The Consignment That Splits Itself

I.The First Known Thing — the good that acquires a nationality

customs law · rules of origin

A shipment arrives at a border and must declare where it is from. This is not a question about geography. Almost nothing of consequence is made in one country any longer, so origin is not observed; it is assigned, by rule, to a good that has passed through several hands. The doctrine that does the assigning is called substantial transformation, and its content is that a good acquires the origin of the place where it was last changed enough. Everything turns on what counts as enough, and enough has been codified.

three forms the codification takes

Three forms recur, often stacked. A change-of-tariff-classification rule says the processing must move the good to a different heading of the harmonised schedule: yarn into fabric counts, fabric into cut fabric may not. A regional value content rule says a stated fraction of the good's value must have been added inside the zone — the figures under the United States–Mexico–Canada Agreement run to seventy-five per cent for passenger vehicles, calculated by prescribed formulae. A specific process rule names the operation itself, as the textile and chemical schedules do. Alongside these sits a de minimis provision, which forgives a small share of non-conforming input: a good that fails its tariff-shift rule may still qualify if the offending material is under roughly ten per cent of its value. Below the line, the input is treated as though it were not there.

the discipline that grew up around the line

What customs law knows, and has known long enough to have built machinery for it, is that a threshold of this kind is not a passive measuring instrument. It is an incentive, and firms restructure production against it — the practice has a name, tariff engineering, and a jurisprudence reaching back to nineteenth-century sugar importers who added molasses to fall into a cheaper heading, and forward to vehicles imported with seats and windows fitted so as to be passenger cars and then stripped to vans. So the rules carry anti-arithmetic doctrine. Cumulation decides when inputs from several places may be added together. Aggregation and absorption rules decide whether related non-conforming materials are counted jointly against de minimis or separately. Anti-circumvention provisions address the splitting of consignments and the interposing of an intermediate step whose only function is to move a line. None of this machinery is about measurement. It exists because the measured quantity is divisible, and a divisible quantity assessed piecewise can be brought under any threshold by partition alone.

II.The Second Known Thing — the metabolite that acquires a dossier

pharmacokinetics · ADME and metabolite safety assessment

Walk to a hall with no borders in it. A small molecule swallowed by a human being is processed: oxidised, conjugated, cleaved, reduced, and circulated as a mixture of itself and its products. The animals in which the compound's toxicology was established process it too, but not identically, and a metabolite abundant in people may be scarce or absent in the rat and the dog whose organs were examined. The discipline's term for such a product is a disproportionate metabolite, and the regulatory question is which of them must be qualified in their own right rather than covered by the safety work already done on the parent.

the line, and what sits on either side of it

The answer is a fraction. The governing guidance — the metabolites-in-safety-testing framework and the nonclinical guideline that absorbed it — sets the trigger at metabolites exceeding about ten per cent of total drug-related exposure in humans at steady state, where human exposure is not adequately covered by the toxicology species. Above the line, a metabolite is in effect a new substance: it may need synthesis, its own general toxicology, genotoxicity assessment, sometimes its own carcinogenicity bridging. Below it, the metabolite has the parent's status and is examined no further. The threshold is approximate by design, argued over in the literature, and applied with judgement — but it is a fraction, and it is assessed metabolite by metabolite.

what the discipline does with the line

The framework is treated in the field as an analytical problem. The published effort goes into measuring the fraction well: radiolabelled mass balance studies, the choice of steady state against single dose, quantification without authentic standards, whether to compare area under the curve or total radioactivity, how to handle a metabolite whose reference material does not yet exist. Reviews of a decade's experience read as accounts of measurement converging. What is not asked, because no one in the hall has needed to ask it, is what the rule does to the molecules — whether the distribution of reported human metabolite fractions bears the marks of a population of decisions made with the line in view.

III.The Bridge

no pharmacokinetics paper cites a customs tribunal

The halls share nothing. Customs lawyers do not read mass balance studies, and the authors of metabolite guidance do not read origin jurisprudence; no review spans them and neither cites the other. But both fields administer the same object: a threshold on a divisible fraction, which confers a separate identity — and a separate obligation to be examined — on whatever crosses it.

X: a fractional threshold assessed piecewise is a rule about partition, not about content

Call it threshold identity. Where a rule confers a new status on a portion of a transformed thing once that portion exceeds a stated fraction, and the fraction is assessed on each portion separately, the rule constrains the arithmetic of division and not the property it was written to police. Two portions of six per cent are, to such a rule, not the same object as one portion of twelve, though the transformed matter is identical and any hazard it carries is identical too. From which follows the half that pharmacokinetics has not ported, because only customs law has had to litigate it: a piecewise fractional threshold is incomplete without an aggregation doctrine, and until it has one, the cheapest way to comply is to redistribute the fraction rather than to reduce it. Trade law reached this conclusion the expensive way, through a century of goods engineered to sit on the convenient side of a heading, and answered with cumulation, joint counting and anti-circumvention. The metabolite threshold is younger, is assessed per metabolite, and has no aggregation rule at all: three related conjugates sharing one structural liability at seven per cent each are, under the arithmetic, three compliant metabolites.

IV.The Question That Becomes Possible

neither discipline would ask this alone

Two things follow, and they must be taken in order, because the second is worthless if the first is false. The first is an observation about a distribution. If the ten per cent line is doing what thresholds are known to do elsewhere, then the reported human metabolite fractions in development programmes should not be smoothly distributed across it. They should pile up beneath it, and the pile should have appeared, or thickened, after the framework was adopted. Customs economics calls this bunching and measures it routinely; pharmacokinetics has never looked, because within the hall there is no reason to expect it.

the partition hypothesis

Does the distribution of reported human metabolite exposure fractions show a discontinuity immediately below the qualification threshold; and where it does, is the mass beneath the line accounted for by metabolic routes divided across several sub-threshold products rather than by genuinely lower total metabolic burden?

why it would matter

Because the two explanations have opposite consequences and are currently indistinguishable. If the mass below the line reflects cleaner molecules, the framework is working and the pile is the point. If it reflects partition — soft spots blocked in ways that split one abundant metabolite into several scarce ones, radiolabel positions and analytical windows chosen so that products are resolved rather than summed, structures advanced because they metabolise diffusely — then a rule written to ensure human-relevant products are examined is selecting for diffuse metabolism, which is not the same property as safety and may be the worse one. It would also supply the missing instrument: a summed fraction over metabolites sharing a structural liability, which is the aggregation doctrine trade law already has and the metabolite rule does not.

V.How It Could Be Proven — or Broken

a candidate must be breakable, or it is merely poetry

An intercognix candidate earns nothing until it names its own test. This one names two, and the first is a precondition for the second.

the honest status

Note the narrowness. The claim concerns small-molecule oral drugs with human mass balance data and a per-metabolite fractional trigger; it says nothing about biologics, nothing about impurity qualification, and nothing about whether any individual programme acted deliberately — bunching is a property of a population, not an accusation against a sponsor, and partition can arise from honest chemistry pursued under a rule that rewards it. Nor does it claim that diffuse metabolism is unsafe, only that the rule as written cannot tell the two apart. And the pile itself is an assertion: unlike a tariff schedule, whose incentives are transparent by construction, this one has to be measured in a distribution that may turn out to be perfectly smooth. Until it is measured this remains exactly what its stamp says: a candidate. A structural echo between two halls of the library, waiting for someone with the tools of both to walk the corridor.