The Mass Budget of the Return: Whether the Small Universes Can Close a "Bounce" and What a Holdback Costs

1. Introduction

It is well known that the Returners' broadcast is treated in the survey literature as a request whose fulfilment guarantees the closure of the main universe, so that the only open matter is whether the small universes will comply (a question of persuasion, in the words of most commentators, and not of physics). That treatment is wrong, and the error is arithmetical before it is cosmological. The mass held by the small universes is a fixed fraction of the closure mass, the main universe stands short of closure by a second fixed fraction, and whether the first exceeds the second is a measured matter that nobody has set out in one place. Let me do so here.

Our vantage is the Galaxy Era, after the broadcast has been received, transcribed and archived. The archive that holds the transcript dates the broadcast to the opening of the Timeline Ending reckoning, and for that reason we print no Galaxy-Era year for the broadcast or for the present writing (the methodological note in Section 2 gives the reason). The survey bodies, whose usage we follow, have not adopted the archive's new reckoning for their own work, and we write in Galaxy-Era terms.

The "bounce" scenario is older than the broadcast. Its first versions treat the closed history of the main universe as symmetric about the moment of greatest curvature (Kesterton, 441 GE), and its recollapse times for universes only slightly above closure were worked out for matter-dominated models long before anyone had a use for them (Tavish and Brandt, 168 GE). The inventory of what the small universes took from the main universe at the close of their construction was compiled earlier still (Pellam, 97 GE), and the most complete census of what remains in the main universe is more recent (Vessey and Orlund, 212 GE). The broadcast itself says that mass withheld by the small universes leaves the main universe unable to close (Varnholt, 1 TE). What none of these works gives is the margin, that is, how much of the lodged mass may be held back before that claim stops holding.

We supply the margin. Section 2 describes the system and the sources, Section 3 derives the closure condition and the recollapse time as functions of the returned fraction, Section 4 checks the inputs against survey data and tabulates the consequences of holdback, Section 5 lists the ways the model can fail, and Section 6 states what we conclude.

Translator's note: this paper was rendered into English from the Chinese original by the Branch translation office, and the credit belongs to Imre, S.

2. System Description

The system has two parts. The closed main universe is the one the survey fleets measure, and the small universes are the pocket universes (Universe 647 is the best known) built from mass taken out of the main universe and now asked to return it. We measure all masses in units of the closure mass, written M, which is the mass inside a comoving volume at which the density of the main universe equals the critical density at the epoch of the survey. A universe with total mass above 1 in these units is closed and, in the matter-dominated model used here, recollapses, and one with total mass below 1 expands for ever.

Three quantities enter. The first is the density parameter Ω of the main universe after the withdrawal, taken from the dynamical census, which weighs luminous matter, dark matter and the diffuse inventory together. The second is the total mass S lodged in the small universes, which the broadcast gives as a single figure of 0.083 together with a tabulation by size class. The third is the returned fraction f, running from 0 to 1, and its complement h = 1 - f, called the "holdback" (the paper's term, used without quotation marks from here on). A loss in transfer would act exactly like a holdback, and we make no attempt to separate the two.

The transcript tabulates the small universes in four size classes, which we number A to D from the largest down. Class A holds 9 universes and 0.31 of the lodged mass, class B holds 71 and 0.29, class C holds 342 and 0.27, and class D holds 716 and 0.13, which makes 1,138 universes and the whole of S. We call this tabulation the "Return Mass Ledger" (our term, and below simply the ledger), and it is our own compilation from the transcript, not a document the Returners issued. The transcript places Universe 647 in class D, with a share too small to print at the transcript's precision, and the largest single entry in class A holds 0.052. The concentration of mass in the largest classes matches the size spectra fitted to the pocket universes before the broadcast (Dalgleish, 380 GE).

A methodological note on dating and vantage. The Returners' broadcast falls far beyond the range of years that Galaxy-Era reckoning was built to carry, and the archive of the transcript (compiled by Varnholt) therefore records it against the opening of the Timeline Ending reckoning, as year 1. We follow the archive and cite the transcript as 1 TE, although the survey bodies, whose usage we follow in everything else, have not adopted the archive's new reckoning for their own work, and we write in Galaxy-Era terms. We write from the Galaxy Era, after the broadcast, and we treat the Luo-Ji / Ye axioms and the whole of cosmic sociology as a historical discipline at this vantage. The census (212 GE) is some 19 million years older than the broadcast, an interval of about 0.1% of a Hubble time, so the drift in Ω over that interval is far below the census error.

3. Analysis / Model

Let the main universe carry a density parameter Ω after the withdrawal, and let a fraction f of the lodged mass S return to it. The total is then Ω(f) = Ω + fS, and closure requires Ω(f) > 1. The returned fraction at which closure is reached is therefore f* = (1 - Ω)/S, and the largest holdback the main universe can absorb is h* = 1 - f*. Both are linear in f and need no assumption beyond the two measured masses.

That is the entire mass condition, and everything below is a consequence of it.

The time to recollapse is not linear, and it is the quantity that decides whether closure matters. For a matter-dominated closed universe the total lifetime from the initial state to the final crunch is t = πΩ(f) / [H(Ω(f) - 1)^(3/2)], where H is the present expansion rate. The factor (Ω(f) - 1)^(3/2) in the denominator means that the lifetime diverges as the total approaches 1 from above (rising margin, rising curvature, rising deceleration, and so a shorter lifetime, with the reverse for a falling margin).

A "bounce" in the sense of the symmetric closed history needs a recollapse that actually arrives. A universe that closes by a hair closes in a time that no civilisation will wait for, so we also report the returned fractions at which the lifetime reaches 50 and 100 trillion years, using 1/H = 14.3 billion years from the survey expansion history (Hadley, 305 GE).

Uncertainty enters through the census value of Ω alone, with standard error 0.012, and we treat S as the broadcast's figure without error (the transcript quotes two significant figures). The lifetimes carry a further uncertainty of ±3% from 1/H (13.9 to 14.7 billion years), and we do not propagate it into the closure probabilities because 1/H does not enter the closure condition. The probability of closure at a given f is then the normal probability that Ω(f) exceeds 1. We use this form because the census error is a sum of many small systematic terms and the normal approximation is the one the census authors themselves adopted.

4. Validation Against Field Data

The census gives Ω = 0.940, 95% CI 0.916 to 0.964. The broadcast gives S = 0.083. The shortfall of the main universe is 1 - Ω = 0.060, so the central estimate of the critical returned fraction is f* = 0.72 (0.723 before rounding) and the largest absorbable holdback is h* = 0.28. Because the confidence interval on Ω carries over to f*, the interval on f* runs from 0.43 to 1.01 (standard error 0.145). The upper end exceeds 1, which means that the data do not exclude a main universe that full return cannot close.

A broadcast figure for S is a claim, and a claim from the party that asks for the mass. I shared, when the transcript was first released, the objection that such a figure should be set aside until someone verifies it, and what changed my mind was the independent census deficit. The census reports the mass missing from the main universe, relative to the inventory at the close of construction (Pellam, 97 GE), as 0.079, 95% CI 0.061 to 0.097. The broadcast figure differs from it by 0.004, which is 0.4 standard errors.

The two determinations share no instrument, and they agree.

Full return gives Ω(1) = 1.023, which exceeds 1 by 0.023 and corresponds to a lifetime of 13.2 trillion years. The probability that full return closes the main universe is 0.97 when S is taken from the broadcast and 0.90 when S is taken from the census deficit (the latter combines the census errors in Ω and in the deficit, 0.012 and 0.009, as if independent, giving 0.015; both come from the same census and are probably negatively correlated, a correlation that would shrink the combined error, so 0.90 is a conservative figure). Closure after full return is therefore likely and not assured, and the margin is of the order of two census standard errors.

Holdback is costly at the scale the ledger allows. A holdback of 0.10 gives Ω = 1.015, a lifetime of 25.6 trillion years and a closure probability of 0.89. The lifetime passes 50 trillion years at a holdback of 0.16 and 100 trillion years at 0.21, and the central estimate fails to close at 0.28. The concentration of the ledger matters as much as its total, since the nine largest universes hold 0.31 of the mass, which is more than the main universe can absorb.

The pattern in the table follows from the model and was not fitted.

Withholding the single largest universe (0.052 of the mass) changes the closure probability little, from 0.97 to 0.94, though it lengthens the lifetime by about a third. Withholding the nine largest (class A) leaves a main universe that closes with probability 0.41, and withholding the 1,058 universes of classes C and D, which is 93% of all the universes, leaves 0.20.

5. Failure Modes

The model is matter-dominated, and the survey expansion history finds a residual vacuum term of 0.00 with a standard error of 0.02 (Hadley, 305 GE). We do not model the term. Geometric closure, Ω(f) > 1, is a condition on curvature and does not depend on it, whereas recollapse does: a positive term would raise the returned fraction needed for recollapse, and a negative one would lower it. Assuming the term is not negative, the thresholds reported here are lower bounds on what recollapse demands, the lifetimes are lower bounds on the time to recollapse, and the probabilities of recollapse are no larger than the closure probabilities we print.

Census error in Ω is the single largest source of uncertainty, and its systematic part is not fully controlled. If the dark inventory has been overcounted, the shortfall is larger than 0.060, and the data that agree on S would then fail to agree on closure. A second failure lies in the assumption that returned mass is equivalent to lodged mass. A conversion loss in transfer of 5% would act as a holdback of 0.05, and full return would then sit near the largest-universe-withheld row (Ω 1.019, lifetime about 18 trillion years).

Timing is a third failure that the model cannot see. The linear condition treats the returned mass as arriving at once and isotropically, while a staged return, with class D coming back first, would shift the curvature of the main universe during the interval and so alter the symmetry on which the "bounce" itself depends (Kesterton, 441 GE). We leave that question to a dynamical treatment of the return, which this paper does not attempt.

6. Conclusion

The finding is that full return of the lodged mass likely closes the main universe, with a probability between 0.90 and 0.97 depending on which determination of S is used, and that the margin is narrow. The returned fraction must reach about 0.72, so a holdback of 0.28 or more forfeits closure at the central estimate, and a holdback above 0.16 delays the "bounce" past 50 trillion years. A small holdback may be tolerated, and the broadcast's own claim, that withheld mass leaves the universe unable to close, is correct in direction and wrong if it is read to mean that any withholding at all is fatal.

A framework declaration belongs here. I write as a partisan of the Lurker framework (the Lurker, also "Spirit" or "Spirit Mother" in some translations, is the cosmogonic agent that shaped the history of the dimensional-reduction war and its aftermath), and for me the Luo-Ji / Ye axioms hold in a local domain only. At this vantage they are also historical, the account of one epoch (the epoch of the dark forest) that the deep-time war-fields have long since replaced as the setting of cosmic politics. Nothing in the mass budget depends on that stance, but the reading of the broadcast offered here does.

One alternative deserves a hearing. The census value of Ω could be biased high by an overcounted dark inventory, in which case full return would not close and the agreement on S would be a coincidence of two determinations. We judge this less likely than the finding, but only because it needs a bias of 0.023 in one direction (1.9 census standard errors). The agreement of the census deficit with the broadcast figure bears on S and not on Ω, and nothing in the Results excludes such a bias, so the alternative remains open.

I close with an analogy that proves nothing and claims no scientific standing. A village that has lent grain to its neighbours for a hard winter does not ask whether the loan will be repaid in principle (it asks how many sacks are in the ledger, how many the granary is short, and who among the borrowers keeps the largest stores). The answers decide whether the next sowing comes in this decade or in a century nobody in the village will live to see.