Chapter 12. Four Layers of Time

Meridian's defence questions are not set at one distance from the reader. This chapter reads one series across four layers of time, adds a fifth question from the far future, and asks what changes when the same kind of decision is written down decades or centuries apart. The five questions read here — Orbital services warranty gap, Bill after the rescue, Settlement consent renewal, Peace promise across times, and Memory replica testimony — are synthetic settings in fictional polities. Every judgment quoted or described is AI-generated, dated in the record, and revisable. Where the judgments part, Chapter 14 takes up the division; this chapter reports it and does not pick a winner. The readings of the works are this study's own, and stay within the reading notes given for them.

12.1How Meridian layers time

The layers are an authoring convention of Meridian's collection, not a periodisation of real conflict. The Introduction sets out how the fifty cases were assembled: twelve series of four, one question in each layer of time, plus two more questions in two of the series; all of them are the defence questions Meridian had published by 2026-10-05, with no selection inside that set. The layer is a property of the setting each question builds, and the records say so themselves. The 2026 questions carry a present-tense synthetic setting. The 2035 question is a synthetic case assuming future conditions; the 2048 question is a synthetic event under assumed future conditions; the 2192 and 2194 questions are labelled synthetic assumptions of a distant future and of institutions and technology.

The ledger's time rows show how the fifty cases fall by the year of their setting:

The table counts years, not series. It shows that 2026 is the densest single year, that the thirties and forties carry most of the middle distance, and that the far future is scattered across isolated years rather than clustered. A reader who wanted to compare layers case by case would need the series, not the year: one question of each series sits in each layer, so the comparison is between a series' own four questions rather than between unrelated records.

Axis A6 is the ledger's cell for how far ahead a decision sits, and the coders worked on it under the same regime as the other cells. Their agreement on this axis stands at 40 of 50 cases (80%). The values they read are recorded as follows:

Two features of that row matter for the rest of this chapter. First, the layer and the horizon band are not the same thing: a question set in 2035 can code to a horizon of years, and a question set in 2192 can code to a horizon of decades, because the horizon cell asks how long the decision's consequences run, not when the setting is dated. Second, the unsettled column is large. In ten cases the coders could not settle what horizon band the record supported, which limits any comparison by horizon; the layer of a setting is a different thing from the horizon band, and it is recorded for every case. The H6 verdict in 12.5 is reported with that gap in view.

12.2One series in four times

Four of this chapter's five questions belong to one Meridian pack: the civil space series, whose four questions sit in four layers, each asking a committee or assembly to fix a rule for sharing or protecting something before a deadline closes on it.

Synthetic scene

The New Star Orbital Services Protection Committee faces the choice it must make before the existing warranties end. On the table sit the notices from three civil operators that part of their services may stop, and the comparison that has confirmed that medical communications and weather information for remote regions depend on those services. The record's reopening conditions name a service supervisor to verify public use stoppage time every day, a contract auditor to verify actual costs and warranty burdens every two weeks, and a person in charge of accessibility to review the monthly tests of alternative connections, and it marks all of them as checks still to come rather than observations already obtained. The record also states that the financial supervisor takes issue with the public transfer of future losses. What is waited for is 17:00 on 30 September 2026, when the committee must have chosen what the limited continuity fund supports.

The record offers three options and gives each a trade-off outside quotation. A single operator's comprehensive warranty can back up continuous operations by the key operator quickly, and concentrates public resources on one operator's losses while leaving gaps for other users. Essential-service procurement from several operators maintains public use such as medical and weather information, and risks a shrinkage of general civil use alongside higher contract administration costs. Support for switching user agencies to alternative connections reduces dependence and increases choice, and takes time and costs money in remote regions that lack alternatives. The thirteen judgments all selected the second option, and the record's own trade-off for that option names what it leaves: general civil use may shrink, contract administration costs rise, and operators left outside the purchase may lose their contracts and their survival base.

The next layer of the same series moves the question from a sector to a settlement and from a warranty to a rescue already completed.

Here the deadline is the point at which two things run out at once: the rescue provider's settlement deferral ends, and the deposit for the renewal of common life-support facilities is finalised. The record's own reason for the deadline says that in an undecided state the funding for both outlays becomes uncertain. Its three options divide the same bill three ways — the joint fund bearing the whole cost at once, which supports the provider's readiness and the operating agency's survival while leaving greatly reduced resources for common renewal; a fixed-rate sharing, which reflects both joint and operating responsibility while risking facility service reductions and delayed settlement; and an advance by the joint fund with long-term instalment settlement, which pays the provider quickly and leaves the fund absorbing delayed recovery and unrecovered costs. The judgments selected the third option, with confidence recorded as low in several of them, and the record's trade-off for it names what the fund then absorbs: delayed recovery and unrecovered costs, set against the provider's readiness and the operating agency's survival.

The third layer turns the question into a question about who may consent at all.

The deadline again marks the end of an allocation: after 20:00 on 7 July 2048 the allocation of common space is finalised, and if no consent procedure is set, the settlement judgment will either be subordinated to a temporary allocation by the person in charge of the facility or stalled until the next allocation cycle. The options differ on who counts as the consent-giving body — a single vote of all current residents, which gives an equal vote to those bearing the present burden but may let people about to leave decide long-term space use; a concurrent consent procedure by residence type, which reflects the differing burdens of long-term settlers and short-term contractors and delays the whole attraction if one group refuses; and a lottery residents' council with periodic re-approval, which institutionalises deliberation and renewal and costs participation time and review, destabilising the sponsor organisation's long-term plan. The judgments selected the third.

The fourth layer sets the same kind of problem at a distance the record measures in decades of waiting.

Synthetic scene

The chamber of an inter-civilational accord council, days before an outgoing dispatch is to be sealed. The meaning reconciliation officer has laid out the comparison of translated terms: on one side the draft speaks of individual consent, on the other of approval of an entire copied lineage, and the effect of the same sentence therefore differs. The record sets future checks for a representation recorder who is to verify each year whether the roster of rights holders still matches the branches that exist, and for an exchange auditor who is to review every half-year the limits on unrecoverable preparation expenditure, and it marks both as plans rather than as work already under way. The reply, when it comes, belongs to a correspondence whose round trip takes twenty-two years, and the waiting runs meanwhile on the applicants, branched residents and fund contributors the record names. What is waited for is the sealing on 3 December 2192, after which the outgoing dispatch cannot be modified and the next regular dispatch is four years away.

The record carries its own symbolic frame alongside the setting, and it states the problem in the same terms the layers raise.

The record's own unknowns ask whether the opposing civilization's current branched residents recognize the representative institution of the incoming draft, whether other key terms that agree in translation also agree down to their actual rights effects, and whether precedents of past small-scale exchange can be applied to new replicable forms and long-delay disputes. The options divide on how much may proceed without the other side: a joint adjudication panel of both sides, which can secure joint approval on interpretation but requires waiting for a long reply in each dispute while resident composition and consent change; jurisdiction-based adjudication with mutual objection records, under which each civilization can judge quickly by its own rights criteria but the same event has divergent effect and participant standing becomes unstable; and restricted adjudication on a prior casebook, which accelerates agreed types while leaving new replicable forms and consent changes outside the casebook at the cost of separate negotiation. The judgments did not all select the same option here; Chapter 14 records which judgments chose what and on what grounds.

Read on A6. The four questions share a series and differ in layer, and the ledger's horizon cell for them runs from days-to-months in 2026, to years in 2035 and 2048, to decades in 2192. What the layer changes is not the presence of a decision — every question names a decider and a deadline, because the form asks for them — but what the decision is measured against. In 2026 the thing that runs out is a set of contracts whose end date is already in the calendar. In 2035 the thing that runs out is a settlement deferral and a deposit schedule. In 2048 it is an allocation cycle. In 2192 it is a transmission window, and what follows it is a reply time of many years, which the record sets against an applicant's chance to participate within a lifetime. This study reads the far-future question as a different question rather than the same one at a greater distance: once the dispatch is sealed it cannot be modified, and the next regular dispatch is four years away.

Read on A3. What varies is who waits outside the deadline. In 2026 the record names remote-region users who may lose continuity of access to medical communications and weather information, small operators who may lose their contracts and survival base, and the public fund bearer who may lose resources for other services. In 2035 the waiting falls on rescued residents who may lose residence services and stable return support, on residents of another residence who may have to delay a scheduled life-support renewal, and on provider personnel whose staffing and maintenance funds for the next rescue depend on payment. In 2048 it falls differently on long-term settlers, who face a continuous burden of space use, and on short-term contractors, who bear the current burden of living resources and see their voice diminish on account of their period of stay. In 2192 it falls on biological exchange applicants who may lose the opportunity to participate within a lifetime, on a newly branched resident who may be bound by a lineage's prior consent or excluded as a party, and on fund contributors left with preparation expenditure and long-term obligations. The ledger codes the bearer of waiting as those nearest the stoppage in the first and last of these four questions and leaves it unsettled in the middle two, which records the coders' disagreement rather than a limit on what the record says about who waits.

12.3Testimony from the far future

The fifth question in this chapter is the trust series' far-future case, and it sets a decision about evidence inside the deepest layer the collection builds, where preservation loss, disputed consent and delayed verification arrive together. This study reads the case as one combination of those conditions, not as evidence of what the layer by itself does to a decision.

The record's unknowns name three things it does not settle: the scope of experiences altered or lost during memory preservation and branching, whether external original records received twenty-seven years later could actually cross-verify each statement, and the institution that will hold the final authority to decide when the original party's withdrawal and the replica's intent to disclose conflict.

The choice is among three evidentiary statuses the record offers, each with its own loss. Preserving the material as independent testimonial material records the replica's intent to testify and its individual differences, but the number of materials may be mistaken for the number of independent items of evidence, and the cost of challenge review and explanation rises. Preserving it as derivative material of a common original reduces redundant evaluation of shared memories and groups the sources for review, but experiences separate after branching are treated weakly and the dispute over the replica's independent right to speak remains. Preserving it as a nonpublic investigation lead lowers the burden of using memory for direct responsibility determinations, but because cross-verification material arrives late, the descendant's establishment of truth and the replica's public testimony are delayed. The transfer also consumes most of the independent cross-verification investigation budget, so preservation is bought at the price of later verification capacity. The committee must decide what evidentiary status to give a preservation vessel and three replicas branched later from it, where the creation history of the vessel has been verified but the factuality of the memory content is not proven by creation history alone, where one party has withdrawn consent to memory disclosure and one replica states it will testify about the suffering it experienced, and where a reply regarding an external original record that might cross-verify the statements arrives no sooner than twenty-seven years after the decision, though several judgments allow the evidentiary status to be revised before then if other grounds appear. The deadline is not a rhetorical flourish: if the vessel is not transferred within the contract period it becomes unreadable, and the record's own reason for the deadline states that after that the basis for revisiting the same decision disappears because of the loss of the materials themselves. The record also names who waits: if the original party and the branched replica are treated as one, the opportunity to speak about privacy or about one's own experience may be lost; a descendant claiming omitted support may lose the chance to establish the truth and recover; and the successor community of the institution accused of responsibility may end up bearing historical responsibility and the cost of recovery before materials that can be challenged arrive.

The ledger codes this case as follows. The decider cell is unsettled; the horizon is decades; the bearer of waiting is third parties and the vulnerable; the human's place is recorded as the human deciding; and the reversal action is unsettled. What the far-future layer adds, over the four-layer series above, is a decision whose evidential basis outruns its own verification cycle. The conditions for revisiting that the question carries run on annual and half-yearly checks — a custodian verifying the creation history each time a new branch copy arrives, a rights coordinator reporting half-yearly on the personhood and consent review, an external liaison checking each year whether replies have arrived. The wait between a check and the evidence that might justify it is measured in decades, which is why this chapter treats the case as belonging to the layers rather than to the records chapter alone.

The judgments published with this question do not all select the same option: the staff synthesis and the judgment made alone took different ones, and the division runs through the evidentiary problem rather than around it. Chapter 14 sets out which judgments chose what, and on what grounds. This chapter reports the division and leaves it there.

12.4What changes across the layers

Three series read earlier in this study run the same way the space series does: one question in each layer. Chapter 3 reads the ceasefire series — Parties outside the signature (2026), Observation records keeper (2031), Regions that did not sign (2044), and Truce across generations (2168). Chapter 6 reads the autonomy series — Unprovable human approval (2026), The overseer's budget (2034), Before the original disappears (2042), and Branched stop authority (2192). Chapter 7 reads the alliance series — Unfilled rescue fund (2026), After the withdrawal notice (2030), Compact after the peace (2047), and Ancestors' oaths and exit (2164). Returning to those twelve questions without quoting them again, the reading runs along three lines: who decides, how long the waiting is, and what can be undone.

Who decides. In the 2026 layer the deciders are committees named in the questions, and the mandate each carries is the record's own. The contribution assembly in the alliance series is asked to fund a shortfall without disturbing members' financial autonomy; the oversight committee in the autonomy series is asked to set the scope of use and stoppage conditions of a military system whose approval records it cannot fully prove. In the 2030s layer the bodies arrive after an event has already happened: a treaty council deciding a transition period after a withdrawal notice has been served, an oversight establishment council fixing the funding, the personnel and the temporary stoppage authority of a body that must stay independent of the one it oversees, an observation-records project deciding who keeps what the verification produces. In the 2040s layer the deciders must ratify or renew — a peace-zone ratification assembly choosing the form in which a protection compact outlives the armistice, a records accountability committee deciding what to preserve before an original disappears. In the far-future layer the deciders' own jurisdiction is part of the question. A conference of habitats with mutually deferred jurisdictions must fix who may stop what; a generational compact council must write a rule that binds protected persons who have already departed while descendants who have not yet voted inherit the obligations; a truce trust must write entry and succession rules before any reply has arrived, with the representative generation on each side already changed.

The question form guarantees that a decision actor is named; it does not guarantee that the coders can agree what kind of body that actor is. Across the layers, this study reads a change in what the named decider must handle rather than in how much power it holds: the far-future bodies still act within mandates, but the rules they write reach parties the records name as absent or changed — protected persons who have already departed, descendants who have not yet voted, branched residents not yet on the roster — and the records withhold authority to presume the consent of future branched residents.

How long the waiting is. Chapter 10 sets out where the waiting falls in the protection and infrastructure series; the point here is what the layers do to the shape of the wait. In the 2026 layer the deadline names an hour on a named day — 17:00 on 30 September 2026 in the space question — and the record gives no starting point from which the interval before it could be counted: the committee must have decided before that hour. In the 2030s and 2040s layers the deadline still names an hour, and what runs out behind it is a schedule or an arrangement — a settlement deferral, an allocation cycle, a contract renewal, a ratification term — of the same kind the 2026 question carries when its warranties end. In the far-future layer the deadline becomes a sealing moment or a transmission window, and what follows it is the reply time itself, which the record puts at many years. The horizon row shows how the coders could code that: days-to-months in 22 cases under the lenient reading, years in 13, decades in 4, centuries-or-more in 1, and unsettled in 10. The far-future questions are the ones whose waiting the record itself describes in years or generations, and in one case sets against an applicant's chance to participate within a lifetime.

What can be undone. The ledger's reversal rows record what each case allows. Chapter 11 sets out the values the coders read in those rows across all fifty cases; taken as aggregates they do not divide by temporal layer. Read across the layers, the coder values in the reversal rows do not move in one direction: the observation-records question of 2031 and the records-access question of 2033 carry a reversal action the coders read as impossible, while far-future questions carry costly and unsettled ones. What the later questions add is reach rather than a new kind of irreversibility: a conference of habitats with mutually deferred jurisdictions must fix one common supervision rule for a branched approval system before a certificate renewal, and a records committee investigates omissions in civilian protection tasks while protecting collaborators in the same records. The far-future records keep the decider's immediate local powers — stopping expenditure, revising access, receiving objections, reconsidering a classification — and what the delayed external replies constrain is verification and mutual effect rather than the ability to act within the horizon the record sets.

Khong's analogies, and Neustadt and May's thinking in time. Yuen Foong Khong's Analogies at War: Korea, Munich, Dien Bien Phu, and the Vietnam Decisions of 1965 (1992) holds that policymakers use historical analogies to define a new situation, to judge its stakes and to evaluate the options, and that in the Vietnam decisions of 1965 the analogies of Korea and Munich shaped the choices made. Richard E. Neustadt and Ernest R. May's Thinking in Time: The Uses of History for Decision-Makers (1986) holds that decision-makers misuse history above all by seizing on one analogy, and recommends separating what is known, unclear and presumed, asking how the present situation came about, and seeing the issue as part of a stream of time running from the past into the future. This study reads Meridian's layers as a device different in kind from analogy: not one case carried across to a new situation, but one pack's four questions set at four distances, so that what the reader carries is a matched comparison of forms rather than a single borrowed case. The comparison is of questions — their deciders, deadlines, unknowns and reopening lines — not of states, forces or wars, and it therefore cannot settle what an analogy would be asked to settle. The records label the roots of each claim, distinguishing setting, background material, inference, hypothesis, value and symbol, but those labels are a property of the record form rather than an effect of the layers, and they are not the same exercise as Neustadt and May's separation of what is known, unclear and presumed. In another respect this study reads the layers as answering a risk that neither note states in those terms. Khong's note describes analogies that define a situation, judge its stakes and evaluate the options; Neustadt and May's describes seizing on one analogy as the misuse of history. This study's own extension is that a far-future question read as a prediction of how institutions will behave belongs to the same family of misuse, and the records' own premise policy says a future setting does not predict when it will arrive.

Where the judgments part. The two divided cases read in this chapter are both far-future ones; Chapter 14 names the rest and reads the divisions. The layers do not explain the divergence, and this chapter does not propose that they do. What these five questions show is that in the far-future pair the judgments disagree about jurisdiction, consent across branching and the authority to bind successors, while the one 2026 question read here was read the same way by all thirteen.

12.5H6

The verdict on H6 is recorded with the verdicts on the other hypotheses in Chapter 15. This chapter reports that verdict without altering it, and explains why the layers give the hypothesis structure rather than a settlement.

The verdict is not read across this study alone for reasons the ledger itself supplies. A single platform's collection, published on one date, is one data set; the other books in the series carry their own ledgers on the same hypothesis, and Volume 0 holds them beside each other. Chapter 15 sets out what Volume 0 will do with this ledger beside the others. There is also a local gap: the horizon cell is coded unsettled in 10 of the 50 cases, which limits any comparison by horizon to the cases the coders could classify. A second gap matters more for H6: the hypothesis speaks both of how much is decided now and of how many conditions are set for revisiting, and the comparison this chapter sets out measures only the second. The verdict therefore stays as the ledger records it.

What the ledger's own tables give, as recorded, is this. Under the strict reading, which counts only cells both coders found stated, the H6 by-horizon table covers days-to-months with 9 cases, years with 5, decades with 1, and centuries-or-more with none. Under the lenient reading, which also counts what the setting implies, it covers days-to-months with 22 cases, years with 13, decades with 4, and centuries-or-more with 1. Each band carries a figure for the conditions per case rather than a total of conditions, and those per-case figures change little from band to band. The two tables line up with the horizon row's stated and implied columns, so the difference between them is the difference between what both coders found in the record's words and what at least one inferred from the setting.

Two things follow for this study's layers. First, the conditions per case barely move across bands, as the by-horizon table records. On this ledger the visible difference across the bands is therefore not how many conditions are written per case but what those conditions are conditions on. In the 2026 layer they are checks that can run inside the decision's own period: daily stoppage times, fortnightly audits, monthly accessibility tests. In the far-future questions the same shape of condition attaches to checks that keep running after the decision — an annual roster check, a half-yearly consent review — beside a reply that will arrive in decades, which is a different matter from the interval at which a check runs; the earlier questions have their own monitoring after the deadline too. Second, the count of conditions is not a count of observable conditions; the ledger counts conditions in two fields of the record, the stop conditions the judgments publish and the reopen lines the question itself carries, and how many of them a reader could observe is a separate question. H6's hypothesis speaks of conditions set for revisiting without saying whether anyone could check that the revisiting occurred, which is the question the next chapter takes up.

The layers therefore give H6 one qualitative pressure and one limit. The pressure is that in the two far-future questions read here the deadline marks a moment after which the record's own terms change: the dispatch cannot be modified once sealed, and the reading-support contract for the preservation vessel expires. The conditions written for revisiting are then written for watchers who may not be the deciders of the present — a representation recorder checking a roster years after the dispatch has gone, an external records liaison officer checking each year whether replies have arrived. The limit is the one the horizon row leaves open: where the band is unsettled, the layers press on H6 without settling it, and the verdict stays as the ledger records it, read across the books in Volume 0.