Revisiting “The Shape Fallacy”: A Response to Ben Jefferies

Note: Because the Davenant Institute published Bray’s initial essay and The North American Anglican published Jefferies’ response, we have both published Bray’s rejoinder.

By Samuel L. Bray

In a response to my recent essay “The Shape Fallacy: Reconsidering the Book of Common Prayer as Text,” the Rev. Ben Jefferies offers hearty disagreement and eloquent rejoinder. These are always welcome.

Jefferies calls my essay “an all-out assault on the Book of Common Prayer 2019.” It manifestly is not. I refer to the ACNA book on occasion (twice in text, once in a parenthetical, and once in a footnote), but my argument is much broader. In support of this narrow and mildly obsessive reading, Jefferies quotes the penultimate sentence of my essay, saying it “reveals” my “opponent.” Yet the crucial words that show the true scope of my thesis are replaced in the quotation with an ellipsis. I said that the Dixian turn “strongly influenced every subsequent prayer book revision, including the BCP 1979 of The Episcopal Church and more recently the BCP 2019 of the Anglican Church in North America.” What I am quoted as saying is that the Dixian turn “strongly influenced. . . the BCP 2019 of the Anglican Church in North America.”

In my essay, I recognized that there are gains from the turn to thinking about the Book of Common Prayer in terms of shape, especially contextualization and freedom for creativity. But I also surveyed what has been lost, and I described six losses: first and paradoxically, the shape of a prayer book service; second, the excellence of the language; third, the stability of the text, with implications for intergenerational transmission and catechesis; fourth, protection for the laity; fifth, the tradition of prayer book manuals and commentaries; and finally a centering function, for the text of the prayer book can provide a settled center for Anglicanism. In addition to this recounting of losses, I asked who bears their burden. (Spoiler alert: it is the laity, not the liturgists.)

And I raised the inevitable objection that the language of the traditional prayer books is too hard, and I pulled apart two different forms of the objection. One is that the language is unintelligible (the comprehension objection). The other is that the language is not how we talk today (the currency objection). Each objection has a different force, and only one of them (I argued) should be granted. I refer anyone who is interested in the language of the prayer book to my discussion of these two objections.

“The Shape Fallacy” was published in the journal of the Davenant Institute, which is called Ad Fontes. Jefferies and I apparently agree in encouraging a return to the sources. Jefferies helpfully refers at numerous points to two prayer books: the one that has a central place in the Anglican Communion, BCP 1662; and the first prayer book for the United States, US BCP 1789. I fully agree with Jefferies in directing the reader’s attention to these sources. Look at their daily offices; look at their calendars of saints; look at their use of the traditional Western eucharistic lectionary; look at the sacraments of Holy Baptism and Holy Communion; look at the Visitation of the Sick and the Burial of the Dead; look at their Psalms. I dare the reader to discover any other dominant theme than continuity.[1] That continuity, though diminishing a little over time, continues strong through the US BCP 1928.

Once the reader has examined these prayer books, I encourage comparison with the late modern ones, such as the US BCP 1979 and the ACNA BCP 2019. They are profoundly and pervasively different. And opinions will vary: some people will favor the earlier books, and some will prefer the late modern ones.

But the differences are not just about a few words here and there, an emending of an epiclesis or a replacement of a collect for a king. The older books that Jefferies cites—whether of 1637 (Scotland) or 1662 (England) or 1789 (US)—offer common prayer. It is true that the text varies in places from book to book, but upon even slight acquaintance with them the reader will find massive continuity. And, more to the point of my argument, each one of these texts—in its daily and sacramental offices—provides the virtues of a common text, not the quite different virtues of a highly flexible and variable shape.

To be clear, neither my essay in Ad Fontes nor this reply is a review of the ACNA book. It has pros and cons. It is a revision of the 1979 prayer book of the Episcopal Church, though at various points there are adjustments to bring it closer to the classic prayer book tradition. But I am not writing a review of the 2019 book.

I am concerned with something bigger than any one late modern prayer book: how the Dixian shift to thinking of the prayer book in terms of “shape” has affected the virtues of the prayer book tradition. I did note in my essay that different people will weigh differently the gains and losses from the late-twentieth-century shift. I am only trying to provide a closer accounting of the losses, and then others can do the weighing for themselves. A more fruitful criticism of my piece would be that I have overlooked gains from modular, option-filled, highly variable prayer books. Perhaps so.

“I am concerned with something bigger than any one late modern prayer book: how the Dixian shift to thinking of the prayer book in terms of ‘shape’ has affected the virtues of the prayer book tradition.”

There is one final substantive point to mention. Jefferies uses the “elitism” label for the use of traditional language in worship, and he is concerned that some who hear such language may associate it with the difficulty of reading Shakespeare. Here are two possible misunderstandings.

First, traditional language versions of the Book of Common Prayer—whether BCP 1662, US BCP 1928, US BCP 1979 Rite I, or the forthcoming traditional language version of ACNA BCP 2019—are nothing like the Bard. As David Martin once wrote, “Cranmer is as simple as: ‘O God our help in ages past’. People talk loosely of ‘beautiful Shakespearian English’ when in fact Cranmer is not at all like Shakespeare and very much more simple.”[2]

Second, the evidence, at least in Anglophone contexts, does not show that the affection for traditional religious language is associated with “elites.” Among all Christians in the United States, it is black Christians who are most likely to use the King James Version.[3] I am told that in Canada the older liturgical texts are most pervasively used by the indigenous peoples. If the point can be extended to gender, a study of the reception of contemporary Anglican liturgies found women more receptive than men to traditional language.[4] And in the first of his Letters to Malcolm, C. S. Lewis notes that linguistic revisions in prayer books can leave behind both the most and the least educated. More generally, in liturgical revision, the clergy are usually applying the gas pedal while the great body of relatively powerless laity are trying to apply the brake, and that holds with respect to linguistic revision. If there is a secret elite affection for the traditional-language hymns of Watts, Wesley, and Crosby, the point has escaped my notice, but it would surely be of interest to those who spend their time tabulating cultural oddities.

One final point. I think it is important not to impugn the thoughtfulness, good will, or mission-mindedness of those who prefer prayer books in either traditional or contemporary language. My essay did not say that everyone should use traditional language. To the contrary, I said: “My argument is not that it must be so in religious speech, but that it may be so.” On this point charity and generosity should prevail.


Samuel L. Bray is a Professor of Law at Notre Dame Law School, as well as a McDonald Distinguished Fellow at the Center for the Study of Law and Religion at Emory University.

[1] To support a claim of discontinuity, Jefferies gives a description of the US BCP 1789 that is not wholly reliable. The first US prayer book did not allow the summary of the law to replace the decalogue (it was a permissible addition). The reader may receive the impression that the Americans inserted “Prayers for the Faithful Departed” based on a Scottish model, but the prayer for the church militant is identical in BCP 1662 and US BCP 1789. And the invocation in US BCP 1789 shows Scottish influence but is pulled back toward the BCP 1662. By any measure it is incorrect that the changes that were “enormous” (emphasis in Jefferies), but the evidence given, such as it is, rests on a conflation of two very different categories: (1) points at which the US BCP 1789 was following BCP 1662, which had in turn taken up an innovation in BCP 1637; and (2) points at which US BCP 1789 followed the 1764 Scottish Communion office. For more careful assessments, see Drew Keane, “Seabury and the Scottish Liturgy,” The North American Anglican (Nov. 11, 2019); Laudable Practice, “PECUSA and the Scottish Myth” (Nov. 14, 2018).

[2] David Martin, “Why Spit on Our Luck?,” Poetry Nation Review, vol. 6, no. 5 (1979), p. 3. Similar points have been developed by others in criticizing the late modern prayer books. Margaret Doody said: “The old liturgy is being rejected not for archaism or dryness but fundamentally because it exerted too much dramatic force on the emotions.” Margaret Doody, “‘How Shall We Sing the Lord’s Song Upon an Alien Soil?’: The New Episcopalian Liturgy,” in The State of the Language (Berkeley: University of California Press, Leonard Michaels and Christopher Ricks eds., 1980), p. 123. And: “In any case, the archaic character of the traditional Prayer Book’s diction has been much exaggerated. A few words are indeed archaic, but not necessarily unintelligible. . . . Actually, the real difficulty in understanding the old Prayer Book is not the presence of a handful of archaic words and forms. The real difficulty lies (1) in the figurative language, which the revisionists have been able to do little about, and (2) in the statements of the deep mysteries of the faith, a problem with which no prayer book in and by itself can satifactorily cope.” Cleanth Brooks, “God, Gallup, and the Episcopalians,” The American Scholar, vol. 50, no. 3 (Summer 1981), pp. 317-318.

[3] Kate Shellnut, “Black Bible Reading Endures,” Christianity Today (January 1, 2019), p. 16 (“The King James Version—the top-selling translation in the US overall—remains by far the most common Bible used by black churches, with 42 percent of African-Americans preferring it and just 11 percent preferring the second-most-popular choice, the New International Version, according to [the American Bible Society].”).

[4] Peter Nicholas Davies, Alien Rites? A Critical Examination of Contemporary English in Anglican Liturgies (Aldershot: Ashgate, 2005).