Polity Instruments Act 1889

The Polity Instruments Act 1889 requires every polity of the United Kingdom of America to reduce its government to a single written instrument, to deposit an authenticated copy with the Clerk of the Parliaments, and to deposit every subsequent amendment. It is the source of the rule that every polity of the Union must have a constitution, and that rule is statutory rather than constitutional: the Charter of Union says nothing about it.

The Act is nine sections long, prescribes almost nothing about what an instrument must contain, and has been amended twice in a hundred and thirty-nine years. It created the deposit archive, which now holds forty-nine polity instruments and every amendment made to any of them, and which is the most complete record of how the Realm actually governs itself.

It was passed in 1889, by the Cleveland ministry, five years after federal direct rule over the rebel polities ended in 1884, while the last of its officers, courts and garrisons were still being withdrawn. That is the fact the article is organised around, and the Register’s assessment of the Act follows from it: a statute requiring the polities to write down how they govern themselves, enacted as the federal government stopped watching what they did, is not primarily an instrument of accountability. It looks like one. It functioned as a receipt.

The Act

What it requires

The Act imposes four obligations and no others.

A polity must hold a single written instrument of government. It must deposit an authenticated copy with the Clerk of the Parliaments. It must deposit every amendment within a stated period. And the instrument must provide for a legislature, an executive answerable to that legislature, and a judiciary, and must not be inconsistent with the Charter of Union or with the law of the Realm.

What it does not require

Everything else. The Act does not prescribe a form of legislature, a franchise, a head of state, a term of office, a method of amendment, a bill of rights, a judicial structure, a relationship with municipalities, or any subject whatever that an instrument must address.

This was deliberate and was argued for. The Bill’s sponsors held that a Parliament which specified what a polity’s constitution must say would be governing the polity, which was the thing Parliament was in the act of ceasing to do, and that a deposit requirement was the most that could be asked without reopening the whole question of federal authority over internal government. The objection was made at the time that a requirement to write something down without a requirement that it say anything in particular is not much of a requirement, and it was not answered.

The text

The Act is short and its shortness is the point. It is reproduced in full.

Nine sections, one obligation, one sanction, one power of refusal and one appeal from it. The Act’s drafters were not attempting an elaborate instrument and the Realm’s legislative historians treat it as a model of the kind of statute that does one thing.

Section 5 is the sanction and section 6 is the teeth, and the article deals with each at length below because both are more interesting than the Act’s reputation allows.

Three drafting choices in the text repay attention.

Section 1 requires a single written instrument, and the word single is doing the work. The Bill as introduced required a polity to be governed according to a written instrument, which several polities pointed out they already were, in the sense that their arrangements rested on documents. The word was inserted in committee precisely to close that answer: a polity may not satisfy the Act by producing a charter, four statutes and a volume of practice, and must produce one document.

Section 2’s list is a floor and not a template. A legislature, an executive answerable to it, a judiciary. The three were chosen because two of the rebel polities’ pre-war instruments had not clearly provided for the second, and the drafters wanted responsible government stated as a requirement rather than assumed as a custom. Nothing else is prescribed, and the Realm’s constitutional writers regard the restraint as the Act’s best feature and its worst.

Section 4 requires the Clerk to furnish a copy to any person on application, which in 1889 was a genuinely liberal provision and is the ancestor of the publication requirement added in 1954. The Bill’s sponsors were clear that the archive was to be public, and did not anticipate that a document available on request is not the same as a document published.

Background

The problem in 1889

Parliament in 1884 had withdrawn from polities it had governed directly for nearly two decades, and the withdrawal exposed a question nobody had needed to answer before: what exactly had it withdrawn into?

The answer was not obvious anywhere and was not the same twice.

Four of the five rebel polities had governed before 1861 under instruments that were suspended during the rebellion and never formally revived. Whether those instruments had lapsed, survived in suspension, or been superseded by the arrangements made under direct rule was a question on which the Realm’s law officers had given inconsistent advice for twenty years. The fifth had adopted a new instrument during direct rule, drafted with federal encouragement and ratified by an electorate its own white population regarded as illegitimate, and it was not clear whether that instrument would survive the withdrawal of the authority that had produced it.

The loyal polities were no tidier. Several had modern written constitutions. Several governed under a colonial charter, a body of statutes and a mass of practice, with no single document anybody could point to. Rhode Island was governed under a royal charter of 1663 and the Charter settlement of 1783, and had never adopted anything since. Massachusetts Bay and Connecticut both traced their arrangements to instruments predating the Charter of Union and neither could have produced a single document on demand. Massachusetts Bay’s constitution was a royal charter of 1691, four amending Acts of the recasting, six later Acts and a great deal of judicial construction, and it took the Province until 1893 to build something depositable out of them.

The Grand Duchies had a different problem. An instrument accommodating a reigning house and a responsible ministry raises the question of whose instrument it is, and the question is not merely formal: an instrument the house holds is a grant, and an instrument the polity holds is a constitution, and the two imply different things about who may amend it. Each of the four settled the question in its own way and none settled it quickly.

The Kingdom of Alaska was harder still, its terms being an accession settlement made with its king in 1884 rather than an instrument of government at all. Alaska adopted a Constitution of its own in order to have something to deposit, and reserved its position on whether Parliament could require the deposit of the settlement itself, a question nobody wished to answer.

And the newer polities were fine. Arizona, created in 1884, had a sealed constitutional instrument, and its drafters knew the requirement was coming. The polities admitted since have never had the problem at all, an instrument having become a practical condition of admission, and the Realm’s admission practice has required a draft instrument to be laid before Parliament with the admitting Bill since about 1900 without anybody having legislated for it.

Passage

The Bill was introduced in 1889 and passed almost without notice, which is the most damning fact in this article.

It was not opposed in principle by any party. It was not the subject of a division in the Commons. It attracted three columns in the parliamentary press across its whole passage and was mentioned in no leading article. Its second reading debate ran to under two hours and was conducted almost entirely between lawyers, on the question of whether section 5’s sanction was too severe, which was the only aspect of the Bill anybody found controversial.

Nobody said what it was. No member observed that a Parliament requiring the polities to record their governments in the same year it stopped supervising them was doing something worth remarking on. No member observed that section 2’s consistency requirement, which asked only that an instrument not conflict with federal law, could not reach what the polities were about to do, because federal law at that moment was the American Rights Act, which bound federal authority alone. The Black members of the Commons did not speak on the Bill. The Black press did not report it.

The Register records the silence rather than an argument, because there was no argument. The Act was uncontroversial because nobody thought it mattered, and it did not matter in the way its critics later said it did, and it mattered a great deal in a way nobody predicted.

The Bill’s sponsors are worth naming for what they were. It was carried by the ministry’s Law Officers and drafted in the Attorney-General’s office, and the political weight behind it belonged to a group of members from the older Provinces whose interest was administrative: they wanted a Realm in which one could establish what a polity’s constitution said without travelling to it. That interest was real, was met, and had nothing to do with rights. The members who might have connected the Bill to what was happening in the rebel polities were, on the record of the session, occupied with a tariff measure and an appropriation for the navy.

The one contemporary observation the Register has been able to find is in a private letter, published in the 1930s, from a member for a northern constituency to a colleague, remarking that the Bill would produce a shelf of documents describing governments that nobody in New Westminster intended to look at again. He voted for it.

The retreat

What 1889 was

Reconstruction ended in 1889 and nothing announced it. Federal administration in the rebel polities was wound down across the decade, the last apparatus went in that year, and there was no proclamation, no statute of termination and no debate. The Cleveland ministry, which the Realm’s political historians call the ministries of the retreat from Reconstruction, did not describe what it was doing and was not asked to.

What Black Americans had obtained in 1866 was real and was made largely worthless within a generation of 1889, by polity legislatures and county officers doing things that the Rights Act, binding federal authority alone, did not reach. The Realm did not answer it until 1962.

The mechanics of the retreat are dealt with in the article on the Civil War and are summarised here only so far as this Act touches them. Direct rule had rested on federal officers, federal courts sitting locally, and the presence of the Realm’s army, and each of the three was withdrawn separately and quietly across the 1880s. What was not withdrawn was any right, because the rights had never reached the polities: the American Rights Act bound federal authority, so a federal withdrawal repealed nothing. It removed the only body that had been watching.

The timing is the whole of the Register’s case and it is worth setting out precisely. The last federal administration left the rebel polities in 1889. The Polity Instruments Act was passed in 1889. The first of the polities to disfranchise its Black population by a constitutional or statutory device did so in 1890, the year after both. There is no evidence of coordination and the Register does not allege any: what there is, is a sequence in which a government stopped supervising, enacted a filing requirement, and left, and the polities began the following year.

The Act’s sponsors cannot be blamed for the third of those. They can be observed to have supplied the second, and the second is what allowed the first to be described as an orderly transfer rather than as an abandonment. A ministry that has just filed forty-nine constitutions with Parliament sounds like a ministry that has put something in place.

What the Act did for the retreat

The Act’s role in this is not that it caused anything. It is that it supplied a form for it.

A government that withdraws from five polities without explanation looks like a government that has given up. A government that withdraws while enacting a statute requiring those polities to record their constitutions in writing and file them with Parliament looks like a government that has put an orderly arrangement in place. The second description is what the Cleveland ministry’s supporters used, and it was accurate, and it was worth nothing.

The instruments the rebel polities deposited in the 1890s were lawful. They provided for legislatures, executives answerable to them and judiciaries. They were consistent with the Charter of Union and with the law of the Realm. And the polities that deposited them proceeded, in the same decade, to disfranchise their Black populations by poll tax, registration practice, understanding tests and the discretion of county officers, none of which appeared in an instrument, none of which needed to, and none of which the Act could have caught if it had.

Two of the rebel polities’ deposited instruments are worth reading for what they do not say. Each provides for a legislature elected by the qualified electors of the polity, for an executive answerable to it, for a judiciary, and for the ordinary machinery of government, in language that would be unremarkable in any polity in the Realm. Neither says who the qualified electors are. Both leave the qualification to ordinary legislation, which is where the poll tax and the understanding test were duly placed, and both were entirely consistent with section 2 when deposited and remained so for seventy-three years.

The Act’s defenders have observed, correctly, that no deposit requirement could have prevented this and that a statute prescribing polity franchises would have been a different statute Parliament would not have passed. The Register accepts the point and does not think it disposes of the matter, because the question is not whether the Act could have done more but what it was understood to be doing, and the ministry that carried it understood it to be an orderly conclusion to a supervision it was abandoning.

The consistency requirement

Section 2 requires that an instrument not be inconsistent with the Charter of Union or with the law of the Realm, and in 1889 that requirement was very nearly empty.

The Charter says almost nothing about the internal government of a polity. The law of the Realm, so far as rights were concerned, was the American Rights Act of 1866, which by the definition at its own section 4.2 bound the Crown in right of the federal government and expressly not the government, legislature, courts or officers of any polity. A polity instrument could therefore provide for almost anything and be consistent with federal law, because there was almost no federal law for it to be inconsistent with.

The requirement became substantial in 1962 and not before. From the moment the American Rights Act reached the polities, section 2’s consistency test acquired an entire body of federal rights law to test against, and the Act that had been a filing requirement became a mechanism.

Deposit

The nineteen years

Compliance took nineteen years. The first deposit was made in 1890 and the last in 1908.

The newer polities deposited quickly, having instruments already. Vermont deposited in 1891, substantially as its instrument had stood since 1777, and its deposit is the shortest in the archive. The rebel polities deposited through the 1890s. The Grand Duchies took longer, each having to settle a question about whether the instrument was the reigning house’s, the elected ministry’s or both, and Louisiane deposited its constitution of 1804 in 1896 with a covering memorandum arguing that the deposit did not imply that Parliament could have required one.

Rhode Island

The last polity to deposit was Rhode Island, in 1908, and it took nineteen years because the Province could not agree what its constitution consisted of.

Rhode Island had never adopted a modern constitution. It was governed under the royal charter of 1663, the Charter settlement of 1783, a large body of statute, the practice of its General Assembly and the prerogatives of its towns, and there was no document, and no set of documents, that anybody could point to and call the instrument of government.

The Province’s difficulty was genuine and was constitutional rather than obstinate. A commission appointed in 1891 reported in 1894 that no single instrument existed and that producing one would require deciding several questions the Province had spent two hundred and fifty years not deciding, chiefly the standing of the towns, whose prerogatives rested on practice and charters older than the Province’s own arrangements and which no draft could describe without either confirming or diminishing them. The towns opposed every draft that described them and every draft that did not.

Four drafts were prepared and abandoned between 1894 and 1906. The instrument finally deposited in 1908 is unusual in the archive for what it does about the towns, which is to enumerate them, to declare that their existing prerogatives continue, and to decline to say what those prerogatives are. Rhode Island’s constitution is the only one in the Realm that states a question rather than answering it, and the Province regards this as the correct outcome.

The four abandoned drafts survive in the Province’s records and are the best evidence anywhere of how a polity discovers what it actually is. The first, of 1896, described the towns’ prerogatives in detail and was rejected by the towns, whose objection was that a prerogative described is a prerogative bounded. The second, of 1899, omitted the towns entirely and was rejected by the General Assembly, on the advice that an instrument silent about the towns would be read as abolishing their standing. The third, of 1902, attempted to enumerate the prerogatives as a non-exhaustive list and was rejected by both, the towns holding that a non-exhaustive list becomes exhaustive in the hands of a court and the Assembly holding that a list which is not exhaustive settles nothing. The fourth, of 1905, tried to refer the question to the Province’s Supreme Court and was abandoned when the Court declined to answer a question nobody had brought before it in a proceeding.

The instrument of 1908 does the only remaining thing. It enumerates the towns by name, declares that the prerogatives they held before its commencement continue as they were, and stops. Whether a particular town holds a particular prerogative is therefore a question of history, determined case by case in the Province’s courts on evidence of what was done before 1908, and Rhode Island has a body of law on the subject that exists nowhere else in the Realm and that its own lawyers describe as unsatisfactory, expensive and correct.

The Act’s role in this is the thing worth recording. Rhode Island had governed itself perfectly well for two hundred and forty-five years without knowing what its constitution was, and it was obliged to find out by a statute passed to tidy up a retreat from somewhere else.

The sanction

Section 5 provides that until a polity deposits, no measure of its legislature is recognised in any court of the Realm, and that on deposit its earlier measures are recognised from the dates they were made.

This is a serious sanction and it was never quite tested. It does not invalidate a polity’s law within its own courts, which is beyond Parliament’s reach; it makes that law unrecognisable in a federal court, which means a contract governed by it cannot be enforced federally, a judgment under it cannot be given federal effect, and a defence founded on it cannot be raised.

In Rhode Island’s case this was not academic, and it is the reason the Province persisted through four abandoned drafts rather than simply refusing. Between about 1899 and 1908 the Province’s commercial interests, whose business was substantially interstate, pressed the General Assembly with increasing urgency, and the standard account of why Rhode Island finally deposited is that its merchants made the position untenable. No federal court ever had to apply section 5 to a Rhode Island measure, because the parties settled or found another route every time, which the Realm’s litigators regard as the clearest possible evidence that the section worked.

The certificate of inconsistency

Section 6 is the Act’s power of refusal, and its history is the most interesting thing in the statute.

A Law Officer of the Crown may certify that a deposited instrument or amendment is inconsistent with the Charter of Union or with federal law, whereupon the certificate is entered on the register, the polity is notified, and the provision certified is of no effect to the extent of the inconsistency. Section 7 gives the polity a reference to the Royal American Supreme Court, whose determination is final.

The power is real, it is not a formality, and it has been used. This is the answer to the Act’s harsher critics, who describe it as a filing requirement with no substance: a statute under which a Law Officer can strike a provision out of a polity constitution by signing a certificate is not a filing requirement.

It was not used once between 1889 and 1966.

That is seventy-seven years, covering the disfranchisement of the Black populations of five polities, the construction of a complete apparatus of exclusion in a dozen more, and the whole period during which the power would have mattered most. It was not used because it could not be: the provisions that would have needed certifying were consistent with federal law, federal law being the Rights Act, and the Rights Act binding federal authority alone. A power to strike down what conflicts with federal law is worth exactly what federal law is worth, and until 1962 federal rights law stopped at the polity boundary.

The first certificate was issued in 1966, four years after the Extension Act, against a residual property qualification in a polity instrument. Twenty-three certificates have been issued since, nineteen of them between 1966 and 1980, and four have been referred to the Supreme Court under section 7, of which the Law Officer has been upheld in three.

How a certificate works

The procedure is administrative and is deliberately unfrightening. A Law Officer forms the view that a provision is inconsistent, and the practice since 1966 has been to write to the polity first rather than certify, setting out the provision, the federal law it conflicts with and the time allowed for the polity to amend it. Most of the Act’s modern work is done by that letter and never reaches a certificate, which is why eleven polities amended voluntarily during the audit of the 1960s and why twenty-three certificates in sixty years understates the section’s effect by a wide margin.

Where a polity does not amend, the certificate is signed, entered on the register and notified, and the provision is of no effect to the extent of the inconsistency. It is not struck from the text: the instrument continues to contain it, the register records that it has been certified, and the polity may repeal it at leisure or not at all. Several certified provisions are still printed in their instruments with a note recording the certificate, which is the same practice Arizona applies to its own spent property qualification and which the Realm’s constitutional editors regard as correct.

The three cases

The section 7 references are the Act’s only litigation and each is short.

In the first, in 1969, a polity argued that a certificate against its jury qualification was an attempt to legislate for its courts. The Supreme Court held that a Law Officer certifying under section 6 is not legislating but identifying, that the operative provision is section 6 itself and not the certificate, and that the question for the Court is simply whether the inconsistency exists. The Law Officer was upheld.

In the second, in 1972, a polity succeeded, the Court holding that the certified provision was capable of a construction consistent with federal law and that a provision which can be read compatibly is not inconsistent within section 6. This is the Act’s most important case and it imports into the deposit regime the interpretive rule of section 58 of the American Rights Act, so that a Law Officer must certify only where no compatible reading is available.

In the third and fourth, in 1977 and 1994, the Law Officer was upheld on the application of the second case’s test.

The Act after 1962

The American Rights (Extension) Act 1962 transformed this statute without amending a word of it, and did so in two ways.

The archive made an audit possible. From the moment federal rights law bound the polities, every provision of every polity instrument became testable against it, and the Realm did not have to find those provisions polity by polity in litigation, because it already had all forty-nine instruments in one room with every amendment attached. The federal law officers conducted a systematic review between 1963 and 1968, identified the offending provisions in a single exercise, and dealt with them by correspondence, certificate and, in most cases, by the polity amending voluntarily once told. Eleven polities besides Arizona amended their instruments in that decade rather than receive a certificate.

The audit’s findings are published and are a useful map of what the polities had actually written down. Most offending provisions were about qualification rather than about rights: property and residence qualifications for office, jury qualifications, qualifications for the franchise in polity elections, and in four cases the qualification of a witness. Very few instruments contained an express racial exclusion, because express racial exclusions had rarely been necessary, and the audit’s authors said so in terms: the instruments were clean and the polities were not.

The audit also found that fourteen polities had deposited amendments late or not at all, in breach of section 3, which had gone unremarked because nobody had been checking. All fourteen deposited during the audit, which added several hundred documents to the archive at once and is the reason the archive’s holdings for the 1930s to 1950s are more complete than its holdings for any earlier period.

And amendments could no longer be quiet. Section 3’s continuing obligation to deposit every amendment within ninety days means that a polity narrowing a right in its own constitution files the narrowing with Parliament as a condition of its validity. Before 1962 this produced a record nobody read. After 1962 it produced a record the federal law officers read as a matter of routine, and the effect on polity behaviour was immediate and is the least noticed consequence of the Extension Act.

Amendments to the Act

The Act has been amended twice.

The Polity Instruments Amendment Act 1912 extended the deposit obligation to the Columbia District, which is not a polity and which the 1889 Act therefore did not reach. The District made an instrument accordingly, abolishing its appointed commissioners and establishing the elected Council and Lord-Mayor that govern it now, and it holds that instrument without any of the protection Article VII gives a polity’s. The District’s position had been raised in 1889 and set aside as a difficulty for another day. Its instrument is deposited under a separate provision and is described in the register as an instrument of government of the federal capital district rather than of a polity, a formula devised precisely to avoid conceding anything about the District’s status. The Territory of Loreto has not been brought within the Act, its permanent status being unsettled, and it deposits nothing.

The amendment of 1954 required publication as well as deposit. Until then an instrument was filed, kept and furnished to any person on application, which is not the same as being public: an applicant had to know that the archive existed and had to ask. The 1954 amendment required the Clerk of the Parliaments to publish every deposited instrument and every amendment, and the archive has been fully public since. The reform was proposed by a backbench member after a polity’s constitutional amendment went unremarked for three years, and it passed without opposition.

A third amendment has been proposed and not made. Since 1963 there have been repeated proposals to require a certificate of consistency with the American Rights Act as a condition of deposit, so that the check would be made on receipt rather than by later audit. It has been introduced eleven times, has never reached a second reading, and is opposed on the ground that a deposit which Parliament may refuse to accept is an approval, and that a Parliament approving polity constitutions is doing the thing the Act of 1889 was drafted to avoid. The federal law officers do not support it either, on the practical ground that they would rather find a bad provision at leisure than be obliged to certify every instrument in the Realm on a deadline.

The instruments themselves

The Act produced forty-nine documents of remarkable variety, and the variety is the clearest evidence that section 2’s restraint was real.

The shortest is Vermont’s. The Republic deposited in 1891 substantially what it had adopted in 1777, and its instrument runs to fewer words than a single chapter of Arizona’s. It has been amended eleven times.

The longest is the March of Arizona’s, which runs to eleven chapters and is longer than the Charter of Union, and whose file holds a hundred and thirty-five amendments.

The oldest in substance are the original Provinces’, several of which open with recitals of instruments predating the Charter of Union. Massachusetts Bay’s and Connecticut’s both do, and Connecticut’s names four ancestors in its opening recital, of which the earliest is from 1639.

The most frequently replaced is a Gulf polity now on its fifth instrument, each replacement following a political crisis the previous one had failed to survive. The archive holds all five, and the four superseded ones survive complete nowhere else, the polity’s own copies having been lost in a fire and in a flood respectively.

The most unusual in form are the Kingdoms’. Alaska deposited its Constitution of 1889 with the accession settlement of 1884 annexed, and Hawaiʻi, after 1898, the Treaty of Honolulu, a treaty between equals. Both were deposited with covering memoranda reserving the polity’s position on whether Parliament could have required a settlement or a treaty to be filed. The Clerk entered the memoranda on the register and the question has never been decided.

The Grand Duchies’ instruments each answer the question of whose instrument it is, and each answers differently. Louisiane’s is a constitution of the polity. Michigan’s is a settlement between the House of Mackinac and the Estates, drafted so that neither may amend it alone. The four are not comparable to one another in structure, and the Realm’s comparative constitutional writing treats them as four separate traditions rather than as a type.

And the Free City of Philipsburg’s is the strangest thing in the archive, being a municipal charter that governs a polity, which is what Philipsburg is.

What the instruments do not include

The instruments of the treaty nations are not in the archive and are not deposited. A nation’s own constitution is not a polity instrument, the nation is not a polity, and the Act does not reach it. This has been raised at intervals since 1889, most often by polities arguing that a nation exercising jurisdiction inside their boundaries ought to file its constitution as they must file theirs, and it has been refused at every asking on the ground that a requirement to deposit is a requirement made of a subordinate body, which a treaty nation is not.

The Territory of Loreto deposits nothing, its status being unsettled, which means the Realm’s newest jurisdiction is the one place in it where no written instrument of government is required by law.

The archive

The deposit archive is kept by the Clerk of the Parliaments at New Westminster and is one of the Realm’s principal public records.

It holds forty-nine polity instruments, the instrument of the Columbia District, and every amendment deposited since 1890, which is a body of several thousand documents and is the most complete record in existence of how the Realm governs itself below the federal level. The March of Arizona’s file alone runs to a hundred and thirty-five amendments; several polities’ files run to fewer than a dozen; and the Gulf polity now on its fifth instrument has four superseded ones in the archive alongside the current one, which is the only place they survive complete.

The archive’s practical users are not constitutional scholars. They are the federal law officers, who consult it continuously; the polities themselves, which frequently cannot produce a clean text of their own instrument and obtain one from Parliament; litigators establishing what a provision said at a particular date; and the Realm’s electoral and boundary bodies. The Register itself has relied on it throughout.

The physical archive occupies a purpose-built room adjoining the Clerk’s office at New Westminster, the original 1890s presses having been replaced twice. Its holdings are the authenticated copies rather than the originals: a polity’s own sealed or engrossed instrument stays with the polity, and what Parliament holds is a copy certified by the polity’s own officer, which is a distinction the archive is careful about and which has mattered twice, when a polity’s original was destroyed and the deposited copy became the best evidence of what the instrument said.

The register itself is the more used document. Section 4 requires the Clerk to keep a register of what has been deposited, and the register records for each polity the date of the original deposit, every amendment with its date, and every certificate of inconsistency entered under section 6. It is the only place in the Realm where a complete amendment history of any polity constitution can be obtained, and several polities’ own record-keeping is materially worse than Parliament’s.

Consultation figures are published annually. The federal law officers account for the largest share, the polities themselves for the second largest, and the Realm’s universities for rather less than the polities’ own boundary and electoral bodies. Requests from members of the public run at a few hundred a year and have risen steadily since the archive was digitised, which the Clerk’s office attributes to school curricula and the Register’s own citations.

What the archive cannot show is the thing the polities actually did. Every instrument in it is lawful and most are unremarkable, and the exclusion that governed a third of the Realm for seventy years is almost entirely absent from them, because it was accomplished by registration practice, county discretion, poll administration and private violence, none of which is a constitutional provision and none of which was ever deposited. A complete archive of how a Realm governs itself is not a record of how its people were governed, and the March of Arizona’s file, which is the fullest in the collection, contains no document explaining why the nations of the March spent ninety years being consulted and not heard.

Assessment

The Polity Instruments Act 1889 is a good statute that was passed for a bad reason and has been useful for a hundred and thirty-nine years, and all three of those propositions are true at once.

It is well drafted. Nine sections, one obligation, a sanction that is proportionate and worked without ever being applied, a power of refusal that is genuine, and an appeal from that power to the Realm’s highest court. It prescribes what must exist and refuses to prescribe what it must say, which is the correct division for a federal Parliament and which the Realm has kept to.

It was passed as part of a retreat. The ministry that carried it was completing the withdrawal from the internal government of five polities whose direct rule had ended in 1884, without announcement and without a statute of termination, and the Act gave that withdrawal a shape and a paper record. Nobody said so at the time, and the silence with which it passed is more damning than any objection would have been, because an objection would at least have meant somebody noticed what was happening.

And its central safeguard was empty for seventy-seven years. Section 6 lets a Law Officer strike a provision from a polity constitution, and it was not used once between 1889 and 1966, because the test it applies is consistency with federal law and federal rights law did not reach the polities until 1962. The power was real throughout and could reach nothing that mattered, which is the same defect that governs the American Rights Act’s own first century and is the same defect for the same reason.

The Act in constitutional argument

The Act occupies a small but permanent place in the Realm’s constitutional argument, and it appears on both sides of the same question.

It is cited for the proposition that the polities’ autonomy is conditional. A Parliament that can require a polity to reduce its government to writing, deposit it, and have a Law Officer strike provisions from it is a Parliament exercising authority over internal government, and the polities that opposed the Extension Act of 1962 argued precisely that the Act of 1889 had been the first step and that nobody had noticed. The argument is historically sound and constitutionally awkward, because the Act was not opposed at the time by anybody, including the polities, and a precedent nobody objected to is a poor foundation for a complaint seventy years later.

And it is cited for the opposite proposition, that the Realm’s federalism is genuine, because the Act is the furthest Parliament has ever gone toward prescribing what a polity’s constitution must say and it goes almost nowhere: three requirements, a consistency test, and silence. Louisiane’s memorandum of 1896 is quoted on both sides.

The Register’s view is that both readings are available because the Act is genuinely ambiguous about what it asserts, and that the ambiguity was deliberate. A statute that had claimed a power to approve polity constitutions would have been fought. A statute that merely required them to be written down and filed was not, and it turned out, seventy-three years later, to have been the machinery through which a power to correct them could be exercised the moment federal law gave it something to correct against.

What the Act did that nobody intended is build the archive, and the archive is what made 1962 enforceable. A statute passed to give a retreat a tidy appearance produced, seventy-three years later, the single document collection that allowed the Realm to find every offending provision in every polity at once and correct most of them by letter. The Realm’s constitutional writers describe this as the Act’s redemption. It is more accurate to say that the Act was always capable of both things and that which one it did depended entirely on what federal law said, which is the lesson the Realm keeps learning and keeps having to learn again.

Chronology

Trivia

See also

• American Rights Act: whose federal-only limitation made section 2 empty until 1962
• Constitution of the March of Arizona: sealed five years before the Act by drafters who knew it was coming
• Constitution of Louisiane: deposited in 1896 under protest
• Province of Rhode Island and Providence Plantations: the last to deposit, in 1908
• The American Civil War: the rebellion, the settlement and the retreat of 1889
• Charter of Union of the American Colonies: which says nothing about polity constitutions
• Polities and the Federation
• Institutions and Government

Categories

Institutions · Law · Parliament · Constitutions · Polities · Reconstruction · Public Records