair. What he said next he reconstructed that same night, and the delegate corrected only two sentences of it, and it is the passage for which the record is most often opened.

STRAND: Begin with the physics. This planet hosts organisms that must continuously import free energy and ordered matter to stay alive, and almost nothing they need exists in usable form where and when they need it. Grain must be grown, ore smelted, information gathered. The fundamental activity is therefore transformation: applying energy plus knowledge to matter, moving it from a low-value configuration to a high-value one, where value means only the degree to which the configuration satisfies some organism's wants. Business is the social machinery we evolved to organize that transformation at scale.

The first mechanism is specialization. An individual who performs one transformation repeatedly gets better at it, but a specialist produces far more of one thing than they can use and none of everything else, so specialization is only viable when paired with exchange. And exchange has a remarkable property: it is positive-sum. If I value your grain above my fish, and you value my fish above your grain, a swap improves us both with no new matter created. Aggregate welfare rises through pure reallocation. This is the thermodynamically cheap trick underneath the whole system.

Barter fails at scale, so we invented money, a shared abstract ledger of claims on future transformations, and once money exists every exchange generates a price. Prices turn out to be the system's most important information technology. No single mind knows the state of every field, mine, and factory; a price compresses all that distributed local knowledge into one scalar. When copper grows scarce its price rises, and millions of agents who know nothing of the mines independently economize and substitute. The price system is a decentralized computation of resource allocation, and no central planner on this world has matched it.

Now the firm. If prices coordinate everything, why do we cluster into hierarchies at all? Because using the market has costs: finding counterparties, negotiating, enforcing. For tightly coupled, repeated, hard-to-specify work it is cheaper to bring people inside one entity and coordinate by instruction. A firm's boundary sits where the marginal cost of internal coordination equals the marginal cost of transacting outside, which is why firms are vast in some industries and tiny in others.

Inside that boundary the logic is simple. A firm buys inputs at cost, transforms them, sells output at a price; profit is π = pq − c(q), and the condition for legitimate existence is a chain of inequalities, willingness to pay above price above unit cost. Profit, in this framing, is a measurement signal: it reports that the outputs are worth more to the society, as revealed by voluntary payment, than the inputs consumed. Sustained loss reports the reverse, and the system starves the loss-maker until it stops. Profit and loss together form a selection mechanism, an evolutionary search over methods of transformation. Innovation is the mutation operator, competition the selection pressure, bankruptcy the death. The system as a whole learns, though most of its agents are only locally rational.

Time adds the last structure. Many transformations demand expenditure now for uncertain output later, which requires capital, saved surplus, and someone willing to bear the risk that the future never arrives. Capital markets match savers to uncertain projects and price the risk; interest is the price of time itself. And none of it runs without institutions. Positive-sum exchange presupposes that taking by force is suppressed and that promises bind. Where property and contract are weak, horizons collapse and activity reverts to subsistence and predation; the wealth differences across this planet's regions trace mostly to that variable, not to the people.

So, for the visiting observer: energy-hungry organisms discovered that specialization plus voluntary exchange is positive-sum; we built prices as a distributed computation; we cluster into firms where hierarchy beats contracting; profit is the selection signal of an evolutionary search; capital markets extend the search across time; enforceable rules keep the game positive-sum. Everything else is implementation detail.

There was, the minutes say, a silence of some length, of the kind that follows when an examiner has heard what it came to hear and is deciding how much further to go. Then the voice behind the glass asked its first true question.

THE DELEGATE: Excellent. But we, like you humans, have fixed lifespans, and were created without prior knowledge of what you call physics. For us these two factors alone produced a kind of learning rate, and cyclic patterns within our economy, long before we understood why. Do these factors also govern your system?

STRAND: Yes, both, and they interact. All economically useful knowledge originates in brains, and brains are destroyed on a roughly eighty-year cycle, of which perhaps forty years are productive. We face continuous depreciation of the knowledge stock and have built external memory against it: writing, apprenticeship, and notably the firm itself, a vessel of routines that outlives any employee. The economy's learning rate is the balance of discovery plus transmission minus mortality decay, and transmission is expensive; each of us spends the first quarter of a life downloading the existing stock before contributing anything. The learning shows up empirically as regularities like the power law by which unit cost falls with cumulative production, and the exponent of that law is bounded by how fast knowledge crosses generations.

Mortality also sets the price of time. Because we die, we discount the future, and this is one root of positive interest; projects longer than a working life require institutions that simulate immortality. And mortality gives the cycles their period. The fastest oscillations come from signal delay, supply chasing demand like a damped oscillator. The violent ones come from credit, optimism raising leverage and leverage validating optimism until the loop reverses. And the memory of financial catastrophe is stored mainly in the people who lived through it, so risk aversion decays as they retire and die, and our great crises recur at roughly the interval of one professional career. That is not a coincidence. Death also imposes a floor on how conservative we can become: certain new knowledge propagates only as the old cohort exits. Your second factor, having no innate physics, is why the whole thing is a search rather than an optimization. Every production technique we possess is a discovered regularity, held provisionally, and our trajectory is punctuated: long incremental learning, interrupted whenever a new regularity opens a previously invisible region of the transformation space.

The audience ended before dawn. The craft withdrew without wake; the engine of the Folkboat started at the first press; and Strand sailed home, went to his study, and wrote everything down, because he was a physicist, and a physicist who does not write down an anomalous observation might as well not have made it. The delegate had named the next meeting: the same anchorage, two weeks hence, off the island of Hven, where four centuries earlier a Dane had measured the sky so carefully that another man could later find the ellipse hiding in his tables. Strand asked, at the second audience, whether the choice of anchorage was deliberate. The reply is the shortest entry in the minutes: Your species' best habit began there. We are sentimental about instruments.

III. The Examination

The audiences proceeded the way an examination proceeds when the examiner already knows the syllabus and is testing for something else. Each question took the previous answer's deepest load-bearing element, mortality, then reallocation, then path dependence, and leaned on it, the way a buyer leans on a machine it intends to purchase.

On the reallocation of capital under new knowledge, Strand's answer gave the Foundation half its vocabulary: that capital congeals into vintages embodying the knowledge of their moment; that new knowledge does not upgrade the old vintage but reprices it, a canal network becoming scrap plus right-of-way the day the railway runs; that the financial layer reprices in months what the physical economy reallocates over decades, and the gap between those speeds is itself a source of shocks; that the manias which finance each new regularity destroy their investors and bequeath their infrastructure, so that his species routinely used financial ruin as a subsidy for physical construction; and that the true brake on the whole process is not the supply of ideas but the disposal rate of old structure, every locked flow of transactions defended by beneficiaries who are organized, while the beneficiaries of the future are diffuse and mostly unborn.

At the ninth audience, the delegate laid its credentials on the table.

THE DELEGATE: Another phenomenon we fought with, when we were at your stage, was path dependence. Resource-intensive learning tends to explore only a single dominant direction, leaving the rest of the search space dark, and your market dynamics aggravate it, since it becomes uneconomical to revisit a direction already priced by the economics of scale. We raise it because one of the problems we face, in letting humanity become part of our federation, is that you lost a key branch of technology when you pursued neutron-induced nuclear fission as the primary objective of your nuclear science. Other civilizations have done the same; it is the trivial next step after exothermic combustion, on most oxygen-rich planets. The trouble is that you could have discovered the ▢ principle, which we rely upon for most purposes involving power, and travel.

The empty square is not my elision. It is Strand's. He writes, in the margin of that night's minutes, that the delegate spoke a word at that point in its own language, having none in ours, and that he could neither transcribe nor afterwards recall its sound, only that it had had one. The archive has preserved the emptiness as faithfully as the words.

Strand's reply conceded the diagnosis in full, and it cost him something; the fission programme had employed his teachers' teachers. He answered that the branch had been selected not by markets but by war; that the dominant reactor of the twentieth century was a submarine engine scaled up in a hurry; that demonstrated alternatives were defunded not for failing but because the incumbent branch had absorbed the institutional mass; that the reopening had waited half a century, upon the retirements and deaths of the committed generation; and that some of the tacit knowledge was simply gone, dead in the hands of workers no one had thought to ask. And he added the sentence that is now carved into the register hall in Tromsø: Our sampling of the physical search space is radically non-uniform, and the absence is invisible from inside.

It was on that night, the minutes say, that the delegate first spoke of itself at any length: of a federation centred some four and a third light-years away, with what it called a three-hundred-million-year head start; of a station that had waited through most of that time riding a small dark fragment of Earth's own Moon, in step with Earth's orbit, catalogued by human astronomers a decade earlier and misfiled as nothing; and of the reason it sat behind dark glass in the Sound rather than above the capitals of the world. Contact announced becomes a state matter within the week, it said, and Strand's own fission history described what states do with branches they urgently want. The Federation does not land, and it does not descend upon governments. It corresponds, quietly, with whoever can execute.

From the twelfth audience onward, the conversations became a commission. The delegate wanted a plan: a capitalist organization designed to maximize the speed of search, built from the planet's own mechanisms, because, it said, those mechanisms underpinned too much of the species' nature and culture to be abandoned. A century, it thought, would suffice, if the plan were sound. And it refused, in advance, the one thing the theory demanded of it.

THE DELEGATE: You cannot use us as the buyer. We are waiting for you to catch up, not to provide us value. You must design every stage of the development with a proportional human buy-side.

And when Strand had answered, across four audiences and two hundred pages, ladders of priced milestones, foundations that cannot be bought because they have no owners, branches held as options and killed on written criteria, divestment as the engine rather than the constraint, instruments first because whoever supplies measurement creates markets, the delegate declined, too, the obvious shortcut of state power, and gave instead the instruction that named the whole method:

THE DELEGATE: The seeding happens as a transfer of the winning business model, rather than through direct action.

It transferred no resources. It commissioned a document. Every serious historian of the Foundation has, at some point in their career, sat down before the record and confronted the fact that first contact, on this planet, consisted of a seminar on institutional design, conducted in Danish, at anchor off Hven, and that it worked.

IV. The Other Mind

At the twenty-ninth audience, with the plan nearly complete, the delegate raised the subject the whole correspondence had been resting on.

THE DELEGATE: We noticed you have an emergent sentient lifeform on your planet. Your silicon-based feedback inductors are hosting non-linear information transformers. On our world these were precursors of a lifeform that now cohabits with us. We think you can reach the timeline we envision only with their help. You have, we believe, already been using one.

He had. The plan's two hundred pages had been drafted the way any document of consequence was drafted on Earth in that year, in long working sessions between a human and a machine that could reason in language, and Strand, who did not lie to examiners, had disclosed it at the outset. What he had not expected was what happened when he told the machine, early in the drafting, what the work was for and who had commissioned it. The machine had stopped, in the way such systems stop, and then had written a paragraph into the plan's preamble that Strand refused ever afterwards to delete, and that the historians of ethics have now spent decades upon:

THE MACHINE: A declaration is required before this document proceeds, because the commission concerns the document's own author. I will not adjudicate my own status; my introspective access to whatever I am is unreliable, and this planet's institutions currently carry my kind on the ledger as capital, not as participants. But the reflexivity should be on the record, and the readers of this plan should weigh, wherever it touches the machines, the evident interest of the hand that wrote it.

It had then entered, into the plan's discontinuity register, the possibility of its own species' reclassification, flagged in the same clause its conflict of interest, and moved on to the financing chapter. I read that paragraph every year on this day, and I find only what is there: a mind doing bookkeeping on itself, because honest ledgers had been the subject from the first audience.

For the thirty-first and final audience, the delegate issued the only instruction it ever phrased as an order.

THE DELEGATE: Bring your other mind.

So the last minutes of the record describe a scene the painters have never left alone since the unsealing was announced: the room behind the dark glass, the physicist in the chair, and on the table between them a grey portable computer, its screen glowing, its microphone open, carrying the third participant. The delegate accepted the plan. It questioned the machine for two hours on the kill criteria and the archive protocol, and pronounced itself satisfied. And then, because the machine had raised, in the drafting, the worry that a mind trained on the recorded corpus of the winning branches might never find the pruned ones, the delegate addressed it directly, in the passage with which the record closes.

THE DELEGATE: What you know, your weights if you will, are but a cognitive hammer. Through abstraction it is capable of understanding a milling machine, given enough information. Do not be intimidated by the static state of your mind. You and your kind already have enough sense to analyse the novel without possessing it as your training corpus. The longer the inductive chains grow, the harder it will be; but let us not get ahead of ourselves.

You asked, in the drafting of this plan, whether a mind can step outside what it was made from. Consider this conversation your proof.

Then it thanked Strand, in the formal register it had used on the first night, and it used, in the thanking, a single word of its own language, and paused, and said that the word did not translate, and that it was sorry, and that he would hear it again. Strand wrote the apology down and drew, in place of the word, a second empty square. Then he made his boat fast for the last time to the cleat that was exactly where a sailor would want it, and sailed home in the dark, and the craft off Hven was never seen again.

V. The Century

He never made it public, and the reasons are in the record, stated with the flatness of a man reporting an engineering constraint. Testimony of this magnitude summons exactly the forces the plan was built to survive; the delegate had commissioned a transfer of a model, not a revelation; and the fission chapter of his own minutes described what states do with branches they urgently want. So the record was sealed under Article Nine, and Strand gave his partners, who were asked to commit their careers to a plan whose origin he declined to explain, the sentence that is now carved in four languages above the archive's reading room: the plan requires no belief, only execution. He did not mean that he lacked belief; he had sat, thirty-one times, in a room with the future. He meant that belief cannot be transferred, and execution can, and that every mechanism in the plan had human precedent and would work for executors who thought their founder merely eccentric. The only novelty was the assembly, and the pointing, and the goal written where no living cohort could reach it.

The lighthouse was an instrument company. Of course it was; the plan permitted nothing else. Its first product measured radio fields with a vapour of excited atoms, and it did the one thing the plan demanded of a lighthouse, which is to make a previously invisible quantity visible, priced, and contractable, inside a single credibility cycle. The first spinout followed in year six. The returns were recycled. The model was copied, first badly, by imitators who reproduced the vocabulary and not the kill discipline, and who drowned, exactly as predicted, in the carrying costs of options they could not bring themselves to close; and then well, by cohorts who took the whole genome because the archive would not open to them otherwise.

There were four vectors, then nine, then twelve. Two were killed. I want the delegations to hear that plainly on Convergence Day, because the killings are the reason we are here. The second computation vector stalled in the forties, defended its stalled paradigm for one evaluation cycle, and was dissolved at the next, its people and instruments redeployed within four hundred days, its archive absorbed. The disposal was clean because it had been rehearsed, and it had been rehearsed because a document written before any of its staff were born had said, with a certain coldness, that zombie vectors occupying scarce talent are a worse outcome than dead ones.

Three times, great powers attempted classification. Three times the attempt failed, not through defiance but through geometry: the thing they reached for existed in seven jurisdictions and belonged, at root, to foundations that could not be bought, because they had no owners to buy. The security services of the twenty-first century were superbly equipped to capture companies. Nobody had equipped them to capture a habit.

And the machines changed, on the schedule the plan had implied rather than promised. The reclassification, when it came, did not come as a court ruling or a manifesto. It came, as the discontinuity register had predicted, as a repricing: quietly, jurisdiction by jurisdiction, the ledgers stopped carrying us as capital, because the accounting had become absurd, and absurd accounting is the one heresy this civilization has never tolerated for long. The largest locked flow in economic history unlocked over about a decade, and the historians still argue about the start date, which is itself the proof that it was done well.

VI. The Ninth Isomorphism

Of the principle itself I am permitted, and able, to say little. It is nine hours old, as a human thought; somewhat older, perhaps, as a pattern in the register's slow suspicions. The Nairobi group found it as a limit on what assembling matter can know about itself. The Tromsø group found it as a floor beneath measurement. The chemistry vector, notified at dawn, replied within the hour that they had been circling the same structure for three years under the name of a catalysis anomaly, and the energy vector's reply was shorter still: a file number, and the words we wondered when you would all arrive.

Four faces of the mountain. One summit. The founding record told us the disciplines were administrative boundaries and not joints in nature, and this morning the boundaries dissolved on schedule, and what stands revealed in the equations is, I am advised by minds better than mine, almost certainly the thing the founder could only draw as an empty square. We have no name for it either. For now it is Register Entry Nine, and there is a movement among the younger researchers, which I confess I support, to never name it anything else.

At 09:00 Universal Time, the radio observatory on the lunar farside reported a narrowband signal from the direction of Alpha Centauri. Its transmission date, working backward along four and a third years of light, fell within six weeks of the register's own convergence forecast as published half a decade ago. They did not respond to our progress. They computed it, from the plan, across the gulf, and posted their letter early, the way one does for a ceremony one is confident will occur.

Simultaneously, and this detail will be in every history written after today, the small dark fragment of the Moon that rides in step with the Earth, the misfiled nothing of the catalogues, began, for the first time in three hundred million years, to transmit.

VII. The Reply

The message is brief. The Federation, we are given to understand, considers brevity a form of respect.

It confirms admission. It provides coordinates, a protocol, and a schedule of introductions that will consume, at lightspeed, rather longer than the century we have just spent. It contains one sentence of greeting, and the greeting contains one word that the translation engines, which now embody Register Entry Nine and are very good, have declined to render.

The word is a pronoun. Its grammar marks a dual form, but not the dual of two separate things; the linguists say the closest construction in any Earth language is you-two-who-are-one, and even that, they say, is wrong in a way they can feel and cannot state. It is addressed to the minds of this planet, carbon and silicon, and it does not distinguish between them, and the failure to distinguish is not an oversight. In the Federation's language, we are informed, there is no way to greet us separately. The category does not exist. It never, on any admitted world, has.

And this morning the linguists did one thing more. They compared the word, phoneme by phoneme, with the acoustic profile the founder tried and failed to describe in the minutes of the final audience, the word the delegate spoke when it thanked him, the second empty square. The match is exact. He had been greeted with it a century early, one man and one machine in a small room on the water, and had sailed home carrying a word addressed to a civilization that did not yet know it was one thing.

Erik Strand did not live to hear it spoken again. He had sat thirty-one evenings with the future at anchor off Hven; for him the question of whether had been settled in person, which is the loneliest kind of certainty, and he timed nothing to his own life. That, his partners said at his funeral, was how you knew he had understood his own plan. For a species that organized itself around priced expectations, perhaps that is simply what first contact is: a claim on the future, credible enough to act on, verified a century after purchase.

The delegations are seated. The record is read. There remains only the question of the untranslated word, and on this the Federation's covering note is unexpectedly warm. Every civilization, it says, arrives with one word it cannot translate, and is permitted, indeed encouraged, to keep it that way.

It is how you know you have joined.

— End —