THE FOURTH VOLUME of The Complete Works of William Billings is an edition of The Continental Harmony, published by the Boston firm of Thomas and Andrews in 1794.1 The last of Billings’s tunebooks, The Continental Harmony is also his longest at 200 pages, the most wholly devoted to large-scale works with seventeen anthems, and the only one of Billings’s collections printed from moveable type rather than engraved plates. Unlike his earlier tunebooks, which he brought out as commercial ventures, The Continental Harmony was published under the sponsorship of a committee of Boston choristers as an act of charity toward Billings and his family.

In the Introduction to Volume III of this edition, The Continental Harmony was described not as a collection of newly-composed pieces but as “a retrospective compendium of [Billings’s] creative achievements.”2 Indeed, some of the pieces in The Continental Harmony are found in manuscript copies dating from the 1770s and early 1780s.3 The musical style of others suggests that they may also date from early in Billings’s career. On the other hand, some of the anthems and more complex psalm-tunes show a deftness at handling harmony, counterpoint, and form that appears in none of Billings’s earlier compositions, suggesting that they were composed closer to the time of their publication. Both the variety of the contents and Billings’s technical mastery of musical elements in The Continental Harmony justify the encomium applied to him in Volume III: “elder statesman” of American psalmody.4

In the editor’s Introduction that follows, two separate essays appear, the first dealing with The Continental Harmony and the second with the performance of Billings’s music. The first touches upon Billings’s personal situation in the early 1790s, his relationship with the Danish immigrant musician Hans Gram, his negotiations with the printers Isaiah Thomas and Ebenezer T. Andrews, and the eventual publication of the tunebook by others on Billings’s behalf. The second examines evidence from Billings’s time—evidence found not only in his own writings but also in those of other psalmodists—on the ideals that governed the performance of eighteenth-century New-England psalmody.

The order of pieces in the volume follows that of the tunebook, except that a transcription of the frontispiece tune, Connection, is included as an appendix. The Title and First Line indexes at the end of the volume are comprehensive, covering all four volumes in the set. The Index of Facsimiles, however, refers only to Volume IV.

With this volume, The Complete Works of William Billings, the first critical edition of the works of an American composer, is concluded, and a project begun in honor of the American Revolution’s Bicentennial comes to an end. Now that all of Billings’s music has been edited and studied, its significance comes more sharply into focus. Like other New-England psalmodists of his generation, but perhaps more fully than they, Billings imagined and created a world of musical sound into which Americans of his day could fully enter as participants and listeners. He composed a peoples’ music, based on religious texts and sung not only in churches, singing schools, and concerts, but also in homes, workplaces, and even, as Irving Lowens has suggested, in inns and taverns.5 Billings’s music served the common folk as both religious expression and popular entertainment. Indeed, it would probably be fair to say that it reached and moved more people in its time and place than the symphonies of Mozart or the cantatas of J. S. Bach did in theirs. In compositional technique, psalmody can be seen to stand in a kind of “prima-prattica” relationship to the harmony-based European music of the day,6 although its small scale, modest forces, and historical position at the very start of an American tradition of composing may make comparisons with eighteenth-century European masterworks seem exaggerated.7 Perhaps it is enough to say that Billings’s music was well-suited to the circumstances for which it was composed and that, as time has passed, it has eventually transcended them.


publication history

For Billings, the 1780s was a time of declining fortune. The relative affluence he seems to have enjoyed during the decade’s early years—of which the purchase of a house and the issuing of several collections of his music are evidence—seems to have melted away. Other than teaching singing schools and selling his tunebooks, we do not know if Billings worked at other occupations. Although he was apprenticed to a tanner as a youth, there is scant evidence that he actually followed this trade in later life. His physical handicaps—a withered arm, one leg shorter than the other, a blind eye—would have ruled out some types of labor. That, in addition to his musical activities, he served as editor of the Boston Magazine in 1783, however briefly, suggests that he aspired more to intellectual pursuits than physical labor.8

The reasons for Billings’s financial problems, although they cannot be documented with certainty, must have been manifold. Certainly, his growing family was a major factor.9 Public taste probably also played a role. Although his earlier psalm-tunes won wide popularity, and The Singing Master’s Assistant (1778) went through four editions, his own collections of the 1780s—The Psalm-Singer’s Amusement (1781) and The Suffolk Harmony (1786)—appear not to have aroused much enthusiasm. Singing schools were seasonal institutions in eastern Massachusetts, usually beginning sometime in late fall after the harvest and ending in early spring around the time of planting.10 Thus, there was only a limited period of activity, and since Billings was not itinerant, so far as we know, the range of his teaching must also have been restricted to Boston and its environs. No notices of Billings’s schools appear in the Boston newspapers of the time. Perhaps arrangements were made in person between Billings and church committees or individuals, another factor that may have limited his range. Finally, although Billings was the best-known Boston psalmodist, he was not the only one in town, and as the next decade passed competition grew more intense.11

To supplement his income from psalmody, Billings held several municipal appointments in the late 1780s. The most enduring of these was as “Sealer of Leather for the City of Boston,” under which office he was to inspect all tanned leather for proper curing and confiscate that which was improperly prepared. Billings occupied this post from 1787 to 1797, except for 1793. He also acquired other minor public posts, including scavenger for Ward 11, inspector of the trade, and perhaps “hogreeve.”12 None of these positions was of long duration, and by 1790 Billings’s financial condition caused alarm among his supporters. A benefit concert for Billings was announced in the Columbian Centinel of December 8, 1790:

A correspondent was very happy in hearing that a number of benevolent characters are determined to bring forward a Concert of Sacred Musick, for the benefits [sic] of Mr. WILLIAM BILLINGS, of this town—whose distress is real, and whose merit in that science, is generally acknowledged.13

Later announcements set the concert for December 21 at 3 o’clock at the Stone Chapel, one of Boston’s principal Anglican churches. The Chapel’s organist, William Selby, perhaps the best trained musician in Boston at the time, probably directed the music.14 We know nothing of the result of the concert: what pieces were performed or how much money was raised for the Billings family. An indication that the outcome was insufficient to alleviate his financial plight is the fact that Billings mortgaged his house for £40 in January 1791.15

In 1791, Billings approached the publishing firm of Thomas and Andrews through an intermediary, Hans Gram, proposing to sell them all of his music—both the pieces he had already published and a considerable quantity still in manuscript.16 Apparently Isaiah Thomas, the senior partner of the firm, declined the offer. Thomas and Andrews did eventually publish Billings’s final tunebook, The Continental Harmony, but under conditions quite different from what Billings probably imagined in 1791.

That Billings approached Thomas and Andrews through Gram suggests a relationship between the two musicians and invites a further look at the latter.17 Gram was the immigrant son of a well-to-do Danish family who had had the advantages of a good education.18 He was born in Copenhagen in 1754, but died in poverty in the Boston Alms House in 1804. Although little is known about his early life, he served as secretary to the governor of the Danish West Indies before he came to Boston, suggesting a background of some social standing. He arrived in Boston in 1785 to settle some family business affairs. Why he decided to stay, marry, raise a family, and pursue the occupation of church organist and music teacher is unknown.

Very likely, in his youth Gram had received orthodox musical training on keyboard instruments and in thoroughbass.19 His music, what little of it was published, shows him to have been a competent composer in the thoroughbass tradition. Choruses such as The Pilgrim20 and Edmund21 are treble-dominated settings featuring greater harmonic variety than American-born composers normally attempted, but their melodies lack sweep and vigor. Hymn tunes such as Worship22 and Zion’s Praise23 have more the stolid character of Lutheran chorales than Anglo-American psalm-tunes. Gram’s anthem, Bind Kings With Chains,24 is segmented like Anglo-American parish-church anthems, but the voices lack the polyphonic independence usually found in these works. We may take Gram’s admiration of the “9 or 10 tunes . . . [he] played on his Harpsichord”25 as recognition of Billings’s talent from someone with a superior musical education. In point of fact, however, Gram and Billings inhabited different musical worlds. Technically, Gram’s music is keyboard oriented, employing voice leadings and harmonic progressions derived from thoroughbass procedures. Billings, for all his efforts to approach a more orthodox harmonic style, never abandoned the additive compositional process he mastered in the 1770s and used throughout his life. Gram may have assisted Billings in some minor technical matters, but whatever their relationship may have been, he did not materially affect Billings’s basic approach to composition.26

Thomas and Andrews, publishers of Billings’s final tunebook, are widely recognized as the most important American publishing house of the Federal era. In 1788, Isaiah Thomas, the firm’s senior partner, who lived in Worcester, established a printing office in Boston under the supervision of Ebenezer T. Andrews, a former apprentice who was then taken into partnership.27 The Boston branch of Thomas’s business soon outstripped his Worcester office in size and importance. Thomas, however, remained in Worcester, where he controlled the Boston branch’s activities through a frequent exchange of letters between the two partners. Much of Andrews’s side of the correspondence is preserved in the Isaiah Thomas Papers at the American Antiquarian Society.28 Although almost all of Thomas’s letters are lost, many of his replies, queries, and decisions can be inferred from Andrews’s comments and subsequent events. References to Billings and his music begin to appear in Andrews’s letters in 1791.

It seems likely that Thomas established a relationship with Billings well before 1791. Although Thomas was not a musician, he was a keen observer of the church music of his day. In 1786, he published The Worcester Collection of Sacred Harmony, a tunebook which for the next decade was unrivalled in influence and popularity in New England.29 In the preface of the first edition, Thomas acknowledged Billings’s preeminence in American psalmody:

For the progress of Psalmody in this country the Publick are in great measure indebted to the musical abilities of Mr. WILLIAM BILLINGS, of Boston: It is but doing him justice here to observe, that he was the first person we know of that attempted to compose Church Musick, in the New-England States; his musick has met with great approbation. Many tunes of his composing are inserted in this work.30

The second edition of The Worcester Collection (1788) contained a tune by Billings that had not appeared in any of Billings’s previous tunebooks. Union, a setting of three stanzas of Isaac Watts’s versification of Psalm 148, shows the earmarks of the style Billings employed in The Suffolk Harmony (1786), suggesting that it was a recent composition.31 Billings’s compositions circulated widely in manuscript, and Thomas may have come across a copy of Union, perhaps through Worcester-area psalmodists like Elias Mann, Joseph Stone, or Abraham Wood,32 found it attractive, and decided to publish it. However, because of Thomas’s high regard for Billings’s ability and the subsequent dealings between them, it seems more likely that he secured the copy directly from Billings. Union was not credited to any composer in the second edition, but when it appeared in the third edition of The Worcester Collection (1791) it carried Billings’s name.

The beginning of the sequence of events that led to the publication of The Continental Harmony is found in Andrews’s letter to Thomas of June 19, 1791. Andrews and his printers were engaged in printing the third edition of The Worcester Collection. Buried amid a list of tunes still to be set for the collection is the following off-hand reference: “Assurance, by Billings—he has given permission in writing and wishes us to purchase some musick he has by him.” Four days later, on June 23, Andrews wrote about the Billings proposal in greater detail:

Billings, thro Gram, wishes us to purchase his musick of him. He has got 200 pages by him, that never was published, 9 or 10 tunes of which Gram has played on his Harpsichord, and thinks very good. Gram supposes he would ask 30 or 40 £ for all his musick (what he has in manuscript, and what he has published) but thinks he would take less, and that it might be worth our while to purchase, as they would serve occasionally to introduce into the Collection. If you have any inclination to purchase, at any rate, I will ask Gram to get Billings’ lowest terms.

Perhaps Thomas and Andrews’s interest in Assurance encouraged Billings to hope that the firm would welcome a chance to buy more of his music. Whatever Billings’s goal may have been, Andrews clearly thought of it as an opportunity to acquire additional music to be introduced “occasionally . . . into the [Worcester] Collection” and not as a step toward issuing a collection of Billings’s music. Thomas obviously showed little enthusiasm for Billings’s proposal. He must have written back to Andrews asking why Billings wanted to sell all of his music, to which Andrews replied on June 27: “Billings, I suppose, is not able to get his musick printed.” Perhaps Andrews did not know Billings’s financial condition, or perhaps Billings’s difficulties did not concern him, for his reply is perfunctory.

There the matter rested for over a year. But the following summer, certain singers and choristers of Boston, some of whom had undoubtedly participated in the benefit concert for Billings in December 1791, took up Billings’s cause. Members of a committee representing the singers approached Andrews about bringing out a collection of Billings’s music as a means of easing his plight.33 On July 1, 1792, Andrews reported to Thomas: “Don’t know but shall undertake to print some of Billings[’s] Musick—The Singers in town have [been] in agitation to get it done and give him the profits.” Two more communications from Andrews trace the plan’s progress. On August 2, he wrote: “We shall very probably print Billings’ book though no contract is yet signed.” Three weeks later Andrews was even more optimistic: “They have been to us today about Billings[’s] musick, and expect to begin it immediately.”

A subscription proposal was placed by a committee of singers in the Boston newspaper, Columbian Centinel, of August 22 and in the August issue of Thomas and Andrews’s Massachusetts Magazine.


A large Committee having been selected by the several Musical Societies in Boston and its vicinity, beg leave to solicit the attention of the publick to the following Proposals for Publishing a Volume of Original American Musick, composed by William Billings, of Boston.

The intended Publication will consist of a number of Anthems, Fuges, and Psalm Tunes, calculated for publick socical [sic] Worship, or private Musical Societies.—A Dialogue between Master and Scholar will preface the book, in which the Theory of Harmony, grounded on Question and Answer, is adapted to the most moderate capacity.—Also an elegant Frontispiece, representing the Aretinian Arms, engraved on Copperplate.


I. The Book shall contain about 200 pages, typographically printed, on good paper.

II. It shall be put to the Press as soon as a sufficient number of copies are subscribed for, to indemnify actual expenses.

III. The price to subscribers shall be 5 s. single and 4 s. 6 per dozen, one half to be paid on the subscribing, the other on the delivery of the books.

address to the benevolent of every denomination

The distressed situation of Mr. Billings’ family has so sensibly operated on the minds of the committee as to induce their assistance in the intended publication.

The Inspection and Revision of the whole is submitted by Mr. Billings to the aforesaid committee, many of whom are deemed of approved knowledge in the science of Musick, and nothing will be offered to the publick but what they recommend and approve of.34

McKay and Crawford remarked upon the ironic contrast between the “bold confidence” Billings displayed in The New-England Psalm-Singer, with statements like “I don’t think myself confined to any Rules for Composition laid down by any that went before me,” and the advertised fact that Billings submitted the music in The Continental Harmony to a committee for its approbation, and “nothing will be offered to the public but what they approve of.”35 It must be remembered, however, that this advertisement spoke for the committee and not for Billings. Perhaps Billings did agree to give committee members a say in the selection of pieces for the collection. However, the contents of The Continental Harmony show no striking variation in musical style from Billings’s earlier works.36 Billings had been moving toward a more regular harmonic style at least since the mid-1780s,37 and nothing appears in The Continental Harmony that was not presaged by the Relly settings in The Suffolk Harmony (1786). In fact, when one considers the problems with prosody in the fuging-tune Great Plain and the unusual modulations in the anthem Variety, Without Method,38 one might question how closely the committee supervised the music. The statement by the committee may have been little more than an effort to assuage doubts and increase the subscription.39

By October 1792, the proposal for publishing the tunebook had been before the public for over a month. The committee, anxious perhaps to complete the project, approached Andrews with another proposal—to take a mortgage on Billings’s house as surety for printing costs not yet raised by the subscription. On October 2, Andrews reported to Thomas:

They want us to go on with Billings’ Musick, and to take a mortgage on his house for security of what money is not subscribed—in addition to which they are willing we should keep the books in our hands. They have not yet ascertained how much is subscribed for—but they suppose there is 500 or 600 subscribed for.

Whether Thomas refused the offer of a mortgage, or whether there were some problems connected with it—the mortgage Billings had already obtained on his house in January 1791 was still outstanding40—is unknown. But the publication effort made no further progress for almost another year.

During the next ten months, the committee secured enough subscribers to go ahead with the project. In July 1793, they again approached Andrews about printing the Billings volume. On July 25, Andrews wrote to Thomas asking whether Thomas expected to need the music types.

[The] Billings matter has again revived, and they appear to be in earnest about having it done—but if you are like to want the types it will be impossible to do it in this season, as Holden’s 2d volume must be completed.41

At this point, the flow of preserved letters between Andrews and Thomas breaks off so that we do not know of the further negotiations between the committee and Andrews. Work on the Billings volume probably began early in the fall. On October 16, Thomas and Andrews advertised Billings’s anthem for Thanksgiving, Universal Praise, which was also to appear in The Continental Harmony.42 Judging by the production time of other tunebooks issued by Thomas and Andrews, it probably took about four months to set and print Billings’s volume.43 The new tunebook was advertised in the Boston newspapers beginning on February 1, 1794, in the Columbian Centinel:


Just published, price 6s single, and 5s3 by the doz.

[Ornamented with a Copperplate Frontispiece]

And for sale, by THOMAS and ANDREWS, Faust’s Statue, No. 45, Newbury Street, Boston, THE CONTINENTAL HARMONY. Containing a number of Psalm Tunes, Anthems, Fuges and Choruses, in several parts; never before published. To which is prefixed, Rules for Singers, and a Dialogue between Master and Scholar. By WILLIAM BILLINGS.

*** Subscribers are requested to apply at the above Store, for their Books, as soon as possible, otherwise the benevolent purpose which they had in view cannot be accomplished.

The same advertisement or one similar to it appeared in the Boston newspapers at irregular intervals over the next year.44 As new tunebooks came from the Thomas and Andrews presses, Billings’s work no longer occupied center stage, but it was usually mentioned along with other tunebooks as being available.45 The Continental Harmony was advertised outside Boston in Massachusetts west to Worcester and Greenfield and north into New Hampshire and Maine.46 It appears not to have been picked up by booksellers in Connecticut and Rhode Island, and no advertisements for it have been located further afield, in New York, New Jersey, or Pennsylvania.47

Throughout the negotiations for the printing of The Continental Harmony, Billings seems to have remained quiet. He is not mentioned directly in any of Andrews’s letters; members of the subscription committee seem to have made all of the arrangements with the publisher. While the recipient of charity may have little say in its disposition, it seems odd that the ebullient Billings remained so detached from an act as important as the publishing of his music.48 Judging from the many typographical errors in the music, it is hard to imagine that he was even involved in proofreading.49

For a popular eclectic tunebook like Thomas’s own Worcester Collection, a press run of between 3000 and 4000 books seems to have been standard.50 Although we do not know precisely how many copies of The Continental Harmony were printed, the anticipated demand for the work must have been less. Slightly more than 1000 copies, bound and in sheets, were still on hand at Thomas and Andrews in 1801; a press run of about 2000 books seems likely. If the committee’s 1792 estimate of “500 or 600 subscribed for” is not too far off the mark, then a total of between $400 and $500 was raised toward the cost of production.51 A modest sale outside the subscription may have raised several hundred dollars more, although whether any of this money ever reached Billings is unclear.52 In any case, it appears that the publication of his last collection of music, undertaken as a means of raising funds, produced little if any additional income for Billings.

As the years passed, The Continental Harmony remained a stock item in Isaiah Thomas’s bookstores—one that rarely moved.53 In his Worcester store, two or three copies were on hand; the branch in Walpole, New Hampshire, stocked about half a dozen copies. In an inventory taken in 1811, Thomas and Andrews in Boston listed fifty copies bound in boards, seven bound in leather, and 600 still in sheets, unbound, the last rated at a twenty-five per cent deduction—the rating given to their least saleable items.54 In 1811, Billings’s tunebook was among nine listed for sale by John Wyeth in Harrisburg, Pennsylvania.55 Even as late as 1847, it was still available in Boston; Alexander Wheelock Thayer remarked in an article in The World of Music that “the Book is not very rare. . . . There are still many copies to be found in Boston by those who would like to own one of Billings’s Books.”56

With the publication of The Continental Harmony, over 300 compositions by Billings had, over nearly a quarter century, come into print. Yet others remained unpublished. The psalm-tunes Germantown and Hacker’s Hall, found in the Waterhouse manuscript,57 must have remained in Billings’s hands. At least one anthem, “I Was Glad When They Said Unto Me,” known to have been performed on March 13, 1785, was never published and has since disappeared.58 A number of variant versions of published tunes, usually with different names and significant alterations most likely made by Billings himself, seem also to have remained in manuscript.59 What other works Billings may have left unpublished—anthems, fuging-tunes, extended psalm-tunes—can only be imagined,60 but it is clear that in The Continental Harmony Billings did not issue all of his remaining music. Although there is no evidence that Billings composed much during the 1790s,61 he seems not to have ceased writing music entirely, though he published nothing after the expanded version of An Anthem For Easter was issued in 1795.62 Whether this was due more to Billings’s poverty than to a lack of productivity cannot presently be told. The music that he did publish, however, stands as a testament both to his industry and to his lively imagination.

description of the tunebook

The Continental Harmony was published in the oblong format in which nearly all American sacred music collections appeared until after the middle of the nineteenth century. At 200 pages, it is larger than most American tunebooks of the day.63 The first thirty-four pages are devoted to the title page, an exposition of musical rudiments, and a dialogue between a scholar (i.e., student) and a master explaining and elaborating upon points covered in the rudiments. The music, which occupies some four-fifths of the book, consists of fifty pieces, distributed almost equally among anthems, fuging-tunes, and psalm-tunes. Unlike The Singing Master’s Assistant (1778) and The Suffolk Harmony (1786), which show some effort by Billings to arrange the contents systematically, The Continental Harmony gives no evidence of any such intent. Anthems and tunes follow each other without any apparent order to their location. A final page is devoted to the index, divided between strophic pieces and anthems.

The subscription proposal had promised “an elegant frontispiece, representing the Aretinian Arms.”64 Instead, the title page is faced by a musical composition—Billings’s Short-Meter psalm-tune Connection—engraved in a circle surrounded by decorations: winged cherubs and vignettes of sheet music superimposed upon musical instruments.65 “This tune is thus disposed,” the frontispiece explains, “to shew that every tune is a Compleat circle; & that what may be deficient in the first barr [which contains just one beat in 3/4 time] is supplied in the last [which contains two].” And all three verses of the text play on the theme of circularity:

We offer at thy shrine

One perfect round complete in sound . . .

In one majestic sound,

We celebrate Jehovah’s state,

In one eternal round.

Just are his ways, then let his praise

Eternally go round.

The title page of The Continental Harmony also invokes the spirit of energetic gusto that marks the frontispiece and that had also marked the beginnings of The New-England Psalm-Singer (1770) and The Psalm-Singer’s Amusement (1781). The decorative border ornaments, variety of type styles and sizes, and neatness and balance of layout bespeak a product designed to attract discriminating buyers with an eye for aesthetic detail. As in many other tunebooks of the day, the title page of The Continental Harmony includes a selection of Biblical verses chosen to sanction and exemplify the use of sacred music. In the selection and arrangement of these verses, Billings chose to raise a controversial subject: instruments in church, which he obviously favored. He also included a verse from the Psalms that gives singers precedence over instrumentalists, another position Billings seems to have advocated. Following the Biblical verses is a six-line hymn stanza by Billings himself, which he had set to his tune America in The Singing Master’s Assistant.66 The title page advertises the music of The Continental Harmony as “never before published.” While this claim is a bit exaggerated—several pieces had been previously issued in abbreviated form—nothing except the tune on the frontispiece had appeared in print in the musical settings found in this tunebook.67 Taken together, the title page and frontispiece must be counted among the most attractive combinations printed in American tunebooks.

The theoretical introduction, occupying nine pages, follows the title page. Intended to supply basic information about musical symbols and to guide the performance of the music, the introduction was copied almost verbatim from the fourth edition of The Singing Master’s Assistant, which Billings published sometime between 1786 and 1789.68 While the presence of a theoretical introduction in a tunebook implies its use as a singing-school book implies its use as a singing-school book implies its use as a singing-school book—which The Singing Master’s Assistant emphatically was—little in the musical contents of The Continental Harmony seems suited for beginners in psalmody. Perhaps the introduction was included as a point of departure for the extended discussion that follows in the dialogue.69

Following the rudiments is “A Commentary on the preceding Rules; by way of Dialogue, between Master and Scholar,” one of the most valuable and revealing documents to appear in any American tunebook. In it, Billings (the master) is led by the student’s questions into a wide-ranging expression of his views on many aspects of psalmody, including its performance and composition. Only about half of the more than thirty questions asked by the student deal with topics mentioned in the rudiments section. The rest are matters of theory and aesthetics aimed at experienced performers. Billings’s writing style throughout is lively and engaging. Sprinkled with Biblical references, humorous anecdotes, and self-deprecating asides, the dialogue imparts many useful ideas in a style designed both to entertain and to inform.

While the dialogue generally follows the sequence of lessons in the rudiments, it adheres to no rigid agenda. Some topics, like syncopation, the importance of which Billings stressed in Lesson VII, are not discussed at all. Others are introduced that were not mentioned in the rudiments: e.g., the difference between the medius and treble voices, the effect of the hold sign, the difference between concords and discords, and the question of what makes an anthem. Based on the length of the discussions, Billings’s major concerns were the transposition of B-mi, the leading-tone or “master” note,70 the distinctions between duple and triple time, the classification of compound meters, and the qualities of the major and minor modes.

Billings may have modelled his discussion format on William Tans’ur’s A New Musical Grammar,71 also a dialogue between a master and a scholar. Tans’ur may also have supplied Billings with historical information, such as his comments on Guido d’Arezzo and the “invention” of the scale. In addition to Tans’ur, Billings quoted from Benjamin Martin’s Philosophical Grammar,72 tunebooks by John Arnold,73 Aaron Williams74 and Thomas Walter,75 Isaac Watts’s preface to his The Psalms of David Imitated, and Milton’s Paradise Lost. Nevertheless, most of the information in Billings’s dialogue seems to come from his years of experience, especially as a teacher of singing schools and a composer of sacred music. Billings’s explanations are generally concise and to the point, avoiding tautology and, on the whole, covering the subjects well from his own distinctive point of view. However, on some subjects, such as his method of composition, the effect of directive words on tempo, dynamics, and specific details of performance practice, he is either frustratingly brief or conspicuously silent.

The music of The Continental Harmony emphasizes variety. Except for a twenty-two page section near the front containing seventeen psalm- and fuging-tunes, the different types of pieces are distributed rather evenly throughout the tunebook. In all, seventeen anthems,76 seventeen fuging-tunes, and sixteen psalm-tunes are found among the fifty pieces.77 In the variety of its contents and emphasis on long, relatively complex anthems, The Continental Harmony takes up where The Psalm-Singer’s Amusement left off.78 Like that tunebook, The Continental Harmony is a book “not designed for Learners.”79 Billings seems to be addressing the same public that he did in The Psalm-Singer’s Amusement—one for whom sacred music was both pious recreation and sacred duty.

Comparing the two tunebooks, one is immediately struck by two differences: in The Continental Harmony Billings’s imagination seems more disciplined and his compositional technique more refined. In The Psalm-Singer’s Amusement, for example, Billings employed unusual key signatures—B major and F-sharp major—but in The Continental Harmony he apparently felt no need to present such challenges to the singers. He goes no further afield than F minor, and that only twice. Almost half of the pieces are in C major, perhaps the most familiar key to the tunebook’s purchasers. Billings’s contrapuntal technique is compact and direct, making its musical point without prolix elaboration.80 In The Continental Harmony one finds no extended choral fuges, like that in “Who Is This That Cometh from Edom.”81 On the other hand, Billings’s earlier tunebooks contain nothing like the structural sophistication of Lewis-Town, an extended psalm-tune in which the third section is a rhythmic recasting of the first, nor like the frequent and unusual key changes in Variety, Without Method, which must have been an experiment in modulation.82

In The Continental Harmony, Billings is able to achieve dramatic effects with a precision and conciseness not seen in his earlier works. In earlier pieces, Billings tended to extend to considerable lengths the “Hallelujah” sections that close many anthems.83 In The Continental Harmony, however, Billings often brings off these sections with a concision bordering on abruptness.84 He also seems more sensitive to the need for textural variety in the anthems. Rarely do more than eight or ten measures pass without a change in choral texture, intermixing chordal writing with sectional solos, duets, and fugal counterpoint. Billings usually saves his most forceful choral effects for the final measures of the work, which are often in a grand and ecstatic mood.85 While The Psalm-Singer’s Amusement can justly be characterized as Billings’s “most flamboyant performance,”86 The Continental Harmony may be considered his most technically secure and musically mature work.

Undoubtedly, some of the anthems that Billings could not publish in The Psalm-Singer’s Amusement (1781) because of the “present extravagant Price of Copper-Plate & Paper,”87 found their way into The Continental Harmony. What refinements Billings may have made in them are unknown, but there are few deficiencies that suggest immaturity.88 All seem to be on an optimum level of technical competence and musical imagination. Several anthems include music known to date from earlier in Billings’s career. “Hark, Hark, Hear You Not,” for example, begins with a section that had once been a fuging-tune called Hadley, composed in the 1770s or early 1780s.89 The ordination anthem, “O Thou To Whom All Creatures Bow,” includes two sections that Billings had published earlier as parts of other works: mm. 144–164, which appear in Retrospect in The Singing Master’s Assistant,90 and mm. 215–239, taken from the independently published anthem Peace.91 “I Charge You, O Ye Daughters” resembles “I Am The Rose Of Sharon” in The Singing Master’s Assistant so closely in style that one could believe both were composed around the same time.92 On the other hand, some works, such as “O Praise The Lord Of Heaven,” Deliverance (“I Will Love Thee”), “When The Lord Turn’d Again,” and Universal Praise (“O Praise God”), are so well integrated and flow so smoothly from one section to the next that one feels they must be later works, composed not long before publication.

Of the seventeen anthems in The Continental Harmony, six are of the refrain type, in which a section of music and text is repeated several times with different music intervening. Billings had employed this approach earlier,93 but the examples in The Continental Harmony are by far his most extensive and imaginative use of it. In Universal Praise, for example, each major section except the final one ends with the refrain, “Then join all your voices, one chorus to raise, ascribing all honour, all glory and praise.” The final section is a “Hallelujah” chorus; a similar paean also ends seven other anthems in the collection. Perhaps his most imaginative use of the refrain format is found in the anthem “Sanctify a Fast.” In it, Billings delays the refrain until well over half the anthem has been sung. The words, “And he will cause to come down for you the former and the latter rain in the first month,” are then repeated three times, each repetition being an expanded and varied version of the previous one.

The fuging-tunes, like the anthems, also emphasize variety. Of the two basic styles of fuging-tune—the integrated and fuging-chorus types94—Billings distinctly preferred the latter. Of the forty-three fuging-tunes that Billings published during his career, twenty-eight end with fuging-choruses. In The Continental Harmony, however, the numbers of each are almost equal—eight are integrated, nine are fuging-choruses. In the fuge, Billings seldom follows the propulsive rhythmic formula of repeated quarter-notes that Irving Lowens exemplified as characteristic of “the typical American fuging-tune.”95 His fugal sections tend to be longer (in Creation, for example, the fuge runs to 31 measures) and his fugal subjects more varied in rhythm, melody, and mood than those of most other American psalmodists.

Not all the fuging-tunes in The Continental Harmony are recent compositions. Several pieces printed in The Continental Harmony as fuging-tunes had appeared earlier as plain tunes. The first fifteen measures of both Creation and Revelation were published as psalm-tunes in Music in Miniature (1779). Weymouth is found as a fuging-tune in three manuscript copies, dating from about 1780.96 Great Plain may also date from early in Billings’s career, for unlike most other pieces he composed after The New-England Psalm-Singer (1770), it shows uncharacteristic difficulties with prosody.97 He admits his lapse in a footnote, saying: “Part of this tune is very badly bar’d, but I will leave it for the observation of the reader.”98

If Great Plain represents an early fuging-tune, then Hopkinton must surely be a late one. It displays compositional features not seen in earlier Billings tunes. It resembles some of Billings’s settings of Relly texts found in The Suffolk Harmony, with their carefully fashioned counterpoint and tonal harmony.99 Perhaps the most distinctive and noteworthy feature in Hopkinton is the contrapuntal exchange of voices, rare in Anglo-American psalmody: mm. 9–16 are a rescored repetition of mm. 1–8, with the treble taking the previous tenor part and the tenor singing what the treble had sung.100 The sturdy quarter-note pace of the fuge, with its minimal textual conflict, provides a sharp contrast with the melismatic figures of the earlier part, forming one of Billings’s most unusual pieces.

The sixteen psalm-tunes in The Continental Harmony are almost exclusively tunes with extension.101 Only one, St. John’s, is a plain tune. Several tunes, such as St. Thomas (which sets four verses of text) and Lewis-Town (which sets three) approach small set-pieces in size.102 The tunes vary greatly in structure and musical sophistication. Tunes like Bellingham, Cross Street, and Rochester show a deftness in handling texture and counterpoint similar to some pieces in The Suffolk Harmony. Others, such as East Sudbury, South Boston, and Thomas-Town, exhibit an archaic simplicity in their settings, suggesting that they are early pieces. Thomas-Town appears in a manuscript copy at the Massachusetts Historical Society (ca. 1780), but under the title “Deearfield.” It is also found incomplete, and set for tenor and bass only, in a manuscript at the New York Public Library, thought to date no later than 1773.103 The first fifteen measures of St. Thomas were published in Music in Miniature as Calvary. Cobham is also found earlier in the Waterhouse manuscript and in a manuscript at the University of Michigan’s Clements Library as Raynham, set for tenor and bass only.104

Perhaps because The Continental Harmony was protected by copyright or, as McKay and Crawford suggest, “a certain standard repertory of psalmody had already taken shape,”105 few of its pieces found their way into other compilers’ collections. Thomas-Town (Deerfield), reprinted more often than the rest, appeared in at least six collections by 1814.106 Four other tunes made a single appearance in other tunebooks: Rocky Nook in Freeman Lewis’s The Beauties of Harmony (Pittsburgh, 1814); Sudbury in Abraham Maxim’s Northern Harmony (Exeter, 1805); and West Sudbury in Jacob French’s Harmony of Harmony (Northampton, 1802). The version of St. Thomas published in Walter Janes’s Harmonic Minstrelsey (Dedham, 1807) differs from Calvary printed in Music in Miniature only by the change of its tune name. The other forty-five tunes in The Continental Harmony were ignored by the succeeding generation of psalmodists.107

In summary, one must affirm McKay and Crawford’s assessment: as an anthology of Billings’s compositional achievements to 1794, The Continental Harmony was distinctly “an artistic success,” but it had little effect on the future of American psalmody, and for Billings’s financial plight, it appears to have been “a commercial failure.”108


One of the most influential musical developments of the recent past has been a growing mistrust of received performance traditions—customs of musical interpretation passed on across generations and, in the belief that they come naturally to a sensitive musician, accepted without question. A generation ago, there seemed little reason to doubt that musical sensitivity and technical competence qualified performers to sing or play almost any western art music, except perhaps medieval and renaissance works, whose stylistic unfamiliarity invited specialized approaches. Today, that belief is more difficult to maintain. Through the combined efforts of singers, instrumentalists, scholars, and instrument-makers, performance practice has become an issue of intense concern and sharp disagreement. “Historically-informed performance”—rooted in the conviction that, as Howard Mayer Brown has written, “music is best served by performances that reproduce as exactly as possible the conditions and conventions current when it was written”109—has now touched virtually all eras of western music-making. The idea of “historically-informed performance” has been widely taken up, not so much because of its appeal to historical “authenticity,” but because it has inspired performances of new freshness and vitality.

Historically-informed performance takes as its starting point the belief that each repertory has its own interpretative conventions, and it is a performer’s duty to discover and put those conventions into practice. Specifically, a piece of music is conceived and notated within an aural tradition that tells knowledgeable performers how to sing or play it. As time passes, however, and the performers of an era disappear from the scene, the aural tradition that originally encompassed that repertory gradually fades from memory. New styles and techniques replace older ones, and the meaning of the notation begins to change accordingly. Inevitably, music that endures beyond its own time comes to be performed differently from the way it was at first. In fact, until quite recently, pieces that survived from one era to the next did so because they proved themselves adaptable to newer performance conventions; their appeal seemingly did not depend upon being performed according to the customs of their time.110

Historically-informed performance has challenged the evolutionary process that, during the nineteenth century, brought earlier music into the realm of that era’s performance conventions, and that has maintained those conventions into our own time. Its champions, through the study of documents and musical instruments of earlier times, and also by searching for surviving or current musical practices that might illuminate the performance of earlier music, seek alternatives to received performance tradition. The growing strength of its influence provides a good reason for a scholarly editor of any “early” music to address performance practice.

Early New-England psalmody—the psalm-tunes, fuging-tunes, and anthems composed by New Englanders in the eighteenth and early nineteenth centuries—fell into disuse after 1820, and it was not until well into the twentieth century that efforts to revive it in performance began. By that time, the aural tradition originally surrounding the repertory was all but lost. But editors and scholars interested in the music were not left entirely in the dark, for there remained a rich source to be mined for information about performance: the introductions to the published tunebooks in which New-England psalmody first appeared. Most tunebook compilers offered advice on performance, and a few attempted a systematic review of the whole subject.111 Certain matters came up again and again. Their persistence in the literature shows that such ideals as pleasing tone quality, well-calibrated dynamics, proper choral balance, and tasteful ornamentation were elusive. Thus, the introductions may reveal the difficulties that eighteenth-century psalmodists found the toughest to overcome. They also provide help for performers who are trying to bridge the gap between the eighteenth and the twentieth centuries. It has been to these introductions that editors and commentators have turned in their studies of New-England performance practice.

William Arms Fisher’s Ye Olde New-England Psalm Tunes (Boston: Oliver Ditson, 1930) was the first work to make early New-England psalmody available to twentieth-century performers.112 Believing that psalmody was “crude” and its composers “untaught,” Fisher rescored the music to make it seem less strange to performers.113 He also recognized the need for some brief “Hints on the Performance of the Music,” which he included in the introduction to his anthology.114 Fisher discussed performance issues in general terms, leaving dynamics and tempo “to the discretion of the leader,” and leaving voice quality and choral balance unaddressed. He suggested that the earlier tunes in the collection—some dating back to sixteenth- and seventeenth-century England—be lined out rather than sung as written. He also advised that “music of the Billings era must not be dragged.” Some of Fisher’s “Hints” are misleading. He wrote, for example, that the “music was written to be sung without accompaniment” but then went on to cite several examples of the use of instruments in his historical discussion. His only mention of ornamentation is the trill, which, according to him, was sung as “an inverted mordent.” Nevertheless, given the state of knowledge about psalmody in his day, Fisher’s treatment of the music and its performance is helpful and sympathetic.

Systematic study of the historical, bibliographical, and musical issues of early American sacred music began shortly after World War II with the work of Allen P. Britton and Irving Lowens. Since then, several important studies dealing with performance practice have been issued, in particular, those of Britton,115 Hans Engelke,116 Crawford and McKay,117 and Sterling Murray.118 Each viewed the issues from a slightly different perspective, contributing to the growing knowledge on the subject.119

Britton’s 1949 dissertation, which summarizes the vast amount of detail in tunebook prefaces, remains today a foundation stone of scholarship in psalmody. The dissertation is concerned largely with the pedagogical aspects of the tunebook prefaces. While he includes only a relatively short section specifically labelled “Performance Practice” (pp. 302–313), Britton goes a long way toward clarifying the meaning of the signs, symbols, and instructions in tunebook prefaces and in the music. Many aspects of performance are discussed, including lengthy sections on voice production and ornamentation, but they are widely scattered and make up only a small portion of the whole work. Such topics as instrumental accompaniment, the disposition of the choir, and the effect of the words upon the musical setting are not covered.

Hans Engelke’s 1960 dissertation is a valuable, though seldom-cited, study of ornamentation in American tunebooks, a topic that has otherwise been underemphasized in discussions of how to perform early American psalmody. Decrying the general lack of attention to graces in performances and performance editions of early American sacred music, Engelke intended his study to be a guide to the proper ornamentation of American psalm-tunes and anthems. He covers the common ornaments, such as the trill, turn, transition, and appoggiatura, as well as those he calls “ornaments of expression”: the accent, mark of distinction, swell, and hold. His Chapter VII is a summary of performance practice as it relates to ornaments in early American psalmody.

With the approach of the American Revolution’s Bicentennial in 1976, attention was focused more intensely than ever before on the music composed and performed in America two centuries earlier. Psalmody stood at the heart of this repertory, and, with the awareness of historically-informed performance of earlier music beginning to grow, several studies dealing with the performance of psalmody were published.

In 1973, Richard Crawford and David P. McKay brought out “The Performance of William Billings’ Music,” which became Appendix II in their biography of the composer.120 In this article, the authors surveyed in detail Billings’s own statements about the performance of his music, offering perhaps the most thorough consideration to that date of performance practice in psalmody. They covered some items not mentioned in earlier studies, like vocal sound quality, the disposition of the choir, the effect of church acoustics upon the sound, and the problems of singing later verses of a hymn to music intended for earlier ones. They also offered guidelines for the organ accompaniment of psalmody, a topic not previously considered in any depth.

In 1976, the American Choral Foundation published a special issue of its American Choral Review, entitled “Performance Practice in Early American Choral Music.”121 Not only did this compendium reprint several important articles previously published, it also included Sterling E. Murray’s “Performance Practice in Early American Psalmody,” a new and perceptive look at the whole range of issues connected with this music. Murray’s survey covered the performance of psalmody from a broad perspective, bringing to it not only Billings’s views, but those of many psalmodists who were Billings’s contemporaries and rivals. Intended as a choir director’s vade mecum, Murray’s study focuses on matters of immediate relevance to performers faced with unfamiliar music, though not including text setting or the disposition of the choir.

The following essay on the performance practice of Billings and his generation is based chiefly on written sources: the tunebook introductions published in America between 1770 and about 1820. Given the number of tunebooks published and the length of time covered—more than 200 titles and 400 editions in fifty years—one might expect difficulty in finding a consensus among those sources. In fact, the introductions show a remarkable consistency in their comments on performance practice.122 While a few compilers conscientiously recast advice in their own words, most were content to pass on the conventional wisdom, often verbatim in the words of their predecessors.123 The goal here has been to identify the prevailing opinions on each facet of performance practice and to present them in the words of the psalmodists themselves.124

But sometimes those words resist ready interpretation. For example, Samuel Holyoke wrote the following in The Columbian Repository (Exeter, [1803]): “Particular directions, when to sing loud and soft, are not always given. In which case, the subject, the music, the occasion, and the judgment [italics added] of the instructors must direct.”125 Changes in dynamics, in other words, are very much a part of the style, even when they are not notated; how and when they are introduced is a matter of personal “judgment.”

Here is where the insights of “historically-informed performance” come into play. The editor of early American psalmody can learn a great deal from the tunebook introductions. But when the introductions were written, those who read them could do so within an aural tradition that informed and directed their “judgment” about issues of performance practice. It was noted above that by the 1930s the aural tradition that in the eighteenth century had “surrounded” New-England psalmody “was all but lost.” In the 1980s, however, the chances of recovering elements of that aural tradition seem brighter than they were sixty years ago. They seem brighter, first, because our growing knowledge of informal American music-making invites us to link certain twentieth-century musical practices with eighteenth-century psalmody, and second, because singers of earlier European music have created an aural tradition in song that differs sharply from modern vocal technique, wedded as it is to the belief that pervasive vibrato is a natural ingredient of good singing.

To take the second point first, when one carries in one’s ear the sound of groups like the Hilliard Ensemble or the Taverner Singers, one reads certain passages in the tunebook introductions—on “the swell,” for example, or on voice quality, or tuning—with a more fully-imagined sense of what they might mean outside the modern received tradition. Such imaginings lead one to think further about southern shape-note singing, where vestiges of New-England practice may persist in living traditions, together with the compositions by Billings, Read, Jenks, and their compatriots that have in fact survived.126 Moreover, certain habits preserved in Anglo-American folk tradition—especially a straight, vibratoless tone and easy execution with raised larynx—may retain echoes of earlier performance styles.127 When the sounds of these practices are taken together with the writings in the tunebook introductions, the guidelines for performing early American psalmody seem ever more closely within reach. It is toward helping present-day singers and conductors find appropriate guidelines—what Don M. Randel has called “the accumulation of habits and experience” that a performer “brings to bear” on a notated musical text128—that the following pages are devoted.

the sound of psalmody

Psalmody was usually composed in a four-part, open-score format with the voices arranged according to their vocal compass: the treble at the top, followed below by the counter, tenor, and bass. Sometimes, particularly after 1800, the parts were reduced to three: treble (or medius), tenor, and bass.129 The principal melody is normally found in the tenor voice in both three-part and four-part settings. However, from the 1790s, an argument raged over whether the melody should be sung by treble voices. Some opted for that method, which they considered fashionably modern, while most others continued with the customary arrangement.130 Whether or not the melody was sung by the treble or tenor voices, it was usually placed just above the bass for the convenience of keyboard players accompanying the singing.131

The assignment of parts to choir members was normally left to the singing master’s or choir leader’s judgment.132 The treble part was sung by female voices in the normal course of performance. The counter part, usually the most limited in both range and rhythmic variety, was often assigned to younger male singers.133 The tenor part was given to men with high voices; the bass to men with lower voices. Billings advocated doubling the tenor part an octave higher by some treble voices and the treble part an octave lower by tenors. He considered the effect to be “beyond expression, sweet and ravishing, and . . . esteemed by all good judges to be vastly preferable to any instrument whatever, framed by human invention.”134

Choral balance in Billings’s day tended to be different from more recent practice, for most tunebook compilers recommended that a large part of the male voices be assigned to the bass part. The proportion of voices on the bass to those on the upper parts varied somewhat. Some singing masters, like Billings, recommended three voices on the bass to one on the upper parts.135 Others said that half the voices should be assigned to the bass with the rest divided evenly among the other parts.136 Simeon Jocelin, one of the more influential of the earlier singing masters, was even more specific: “Three upon the bass, one upon the tenor, one on the counter, and two upon the treble (in general) is about the proportion required by the laws of harmony.”137 The upshot of these recommendations is a bass that is solid and audible, for as Joel Harmon noted, “Great care ought to be taken, that the upper parts do not overpower the Bass.”138 The counter, on the other hand, should not be prominent. Oliver Holden wrote that “In singing the counter, great attention is necessary, that every note be touched soft, and smooth, otherwise the counter will predominate, which is very inconsistent with the principles of music.”139 Ultimately the choral blend should be such that “no one [voice] might be distinctly heard from the rest.”140

The choral sound in psalmody seems to have been different from what choir directors are accustomed to today. McKay and Crawford suggest that “the nineteenth-century Italianate resonance that results from modern vocal training is out of place. . . . [The] tone is straight—or vibrato is at least involuntary rather than cultivated—and nasal, [the] dynamic range small, and [there is] a sense of flexibility and ease.”141 Eighteenth-century descriptions of the ideal choral sound tend to be more fanciful. Simeon Jocelin wrote:

Let the voice be clear and smooth as possible, neither forcing the sound through the nose, nor blowing through the teeth with the mouth shut;—a trembling in the voice is also carefully to be avoided. . . . And let all be done with ease and freedom, endeavouring to cultivate a musical voice; observing for imitation, the sweet sound of the violin, the soft melody of the flute, and the tuneful notes of the nightingale.142

Excessively nasal singing, as Jocelin and others noted, was to be avoided, but Asahel Benham warned against an equally serious failure: the lack of resonance. “Most authors direct to avoid sounding through the nose. I would by no means recommend it however; for those who endeavour to avoid it fall into the other extreme, and sing as if they had no passage through the nose.”143 Enthusiastic singing by a few members of the choir was also discouraged. As late as 1815, Edward Hartwell noted that “Some persons suppose they sing well only when they exert the whole strength of their voice, but this is rather bawling than singing, and precludes all delicacy of taste and expression.”144

The vocal quality of each voice part was often described in colorful if imprecise detail: “Let the bass be sung bold and majestic, counter clear and lofty, tenor firm and manly, the treble soft and delicate.”145 Most writers recommended that “high notes in all parts should be sounded soft and clear, but not faint. The low notes full and bold; but not harsh.”146 These descriptions suggest a choral sound in which male voices predominated—a firm melody in the tenor supported by a strong bass line—with the female voices (and some younger males) on the accompanying treble and counter parts distinctly in the background.147

Many writers recommended “soft singing” as a means to achieve a range of musical expression in the singing company and also to improve individual voices. By soft singing, writers seem to mean a light, smooth, controlled vocal production, kept well within the range of vocal compass and dynamic levels. Thomas Atwill summed up the sentiments of many singing masters when he wrote:

Teachers of music should assiduously inculcate soft singing; for persons who practice soft and smooth singing, retain the power of hearing and conforming to other voices, and of becoming masters of such gestures and expressions as reason and propriety would dictate. Soft singing is in fact the best expedient for refining the ear and improving the taste.148

Oliver Holden’s description of this typical choral sound of his day may be taken as a summary of what enlightened singing masters struggled to overcome.

Harsh singing is attended with convulsive motion, bad pronunciation, misapplied accent, and disgustful jarring. [This] is too just a description of the present mode of singing, occasioned in a great measure by a mistaken idea, which many entertain, that good music, consists principally in great quantity of sound; but, just the reverse of this is the case.149

the words of psalmody

The tunebook compilers took sacred words with the utmost seriousness. Indeed, many writers considered words to be the most important aspect of sacred singing, controlling nearly all aspects of the music. Asahel Benham seems to have had this in mind when he wrote:

The perfection of singing, is to pronounce the words, and make the sounds as feelingly as if the sentiments and sounds were our own. If singers, when performing a piece of music, could be as much captivated with the words and sounds, as the Author of the music is when composing it, the foregoing directions would be almost useless.—They would pronounce, accent, swell, sing loud and soft where the words require it; make suitable gestures, and add every other necessary grace.150

The primary function of a psalm-tune is, of course, to carry the sacred text through time. The range of influence of the text, as Benham suggested, should be wide and pervasive. The subject of the text often determined the music’s tempo, dynamics, and mood, while the meter of the poetry governed its rhythm and phrasing. According to Benham, “all psalm tunes ought to be varied in quality of sound and in movement of time, according to the subject of the Psalm; and even verses in which they are sung this way, words will be more suitably expressed, and the same tune appear like different music.”151 According to this, the pace of a tune might be faster or slower in each verse depending upon the needs of the text.152

The sound of the words, as well as their meaning, plays an important part in the performance of psalmody. New England, like many other areas in Great Britain and America, developed regional dialects with distinctive pronunciation.153 Tunebook compilers were aware of the importance of pronunciation, and many included advice on the subject in their tunebook prefaces. In some cases, these were only admonitions, such as “let the pronunciation be distinct, without drawling or affectation.”154 Others were more specific. Oliver Holden wrote:

Many singers who read and speak tolerably well, are extremely erroneous in their manner of pronouncing some particular words, when they are set to music; for instance, the words reason, hearken, token, &c. which have but one accent, are often sung rea-zon, hear-ken, to-ken, [i.e., with an accent on the last syllable] which is very disagreeable and improper. Words which end in ple, ble, &c. are often falsely accented on the last syllable, which renders them thus, pel, bel, &c. Words ending in y, with a few exceptions should be pronounced as ending in e, or short i. The l should be silent in the words walk, talk, calm, &c. In fine, the best direction which can be given is this; let every word be sung distinctly, smoothly, and gracefully every way conformably to the best rules of speaking.155

Other compilers dealt with other details of pronunciation. Samuel Holyoke noted: “Words beginning with a vowel ought not be pronounced, as if they began with a consonant.”156 In Simeon Jocelin’s opinion,

Many words which end in y, should be pronounced as ending in e, or the short i, as lofte, eternalle, &c. but not in every instance; the words sanctify, magnify, justify, glorify, are exceptions, being pronounced as they are spelt. Hallelujah should be pronounced halleuiah.157

Stephen Jenks reminded his readers that “vowels not sounded in speaking, must not be in singing.”158 Jacob French added that “to should be pronounced two, and a should be pronounced broad; the should be pronounced as thau; Fa, should be pronounced as in Father; and La, as in Lather, except the low notes in the Bass, which should be pronounced as Faw and Law.”159 Daniel Read recommended Noah Webster’s Grammatical Institute as a source of correct pronunciation.160

The compilers’ recommended pronunciations were those of “genteel,” educated speakers.161 That the common people followed a different standard of pronunciation was noted admiringly by Samuel Gilman, who wrote of “four ancient men” who substituted one Sunday around 1800 for the choir in a New Hampshire meetinghouse: “their very pronunciation had in it something primitive and awe-inspiring. Their shall broadened to shawl, do was exchanged for doe, and earth for airth.”162 Gould, who wrote around the mid-nineteenth century, remembered that the singers of his early days “[reserved] consonants, such as m, n, d, st, t, &c. at the close of a word . . . and applied [them] to the word that follows.”163 He gives the following example:










As sung,









The historically-informed performer will want to consider pronunciation carefully, for the color of the vowels and the strength and placement of the consonants can mark a vocal performance just as surely as do the pitches and rhythms of the music.164

Most pieces printed in American tunebooks of Billings’s day were supplied with only a single verse of text. Usually the text was taken from one of the psalms and hymns of Isaac Watts, the New Version of Nicholas Brady and Nahum Tate, or from one of several collections of hymns available at the time.165 Many composers and compilers were careful to credit the author of the words so that performers could find and sing the other hymn verses.166 While single verses of text might suffice for learners in singing schools,167 there is no reason to think that choirs in a church service or musical societies singing for their own pleasure limited themselves to the single stanzas of text printed in most tunebooks.168

Occasionally, the extra verses of text may need adjustment to fit the music set to an earlier stanza. Small misalignment of textual and musical accents apparently did not bother most composers and singers. For example, Billings set the opening line of Isaac Watts’s Psalm 5 in Phoebus—“Lord, in the morning thou shalt hear”—with the primary accent falling on “in,” and “Lord,” the stressed word in the phrase, set as a pick-up.169 Similar situations, where a trochaic foot is substituted for the normal iambic foot, are fairly common in the psalms and hymns of Watts and others, particularly when the name “Jesus” begins a line.170

Accent was a subject important to the psalmodists, and they frequently discussed it at some length. Most carry statements similar to that of Asahel Benham: “If the poetry be good, and the music skilfully adapted, the important words will fall upon the accented parts of the bar.—Should emphatical words happen on the unaccented part, the music should bend to the words.”171 This suggests that accents in the music should shift to match accents in the words, even if rebarring the music is sometimes the result.172

Even grosser mispronunciations, such as might occur when a portion of text was repeated, apparently did not bother many singers. John Cole,173 a Baltimore psalmodist who adopted a reformist attitude toward psalmody early in his career, commented upon the problem:

In singing the above tune [Bath Chapel] the third line of each verse may be sung throughout without repetition, by attending to the slurs, as marked in the AIR. This attention is absolutely necessary, for there are many words which would appear truly ridiculous when divided; e.g. How absurd it would be to sing, as was once done [emphasis added] the second verse of Hymn 114 Dr. Dwight’s collection, and divide the line thus: “Upon a poor poll, upon a poor poll, upon a poor polluted worm.”174

This situation was particularly prevalent in fuging-tunes, where words and phrases were often repeated to realign the declamation between the parts. Undoubtedly, on many occasions the singers split words so that the phrases made no sense, but by careful consideration of the text, it is usually possible to fit word repetition to the music so as to avoid situations like the one mocked by Cole.175

Another problem encountered when singing a whole hymn to the same tune is less easy to overcome. This occurs when later verses fail to match the mood of the music to which the earlier ones are set. Some hymns begin gloomily with meditations on death and judgment and end triumphantly with praise of God.176 Such a text would normally be sung to a minor-mode melody in accordance with the standard admonition of the day to set melancholy words to minor keys and cheerful words to major keys.177 Scant attention apparently was paid by composers to the later stanzas and by performers perhaps even less. A few composers, such as Elias Mann, solved such difficulties by writing set-pieces changing mode and mood as the emotions of the text changed.178 Most others, however, seem more concerned with setting the affections of the opening stanza or stanzas than considering the whole hymn. Although he followed his texts’ moods more faithfully than most psalmodists, Billings occasionally was guilty of the same thing. For example, in his setting of a John Relly text in West Boston, he employs a minor-mode tune, which fits the first six verses well enough but hardly expresses the sentiments of the final two.179 For texts of this kind, perhaps it is best simply to admit that some psalms and hymns exceed the range of ethos that strophic singing provides.


Musical pace was governed by the time signature in which each piece was set. Nine time signatures were normally employed in psalmody, each carrying an implied tempo set in relation to some note unit equal to a second of time.180 A precise tempo could be fixed by comparison with a pendulum swing of a certain length.181 Some composers, including Billings, believed firmly in a steady tempo set by a pendulum swing.182 Others, such as Andrew Law, Samuel Holyoke, and Oliver Shaw, divorced the time signature from tempo, relying instead on directive words and the subject of the text to suggest how fast or slow a piece should go. Joel Harmon remarked upon this situation: “Many Teachers of Music have been governed wholly by the Pendulum, notwithstanding the variety of subjects. But this is a gross error, and discovers want of taste. Whatever the operating mood of time may be, the subject ought to govern the time.”183 In a similar vein, Samuel Holyoke noted:

The Terms Andante, Moderato, Piano, &c. are called directive, because from them we discover the character and movement of a piece of music. Many singers pay no attention to these terms, but decide the velocity of a movement from the signs of the measure, C, 3/2, &c. which are inserted at the beginning of the staff; whereas those signs signify no more than the measure, or contents of the bars. Wherever any directive words appear, an invariable adherence to them is indispensably necessary. At the same time the subject ought to be consulted, especially, when no directive words are found. Then, and then only, may the performer suppose that he has a tolerable idea of the design of the piece.184

Some composers employed directive words, such as “slow,” “fast,” “Andante,” or “Presto,” in addition to relying on the moods of time to determine tempo. These normally affected the basic tempo implied by the time signature, increasing or slowing it a certain degree. Although directive words are infrequent in Billings’s music, he had a particular relationship in mind when he used them. In the dialogue section of The Continental Harmony he wrote: “I don’t know what other authors may intend, but I should be glad to have such strains [over which directive words appear], performed one fourth part quicker or slower.”185 Other writers differ in their recommendations about the effect of directive words on the established tempo. Daniel Read wrote: “When the term slow occurs, let the musick be performed about one sixth slower than the true time [i.e., that established by the time signature and pendulum], and when the term very slow occurs about as much slower still, and the contrary for the terms quick and very quick.”186 Thomas Atwill asserted that directive words “make 1/3 difference in the time or sound & should be strictly attended to.”187

Even though a composition’s tempo was supposed to be governed by the time signature and pendulum swing, choirs generally sang the music faster. Oliver Holden cautioned singers on this point:

The Author wishes that the time in general might be slow, and the strains soft. Doubtless singing Choirs, in general, are too inattentive to these important parts of Music. By hurrying a piece of Music, performers are more likely to sing harsh; in consequence of which, good pronunciation is lost.188

In another tunebook, Holden goes on to say that “no strain should be sung any faster in the quickest mood of time, than will admit of plain, distinct pronunciation.”189 Six-eight time seems to have been particularly troublesome; the compiler of The Salem Collection noted that “some performers also are apt to mistake a Siciliano movement for a Jig, which leads them to perform that music in a merry style, which requires a slow and affecting one.”190

The choir leader did not conduct the choir from in front of it, but apparently led from within the group, more by voice than gesture.191 Time was beaten by each performer raising and lowering the hand in a specified pattern. Billings and most other singing masters were quite precise in describing these patterns for the various moods of time.192 All indications are that a steady tempo was maintained throughout a piece in accordance with the time signature or directive word, and such alterations as ritardando or accelerando seldom appear. Since performers kept their own time, it was vital that it be exact.193 Deviations in the tempo could cause ensemble problems, such as late entrances, ragged attacks and releases, and unsynchronized declamation.

On the other hand, the hold sign was employed by some composers. It appears in American collections as early as John Stickney’s The Gentleman and Lady’s Musical Companion (Newburyport, 1774), where he states that the note should be held “something longer than its usual time.”194 Most singing masters who discuss the hold are unspecific about its length, leaving it to the choir leader’s judgment to indicate the resumption of time beating.195 Others, perhaps bothered by the inexactness of the directions, attempted to specify an actual length for the hold. Walter Janes wrote:

The propriety of the use of this character has been disputed by many, as it does not determine the time of continuation. I have thought [it] proper to direct, in this book, that one third be added to the time of the note over which it is placed, and that the note be beat one third slower. When placed over a double bar, or end of a strain, the parts are to rest the time of one second.196

A few composers, Billings among them, rejected the symbol completely on the grounds of its ambiguity. In the “Dialogue” in The Continental Harmony Billings wrote:

It is a matter of very great surprize to me, that any author should give license, and such room for dispute, as may (and to my certain knowledge does daily) arise from such a paltry insignificant thing; which is so far from being any benefit, that I have known a company of musicians to break off in the middle of a piece, because they were divided in sentiment, at the occurrence of a Hold. . . . I think it so absurd, that it is best to take no notice of it; for my own part I never observe it, and I find upon enquiry that most judges of music are of my opinion.197

Some writers recommended that the hold be used in some instances for an improvised fioritura at the cadence of a piece. Samuel Holyoke noted that the hold “signifies an unmeasured pause, or suspension, that room may be given for a peculiar expression; or for introducing voluntary graces, as may suit taste and fancy.”198 Although Holyoke does not suggest who should sing the “voluntary graces,” the introduction to The Salem Collection states that it should be the principal performer in the choir.199


Most strophic pieces in Billings’s day were printed without any dynamic markings at all. In the rare cases where markings did appear, they are often misplaced, nor is it clear how long they are to remain in force.200 Such desultory use of markings might seem to suggest that dynamic contrast was unimportant to the psalmodists, but this impression is contradicted by many statements in the tunebook introductions. As noted earlier, Samuel Holyoke advised that when written dynamic marks are missing, “the subject, the music, the occasion, and the judgment of the instructors must direct.”201 In other words, dynamic contrast was introduced even when no directive terms appeared. How might Holyoke’s hints be applied to present-day performance? To answer that question, we must first understand that the psalmodists composed in musical, sections that were referred to as “strains.”202

The length of a strain varied from piece to piece. In plain psalm-tunes a strain might be no longer than the setting of a single line of text; in fuging-tunes and tunes with extension, it could be two or even four lines long. In anthems, the strain is normally longer still because of word repetition or longer sentences in the prose text. But strains were usually marked off by cadences, changes in tempo or texture, or pauses in the music.203 Contrasts within a musical setting usually occur between strains, rarely within them. A change in scoring, for instance, between a four-part setting and a duet texture, usually occurs between strains, and the new scoring remains in effect throughout the strain.204

Dynamic contrast in psalmody seems usually to have been terraced, with all voices changing level at the same time. And it was at the beginning of a new strain where a new dynamic level most often took effect. Where they appear, dynamic markings normally consist of a single sign placed above the treble line but intended to apply to all voices singing at the time. Even though a dynamic mark may be misplaced by a measure or more, it is still intended to take effect at the beginning of the strain.205 Normally it lasted the length of the strain, or as long as the musical situation that called for the dynamic level remained in effect.

In tunebook glossaries, the terms for dynamics—piano, forte, pianissimo, fortissimo—are defined with their usual meanings of loud and soft. However, they are relative terms: how loud was “forte”; how soft “piano”? Oliver Holden gave some indication when he wrote: “In all common strains, a medium should be constantly observed, neither so faint as to dwindle into indifference, nor so loud as to admit of a single harsh tone.”206 “Common strains,” in other words, were sung at a level between piano and forte, which today might be termed “mezzoforte.”207 The relationship of loud and soft to a standard level was specified more precisely by Samuel Holyoke: “When the word Soft is placed over a tune, the sound should be about half as strong as common. When the word Loud occurs, the sound should be full, but not harsh.”208

Dynamics may sometimes have been omitted because certain musical situations were understood to call for certain levels of sound. For example, Simeon Jocelin recommended that “single tunes [i.e., tunes setting only four lines of text] call for the piano on the third line, and double tunes [i.e., tunes with eight lines, usually setting two stanzas] on the two lines that precede the two last.”209 When a solo or duet passage appeared in the music, that section was to be sung softer than the music surrounding it. Daniel Read noted that “a solo should generally be sung softer and the chorus which follows louder than the rest of the music.”210 Solo, in New-England psalmody, means not a single voice but the section of the choir in which the “solo” appears.211 To achieve the reduced sound level of a proper solo or duet section, Billings recommended reducing the number of voices singing:

In my opinion two or three [voices] at the most are enough to sing it well, it should be sung Soft as an Eccho, in order to keep the Hearers in an agreeable suspense till all parts join together in full chorus, as smart and strong as possible.212

When the words of a psalm- or fuging-tune were repeated, which often occurred at the end of a tune, they were to be sung louder the second time. As Walter Janes observed: “Where there is a repetition of the same verse, or line, it should be sung soft the first time, and the next time a little louder, increasing with every repeat, unless there are musical terms to direct otherwise.”213 Similarly, Oliver Brownson noted that “either tune or words repeated should be sounded somewhat louder so as to give fresh life to the music.”214 Occasionally, individual words are repeated, calling for an increased dynamic level with each reiteration. According to Asahel Benham, ejaculatory words, such as “Oh,” “Ah,” and “Hark,” should be sung with an accent similar to what European musicians of the day termed sforzando: “In some instances, the voice should strike suddenly on a note, and diminish to the end, especially where words of interjection occur, as hark! oh! etc.”215

The fuge of a fuging-tune called for special treatment. In performing the fuge, writers agree that the leading part should start softly and, as the other parts enter one by one, the dynamic level of the whole ensemble should increase. As Walter Janes expressed it: “In fuging music, the part that leads should be sung very soft, but distinctly; gradually increasing as the rest of the parts fall in.”216 The effect is one of crescendo, and, since the fuge is normally repeated, the voices should drop back to a softer level (but perhaps not as low as at first) and repeat the crescendo a second time.

The terms “crescendo” and “diminuendo” are occasionally found in tunebook glossaries. As the performance instructions for the fuging-tune demonstrate, the concept of gradual dynamic change was known and practiced, but these directive terms are seldom found in the tunebooks.217 A section marked piano at its outset usually would not increase gradually in volume to a forte strain following it but change suddenly to the new dynamic.218 The effect of the directive words, however, when they are used, operated to the end of the strain, as Jacob Kimball observed:

Crescendo: This implies that the force of the voice must increase gradually till the strain is ended. Diminuendo: Means the reverse of the foregoing and is sometimes set in opposition to it; when properly performed they make no trifling addition to the beauties of music.219

The swell, which most writers considered an ornament, was a fundamental part of the vocal technique of the time. Singers were taught to begin a note softly, crescendo to its mid-point, and diminish to the end.220 In a dialogue between a master and scholar, Daniel Read wrote:

S. What is a Swell?

M. There is a character used by some Authors called a swell which when set over any note implies that it should be begun soft and ended full.

S. Should no notes be sounded with a swell but those marked with that character?

M. Yes, all notes, but more especially those of considerable length ought to be sounded with a gentle swell, and diminish both beginning and ending soft.221

Most writers observed that a swell was most effective on larger note values,222 such as the semibreve (whole-note); yet all notes were to be swelled in proportion to their length. Asahel Benham remarked that

a semibreve admits of a more extensive swell than a minim; a minim [more] than a crotchet; a crotchet more than a quaver, &c., which is perfectly consistent: For, if quavers were to be sounded as full as semibreves ought to be, it would be more like coughing than singing.223

The swell was to be used on notes marked with a hold,224 and perhaps also on melismas. Most writers recommend that the melisma be sung softer than single notes,225 but according to Little and Smith, “notes tied to each other, should be sung softer than when one note answers to a syllable, and should be swelled in the throat.”226


Ornamentation in psalmody is one of the most uncertain areas of performance practice, primarily because writers on the topic are inconsistent in their discussion of it. On the whole, one must agree with Hans Engelke’s assertion that “any performance of the music contained in eighteenth century American tunebooks, which gives no consideration to the use of ornaments, must be interpreted as misrepresentative of its time.”227 Ornaments were used widely at the time in both secular and sacred music, but their application in psalmody apparently varied with the performer and venue. Some writers acknowledged the performer’s right to ornament the music, but warned against improprieties.228 Others limit ornaments to sections of reduced scoring, noting that “solos are the proper field for their full display.”229 Still others, like Nehemiah Shumway, omitted them altogether, saying that they were “of little use, especially to learners.”230

Many different graces are mentioned in the New-England tunebooks of the years 1770–1820. Some, like the swell and accent, are generally thought of today as dynamic devices rather than ornaments. Others, like the “forefall,” “backfall,” “beat,” and “turn,”231 were apparently borrowed from instrumental music and generally play little or no part in the notation of psalmody. Still others, like the appoggiatura, were relatively late arrivals on the scene, reaching wide acceptance in psalmody only after 1800.232 This discussion will concentrate on two ornaments that are far and away the most common ones found in American tunebooks: the grace of transition (including its relation to the mark of distinction) and the trill.

Transitions are improvised portamentos supplied by the performer when singing intervals larger than a second. Rather than performing an interval with a clean, distinct leap, an accomplished singer was more likely to slide quickly from note to note, touching lightly upon intermediate scalar pitches to provide a graceful connection between primary tones. The transition may well have been the most widely used embellishment of the day, applicable to all voices in a variety of melodic contexts. Billings admitted that he was “very fond of the notes being graced by sounding the intermediate note” when the length of the principal note would permit it.233 Transitions were sung in a dotted quarter- and eighth-note rhythm when gracing half-notes and in a dotted eighth- and sixteenth-note pattern for quarter-notes.234 The grace was usually limited to the third, either ascending or descending, and “notes descending more than a third, should, generally, be sounded plain.”235 However, one occasionally encounters a notated transition ascending or descending a fourth, but these are rare.236 Shumway warns that transitions are “better omitted than badly performed.”237

The only limitation on the grace of transition, other than interval, was when notes were given a “mark of distinction.” Then, according to Billings, the interval should be sung plainly, without the intervening portamento.238 In form, the mark of distinction resembles a wedge-shaped staccato mark in present-day notation. In performance, however, it seems to have functioned as an accent. Most writers say that notes over which it is placed should be “sung distinct and emphatic.”239 Others wrote that the words should be “pronounced” distinct and emphatic.240 Some included both verbs in their instructions.241 In none of this is there a suggestion that the notes should be performed shorter than their notated value. Engelke notes that “the notes beneath the sign are not to be performed as short, but simply more sharply. While this sharpness may well result in a shorter tone, no complaints against such effects are recorded.”242

Engelke suggests three uses for the mark of distinction in American compositions: to stress musical climaxes, to emphasize off-beat phrases, and to add variety to the music.243 When a long note, like a whole-note, is given a mark of distinction, the swell that it normally received is abrogated, and the note is sung full and steady.244 It is, on the whole, a rather rarely encountered grace, but one whose effect, when come upon, makes a strong impact.245

The trill is the most widespread ornament printed in American tunebooks. Perhaps because it was so common, Engelke notes that few writers “bothered to go into great detail regarding its method of performance and the proper places of deployment.”246 Read described the trill as a “graceful shaking of the voice while sounding the note over which it stands,”247 and other compilers offered similar descriptions. Billings gave examples in The Singing Master’s Assistant of what he termed a “Single Trill” and a “Double-Trill,” which appear to be no more than an inverted mordent and a turn respectively. At the same time, however, he warned:

Many ignorant Singers take great licence from these Trills, and without confining themselves to any rule, they shake all notes promiscuously, and they are as apt to tear a note in pieces, which should be struck fair and plump, as any other. Let such persons be informed, that it is impossible to shake a note without going off of it, which occasions horrid discords; to remedy which evil, they must not shake any note but what is marked with a Trill, and that according to rule which may be easily learned, under a good master.248

Billings’s warning against shaking “all notes promiscuously” was meant to counter the excessive use of graces. Trills were often notated in the music; but they were also supplied extemporaneously.249 Daniel Bayley provided guidelines for adding trills. They “ought to be used on all descending pointed Crotchets [i.e., dotted quarter-notes],” he wrote, “and generally before a Close; also on all descending sharped Notes and Semitones, but none shorter than a Crotchet.”250 With the support of tradition and such invitations to grace notes at will, it is doubtful whether Billings’s admonition had much effect. One point that he made, however, does bear further elaboration. The trill was considered a difficult grace to master on one’s own. Jacob Kimball noted that “the trill is a very beautiful grace, but . . . is very difficult to be acquired.” He recommended a modified turn as a substitute “till it can be perfectly learned.”251 Kimball described the trill as “a quick and alternate repetition of the note over which it is placed, and the note immediately above it (from which the trill begins) so long as the time will allow.”252 Others advocated a slight prolongation of the first note of the trill,253 and still others suggested ending it with a graceful embellishment.254 Most agree, however, that it “had better be omitted than unskillfully attempted.”255

As mentioned above, the appoggiatura and grace of transition were often confused in terminology by American writers on psalmody. The actual appoggiatura, or “leaning note,” as it was sometimes called, entered fairly late into the American repertory of embellishments. It was not until the 1790s that it was separated from the transition and an interpretation assigned to it that squares with European practice of the times.256 It is an expressive dissonance sung on the beat in place of the principal pitch to which it is attached. Its resolution is delayed according to the rhythmic value of the primary note. Samuel Holyoke distinguished between what he called the “common” and the “large” appoggiatura. The former, he said, divides “the length of the principal [note equally] with the appogiature.” The latter, used with dotted notes or notes followed by a rest, “makes the appogiature as long as the principal, and [fills] the place of the point or rest with the sound of the principal.”257 Unlike the transition, which was usually improvised, the appoggiatura was always notated.258 Appoggiaturas are frequently found at cadence points and in preparation for a trill, but they are also used elsewhere in a composition to decorate words with an expressive dissonance.259

While Billings may have been careful to notate the ornaments he wanted in his music and objected to performers adding others, it was common knowledge that singers would follow their own fancies. Ornaments were used, and perhaps also abused, but they undoubtedly added spice and flavor to the music. It is then perfectly proper to supply ornaments in modern performances of Billings’s music and that of his contemporaries. Hans Engelke advises:

If one were to attempt a performance of this music with the ornaments it would be natural, from the present day point of view, to perform the embellishments precisely as they were given in the introductions of the tune books. This would be perfectly correct, since we are certain that at least one segment of the New England population interpreted them in this manner. However, it should be understood that this is not the only correct way. We, of course, can hardly be expected, nor are we able, to perform the graces as the “country people” did. But we should be conscious of the fact that a rigid adherence to the rules constitutes only one way of interpreting them and not the only way.260


As noted earlier, psalmody was published as an unaccompanied choral music for mixed voices, and undoubtedly it was frequently performed that way. However, it could also be accompanied, and there is no reason for present-day performers not to use instruments to support the voices as was done in Billings’s day. Instrumental accompaniment should be carefully handled, however, and especially when the instrument is the organ. Since few New-England churches had organs in the late eighteenth and early nineteenth centuries,261 accompaniment by a few wind and stringed instruments may be more faithful to the practice of Billings’s day.

If an organ is used with a piece of psalmody, certain cautions should be observed. McKay and Crawford list six guidelines for accompanying Billings’s music with the organ.262 These are equally applicable to the music of other composers in the tradition. Briefly, their recommendations may be summarized as follows:

  1. 1. Do not avoid using the organ for fear of being inauthentic.
  2. 2. The organ’s principal role should be to reinforce the lines written by the composer, rather than to add anything new.
  3. 3. The organ should double the bass line, bring out the melody if it is weak, and support other voices as needed.
  4. 4. The organist should not treat the bass as an unfigured continuo part, but play from the choral score or a keyboard reduction. To do otherwise will most likely introduce conflicts with the voice parts, which were not composed according to thoroughbass harmonic procedures.
  5. 5. The organist with a thorough knowledge of the style might be encouraged to interpolate occasional brief interludes between the sections of larger pieces.
  6. 6. The organ should not be introduced for the sake of authenticity if the choir is capable of giving a satisfactory unaccompanied performance.

To these six guidelines a seventh might be added: the organist might be encouraged to play over a psalm- or fuging-tune alone before the choir joins in. All of these recommendations are in line with documented practices in the performance of psalmody during Billings’s time.

More common than organ accompaniment, perhaps, was that of wind and stringed instruments. During the eighteenth century, Puritan prejudice against instruments in the meetinghouse weakened among New England Calvinists but did not entirely disappear. By the 1790s, however, the flames of opposition were burning low, and William Bentley noted in his diary: “The fondness for instrumental music in churches so increases that the inclination cannot be resisted.”263 This situation was seen by some observers as a matter of necessity; choirs in many congregations sang so poorly that the minister was faced with only two choices: abandon music altogether or try to support it by using instruments. Bentley and many others chose the latter alternative.264

The first instrument to gain acceptance in the choir loft was the bass viol (violoncello), which apparently had been used in singing schools since about 1770.265 On October 28, 1795, Bentley made a diary note that he had “sent & purchased at Boston a Bass Viol for 21 dollars.”266 About a month later he remarked that “in all of our societies [i.e., Congregational churches] the Bass viol has been used, having been introduced about two years since.”267 Gould later called it “the grand entering-wedge that opened the way for all other instruments.”268

Other instruments used to accompany psalmody included the flute, clarinet, bassoon, violin, and viola (tenor-viol). The same diary entry in which Bentley recorded the purchase of a bass viol noted that he had “applied to Mr. Gardner to assist the Counter with his German Flute.”269 Some weeks later Bentley reported the use of the violin, viola, and the clarinet.270 The instruments were used to double the voice parts, to maintain the pitch of the music, to steady the rhythm, and perhaps to fill in for any missing voices. Some, such as the flute, may have played the melodic lines an octave higher, particularly the tenor and treble, but probably not the counter, which was not supposed to be prominent.

While instruments were used largely for accompaniment and to reinforce the vocal parts, they sometimes assumed an independent role. In his description of the Thanksgiving Day service at his church in Salem, Massachusetts, on November 30, 1795, Bentley wrote that “the order of service was, An Air—Hymn 73, the instruments going over the tune, before the vocal music joined.”271 It is also possible that instruments played preludes and postludes alone, without vocal additions.

We may gain clues to how the instruments functioned with the voices by observing the recommended doublings in Jonathan Huntington’s The Apollo Harmony (Northampton: Horace Graves, 1807). Huntington provided fingering charts and brief instructions for playing the German flute and bass viol in the theoretical introduction of his tunebook, and for several of the tunes he indicated doublings. In Archdale (p. 50), he asks that the melody (in the treble) and the tenor be doubled by two flutes (probably at the octave because of range) and that the bass viol double the bass part. In Resurrection (p. 54), the three upper voices are doubled by flutes (again probably an octave above) with the bass viol on the vocal bass line. Other instruments, of course, may be substituted, for “the band,” as it was usually called, had no regular instrumentation—just whatever was available in the congregation and whoever was willing to play along.272

Questions of historically-informed performance aside, the addition of accompanying instruments adds a pleasant variety to the timbre of the music. A violoncello, for instance, doubling the vocal bass in a small choir can add a depth to the vocal resonance that may more closely approach what Billings and others had in mind when they recommended a two-to-one or three-to-one ratio of bass singers to upper parts.273 A flute or clarinet on the treble can ameliorate the shrillness often heard in high-ranging vocal lines. The smooth sound of a violin, with careful articulation, can often add color and vitality to a performance that may be difficult to achieve with voices alone.


The most important consideration in performing early American psalmody generally and Billings’s music in particular is that the performance seek to capture and convey the music’s spirit. Howard Mayer Brown has written that

Reproducing as closely as one can the techniques and timbres known to be appropriate to a given period can never replace performances that are musically convincing to the audience; and yet the means and style of performance imagined by a composer are so indissolubly bound up with the whole musical fabric that he has set down, that the communication and impact of the composition are seriously impaired if the sounds he imagined are not at least kept in mind when preparing modern performances.274

There are no hard and fast rules in the performance of psalmody, just guidelines, indications, suggestions, and inferences, giving the performer wide interpretative latitude. The way the music was performed in one church in one place and time may have differed radically in sound, tempo, ornamentation, accompaniment, and singing style from another performance in that same church a decade later or in another church across the state at the same time.275 Psalmody was a people’s music, and they set its standards according to their own tastes and desires. Today’s performers should keep in mind that New-England psalmody was vital, expressive, and adaptable to a variety of performance situations. While attempts at historically-informed performances are laudable, Brown’s advice that performances must also be “musically convincing to the audience” should also be carefully considered. That element brings psalmody to life and moves performers and listeners alike.


The Complete Works of William Billings began as a personal project of Hans Nathan which was adopted by The American Musicological Society as its contribution to the American Revolution’s Bicentennial. General editorial policy for the set was developed by Nathan in editing Volume II. Some flexibility has been observed in later volumes to take into account the idiosyncrasies of each of Billings’s tunebooks, but for general criteria, the user is referred to the introduction to Volume II (reprinted in this volume on pp. lxix–lxxii).

The Continental Harmony presents special problems for the editor, for this work, the only tunebook by Billings printed from moveable type rather than engraved plates, contains more errors than any of his other collections. It is true that the earlier collections—especially The New-England Psalm-Singer—art somewhat cruder in appearance than The Continental Harmony, the product of a professional print-shop. But the music in them, for the most part, is accurately engraved. In The Continental Harmony, on the other hand, symbols are so often omitted, transposed, or inverted that a close, questioning editorial scrutiny has been required to verify that each note makes melodic, harmonic, and rhythmic sense. As in earlier volumes, corrections of errors in both music and text are noted in the Commentary.276

Metronome markings have been supplied according to criteria outlined by Nathan in his introduction to Volume II of this edition. However, in some pieces (e.g., the anthem “I Charge You, O Ye Daughters”) it appears that Billings set aside his usual practice of linking the time signature to a particular tempo, for here he moves freely between time signatures that ordinarily call for different tempo. In such cases, a single tempo has been suggested by the editor.

The purpose of this edition is to present the music and text for performance and study as clearly and accurately as possible, according to present-day editorial and notational standards. Certain symbols used by Billings have lost their meaning today, and modern equivalents have been substituted. Instead of reprinting Billings’s verbal directions in some anthems that certain sections be repeated, for example, these sections have been written out in full, as noted in the Commentary. Or instead of marking the repetition of a phrase or section, as Billings does, by placing the sign :S: over the first and last notes of the section to be repeated, in this edition the standard modern symbol, the dotted double bar, replaces that sign. Occasionally, repeat signs have had to be repositioned and first and second endings added to correct the notation. This need usually results from Billings having changed time signature (such as from 3/2 to 2/2) during a repeated section. Such alterations are noted in the critical commentary. Billings employed the alto C-clef for his counter voice. Here it has been transcribed into the treble clef. The tenor voice in Billings’s original is notated in the treble clef, but with the understanding that it should sound an octave lower. The octavating treble clef has been used throughout this edition.

In several compositions (e.g., the anthems “Hark, Hark, Hear You Not” and “Hear, Hear, O Heav’ns”) Billings verbally directs that the work be performed in a lower key than that in which it is printed. In this edition, the original notation is retained and the note about performance pitch entered in the Commentary.

In eighteenth-century psalmody, it is not uncommon to find a group of repeated notes under a slur. Billings seems to have used this notational device for two purposes: as a substitute for a tie in which the repeated notes should be sung as a continuous sound, and as a vocal ornament in which each note should be articulated. The determining factors in interpretation are the length of the note values involved and the availability of other notational symbols. The tie substitute usually involves half- and whole-notes with one or more bar lines intervening, or smaller note values within a longer melisma for which no single notational symbol is available (e.g., a note length of five eighth-notes). Occasionally, in Billings’s earlier works, it was also used to correct an error in engraved music. The vocal ornament, on the other hand, usually involves smaller note values, such as quarter- and eighth-notes, for which a larger note value was available had Billings wanted a continuous sound. That he did not use the alternate symbol implies that he intended each note under the slur to be articulated. This vocal ornament is often used as a motive, passing from voice to voice several times in a section. A note has been made in the Commentary for repeated notes under a slur that are to be articulated, but ties between individual notes have been supplied without comment when the repeated notes are considered to be continuous sounds.

Ties that appear between only the upper or lower notes of diads (or choosing notes) in the individual voices are assumed to apply to the other notes as well and are supplied without comment.

By the 1790s, Billings was usually both careful and consistent in applying accidentals to his music. However, in certain places he seems to have overlooked an accidental sign, or the notational conventions of his day required none. In such cases, accidentals have been supplied editorially and are noted in the Commentary. Occasionally, Billings supplied accidentals unnecessary by modern notational standards. They have been tacitly omitted in this edition.

In 6/4 time, Billings used two dotted half-notes tied together to indicate the length of a dotted whole-note. (He used the dotted whole-note, however, in 3/2 time.) In this edition, the dotted whole-note has been tacitly substituted. Similarly, in 6/8 time, Billings notated the dotted half-note by employing two dotted quarter-notes tied. In such cases, the dotted half-note has been used instead.

Criteria outlined by Hans Nathan in the introduction to Volume II for the selection, spelling, and underlay of texts have been followed here. Billings provided texts or text incipits for all the music in this volume. However, because Billings appears to have had no hand in the production of The Continental Harmony, each word in the text has been compared with the original text source. Where deviations from the original appear to be intentional and follow Billings’s usual practice—for example, in Biblical anthem texts, where Billings often substituted words and phrases—they are followed. But where deviations appear to be typographical errors, and especially in poetic texts, where Billings rarely altered words, the original word has been substituted. In either case, a note has been made in the Commentary.

In second and succeeding stanzas of text, left margins have been justified and the first letters of each line capitalized. Following eighteenth-century practice, the first letters of nouns have been capitalized. Billings’s own occasional capitalizations of pronouns, adjectives, verbs, &c., have not been retained, except when they refer to the Deity. In the original sources, quotation marks usually begin each line of a stanza including quoted words. In this edition, quotation marks have been applied only to the first line of each stanza and omitted from second and subsequent lines without comment. Eighteenth-century printers usually printed personal names and proper nouns in italic type. Here these have been printed in roman type without comment.

Where words such as “Saviour” and “glorious” are spelled with a contraction (e.g., “Sav’our,” “glor’ous”), the missing letter has been restored and the apostrophe omitted; however, when restoration of the omitted letter would affect the pronunciation, occasional contractions have been retained.

In some hymns by Watts, certain stanzas are enclosed in square brackets, indicating that they could be omitted when the hymn was sung. In Volumes I and II of this edition, the square brackets were replaced by angled brackets, but in this volume all brackets have been omitted, and a note has been made in the critical Commentary indicating which stanzas were bracketed. Also in Watts’s hymns, the notation “Pause” appears between some stanzas indicating that the singing could briefly stop at this point. While these notations have not been retained in this volume, their location is noted in the Commentary.

Punctuation of the text generally follows that of the text source. However, here and there punctuation marks have been supplied or altered without comment where the meaning of the text seemed to demand clarification.

By the 1790s, Billings had mastered the setting of non-metrical prose. His music follows the accents of the text with a fluency that is both natural and musical. Occasionally, however, one finds passages in which the music diverges slightly from the textual accents. These occasions are both rare and momentary, and, since they usually create no gross distortion of the textual accent, it has been considered best to retain Billings’s own notation. In three compositions, however, it was considered necessary to adjust the textual and musical accents. Several measures in the middle of the anthem “We Have Heard With Our Ears” and near the beginning of the anthem “Hark, Hark, Hear You Not” were rebarred to align accents. In the fuging-tune Great Plain, Billings appears to have had unusual problems, some of which defy solution. The first half of the tune has been rebarred to eliminate accentual irregularities in mm. 13–16; however, short of barring each voice individually, there was no way to align textual and musical accents in the fuge. Rebarring in the anthems and in Great Plain has been done in accordance with criteria outlined in Volume I of this edition (pp. lxi–lxii), with the original barring indicated above the treble part.


Now that the final volume of The Complete Works of William Billings has been finished, it remains to thank those who have contributed to its completion. At the head of any list of its patrons must stand the names of Richard Crawford and Cynthia Adams Hoover. They, along with Hans Nathan, were among the project’s originators, and they stayed the long course of its production. Although his name appears on the title page only as “Editorial Consultant,” Crawford’s role was more that of co-editor, checking every note of the music, weighing every idea, every sentence, every word that the editor sent him, and offering his well-considered advice and criticism. Hoover was the facilitator, making arrangements with the press, the sponsoring societies, and the funding sources. Her active participation relieved the editor of many responsibilities extraneous to editing. It has been a distinct pleasure to have worked with these two remarkable people.

I am grateful to The American Antiquarian Society and The National Endowment for the Humanities for grants that allowed me to pursue research for the writing of this volume’s Introduction, and to The University of Colorado at Boulder for its material support of my work. I thank my colleagues Nicholas Temperley, Nym Cooke, William Kearns, Wallace MacKenzie, David Warren Steel, Gillian Anderson, and Susan Porter, for their ideas, counsel, and assistance so freely given. And to Alvin Johnson and Frederick Allis, executive secretaries of the two sponsoring societies, I express my gratitude for their continued interest and support. A special word of thanks to the American Antiquarian Society for supplying the copy of The Continental Harmony from which the facsimiles in this volume were made, and to Paul Hillier for his advice on performance practice issues.

Finally, to my wife, Marie, my sternest critic and my strongest supporter, I dedicate this volume with love and gratitude.