A NOTE ON THE PRAYER BOOK TRADITION

The Tests of a Prayer Book

Five ways of judging a revision — where each comes from, what it can settle, and why they do not always agree.

Arguments About Books Are Arguments About Rulers

When two people disagree about whether a prayer book is legitimate, they very rarely disagree about the facts. Both know that the American Prayer Book of 1789 took its Prayer of Consecration from Scotland; both know that the Thirty-Nine Articles stand at the back of the 1928 Book; both know what the 1979 Book did to the shape of the Communion service. What they disagree about is which of these facts is decisive — that is, which test a prayer book has to pass.

Five such tests are in common use. They are seldom named, and almost never named together, which is why arguments about prayer books so often go in circles: each party is measuring with a different rule and assuming the other party has simply miscounted.

Naming them does not settle any quarrel. It does something more useful. It lets a reader hear an argument and know at once what is being claimed, what would count against it, and what has not been addressed at all.

I. Descent — The Test of the Text

What it asks. Is this book an amendment of the book before it, or a new composition that happens to serve the same purpose?

The test comes out of the way the English books were in fact made. The Book of 1552 is the Book of 1549 revised; 1559 is 1552 with alterations; 1662 is 1559 gone over clause by clause after a century of quarrel. Nobody in that line began again. Each reviser had the previous text in front of him and worked upon it, so that a man who knew one book could find his way about the next.

The American books were made the same way. The Book of 1789 keeps the framework of 1662 and inserts a single new prayer into it. The Book of 1892 is a conservative revision that adjusts rubrics and encourages more frequent Communion. The Book of 1928 rearranges more boldly — it restores the Lord's Prayer to follow the Canon, moves the Prayer of Humble Access to stand immediately before the Communion, and brings back matter that had been absent since 1552 — but it is still an amendment of an inherited text, and a reader of 1892 opening it in 1929 was not lost.

The test has an edge, and it is worth stating plainly, because the sharpest thing that can be said about the Book of 1979 was said by the men who made it. One of its principal architects concluded by the early 1960s that Cranmer's work was no longer adequate and that they would have to start from the beginning. That is not an accusation from outside. It is a description from within, and it is what the test of descent is designed to detect.

Where the test is weak. Descent judges method, not truth. A book can be impeccably descended from its predecessor and doctrinally poor, or badly descended and devout. The test is also silent about the first book in any line: it can tell you that 1552 amends 1549, but it cannot tell you whether 1549 should have existed. And it can be made to prove too much, since every revision alters something, and a sufficiently strict reading of descent would forbid revision altogether.

Applied to 1928. It passes. This is the strongest ground on which the book stands, and the ground its defenders are usually standing on when they call it the last book of the old family.

II. Doctrine — The Test of the Formularies

What it asks. Does the worship of this book agree with the doctrine the Church has stated elsewhere?

Anglicans have no confession of faith in the Lutheran or Reformed sense. What they have instead are formularies — the Prayer Book, the Articles of Religion, and the Ordinal — which are meant to stand together and interpret one another. The test of doctrine takes the Articles as the statement and asks whether the Prayer Book's prayers can be said honestly by a man who holds them.

Two Articles do the most work here. Article XXVIII, Of the Lord's Supper, denies transubstantiation and says that the Sacrament was not by Christ's ordinance reserved, carried about, lifted up, or worshipped. Article XXXI, Of the one Oblation of Christ finished upon the Cross, holds that the offering of Christ once made is the perfect redemption for all the sins of the whole world, and that there is no other satisfaction for sin but that alone.

Applied to 1928 — and here the book strains. The 1928 Prayer of Consecration contains an explicit offering of the bread and wine and an invocation of the Holy Spirit upon them, and the book restores prayers for the departed that the American books had not carried since they followed 1552 on that point. Whether those things can be reconciled with Articles XXVIII and XXXI is a real question, and it has never been squarely answered. It was not answered in 1928: the Convention that ratified the book also declined to remove the Articles from it, so the two were bound together without any ruling on how they were to be read. It has not been answered since.

That is worth saying flatly rather than defending. The honest account of the 1928 Book is that it carries an unresolved tension in its own covers, and that a reader who feels the tension is not imagining it.

Where the test is weak. It presumes the formularies are themselves settled and binding, and in the American Church they never clearly were. The Constitution of 1789 named the Articles among the things that should be used once established; the Convention of 1801 established them without saying what "established" meant; a proposed canon of 1804 that would have required clergy to subscribe them was voted down on the recorded impression that the ordination promise already covered the matter — a reasoning that lives only in the minutes and was never enacted. Candidates were examined on the Articles from 1871, and the requirement was dropped in 1904. In 1907 a committee of the Church proposed removing the Articles from the Constitution altogether, reporting in so many words that they were antiquated without being ancient and a bar to Church unity at home and abroad; the proposal failed only because the House of Bishops declined to concur.

So the test of doctrine is the sharpest of the five, and it rests on the least secure footing. That is an uncomfortable combination, and it is why arguments conducted on this ground tend to become heated without becoming conclusive.

III. Antiquity — The Test of the Early Church

What it asks. Does this book recover the practice of the Church before the divisions?

The appeal is as old as the tradition itself. Cranmer opens the Prayer Book by appealing to the godly and decent order of the ancient fathers, and claiming that what he offers is nearer to their mind than what it replaces. Two centuries later Bishop Seabury made the same appeal his own condition. When the Scottish bishops asked him to consider their Communion Office, he undertook to look at it seriously and to give it his sanction — but only, in his own words, if it were found agreeable to the genuine standard of antiquity. He did not promise to adopt it because they wished it. He promised to test it.

The test has done real work. The invocation of the Holy Spirit in the American canon is defended on this ground and on no other; so is the offering of the elements. Neither is defended as a recovery of 1662, because neither is in 1662.

Where the test is weak, and this is the important part. Antiquity is the most easily abused of the five, because every generation's scholarship discovers a different antiquity, and a book built on the antiquity of its own decade ages with that decade.

The clearest instance is within living memory. In 1945 a fourfold shape was proposed as the structure underlying all the ancient eucharistic liturgies — take, give thanks, break, give — with the claim that the unity of the early Church lay in that shape rather than in any form of words. For a generation this was treated as bedrock, and revisers across several communions aimed at it. It is now heavily qualified. The document on which much of the reconstruction depended was confidently ascribed to a third-century Roman writer and is now generally held to be a composite of uncertain date and provenance, and the tidy fourfold pattern is no longer thought to describe the ancient rites as neatly as it seemed to.

None of that is an accusation of bad faith. It is what happens when a church builds on a moving foundation. The lesson cuts in more than one direction: it is a caution to anyone revising a book today, and it is also a caution to anyone defending a feature of an older book solely on the ground that the Fathers did it.

Applied to 1928. Mixed. The book passes on its own terms — its restorations were made deliberately, on stated patristic grounds, by men who had read the sources available to them. But it shares the vulnerability, and its defenders are not entitled to use the antiquity argument for the epiclesis while refusing it to anyone else.

IV. Authority — The Test of the Making

What it asks. Was this book made by the body competent to make it, in the manner that body's own law prescribes?

This is the most mechanical of the five and the easiest to settle. In the American Church the machinery is deliberately slow: a change to the Book requires the assent of two General Conventions three years apart, in both Houses. The slowness is the point. It is the same machinery that defeated the attempts to remove the Articles, which had to clear the same two readings and never did.

By this test the Book of 1928 is unimpeachable. It was proposed, debated at four successive Conventions, amended, and ratified by the body that alone had the power to ratify it. It is the lawful book of the Church that made it.

Where the test is weak. Authority is jurisdictional, not doctrinal. It establishes that a book is lawful; it cannot establish that it is good. This limitation is not academic. The Act of Uniformity of 1662 gave the English Book the fullest legal authority any prayer book has ever had, and upwards of two thousand clergy left their livings rather than conform to it. Lawfulness settled the question of what the book was. It did not settle whether those men were right.

And there is a harder case, which anyone in a Continuing church has to face squarely. The Book of 1979 passes this test completely. It was made by the proper body, in the proper manner, through the full constitutional process. Those who declined to receive it were not appealing to authority; they were appealing to the other four tests over it — to descent above all, and to reception. That is a coherent position, and it is the position the Continuing churches in fact occupy. But it should be occupied knowingly. A man cannot invoke lawful authority when it favours his book and set it aside when it does not, and expect to be taken seriously either time.

Applied to 1928. Passes cleanly.

V. Reception — The Test of Use

What it asks. Do the faithful actually pray it?

This is the oldest test of all and the least formal. A prayer book is not a treatise; it is a thing said aloud, week after week, by people who mostly did not vote on it. Whether it takes root in their mouths and memories is a real fact about it, and no committee can legislate the outcome.

Two episodes show the test working independently of authority, in opposite directions. The English revision of 1928 was rejected by the House of Commons, twice — in December 1927 and again in June 1928 — and never obtained legal sanction. Its use nevertheless spread widely through the middle of the century. Authority said no; reception said yes.

The American case ran the other way. The Book of 1979 obtained every lawful sanction, and at the very Convention that approved it the Church authorized bishops to permit continued use of the 1928 Book. Nearly fifty years later that book is still in use, in this jurisdiction and others. Authority said yes; reception, in part, said otherwise.

Who administers it. This is worth pausing over, because it is what distinguishes reception from the other four. Descent, doctrine, antiquity and authority are tests for clergy and scholars: they require a man to have the texts open in front of him, to know what the Articles say, to have some idea what the fourth century did. Reception requires none of that. It is administered by the whole congregation, and it is administered by praying — week after week, for years, until the words are in the memory and come unbidden at a graveside.

Most people who love a prayer book have never argued about one, and are under no obligation to start. They are not standing outside this framework. They are operating its fifth instrument, which is the slowest of the five, the least articulate, and the only one no committee can overrule.

Where the test is weak. It is slow, so it can say nothing at all about a book that is new — which is exactly when a judgment would be most useful. It can mistake familiarity for fitness, and habit for conviction. And it is not self-interpreting: a book kept in use by a small and determined body and a book kept in use by a whole province are both, in some sense, received.

Applied to 1928. Passes, and this is the book's plainest claim. It was prayed for half a century by an entire church and has been prayed for another half-century since by those who would not give it up. Whatever else is disputed, that is not.

What the Tests Do Not Reach

Before the results are set side by side, a limit has to be drawn, because these five instruments are easily turned to a use they were not made for. They judge books. They do not judge parishes.

That distinction is not a courtesy. It is the settled doctrine of the formularies themselves, stated twice over in the very book this parish uses.

Article XXXIV

The Articles take it up directly, under the title Of the Traditions of the Church:

It is not necessary that the Traditions and Ceremonies be in all places one, or utterly like; for at all times they have been divers, and may be changed according to the diversity of countries, times, and men's manners, so that nothing be ordained against God's Word.

And the Article's closing sentence gives both the permission and its purpose:

Every particular or national Church hath authority to ordain, change, and abolish, Ceremonies or Rites of the Church ordained only by man's authority, so that all things be done to edifying.

The phrase at the end is Saint Paul's, from the fourteenth chapter of the first Epistle to the Corinthians, where he is regulating a congregation whose worship had become disorderly: let all things be done unto edifying, and a few verses later, let all things be done decently and in order. The liberty is real, and it is given for a purpose. A parish may order its worship as best builds up the people who are actually in that room. It is not licensed to do as it pleases and call the result adaptation.

The Preface to the American Book

The framers of 1789 made the same point the first thing they said. The Preface printed in every American Prayer Book since, including the one on the shelf in front of you, opens like this:

It is a most invaluable part of that blessed "liberty wherewith Christ hath made us free," that in his worship different forms and usages may without offence be allowed, provided the substance of the Faith be kept entire; and that, in every Church, what cannot be clearly determined to belong to Doctrine must be referred to Discipline; and therefore, by common consent and authority, may be altered, abridged, enlarged, amended, or otherwise disposed of, as may seem most convenient for the edification of the people.

Three things are being said at once, and they are worth separating. Different forms and usages are allowed without offence — that is the liberty. The substance of the Faith must be kept entire — that is the limit. And what is not clearly a matter of Doctrine is a matter of Discipline, alterable for the edification of the people — that is the rule for telling which is which.

Where the liberty ends

It ends in the same documents that grant it, and the limits are as plain as the permission.

Article XXXIV rebukes anyone who, through his private judgment, wilfully and openly breaks ceremonies which are not repugnant to the Word of God and have been ordained by common authority. The liberty belongs to the Church ordering its own worship; it is not a private licence.

And the Preface, describing the English revisions it is following, names what must not move: through all the alterations made in several reigns, the main body and essential parts of the Liturgy — in the chief materials and in the frame and order of it — have been continued firm and unshaken. The framers commend seeking a happy mean between too much stiffness in refusing variations and too much easiness in admitting them.

So the doctrine is not that anything may be done. It is latitude in ceremony and usage, constancy in the text and frame. And that distinction is exactly what keeps the five tests worth applying: if "services need not be everywhere the same" extended to the words of the rite, there would be nothing left for descent, doctrine, antiquity, authority or reception to measure.

What this means in practice

Two congregations may both use the 1928 Book and look very different from one another. In one the introits are sung and the ceremonial is full; in another they are said plainly and the ceremonial is spare. One has a robed choir and a procession; another has a said service at eight o'clock with six people. Neither is in default. Neither is evidence about the other's fidelity. The book is the same book, and the Articles say in so many words that it was never required to look alike everywhere.

Which yields a rule for using everything in this study: a man who takes these five tests and applies them to another parish's ceremonial has misused them. They were built to weigh books, and against a book they will bear real weight. Against a neighbour's manner of celebrating they have no jurisdiction at all, and the formularies say so.

VI. Why the Tests Disagree

Set the results beside one another and the shape of every argument about the American books becomes visible.

The 1928 Book passes descent, authority and reception; strains on doctrine; and is mixed on antiquity. That is not a verdict of innocence and it is not a verdict of guilt. It is the actual position, and it explains why the book has been argued about for a hundred years by people who were all, within their own terms, arguing correctly.

Someone who reads the table across a row is doing history. Someone who reads it down a single column and calls the result a verdict is doing something else.

The part that should give a champion of 1928 pause

There is one more thing the record shows, and it is not comfortable.

In 1907, a committee of the Episcopal Church reported that the Articles of Religion were antiquated without being ancient and stood in the way of unity, and proposed to strike them from the Constitution. In February 1928, a gathering of Liberal Episcopalians in Philadelphia warned of outright schism and accused the Anglo-Catholic party of a campaign to seize the Church. Later that year a petition claiming better than thirty-four thousand signatures was laid before the General Convention, begging that the Articles be kept where they were. Not one of the competing proposals carried, and the Articles stayed.

The Book of 1928 was the accused. The arguments now made against later revisions — that a party has captured the machinery, that the received standard is being quietly set aside, that what is offered as development is really rupture — are the arguments that were made against the book this parish uses, by men who believed them as sincerely as anyone believes them now.

That does not make those arguments wrong. Some of them were right then and some are right now. It does mean something about how they should be handled. A man who takes up these tests should know that they were once turned on his own book, and should apply them to another man's with the care he would have wanted applied to his.

VII. What This Is Good For

Not for settling anything. The tests conflict, and no ranking of them can be derived from the tests themselves; where a man puts descent above authority, or reception above doctrine, he is making a judgment that the instruments cannot make for him.

They are also not for everybody, and it would be dishonest to pretend otherwise. Four of the five are instruments for people who work with these texts — priests, students, anyone who has been handed a claim about a prayer book and wants to know what would count against it. A parishioner is not worse off for never having heard of them. He already holds the fifth.

What they are good for, for those who want them, is understanding — and it is a practical kind. A parishioner who hears that some new book is or is not in the tradition can now ask which tradition is meant, and by which measure. A reader working through the history of the American books can see why 1789 moderated what it received, why 1892 changed so little, why 1928 was fought over so bitterly, and why 1979 was received as a break rather than a step. A priest asked why we use this book and not another can answer with something better than preference.

And the answer for this parish, given honestly, is not that the 1928 Book passes every test. It is that it passes most of them, that we know precisely where it strains, and that we would rather pray a book whose weaknesses we can name than one whose strengths we have only been told about.