Kenneth R. Hull

The Conrad Grebel Review 20, no. 3 (Fall 2002)

The word “hymn” has traditionally been employed to denote the text of a
congregational song. The music to which the words of the hymn are sung, on the other hand, has most commonly been called its “tune,” even though we usually mean by that not just an unaccompanied melody, but a melody harmonized with an additional three singable voices. We often read about the relationship between texts and tunes in terms of such questions as, “Does the tune fit the meaning of the hymn?”

It is instructive to reflect on what is implied by this question, and by the
way the terms “hymn” and “tune” are used. This way of speaking about
congregational song clearly privileges its textual (i.e., verbal) component, because the term “hymn” is applied to it, while the “tune” is subsidiary, “fitted” to the text. In fact the very word “tune” is minimizing, implying as it inaccurately does, an unaccompanied melody. The phrase “the meaning of the hymn” also seems simplistic. Poetic texts are commonly understood to have multiple layers of meaning. Critics and others who explicate one of these meanings are said to be offering a “reading” of the poem. To ask whether the tune fits the meaning of the hymn, then, not only privileges the text and minimizes the importance of the music, but also assigns to the hymn text an objective, informational character, rather than a more genuinely poetic, multilayered one.

Why has our language about congregational song been so biased in
favor of the text? Probably because we have understood the meaning of the hymn to be situated exclusively in its text. The music to which hymn texts are sung has been understood as having an expressive character, certainly, for it is important that the music “fit” the tune not only metrically but also in terms of expression. The music of the hymn has also been seen as enhancing or deepening the meaning of the text, but not as contributing any element of Kenneth R. Hull is Associate Professor of Music and Director of the Institute for Worship and the Arts at Conrad Grebel University College in Waterloo, ON. meaning distinctly its own. It is only recently that we have begun to ask whether the music of hymns might not contribute aspects of meaning not found in the text alone.

This question has been put most recently by David Cole in his essay,
“Singing the faith.”1 In it, he asks whether the meaning of a hymn is indeed carried primarily by the text, or whether it might not be that music not only brings additional meanings, but might even be the more dominant partner. “[I]s it,” he asks, “the words which give added meaning to the music, or the music which gives added meaning to the text, or does some other process come into play?” Later, he concludes that “in hymns we must take seriously the text, the music, and the new art form which comes from combining these two art forms.”

My purpose is to offer a model for understanding how the text and music of hymns interact with one another to produce meaning, the music by
creating a reading (not just a “setting”) of the text, and the text by specifying a “hearing” of the music. Text and music each do this by providing a context within which the other is perceived and understood. The framework for the model I propose is adapted from the work of music theorist and critic Edward T. Cone, particularly his seminal book, The Composer’s Voice.2

The Composer’s Voice has been a foundational work for many scholars writing in the area of musical meaning and expression.3 Cone calls his book “a theory of musical utterance” — an answer to the question, “If music is a language, then who is speaking?” His answer to this question considers especially the genre of the art song, in which the words of a poet are set to music for solo voice and piano by a composer. Cone regards his work not as offering a theory of musical meaning, but rather as “prefatory to any theory of musical meaning or musical expression,” though it is probably true to say that such a theory of musical meaning is clearly implied throughout the text. In any case, Cone outlines the theory of musical meaning implicit in the main body of the book in a lengthy epilogue entitled “Utterance and Gesture.”

Music, unlike language, is non-referential; it has neither a single
translatable meaning, in the denotative sense, nor multiple meanings in the
Text, Music, and Meaning in Cong regational Song 83 more connotative poetic sense. What music does share with language, however, is its gestural aspect — the dimension of language that relies on inflection and
context to convey meaning. The word “Oh!,” for example, will be spoken in
many different ways and convey as many different meanings depending on
whether it is a response to — for example — the news of a death, an implausible excuse for not completing homework, the news that one’s daughter is pregnant, or finally figuring out why a circuit breaker keeps tripping. The context of the utterance will determine which gestural shape (and which meaning) is appropriate. It is this gestural aspect of utterance that is “simulated, and symbolized, by music.”

But the gestures of musical utterance remain only potentially expressive
unless they have a context within which to resonate. The contexts of a musical gesture are two: the complete network of gestures which makes up the musical composition as a whole; and the context of human activity and experience which each listener brings to his or her encounter with the music, and which may also be at least partially provided as part of the musical work itself, as in the text of a song, or the program of a symphonic poem.

It is sung text which most directly provides a context for interpreting
the expressive potential of the music. “When the gestures of the music are closely analogous to those implied by the words,” then the effect created is that “the music expresses the emotion, mood, activity, or attitude revealed by the text.” But the expressive potential of a piece of music is not limited to the interpretation suggested by a single text. Consider the strophic song (of which the hymn is of course an example):

The fact that a given musical setting can be applied to a number of different stanzas need not mean that the music is expressively neutral, since for any of the Schubert examples [or for any hymn tune] it would be easy to find stanzas that would fit metrically but would not work because of expressive disparity. What strophic song suggests is that a piece of music allows a wide but not unrestricted range of possible expression: this is what I call its expressive potential. A given
text specifies one possibility, or at most a relatively narrow range of possibility, its verbal formulation providing the immediate context that renders the musical gestures emotionally, etc., expressive.4

The meaning of a song, then, “is not revealed by the words alone but
by the quasi-metaphorical relation between the words and music.” Its expressive content “emerges from the mutual relations of words and musical gestures, and from the light they throw on each other. A song is thus a kind of metaphor, an equation whose significance consists, not in what it states about either of the two members, but in the coupling itself. . . .”

If it is true that various poetic stanzas set to the same music bring out
different aspects of that music’s expressive potential (i.e., create different
“hearings” of the music), it is also true that different music employed as the setting for the same poetic text will likewise produce various “readings” of the text. Of course, as Cone points out, there must be a sufficient degree of similarity between the expressive potential of the music and the potential readings of the poetic text for this process to take place. This might fail to occur either because the expressive character of text and music are too different from one another, or because the music is too neutral or weak in expressive character to evoke much of a reading at all.

Let us turn now to Cone’s theory of musical utterance itself. Briefly
put, his view is that whereas “[i]n the poem, it is the poet who speaks . . . , [i]n the song, it is the composer who speaks, in part through the words of the poet.”5 When composing an art song, the composer does not “set” the poem itself, but rather “appropriates his own reading of a preexisting poem in order to use it as one component of a new work of art . . . [W]hat we then hear in the words is less the poet speaking than the composer through the poet.”6

Cone arrives at this formulation by what he calls a “dramatistic” analysis.
When we read a poem, we understand it as being uttered by a poetic persona whose character may or may not closely resemble the poet himself. Composing a song involves the creation of at least two other personas: the vocal persona, portrayed by the singer and consisting of the poetic text set to the vocal line; and the complete musical persona (or “composer’s persona”), made up of vocal persona and accompaniment together. This distinction between vocal persona and complete musical persona is an important one, because the singer is provided with only some of the musical material. The accompaniment provides information about aspects of the vocal persona that may be outside that vocal persona’s awareness.

It follows from this analysis that “in most encounters between poetry and music, poetry can become the more powerful of the two only by the intentional acquiescence or the unintentional incompetence of the composer.”7 One of the consequences of this reality is that a wonderful musical setting can redeem a mediocre poetic text. Brahms, for example, was notoriously drawn to weak poets, yet managed to create masterful songs using their poems as a starting point. The reverse, however, is not true: a great poem will seldom save a mediocre musical setting, but will be dragged down with it.

Cone’s analysis was undertaken especially with the art song of the
nineteenth and twentieth centuries in mind, beginning with Schubert, who was the first composer of genius to regard the poem as raw material for the creation of a new work of art. Before this time, the German art song, or Lied, had been dominated by another conception, one which prevented it from growing into a fully mature art form. This opposing view was championed by no less powerful and influential a figure than Goethe, himself the author of many lyric and narrative poems set to music by Lied composers. Goethe’s view was that the Lied should be text- rather than music-dominated, its poems set to simple, strophic music with optional accompaniment that disturbed the natural rhythms and inflections of the poetry as little as possible. His interest, in other words, was in having the poet’s persona continue to speak, in part through the music of the composer, rather than in providing poetic material from which the composer might create something fundamentally new. Among Goethe’s
favourite Lied composers were C.F. Zelter and J.F. Reichardt — names whose pallid musical settings are little remembered today. Goethe had a low opinion of Beethoven’s settings of his texts, and returned a parcel of Schubert’s songs to the composer without comment.

So long as Goethe’s view of the Lied prevailed, composers of genius
were not prepared to devote significant energies to it. The songs of Mozart, Haydn, and even Beethoven are among their lesser achievements (with the exception of Beethoven’s song-cycle “An die ferne Geliebte,” which attempted to combine elements of Lied style in individual songs with a more musically sophisticated overall structural design).8 It was Schubert more than anyone who redefined the Lied by his genius for lyric composition and his radically new approach to song composition.

Of course, there are significant differences between the mature art song
and the congregational hymn. The art song is sung by a soloist to an audience; the hymn is sung by a group, normally by all present — there is no passive audience. Hymns are addressed to someone or something, normally either to God, or to oneself or to other members of the congregation; art songs rarely have an actual addressee. In their traditional four-part texture, hymns do not require an instrumental accompaniment; if they do have one, it is normally simply a doubling of the four vocal lines, although an alternate harmonization or a descant may be added, especially for the final verse. In the art song, a fully developed accompaniment containing distinctive musical material is essential to create a larger musical context for the vocal line. Hymns must be rigidly strophic in both textual and musical structure; the structure of the art
song is potentially more flexible, ranging from strophic to through- composed.

Two questions naturally arise in light of these differences between the
hymn and the art song: How far can the insights of Cone’s analysis be applied to hymnody? and, Is hymnody more like the earlier text-dominated conception of the Lied or the later music-dominated one?

Cone does actually make a few comments about hymns. He calls them
an instance of “functional song,” to which he believes his dramatistic analysis does not apply. The reason he gives is that in hymns (and other functional song), the vocal persona is not a dramatic character, but “an aspect of the actual singer[s] at the time of singing. In functional song, the singer expresses himself directly as a member of a specific community.”9 The implications of this observation for hymn singing, however, are perhaps more far-reaching than Cone takes note of. It means that the hymn, rather than being an aesthetic object, experienced at a certain critical distance by an audience, is identified with directly by the congregation, so that while they sing, the persona implied by the text and music is taken on by the singers. The singer imaginatively becomes the vocal persona. The singer’s act of dramatic impersonation in the
case of the art song is, in a hymn performance, undertaken by all. Whatever power the text and music have is amplified by being experienced “from the inside.”10

But how much is the hymn like the mature art song? Is it closer in
character to the best of Schubert’s Lieder, in which the text is absorbed by the music in the service of creating a new art form? Or is it more like the worst of the early Lieder, in which the music is too neutral or too weak in expressive character to evoke much of a reading of the text at all?

The answer, perhaps unsurprisingly, is that the hymn occupies a space
along a continuum somewhere between these two extremes. That space is
itself something of a continuum, spanning relatively stronger and weaker hymn tunes. The difference between the early and the mature Lied lies primarily in the strength of its music. But even after the emergence of the art song as a mature musical genre, weak, unsuccessful songs continued to be written. The musical weakness of the early Lied was due not only to the “intentional acquiescence” of the composers to the text-dominated ideal, but also to their limited compositional gifts. Beginning with Schubert, great composers regarded the art song as a viable form, and contributed masterpieces of the genre to the repertoire. But although weak musical settings were unlikely now to be produced due to intentional acquiescence to the text-dominated conception, many were composed due to the limited talents of composers.

So too with hymn tunes: their musical strengths vary. Some are rather
neutral expressively, but most are sufficiently strong to contribute significantly to a text-music complex that creates new meaning from the interaction of elements.

It must be admitted that in some respects, the hymn is like the early
Lied: the music is rigidly strophic, the text setting is almost exclusively syllabic, and the accompaniment usually does not have a distinctive identity. But within the very real constraints of the hymn tune form, the repertoire of hymn tunes is stronger musically than that of the early Lied. The harmonic vocabulary is richer, the four-part texture is fuller, and the character of the repertoire is both more serious and more varied. The hymnic repertoire is also richer because it embraces so much wider a variety of styles, having been composed over a much longer period of time.

Another important difference between the repertoire of hymns and the
art song, whether early or later, is that it consists of a body of texts and tunes, classifiable according to a relatively restricted number of metres, which can be combined in hundreds of ways. In effect, this “modular” aspect of hymns leaves part of the process of composition to the hymn-book editor, worship planner, or performer. The particular interaction of text and tune which makes up a given hymn is ultimately beyond the control of the poet and the composer. This characteristic of hymns might seem at first to be a weakness. If texts and tunes can be combined so freely, how musically strong could the tunes possibly be?

But a practical test demonstrates otherwise. Chose any hymn stanza in
Common Meter (CM) and sing it to all the CM tunes found in a current
denominational hymnal (probably 40 or 50 tunes). Most of them will be
inappropriate. Frequently the tune will be obviously wrong for the text,
sometimes hilariously so. But there will be at least a few and perhaps even as many as a dozen that produce a reasonable hymn. In other words, there is sufficient musical strength or character in most hymn tunes to prevent random combinations of texts and tunes.

Our common repertoire of hymn texts and tunes, then, as constituted
by the aggregate of our various denominational hymn books, provides a
laboratory for the testing of our model of how hymn texts and tunes act on each other to create a distinctive entity, the hymn. Many of the hymn texts in our common repertoire are sung to differing tunes in different denominational traditions. And many of them have become associated with a single tune that has been more or less universally adopted by the English-speaking church. But even where there is general agreement about the association of a particular text and tune, these associations are contingent, not necessary. Many of them may well represent the best possible union of text and available tune. But even in these cases, the singing of the text to that particular tune creates only one possible reading of the text. Other readings lie dormant in the text, waiting to
be activated by a tune with different musical characteristics.

I want now to test the model I have been proposing on a well-known
text, “Amazing Grace,” one that has come to be universally associated with a particular tune, NEW BRITAIN. The particular text is chosen pretty much at random, but also because the association of text and tune has become so deeply ingrained. Even in the case of such a familiar and well-loved hymn, different tunes create new readings of the text, provided we are able to set aside temporarily the familiar association of text and tune. One of the effects of always singing a hymn text to the same tune, especially if the text is one that has not, for the singer, had an independent existence as a poem, is that it becomes very difficult to experience the text apart from its musical associations. Because the familiar hymn tune creates a particular reading of the poem — in effect acting as a filter through which we hear the text — we have to make a
conscious effort to “hear” the text apart from any music before we can “hear” it set to a variety of tunes.

Here is the text of “Amazing Grace” as it appears in most North American
hymn books:

Amazing grace, how sweet the sound,
That saved a wretch like me!
I once was lost, but now am found,
Was blind, but now I see.

’Twas grace that taught my heart to fear,
And grace my fears relieved;
How precious did that grace appear
The hour I first believed!

Through many dangers, toils and snares
I have already come;
’Tis grace has brought me safe thus far,
And grace will lead me home.

The Lord has promised good to me,
His word my hope secures;
He will my shield and portion be
As long as life endures.

When we’ve been there ten thousand years,
Bright shining as the sun,
We’ve no less days to sing God’s praise
Than when we’d first begun.

This is not John Newton’s original text, however, nor is it even all by Newton. The last stanza of this version originally appeared anonymously in a collection published in Richmond, Virginia, in 1790, and was not attached to “Amazing Grace” until 1910, in an American collection called Coronation Hymns.11 In the version printed above it replaces Newton’s original two final stanzas:

Yes, when this flesh and heart shall fail,
And mortal life shall cease:
I shall possess, within the veil,
A life of joy and peace.

The earth shall soon dissolve like snow,
The sun forbear to shine;
But God, who called me here below,
Will be forever mine.

Time and familiarity have made the replacement stanza seem a natural
conclusion to the poem. But a closer look reveals how different it is in tone, style, and perspective from the preceding stanzas. Newton’s text is written in the first person singular: “I”, “my” and “mine” appear frequently throughout. The transition to the anonymous stanza is awkward.12 The new stanza’s content is similar to the two stanzas it replaces, but the movement from “as long as life endures” to “When we’ve been there ten thousand years” requires the reader to infer where “there” is (and to adapt to being included suddenly in the shift from “I” to “we”). All of Newton’s verses are carefully constructed in two halves, with a strong ABAB rhyme scheme. The added stanza, however, is a single statement that runs through the four lines, and only its second and fourth lines rhyme. There is also an internal rhyme in the third line (“days”/ “praise”) which has no parallel in Newton’s text. Nor is the language of the anonymous stanza the equal of Newton’s: it lacks both his strong syntax and colorful vocabulary.

The first stanza of the poem announces grace as its subject, contrasting
the author’s former state with his present one. The remainder of the poem is ordered chronologically. It speaks of the action of grace in the past, both before and after coming to belief (stanza 2), returns to the present (stanza 3), then turns to the future. Stanza 4 considers the remainder of the author’s life on earth; the original stanza 5, the transition to life after death; and the sixth stanza, the final judgment. All of this is related from the perspective of the author’s experience, either actual or imagined. The replacement stanza also speaks of the experience of life after death, more in the manner of stanza 6 than of stanza 5.

In his discussion of Newton’s hymn texts, J.R. Watson notes the poet’s
placement of crucial words at the ends of stanzas and half-stanzas, a device that is much in evidence here.13 Newton delights in the employment of multiple images, especially when he can juxtapose extreme or dissimilar ones: “’Twas grace that taught my heart to fear, and grace my fears relieved,” “shield and portion,” “dangers, toils and snares,” “dissolve like snow”/“forbear to shine.” Above all, Watson identifies what he calls Newton’s “unabashed concentration

New Britain

New Britain music staff

on the self” (285), by which he means his fascination with “a certain kind of religious experience” (286). This is a text that is not only written from the perspective of the first person singular, but which views even the most unimaginably cosmic events — the end of creation — through the filter of personal experience. By the time we reach the final stanza of Newton’s original text, this orientation has become almost bizarre in its self-centeredness. Perhaps it was a sense that the juxtapositions of Newton’s last stanza were too extreme that contributed to its being replaced by the anonymous “wandering stanza” now commonly sung.

Like many hymn texts, “Amazing Grace” has become almost universally
associated with a particular tune.NEW BRITAIN, shown in Example 1, probably originated as a folk tune, and was first published in 1829. It first appeared as the setting for “Amazing Grace” in William Walker’s Southern Harmony, in 1835 (with all six of Newton’s original stanzas). The tune is pentatonic, and the effect of this on contemporary singers and listeners, as with many such tunes, is a slightly rugged and primitive one. Because of this, it is well-suited to texts whose language is strong and direct

Harmonizations of this tune vary somewhat from hymnal to hymnal,
but are generally limited to four or five chords, and have a slow harmonic
rhythm, with two or three measures of uninterrupted tonic in a few places.
The tune is in triple meter, with an almost unvarying rhythm of Ï | ú ,
occasionally modified to Ï Ï | ú or Ïj Ï Ï | Ï .

This rhythm has a couple of effects on the text. One is to create the
simplest possible declamatory rhythm, with half notes falling consistently on stressed syllables: “a-MA-zing GRACE, how SWEET the SOUND. . . .” The other is to allow the singer to savor each of the stressed syllables as it goes by: “grace,” “sweet,” “saved, “found,” “see. . . .”

David Douglas has written that to sing this text to NEW BRITAIN is to
experience “the comfort of salvation,”14 and many of the characteristics of this tune invite just this experience. The uninterrupted rocking rhythm of the tune suggests security, predictability, a kind of luxurious warmth.

Moving from the level of individual measures to the phrase, we see that
three of the four phrases begin identically (1, 2 and 4), and that the fourth phrase is an abbreviated version of the first: the comfort of predictable repetition Text, Music, and Meaning in Cong regational Song 93
is here as well. Three of the four phrases (1, 3 and 4) are arch-shaped, each beginning with an energetic upwards movement, and then settling more gently down onto a cadence. The second phrase departs from this pattern by ending on a high C rather than low one, imbuing this phrase with a sense of exultation the others lack. Finally, this high ending of the second phrase helps to create an overall arch shape for the entire tune, with its highest notes occurring in the middle two of the four phrases, and the outer two phrases peaking a third lower.

The cumulative effect of these rhythms and melodic shapes is oceanic,
with the regular lapping of measure-long rhythms subsumed into the larger
crests of the phrases, and these in turn parts of a yet larger swell. For the
singer, the tune is unchallenging: smooth, uncomplicated, regular and
symmetrical in construction, phrases requiring the most breath at their beginnings when most breath is available, and tapering off nicely as the breath itself wanes. It is a sensual pleasure to sing this tune.

What kind of reading does the interaction of this particular tune with
this particular text create? The subject of the text is the personal experience of salvation, and the qualities of the tune certainly allow those personal and experiential dimensions to come through. In the first stanza, the words “sweet,” “found,” and “see” come particularly alive, while “wretch” and even “amazing” seem softened. The singer seems to sing from a place of security, warmth, absolute assurance. Life before encountering grace is mentioned, but not recalled in either feeling or imagination. The spaciousness of the tune, the ease with which its first, third, and fourth phrases descend stepwise to their cadences makes the phrases “as long as life endures” and “grace will lead me home” seem very real. The text and tune seem less well suited to each other in the final (replacement) stanza. Here the words seem to call for some sort of intensification, perhaps because of the change of perspective to collective
activity — this is the first time the text speaks of the singer actually doing
something. Perhaps it is this shift of tone which has probably led to the
frequently encountered practice of singing the final stanza up a semitone.

So, there is good reason for this text and tune to have become so closely linked with one another. NEW BRITAIN captures something essential about “Amazing Grace” — probably its most essential feature — that “certain kind of religious experience,” the comfort and assurance of personal salvation which Newton juxtaposes with past, present, and future circumstances in almost every stanza of his text. But even here, where text and tune seem so perfectly “matched,” the interaction of the text with other tunes produces other possible readings of the poem.

My purpose here is not a practical one — the tunes we are about to
look at are not likely to be used as alternatives to NEW BRITAIN in congregational singing. For one thing, the associations both with the text and with some of the tunes I will suggest are too strong for most people to adjust easily to a change of tune. The reader will need to do his or her best to set aside pre-existing associations with these tunes. Nor am I necessarily suggesting that these alternative tunes are equally successful settings for “Amazing Grace.” NEW BRITAIN will probably never be surpassed in that regard. But as we have seen, no single tune can capture all of the possible connotations of a text, and other readings are bound to emerge as we consider alternative tunes.

My purpose here is not a practical one — the tunes we are about to
look at are not likely to be used as alternatives to NEW BRITAIN in congregational singing. For one thing, the associations both with the text and with some of the tunes I will suggest are too strong for most people to adjust easily to a change of tune. The reader will need to do his or her best to set aside pre-existing associations with these tunes. Nor am I necessarily suggesting that these alternative tunes are equally successful settings for “Amazing Grace.” NEW BRITAIN will probably never be surpassed in that regard. But as we have seen, no single tune can capture all of the possible connotations of a text, and other readings are bound to emerge as we consider alternative tunes.

So let us turn now to another very familiar and popular tune, ANTIOCH, sung universally to the text, “Joy to the World!” Lowell Mason composed ANTIOCH in Handelian style, claiming only to be the “arranger” of certain short phrases from Messiah, though this claim is too modest. Although it was first published at about the same time (1836) as NEW BRITAIN, it is an example of very different musical style and aims. ANTIOCH  is an exuberant tune with a strong rhythmic character,  elementary harmony (I-IV-V), and a rather uninteresting melodic outline. Its structure is irregular: it is classified as “CM with repeat” rather than simply “CM” (like both “Amazing Grace” and NEW BRITAIN) because of multiple repetitions of the fourth and final line of each stanza:

Joy to the world! The Lord is come:
Let earth receive her King;
Let every heart prepare him room,
And heaven and nature sing,
And heaven and nature sing,
And heaven,
And heaven and nature sing.

Or, if we include the dialogue between women’s and men’s voices in the
second half of the tune:


Antioch music staff

Joy to the world! The Lord is come:
Let earth receive her King;
Let every heart prepare him room,
And heaven and nature sing,
(And heaven and nature sing)
And heaven and nature sing,
(And heaven and nature sing)
And heaven,
And heaven and nature sing

These repetitions strongly emphasize the final line of each stanza. As
shown in Example 2, the repetition works well for most of the stanzas of
“Amazing Grace,” either because the final line expresses the central idea of the stanza, or because, given the psalm-like parallelisms Newton typically employs, the last line captures the sense of the entire stanza.

But more than just adding emphasis, the repetitions built into the tune
are an expression of joyful exuberance. The repetitions aren’t textually
necessary: they are there for the pleasure of singing them, and because the joy of the singer is too full to be contained by a regular CM structure. It is as if the tune simply cannot contain itself.

The dialogue between women’s and men’s voices is part of this mood
of celebration, and creates the sense of a community interaction. So it is the final (added) stanza that benefits most from being sung to this music, because the poetic voice is now “we,” not “I,” and also because the tune illustrates the text so aptly: “We’ve no less days to sing God’s praise,” and so we sing the words “than when we’d first begun” over and over. Singing God’s praise is the culminating activity of both the song and our life, the end for which we were created. (Incidentally, the musical repetition at measures 8-11 nicely parallels the internal rhyme “days”/“praise” of line 3 of the last stanza).

The dialogue between women’s and men’s voices is part of this mood
of celebration, and creates the sense of a community interaction. So it is the final (added) stanza that benefits most from being sung to this music, because the poetic voice is now “we,” not “I,” and also because the tune illustrates the text so aptly: “We’ve no less days to sing God’s praise,” and so we sing the words “than when we’d first begun” over and over. Singing God’s praise is the culminating activity of both the song and our life, the end for which we were created. (Incidentally, the musical repetition at measures 8-11 nicely parallels the internal rhyme “days”/“praise” of line 3 of the last stanza).

Hymn tunes are typically 16 to 32 measures long, too short to accommodate internal contrast. They have a homogenous texture and therefore a single expressive character from beginning to end. This musical character acts like a lens or filter through which the text is experienced, amplifying certain words and phrases, muting the effect of others. Where NEW BRITAIN tends to amplify words that express sensory experience, like “sweet,” or to invite us to experience sensory associations of words, like “precious” and “home,” ANTIOCH adds no special flavor to these words. Instead, ANTIOCH’s enthusiastic rejoicing is more attuned to phrases that describe actions: “saved a wretch like me,” and “grace my fears relieved,” for example.

A hymn tune’s musical character also creates or implies an experiential
state from which the words are being sung, and within which they are entered into imaginatively by the singer. The singer participates imaginatively in the subjective space created by the tune and experiences the text from within that space. Of course, there must be sufficient congruity between the meanings expressed by the text and the subjective space implied by the tune in order for meaningful interaction between the two to take place. We have already noted that the character of ANTIOCH may be described with words like “jubilant,” “extroverted,” “unreflective,” “enthusiastic,” and “declamatory.” How does a subjective state described by these sorts of adjectives interact with the text of “Amazing Grace”?

We might imagine that the person singing these words to this particular
tune is closer in time to the life-changing experience of grace than the singer of NEW BRITAIN. Where the singer of NEW BRITAIN seems to have had time to absorb and integrate the significance of grace in his life, the singer of ANTIOCH still seems overwhelmed with excitement and even some incredulity at the newness of the experience. In this he is perhaps closer to the man born blind of John 9:25 to whom Newton alludes: “He answered, ‘I do not know whether he [Jesus] is a sinner. One thing I do know, that though I was blind, now I see.’” The way the juxtaposition of opposites of the first stanza is expressed in the music creates the impression that the singer is trying to convince himself, and can do little more than to repeat what he knows must be true, but seems too wonderful to be grasped: “I once was lost/but now am found,” set to the
same phrase of music accompanying the parallel syntax, thereby heightening the opposition of meaning of the two text phrases; and the delirious repetition of:

Was blind but now I see,
(Was blind but now I see)
Was blind but now I see,
(Was blind but now I see)
Was blind,
Was blind but now I see,

with emphasis given to the last three words by the return of the more declamatory rhythm of the first measures of the tune. ANTIOCH also seems to imply a more social context than NEW BRITAIN, in part for the reasons already mentioned above, but also because of its extroverted, declamatory style. NEW BRITAIN is more ruminative, and could well be heard as being sung to oneself. But ANTIOCH is bursting with good news, as if the story must be told, and not only told, but heard as well.

In Example 3 we find a sixteenth-century tune by the great English
composer Thomas Tallis, one of nine he wrote to accompany Matthew Parker’s The Whole Psalter Translated into English Metre of c.1567. It is not a CM tune but CMD (Common Meter Double), and therefore requires an even number of four-line stanzas to fill out the music. Accordingly, we will consider it with Newton’s original six stanzas, rather than with the added stanza, which in any case does not suit the tune as well.

The tune is a modal one, on E, with an unstable third scale degree that
oscillates between Gn and G#. While it has a steady pulse, its meter is irregular; the only bar lines occur at the end of lines of music (that is, after every two lines of text). The harmonic rhythm is slow, usually with several quarter-note beats per chord. Chords on E are held the longest, sometimes e-g-b, and sometimes e-g#-b, because of the changeable third degree of the modal scale. The two halves of the tune are somewhat different in character. The first two lines create a rather static effect: most of the melody notes are repeated several times, and the movement is primarily stepwise around B, which serves almost as a reciting tone. The range is only a sixth, and the harmony is mostly a triad on E. The second line is an almost exact repetition of the first. In its second half, the tune becomes more animated: the range expands to a ninth, repeated notes

The Third Tune

The Third Tune music staffare fewer, harmony is more varied. There are a few dotted rhythms,
and a short sequence at the beginning of the last line. The last two lines are dissimilar from one another. The third line rises to an early climax on E, the high point of the entire tune, and moves gradually down through an octave to a cadence on the lower E. The last line begins with a descending sequence and again cadences on the lower E.

This music has a static, somewhat floating quality that is created by a
number of features: the absence of regular meter, repeated notes and chords, and the modal flavor with its G/G# instability. Affectively, these features produce an expression of awe or mystery. The absence of a strong meter, particularly, so unlike the other two settings, creates a sense of other-worldliness that contrasts with the strongly embodied character of NEW BRITAIN and ANTIOCH. Accordingly, the music de-emphasizes what is so strongly present in the other settings, the experience of grace. Instead, singing “Amazing Grace” to THE THIRD TUNE invites contemplation of grace itself: awesome, a mystery, “amazing” in the traditional sense. Sung to this music, the first person singular
language of the poetry is humbled and the sense is almost of unworthiness. Instead of singing about grace as experienced within a human frame of reference, we find ourselves considering the paradox of the unfathomable mystery of God’s grace as an unknowable spiritual reality. The further the text moves forward in its description of the grace’s gifts to me, the deeper the experience of awe becomes. Who am I to receive such lavish generosity?

The first stanza is sung to the more subdued first half of the tune.
Beneath the tune, the shift between Gand G# delicately underscores the
oppositions between lost and found, blindness and sight. The tune expands into its second half for the second stanza. The climactic phrase accompanies the words, “’Twas grace that taught my heart to fear,” while the music to the answering text phrase “and grace my fears relieved” is much less intense, illustrating something of the experience of these two processes. Then the more lyrical sequence at the beginning of the last line of music adds a poignancy to the lines, “How precious did that grace appear.”

Of course, the same succession of musical gestures undergirds verses 3
and 4, and 5 and 6. We will note only the aptness of the music that accompanies Newton’s sixth and final stanza. In this instance, the intensity and relaxation of the third line of music heightens the apocalyptic lines, “The earth shall soon Text, Music, and Meaning in Cong regational Song 101 dissolve like snow, the sun forbear to shine,” while the sweeter character of the sequence that follows is associated with, “But God, who called me here below, will be forever mine.” The otherworldliness of the music suggests that I will be as much possessed by God as God will by me. (Contrast this with the more mundane sense of possession implied by NEW BRITAIN.)

Finally, we turn to Example 4, a Scottish psalter tune, LONDON NEW,
first published in 1635, in the century after THE THIRD TUNE. Like NEW BRITAIN, it is a straightforward CM tune, most closely associated with another text from Olney Hymns, “God Moves in a Mysterious Way,” by Newton’s collaborator William Cowper. LONDON NEW illustrates the effect of a tune of more neutral expressive character than those we have been considering, one that has a weaker impact on its poetic text.

London New

London New music notes

This relative weakness is not so much a characteristic of the tune itself
as it is of the Scottish psalm-tune genre, at least in the form in which it is
represented in contemporary hymnals. For one thing, tunes of this type tend to resemble one another. Substituting the tune DUNFERMLINE or CAITHNESS for DUNDEE will have a minimal impact on whatever text it accompanies. But this in itself does not mean these tunes are musically weak — a pair of identical twins may each be beautiful; the fact that there are two of them does not diminish their individual beauty. Rather, it is the fact that their similarity to one another arises from their identical and uninteresting rhythm, whether in the versions with “gathering” notes at the beginning of each phrase:

ú | Ï Ï Ï Ï | Ï Ï ú ú | Ï Ï Ï Ï | ú ú | Ï Ï Ï Ï | Ï Ï ú ú | Ï Ï Ï Ï | ú
or without:
Ï | Ï Ï Ï Ï | Ï Ï Ï Ï | Ï Ï Ï Ï | ú. Ï | Ï Ï Ï Ï | Ï Ï Ï Ï | Ï Ï Ï Ï | ú.
This rhythmic strait-jacketing (which was not a feature of most of these tunes in their earliest recorded versions), assists unaccompanied ensemble singing, but is both dull and predictable. While the best of these tunes exhibit great variety and beauty in their melodic construction, their rhythmic monotony seems to inhibit significantly their ability to infuse their texts with much energy. If anything, their steady and even rhythmic motion tends to focus attention on the words themselves.

As an example of the Scottish psalm tune, LONDON NEW is remarkable
for its movement primarily by leap, and leaps of all sizes: ascending and
descending 2nds, 3rds, 4ths, and 6ths, and ascending 5ths. Three of its four phrases span an octave, while its third phrase moves in smaller intervals in the upper half of the tune’s range. The melodic intervals tend to alternate direction — up, down, up, down. The shape of the tune this movement creates is unusual, rugged and jagged, and requires energy to sing. (It is easy to imagine that it was this aspect of the tune that suggested its association with “God moves in a mysterious way.”) Here, for the first time, we have a tune with a fast harmonic rhythm, a change of harmony with almost every melody note. The effect of such a harmonic rhythm is to give weight to each note, and to suggest a slower tempo than a tune with a slower harmonic rhythm. The tune Text, Music, and Meaning in Cong regational Song 103 does not so much move fluidly through the lines of text as weigh each syllable carefully.

So the rhythmic flatness of the Scottish psalm tune style is counteracted
to some extent by the angular movement of the melody. But this tune is much less an expression of feeling associated with some aspect of the text, and more an invitation to take in the words as poetic text — an effect more like the earlier art song, in fact.

What LONDON NEW does bring to the text of “Amazing Grace” is a
sense of travel, of journey. “I once was lost but now am found,” and “Through many dangers, toils and snares I have already come; ’tis grace that brought me safe thus far, and grace will bring me home” — these are phrases that seem to be directly illustrated by the contours of this angular tune. In this version of the hymn, the singer is still in the midst of an active journey, and grace is leading him from somewhere on ahead. We have exchanged the warmth and comfort of NEW BRITAIN, whose impact on the text creates more of a focus on the end of the journey, for the vigor and strength of a journey still in progress.15

The influence of music — even relatively weak music — on how the text it
accompanies is perceived and understood is inescapable. When we experience a text only in association with a particular musical setting, we may be utterly unaware of how the music shapes and delimits our perception of it. But the filtering effect of the musical setting is nonetheless present, probably most powerfully in precisely those cases where the text has come to be associated with a single musical setting. Having more than one musical option for singing a text means that the poetry has not become so thoroughly fused by familiarity with a particular tune.

A hymn, then, is neither text nor tune alone, but the product of the
reciprocal interaction of text and tune on one another. We need to think of the hymn text more like an opera libretto or a screenplay — the foundational component for the creation of a larger work of art, but one which does not and cannot remain inert, impervious to dynamic interaction with the other media that contribute to the total finished work.

Thinking of hymns in this way has a number of both practical and
theoretical consequences. I can do little more than enumerate them here. One of the significant practical consequences of thinking of hymns in this way lies in the area of worship planning. In recent decades, the biblical index has become an essential part of the denominational hymn book, permitting the selection of hymns to be sung on a given Sunday to be coordinated more closely with the lections of the day. This is certainly a good and important development. But as we have seen, a congregation does not experience these words directly, but as interpreted through the medium of the musical setting. Worship planners need to attend to the reading of the hymn text which will be received by the congregation, not just the text in isolation.

We have explored how the contextualizing effect that text and music
have on one another in congregational hymns creates new meaning. But of
course this is not the whole story. The reading created by the interaction of a text and tune is not experienced in a vacuum, but within liturgical, cultural, social, and other larger contexts, each of which modifies how we perceive and understand the musical-textual reading. For example, the placement of a hymn (or any other musical or textual item) within the liturgical structure will affect how it is heard. Whether “Amazing Grace” is sung at the beginning of a service, as a response to a reading, or during communion will have an influence on the total meaning of the hymn as experienced by the congregation. The nature of the service itself can also create a powerful context which colors how we understand the meaning of a congregational song. To sing “Amazing Grace” at a baptism, a funeral, or a regular Sunday service — each of these experiences will differ considerably from one another.

It is not only hymn texts but any liturgical text whose meaning will be
affected by its musical setting. Indeed, where a text is more general in content and more familiar to the singer from routine repetition (e.g., “Lord have mercy,” “Holy, holy”), the role of music in creating meaning will be so much greater, the experience of the text much more about the expressive character of the music. Our perception even of purely instrumental music is guided by the liturgical context in which we hear it, for, as Cone observes, an accompanying text is only one way of specifying a “context of human experience and activity.” The very fact that a piece of music — whether texted or not — is heard in the context of worship and not in the concert hall already limits and directs the range of human experience and activity which will be brought to bear on the experience of the music.

The meanings created by the texts and music of our worship do not only express the spirituality of a congregation; they form it as well.16  We readily recognize the formative power of the texts we use in worship, carefully weighing the pastoral appropriateness of the words we chose, testing affirmations of faith for theological correctness. The formative power of music is less often considered, and yet is potentially much stronger than that of the text. Music is a non-rational, non-verbal medium. It is perceived with a sensual and imaginative immediacy that penetrates more readily to the deeper layers of mind and spirit, more easily slipping past the critical, judging function that words are naturally subject to. And it is music that guides and shapes our perception of those textual meanings whose appropriateness we take so much care over. How much more consideration might we give to this potent agent of meaning, not only in our hymns, but in every aspect of our worship?


1 David Cole, “Singing the faith,” The Hymn 51:3 (July 2000), 24-27.
2 Edward T. Cone, The Composer’s Voice (Berkeley: University of California Press, 1974).
3 In 1989, on the fifteenth anniversary of its publication, a session of six papers responding to the book was held at the annual meeting of the College Music Society. These were published, together with Cone’s response, in College Music Symposium 29 (1989).
4 Ibid., 166-67.
5 Ibid., 19.
6Ibid., 41.
7 Ibid., 45.
8 See Joseph Kerman, “An die ferne Geliebte,” Beethoven Studies 1, ed. Alan Tyson (New York: W.W. Norton, 1973), 123-57.
9 Cone, The Composer’s Voice, 49.
10 Philip H. Pfatteicher has also noted this dimension of the experience of hymn-singing. See his “Hymns in the Life of the Church,” Cross Accent 2 (July 1993): 7-15, esp. 8 and 15.
11Psalter Hymnal Handbook, ed. Emily R. Brink and Bert Polman (Grand Rapids: CRC Publications, 1998), 628. The added verse is an example of “a ‘wandering’ stanza in CM that appears at the end of a variety of hymns in nineteenth-century hymnals.” (The Hymnal 1982 Companion, ed. Raymond F. Glover, vol. 3B [New York: Church Hymnal Corporation, 1994], 671).
12 In some recent hymnals, the order of verses 3 and 4 is reversed to ease the transition.
13 J.R. Watson, The English Hymn: A Critical and Historical Study (Oxford: Clarendon Press, 1997), 282-88.
14 David Douglas, “Amazing Grace: A Journey in Time and Faith,” The Hymn 49:3 (July 1998), 12.
15 In the context of considering the steady rhythmic character of this tune, it is interesting to look briefly at ST. PETER, not a Scottish psalm tune, but an early nineteenth-century English tune. It also moves in even quarter notes (like the second rhythm given above) and has a fast harmonic rhythm, though it is much smoother and stepwise in melodic motion. The text to which it usually joined, “How Sweet the Name of Jesus Sounds,” is also by John Newton, and shares features with “Amazing Grace”: the image of “sweet sound” in the first stanza, and the “unabashed concentration on the self.” But it is much easier to imagine singing it to NEW BRITAIN than it is singing “Amazing Grace” to ST. PETER.
16 See, for example, Edward Foley, “Musical forms, referential meaning, and belief,” Ritual Music: Studies in Liturgical Musicology (Beltsville, MD, 1995), 145-72; Don E. Saliers, “Liturgical musical formation,” Liturgy and Music: Lifetime Learning, ed. Robin A. Leaver and Joyce Ann Zimmerman (Collegeville, MN: Liturgical Press, 1998), 384-94; and Saliers, “Sound spirituality: on the formative expressive power of music for Christian spirituality,” Christian Spirituality Bulletin 8:1 (Spring/Summer 2000): 1, 3-5.