The Amendment Atlas: A Proposal

Visiting paper, circulated for comment under the Archive's working-paper conventions. The Institute's own conventions also apply: this paper states what it proposes to measure, publishes in advance what would count as failing to find it, and asserts no result it has not yet obtained — which, since the atlas is unbuilt, is all of them.

---

The longest continuously maintained dataset in settled space is not a survey series. It is the constitutional register: the Concord compact's founding text and every amendment enrolled against it since 2543 UCSD — each one dated, attributed, argued, and filed in difference form, beside the proposals that failed, for very nearly three centuries without a gap.

We have treated this record as law, which it is, and as history, which it is. My proposal is that it is also an instrument log — the output of a slow instrument that has been pointed at one question since Tycho, namely what settled space believes it owes its people — and that nobody has ever plotted it. Difference form makes the plotting honest in a way no other legal corpus allows: an amendment states exactly what moved and when and at whose urging, so the register can be walked by procedure rather than interpretation, clause by clause, strike by insert, from the founding enrolment to this morning's consolidation. The law, almost uniquely among human productions, comes pre-reduced.

I propose an atlas of four sheets.

Sheet one: the boundary of the person. Take the personhood clauses alone and plot every enrolled change to them against time — each widening, each qualification, each definitional repair, as a dated event on one axis. The folk belief is that this line only ever widens. I decline to assert the line is monotone before drawing it, and I note that the folk belief is precisely the kind of claim an unplotted record lets a civilisation hold comfortably. If the line has ever narrowed, the atlas will show where, and we will owe the register thanks for the discomfort.

Sheet two: the lag. For every correction the compact has enrolled, the register also holds the correction's first filing — the defeated proposal, the withdrawn draft, the difference returned at certification and redrafted across a generation. The distance between first argued and finally enrolled is a duration, and durations can be charted. Call it the lag: the time the law spends catching up to a case that was already in front of it. I expect the lag to vary by subject matter, and the variation to be the finding.

Sheet three: the still core. Some clauses of the founding enrolment have never been amended — not once, under any of the regimes that enforced the compact, each of which had its own reasons to want to. Chart the compact by stillness and the unmoved clauses form a core sample of what three centuries of otherwise-restless law has treated as settled. A map of what never changed is as much a portrait of a civilisation as a map of what did, and it is the cheaper of the two to verify.

Sheet four: the weather of amendment. Enrolment density over time — the quiet decades, the corrective bursts, the pattern of activity around each change of regime. I make no prediction here at all; the Imperium's sixty-six years alone could embarrass any prediction I drafted, and the honest form of this sheet is a question laid against an axis.

One worked figure, to show the method's teeth. The compact's current personhood standards accommodate plural registration; the older frameworks that filed plurality as a disorder held institutional force within living archival memory. The register holds the whole road between those two states — every strike, every insert, every defeated draft along it, each with its date and its filed reasons. Sheet one renders that road as a line, and sheet two renders its cost as a duration. And here I state the method's limit in the same breath as its power, because the limit is the finding's correct caption: the line is the record's line, not the people's. The people plural registration was ever true for did not change across those centuries. The paperwork did. An atlas of amendments is a chart of the law's opinion moving, and any sheet that lets a reader mistake the motion of the opinion for the motion of its subjects has failed as an instrument, whatever it achieves as a picture.

The register's habit of keeping its defeats is what makes the atlas possible at all. Plot enrolments alone and you have drawn the winners' history — a chart with its null results deleted, which my Institute would decline to publish as a matter of house rule and the Archive would decline to shelve as a matter of doctrine. The atlas takes the whole register or it takes nothing.

I am told a version of this proposal is raised at the Archive roughly once a generation, and dies each time of scope. I have filed this one with a scope: four sheets, one compact, the personhood clauses first. Comments to the usual shelf.

— I. Corren, for the Institute, which as usual takes no position on what any of it means.

---

Shelf comment, filed beside the paper: The proposal is sound archival practice and the Archive should assist it; walking a holding from first claim to present standing is what this building is for, and the Fellow has noticed that a civilisation's law is a holding. Two conditions, both of which she has in substance already conceded. First, every sheet carries her own caption — the record's line, not the people's — printed on the sheet and not in an appendix, because the voice that summarises may not revise, and a chart is a summary wearing its best clothes: where a sheet and the register disagree, the sheet is wrong, always. Second, the atlas is drawn from the register and never becomes it. The day a tribunal cites a sheet instead of an enrolment, the atlas has stopped being an instrument and started being an authority, and we have enough of those. Comment favourable. Conditions standing. — Sen, Senior Archivist.