The Shape-Fallacy Fallacy

While reading Samuel Bray’s recent assessment of 20th century Prayerbook revision I was reminded of a poem by Tony Hoagland

[they]…casually dropped his name

the way pygmies with their little poison spears
strut around the carcass of a fallen elephant.
“O Elephant,” they say,
“you are not so big and brave today!”

It’s a bad day when people speak of their superiors
with a contempt they haven’t earned,
and it’s a sorry thing when certain other people

don’t defend the great dead ones

—from ‘Lawrence’ by Tony Hoagland

Bray’s claims stand in need of correction at multiple levels; I wish to defend ‘the great dead ones’ (and some living ones, too).

DEFENDING DIX

To start with, there’s the inimitable Dom Gregory Dix. Now, there is no question that Dix asserts his liturgical theses with a confidence and a certainty that historical data can never warrant, e.g. “In that form and in that order these four actions constituted the absolutely invariable nucleus of every eucharistic rite known to us.”^[1]

But that there was a shape to the liturgies of the early centuries that was something like what Dix outlines is patent. It is noteworthy that among the handful of sources of what ante-Nicene liturgy looked like in the Church, the fact that a prayer is made by the celebrant (in the ante-Nicene church, almost always the bishop) ex tempore, rather than from a fixed text is strongly suggestive of the notion that shape initially took priority over text. For instance, the famous passage in Justin Martyr’s First Apology:

There is then brought to the president of the brethren bread and a cup of wine mixed with water; and he taking them, gives praise and glory to the Father of the universe, through the name of the Son and of the Holy Ghost, and offers thanks at considerable length for our being counted worthy to receive these things at His hands. And when he has concluded the prayers and thanksgivings, all the people present express their assent by saying Amen.^[2]

This testimony is corroborated by the outline-nature and “living text” quality of the Eucharistic rite contained in the Apostolic Tradition ascribed to Hippolytus. Much hay has been made in recent years over the Apostolic Tradition, often over and against Dix and his comrades. And the scholarly points have been well taken. Contrary to the impression left by Dix’s lucid history-telling, there clearly is historical confusion and conflation over which Hippolytus is who, and there is little confidence in “Hippolytus” (whichever one) being the author of the Apostolic Tradition, still less that it definitively reveals what the liturgy looked like in Rome in the late second century.

However, Bray makes it sound like the testimony of the Apostolic Tradition is of no merit or weight in revealing early Christian liturgy, and that Dix et al. were loons for thinking so. Bray cites the famous “take-down” works of Spinks, Bradshaw, Johnson and Baldovin. Here’s Bray, supposedly giving the status quaestionis:

His idea of an invariable shape to the primitive eucharist and his treatment of the Apostolic Tradition…have been demolished by a number of liturgical scholars who are far more careful….Dix’s fundamental claim, after all, was not really a historical one—the now thoroughly debunked claim about a universal shape of the primitive eucharist.

Two things are lacking in this analysis.

One, the scholars Bray cites expend most of their labors seeking to distance the text known as the Apostolic Tradition from Hippolytus (of Rome) and the city of Rome. They do not discount the AT as a valuable insight into early liturgy, as Bray implies by saying Dix’s claims have been rejected (“demolished”). Dix’s mono-history has been complicated, but the animating idea that he traced out remains. Instance Baldovin, who concedes about the Eucharistic prayer from the Apostolic Tradition:

[T]hat some of the language, e.g., puerum (child) or angelum voluntatis tuae (angel of your will) do suggest a second-century origin for at least parts of the prayer since these words are not found in later literature.

This is no small concession from a skeptic! The Anaphora of “Hippolytus” may not have been the Anaphora of the Early Church, nor of Hippolytus, but it certainly has a structure and a feel that is of a piece with ante-Nicene liturgy, as revealed in the next strata of preserved Eucharistic texts: The liturgies of Serapion, Addai and Mari, and St. Basil, all of which have an anaphora that takes a similar shape to that of “Hippolytus” (It is also worth noting here that the “Renewed Ancient Text” of the BCP 2019 is far from being a mere translation of Hippolytus, but rather seeks to capture the spirit of early Christian anaphorai).

Two, the impulse to be hyper-critical of the Liturgical Movement’s appraisal of the Apostolic Tradition seems to have peaked in the late 1990s and early 2000s. It is noteworthy that Bradshaw and Johnsons’ seminal work on the AT is in the Hermeneia series — a commentary series that has distinguished itself in taking historical skepticism to new heights — often arguing the absence of evidence as the evidence of absence, and constructing the past as a farrago of disparate, competing traditions. But having vented their grief, the anti-Dix school gentled, and Bradshaw’s 2009 work Reconstructing Early Christian Worship is more temperate. Bradhshaw utilizes the AT text itself as a window into early liturgical development and happily concedes that:

It [The Eucharistic Prayer of the Apostolic Tradition] probably attained its final form around the middle of the fourth century and that some version of the prayer was in existence from quite early times.^[3]

This is in line with other more recent appraisals (which Bray also does not cite) which have been much more moderate and synthesizing of the historical data, like Neil O’Donoghue’s The Shape of The History of the Eucharist (New BlackFriars, 2011) which makes a theological case for reading “Hippolytus” once again in a Dixian way.

Suffice to say, taking the Anaphora of the Apostolic Tradition seriously was not just a mid-20th century fad. Discerning a rough shape common to (almost) all early liturgies was not a gross imposition on the texts themselves. What Bray calls “The Shape Fallacy” — which he uses to dismiss the 20^th—century Liturgical movement in toto — appears to itself be a fallacy of a certain school of liturgy aficionados.

But what moves me to write is not a die-hard devotion to Dix, but the way Bray uses his critique of Dix to launch an all-out assault on the Book of Common Prayer 2019. Not just the Renewed Ancient Text, but the whole thing (though, it seems he has Holy Communion chiefly in his sights).

BRAY VS. THE BOOK OF COMMON PRAYER

I take umbrage with Bray’s assessment on four fronts:

1. It fails to account for liturgical development across the divide of the Reformation
2. It grossly understates the degree of Prayerbook revision prior to the 20^th century
3. It misrepresents how much language from the 1662 BCP remains in the 2019 BCP
4. Lastly, Bray seems to fundamentally misunderstand the role of text in Divine Worship

Failure to account for liturgical development across the divide of the Reformation

Putting aside Bray’s occasional confusion between “actions” (in the Dixian sense) and Rubrically-directed gestures like kneeling, Bray’s fundamental assertion is that, “Dix’s idea that liturgy is about a sequence of actions is fundamentally foreign to the prayer book tradition.”

And yet, if we examine what Cranmer’s prayerbook of 1552 (which, as Bray rightly notes, differs almost insignificantly from the 1662) with his source material. When the 1552/1662 is compared to medieval liturgies like that of Sarum, it is quite clear that — participating in the Reformation ressourcement of patristics — Cranmer sought to discern an ancient “shape,” and rearranged and edited his received liturgical material accordingly.

Here is the sequence of prayers of the medieval (Sarum) liturgy for Holy Communion, side by side with the 1662:

The Sarum rite, through the centuries of medieval additions, constantly oscillates back and forth between personal devotion and deprecation, intercession, and acts of oblation. The focus is on the elements of bread and wine almost the entire time — before and after consecration.

Cranmer cut and paste all of this into what, by comparison, is a very orderly “shaped” liturgy — with clearly distinguished elements of preparation, Word, Intercession, Confession, and then, turning attention to the Holy Table, a brief anamnesis and recitation of the Institution, after which Communion is immediately received, and prayers of thanksgiving are offered as a sacrifice. This is not a unique analysis to me. Here is evangelical R.T. Beckwith commenting on the 1552 revision of the prayerbook:

In 1552, the structure and content of the service were changed much more strikingly than in 1549…The new structure was impressively clear and purposeful, leading up to a climax.^[5]

Substitute the near-synonym “shape” for “structure” and Cranmer and Dix appear to be confreres when it comes to reforming the liturgy.

For Dix to take the Cranmerian liturgy and run it through the patristic pot-still one more time, far from being “fundamentally foreign to the prayerbook tradition” is actually of a piece with the prayerbook tradition, when we situate it within its larger ecclesial context.

Gross understatement about Prayerbook revision prior to the 20th century

Bray gives the impression that all of the revisions to the prayerbook prior to the 20^th century were of almost no consequence:

Outside of England, the various national Anglican churches began producing their own BCPs, beginning with Scotland in 1637 and the United States in 1789, but picking up steam in the late nineteenth and early twentieth centuries. Yet there was still remarkable continuity.

Which he claims is in marked contrast to the 1928, 1979, and 2019 revisions, “But today that is no longer true of the prayer books…” This claim is misleading on both sides.

In the first place, the additions of the Scottish book of 1637, carried forward into the first American prayerbook (1789), were, liturgically speaking, enormous, and conscious “back-steps” to the patristic and pre-1552 liturgies that Cranmer so earnestly worked to distance the Anglican liturgy from.

The 1789 BCP includes:

• The Summary of the Law may replace the Decalogue

• The prayer for the King following the Decalogue is replaced with a for personal sanctity and obedience

• Prayers for the Faithful Departed among the prayers for the whole state of Christ’s Church (previously excised by Cranmer in 1552)

• Placing the Prayer of Oblation immediately after the Words of Institution. This resembles patristic liturgies in its location and content: strongly suggesting that the Sacrament itself is a part of the Church’s offering to the Father. (Cranmer removed these words to after the Distribution of Communion, and made them optional, to try and make sure that this meaning would not be communicated)

• The inclusion of an Epiclesis — the calling down of the Holy Spirit to “Bless and Sanctify (also cut by Cranmer in 1552, because too suggestive of a Real transformation).

It is hard to see how these additions can be called “continuous” (Bray’s word) when they run contrary to the express wish of Cranmer and 1662; when they restore things that had intentionally been omitted.

With this in view, it is also plain overstatement to say that 20^th—century prayerbooks are substantially less continuous. Since the 2019 BCP is foremost in Bray’s sights, let’s look at what the 2019 adds or alters structurally to the 1789. I have marked with an asterisk (*) the elements of the 2019 rite that are new introductions and a carat (^) those which are mere re-arrangements (compared to the 1789; all of them have precedent in previous 20^th—century revisions, as indicated):

The two biggest differences are:
(1) A more definite turning towards the Holy Table, as indicated by the re-insertion of The Peace and the new location of the Offertory Sentences. This under-scoring of the distinction between the hearing of the Word and the ministry at the Table is clearly continuous with the vector created by Cranmer vis-a-vis the medieval liturgies. And,

(2) the inclusion of more definite devotion following the Words of Institution — all of which is rubrically optional in the 2019 — which is clearly continuous with the vector created by 1789 vis-a-vis Cranmer.

It would appear Bray is either under-stating his disapproval of 1789 (which could hardly be said to be under the influence of Dix!) or is merely prejudiced against the 2019 BCP.

Misrepresentation of how much language from the 1662 BCP remains in the 2019 BCP

Bray lays great stress on the value of continuity with the past. And this is a principle that the Liturgy Task Force also greatly valued. What Bray seems to forget is that the 1979 revision to the BCP is also a part of our real past, and has been the liturgical lingua franca for the vast majority of Anglicans in the United States for the past forty years. To totally ignore the real existence and formative impact that the 1979 has had on the hearts and minds of Anglicans would mean to be guilty of the very thing Bray is concerned about: prayerbook revision entirely dislocated from its predecessors.

When it comes to the 2019 BCP, I sometimes wonder if we are reading the same book. The only things that are “new” to the 2019 BCP compared to the 1979 BCP are elements from the 1662 tradition that had previously been cut and have been re-introduced!

Bray writes as if the attention to the “shape” of the liturgy is necessarily exclusive to continuity of the text:

…the point is simply that the widely praised language of the prayer book is in the text, not in the shape….

the main way, the defining way, that Anglican worship has traditionally indicated the liturgical threshold […] is with the words of the BCP. And this can be lost when we move from text to shape…

If the text is constantly changing, stability and continuity will prove elusive.

If this were an analysis of 1979-Eucharistic Prayer C, I would be inclined to agree with him, but it is not. The penultimate sentence of his paper reveals his opponent:

The Dixian turn to thinking of liturgy in terms of shape was a mistake. It was also momentous, for it has strongly influenced…the BCP 2019 of the Anglican Church in North America.

Let us examine the plaintiff side by side with the 1662, to see how much the text has suffered by according to the influence of “shape.” Here are some selections from the 1662 Communion liturgy and the 2019 Anglican Standard Text:

Have the formative, memorable, pious phrases of the 1662 been mutilated beyond recognition in favor of the Dixian penchant of modern liturgists? I leave it for you to decide for yourself, dear reader.

Misunderstanding the role of text in Divine Worship

Ultimately, my main concern about Bray’s analysis is the fundamentally flawed understanding about the relationship between a text and worship.

The text of the Book of Common Prayer (or any liturgy) is not an end — but the means to an end. The end is the soul’s worship of the one true God. Bray’s plea for the sine qua non of 1662 language reminds me of those literature professors who — to use Auden’s memorable phrase — read the Bible for its prose. The King James Bible doesn’t exist to perpetuate its high language — it exists to point people to God. Same thing with the prayerbook. To fixate on the letter is deadly, it is the spirit that gives life. In the abstract, Bray might agree with this principle, but to be so aggrieved over the tiny fraction of different wordings (see the side-by-side tables above) reveals an over-attention to the letter.

Moreover, the Elizabethan grammar and pronouns in which the 1662 is written, with its long sentences and “thees” and “thous,” presents the liturgy to any contemporary listener as a museum piece. While it is the case that the sensus fidelium has in many places preserved archaic language, Bray’s argument that “liturgical and biblical texts have tended to be read in a decidedly older version of the language” is not strong. In the first place, many of the instances Bray adduces actually prove the felt need for contemporization of language: the LXX, the Vulgate (vulgar!), etc. (to which others could be added: The targumim, the Wycliffe New Testament, etc.) In the next place, the fact that Luke 1 or Ephesians 1 are not casual street-speech does not prove his point. In the ancient world, Rhetoric partially occupied the place of entertainment, and people would listen to rhetors in something like the manner with which we watch a television show (this is what St. Paul is railing against in 1 Cor 1-2). Nobody actually speaks like the characters on a sitcom, but that doesn’t imply incomprehensibility. The heightened tone of Luke 1 would be a verbal clue as to the species of text that was to ensue, but it would not have communicated a “churchy” air. Moreover, it is telling that immediately after his introduction, Luke reverts to koine street-speech — the register of the majority of the NT.

And here we come to the real lacuna in Bray’s analysis. Bray presents a straw-man of what he describes as the ‘currency objection’, namely, “that the language of the BCP is not how we talk.” This is not the actual problem that prompted the language-polishing that went into the BCP 2019. The problem is the atmosphere that a text creates. At the level of reception — and as parish-priests most of us, on the Liturgy Task Force, we have plenty of focus-group material on which to draw — In our decadent, Netflix-binging age, Elizabethan English doesn’t create an atmosphere of reverence and piety, but elitism. The ACNA is trying to bring the Gospel to Americans. Most Americans are middle-class. Most middle-class people don’t think they can read Shakespeare (even if they actually could) because they think it is for the elites.

Therefore, 1662 language would present an obstacle to the stated task of divine worship: Worship. How can the soul enter into and pick up and utilize text that, out of the gate, feels like it shouldn’t belong to her? It might be possible, with lots of time and patience and familiarity, but is there not a better, more moderate way? Could not the great language of the 1662 be polished in such a way that it could retain its heightened language, while losing some of the dusty elite/museum feel? Could weighty words (JI Packer’s phrase) be used, that would have a similar gravity to those used in 1662, and which were well beyond “street-speech”, which nevertheless were a league closer to home than the lexemes and syntax of the 16^th and 17^th centuries? This was what the BCP 2019 attempted, and I believe successfully accomplished.

To oppose using the 1662 text (which, is ironic if not impossible for any American, given the manner in which fealty to the King is an inextricable element of its piety) does not mean that liturgy should be plebiscite, it only means — in the case of the 2019 — that a slight contemporization might be a better vehicle for presenting the spirit of the four-centuries-old Anglican prayer tradition, for actual use by actual North Americans today as they seek to worship. Worship is an action, not a text. To call out a shape-fallacy when speaking of the BCP 2019 is itself fallacious.

1. Dix, Dom Gregory The Shape of the Liturgy (London: Dacre, 1945), 48 (emphasis mine) ↑
2. Justin Martyr, First Apology 65 ↑
3. Bradshaw, Paul Reconstructing Early Christian Worship (Pueblo Reprint, 2010), 51 ↑
4. Even though 1552 not 1662 was Cranmer’s last work, 1662 so closely resembles 1552, and is the foundational prayerbook to which Bray is appealing, and therefore more useful to analyze in detail ↑
5. Beckwith, R.T. “Th Anglican Eucharist: From the Reformation to the Restoration” The Study of Liturgy Eds. Jones, Wainwright, Yarnold (OUP, 1979), 267. ↑