{"slug": "ninety-and-nine", "name": "THE NINETY AND NINE", "meter_label": "Irregular", "meter": {"pattern": [10, 7, 9, 7, 9, 9, 9], "doubled": false, "refrain": false, "alleluia": false, "irregular": true}, "composer": "ira-david-sankey", "composer_date": "1874", "canonical_key": {"tonic": "aes", "mode": "major"}, "preferred_harmonization": "ninety-and-nine-gospel-hymns-1883", "provenance": {"layer": "tune", "source": "gospel-hymns-sacred-songs-1875", "source_kind": "hand_encoded", "pd_basis": "pub_pre_1931", "pd_evidence": "Ira D. Sankey composed this tune extempore at the Free Assembly Hall, Edinburgh, in 1874, to Clephane's already-printed poem; it is through-composed and irregular \u2014 the seven lines of stanza 1 run 10.7.9.7.9.9.9 and the other stanzas do not scan the same way. First printing read for this record: Gospel Hymns and Sacred Songs (P. P. Bliss & Ira D. Sankey; New York: Biglow & Main / Cincinnati: John Church & Co., 1875), No. 6, printed page 8, headed \"The Ninety and Nine\" with \"Elizabeth C. Clephane. 1868.\" at the left and \"Ira D. Sankey, by per.\" at the right \u2014 archive.org gospelhymnssacre00blis_0, leaf n11. US public domain: published 1875, well before 1931. The tune carries no printed NAME on that page (Gospel Hymns numbers rather than names its tunes); \"THE NINETY AND NINE\" is the name it goes by in the later tune indexes and is used here. Fingerprinted against all 275 corpus tunes before engraving \u2014 best match 0.53 (ST. FLAVIAN), so it is not a duplicate under another name.", "pd_jurisdictions": {"US": true}, "hymnary_ids": {"tune_auth": "there_were_ninety_and_nine_that_sankey", "total_instances": 223}}, "harmonization": {"slug": "ninety-and-nine-gospel-hymns-1883", "tune": "ninety-and-nine", "music_file": "ninety-and-nine-gospel-hymns-1883.ly", "canonical_key": {"tonic": "aes", "mode": "major"}, "time": "6/8", "pickup": "8", "system_breaks": [11, 19, 30, 39, 50, 61], "breaks_manual": true, "final_barline": "|.", "amen": false, "verse_variable_melisma": true, "provenance": {"layer": "harmonization", "source": "gospel-hymns-consolidated-1883", "source_kind": "hand_encoded", "pd_basis": "pub_pre_1931", "pd_evidence": "Ira D. Sankey's own four-part setting, read off the page: Gospel Hymns and Sacred Songs, Consolidated Nos. 1-4 (P. P. Bliss & Ira D. Sankey; New York: Biglow & Main / Cincinnati: John Church & Co., 1883), No. 6, printed page 8 \u2014 archive.org gospelhymnsconso00blis_202606, leaf n9 \u2014 headed 'The Ninety and Nine' over the epigraph 'Rejoice with me, for I have found my sheep that was lost.' \u2014 Luke 15: 6, credited 'Elizabeth C. Clephane, 1868.' and 'Ira D. Sankey, by per.', with stanzas 1 and 2 underlaid and 3-5 as text below. It is a four-part SHORT SCORE (soprano and alto on the treble staff, tenor and bass on the bass staff): A-flat major, 6/8, a two-sixteenth pickup, fourteen measures, 72 soprano notes. US public domain \u2014 published 1883, well before 1931 \u2014 and no modern arranger stands anywhere in the chain.\n\nWHY THIS PRINTING AND NOT THE FIRST. The same setting is in the first printing, Gospel Hymns and Sacred Songs, 1875, No. 6, page 8 (archive.org gospelhymnssacre00blis_0, leaf n11), which was read bar by bar first and agrees with the 1883 plate in every pitch. (NOT in every rhythm, and this record used to claim otherwise: at mm. 12 and 14 both plates print a held quarter in the lower staff under the soprano's slurred pair, where this encoding restrikes. That is the corpus's universal convention \u2014 278 of 278 other harmonizations are homorhythmic \u2014 but it is a deviation from the page and is now stated as one, in the .ly header and here.) But this setting's whole character is four chromatic accidentals, and the 1875 scan does not resolve them: the alto's E natural at the head of measures 9 and 10 ('way on the' / 'wild and bare'), which with the soprano's c'' and a bass c lifts a plain C-major triad into the middle of A-flat, and the alto's D natural on the last chord of measure 13 against the tenor's des', answered by the alto's D flat at the head of measure 14 ('from the ten-der Shep-'). All four are printed, each with its own accidental before the LOWER notehead, on the 1883 leaf and can be read there; on the 1875 leaf they are not legible. So the 1883 plate is the witness of record \u2014 the page these chords were actually taken from. A third dated printing, Christ in Song (rev. and enl., 1908), No. 61, page 61 (archive.org songfora00beld, leaf n76), carries the same notes with four verses underlaid.\n\nWHAT THE PLATE PRINTS AND THIS RECORD DOES NOT. (a) Fermatas, omitted per house style as the Bach chorales in this corpus omit theirs: 1883 prints them over the final chord and at two interior cadences, 1875 prints only the final one. (b) The 'rit.' over measure 9, which 1883 and 1908 print and 1875 does not. (c) The plate slurs several note-pairs that stanza 1 sings SYLLABICALLY \u2014 the two pickup notes, the first pair of measure 1, the last pair of measure 5 \u2014 because the slurs are drawn for the union of all five stanzas' melismas; in this corpus a drawn slur means the reference stanza's melisma, so only stanza 1's twelve are drawn here and the others live as '_' in the setting. \n\nNOT the Cyber Hymnal. Its NoteWorthy file (TheNinetyAndNine.nwc) was decompressed and read as a transcription scaffold only, and its |SongInfo| Comments names the edition it was copied from \u2014 'Trinity Hymnal (Philadelphia, Pennsylvania: Orthodox Presbyterian Church, 1961), # 137', a POST-1930 book, so it can be no part of the public-domain basis. It differs from the dated plates in exactly two places, and in both the plate was followed: (1) measures 4 and 8 ('fold,' / 'gold,') print a dotted quarter, a QUARTER REST and then the two sixteenths that pick up the next line, in all four voices \u2014 the Cyber Hymnal has no rest and ties a quarter on instead; (2) the plate ends on the quarter of measure 14 under a fermata, five eighths long, exactly what the one-eighth pickup owes back \u2014 the Cyber Hymnal ties that quarter across a further barline into a half plus a quarter rest, a fifteenth measure that appears in no dated printing.", "pd_jurisdictions": {"US": true}, "encoding_origin": "Hand-encoded from the page images of archive.org gospelhymnsconso00blis_202606 leaf n9 (Gospel Hymns and Sacred Songs, Consolidated Nos. 1-4, 1883, No. 6, p. 8), read staff-position by staff-position and render-diffed against that leaf and against archive.org gospelhymnssacre00blis_0 leaf n11 (the 1875 first printing); the Cyber Hymnal NoteWorthy file TheNinetyAndNine.nwc (decompressed from byte 6) served only as a scaffold and was corrected against the plate where it differs.", "encoding_license": "PD-dedication"}, "music_ly": "%% THE NINETY AND NINE (Irregular) \u2014 Ira D. Sankey's own setting, 1874.\n%% Four SATB voice streams, canonical key A-flat major, 6/8, \\partial 8 (the\n%% pickup is TWO SIXTEENTHS, not one eighth); 14 measures, 72 soprano notes.\n%% RULES (CI-enforced): exactly the variables soprano/alto/tenor/bass/global;\n%% absolute octaves; each voice rhythmically complete; NO chords <>, NO <<>>,\n%% NO \\secondShift/\\unisonWholeShift/tightSpacing, NO \\break, NO \\bar, NO lyrics.\n%%\n%% SOURCE OF THE NOTES.  The witness of record is \"Gospel Hymns and Sacred Songs,\n%% Consolidated Nos. 1-4\" (P. P. Bliss & Ira D. Sankey; New York: Biglow & Main /\n%% Cincinnati: John Church & Co., 1883), No. 6, printed page 8 \u2014 archive.org\n%% gospelhymnsconso00blis_202606, leaf n9.  The page is a four-part SHORT SCORE:\n%% soprano/alto share the treble staff, tenor/bass the bass staff, one rhythm per\n%% staff.  The FIRST\n%% printing, \"Gospel Hymns and Sacred Songs\" (1875), No. 6, page 8 (archive.org\n%% gospelhymnssacre00blis_0, leaf n11) was read bar by bar first and agrees in\n%% every PITCH \u2014 but its scan does not resolve the four accidentals\n%% below, which the 1883 leaf prints legibly, so the 1883 leaf is what these chords\n%% were taken from.  A third dated printing, \"Christ in Song\" (rev. & enl., 1908),\n%% No. 61, page 61 (archive.org songfora00beld, leaf n76) carries the tune with four\n%% verses underlaid, but as a DIFFERENT arrangement bearing \"Copyright, 1904, by\n%% Ira D. Sankey\" (still public domain) \u2014 it is a corroborating witness to the tune,\n%% not to these chords.\n%%\n%% ONE DELIBERATE DEVIATION FROM THE PLATE, recorded because the header used to\n%% claim there was none.  At mm. 12 and 14 the lower staff prints a HELD QUARTER\n%% across the melody's slurred pair \u2014 m.12 has five treble noteheads over four bass\n%% onsets (tenor E-flat/bass E-flat sustained while the soprano moves bes'->aes'),\n%% m.14 four over three (tenor G/bass E-flat held under bes'->c'').  Both the 1875\n%% and the 1883 leaf agree, so it is not a scan ambiguity.  This encoding RESTRIKES\n%% instead (ees'8( ees'8) / g8( g8)), because that is this corpus's universal\n%% convention: all 278 other harmonizations are strictly homorhythmic across the\n%% four voices, and at all 816 moving-soprano slurs corpus-wide the lower voices\n%% restrike and never hold.  Encoding the holds would be the corpus's first\n%% voice-independent rhythm and needs an engine check and an owner decision \u2014 so\n%% the deviation is stated here rather than hidden behind a \"note for note\" claim.\n%%\n%% NOT the Cyber Hymnal.  Its NoteWorthy file for this hymn (TheNinetyAndNine.nwc)\n%% names its own edition in |SongInfo| Comments \u2014 \"Trinity Hymnal (Philadelphia:\n%% Orthodox Presbyterian Church, 1961), # 137\" \u2014 i.e. a POST-1930 book, so it can\n%% be no part of the public-domain basis.  It was decompressed and read only as a\n%% transcription scaffold, and it differs from the dated plates in exactly the two\n%% places recorded below; in both, the plate wins.\n%%\n%% THE TWO BREATH BARS (measures 4 and 8, \"fold,\" and \"gold,\").  The plate prints,\n%% in ALL FOUR voices, a dotted quarter, then a QUARTER REST, then the two\n%% sixteenths that pick up the next line \u2014 3 + 2 + 1 = 6 eighths.  The Cyber\n%% Hymnal file has no rest there: it ties the dotted quarter to a quarter instead.\n%% The rest is what both dated printings show (the old-style quarter rest, drawn\n%% as a reversed eighth rest, is plain on leaf n9 and on leaf n11), and it is the\n%% reason the singer breathes before \"But\" and before \"A-way\".  Kept as printed\n%% rather than collapsed: the house collapse convention is documented for 4/4 only,\n%% and here the rest carries no lyric slot at all.\n%%\n%% THE LAST BAR.  Measure 14 is FIVE eighths (three eighths + a quarter), which is\n%% exactly what the one-eighth pickup owes back.  The plate ends on that quarter,\n%% under a fermata, with no rest after it and no further bar.  The Cyber Hymnal\n%% file instead ties that quarter across a bar into a half plus a quarter rest \u2014 a\n%% fifteenth measure that appears in no dated printing; it is dropped.\n%%\n%% FERMATAS, omitted per house style (as the Bach chorales in this corpus do), and\n%% recorded here instead: 1883 prints one over the final chord (soprano note 72) in\n%% both staves, and two more at interior cadences; 1875 prints only the final one.\n%% 1883 and 1908 also print \"rit.\" over measure 9; 1875 does not.\n%%\n%% CHROMATICS \u2014 the fingerprint of this setting, all four printed on leaf n9 with\n%% their own accidental before the LOWER notehead.  (a) The ALTO's E natural at the\n%% head of measures 9 and 10 (\"way on the\" / \"wild and bare\"), against the\n%% soprano's c'' and a bass c: the plain C-major triad Sankey lifts into the middle\n%% of A-flat.  (b) The alto's D natural on the last chord of measure 13, against\n%% the tenor's des', answered by the alto's D flat at the head of measure 14\n%% (\"from the ten-der Shep-\").  The plate's D flat is a COURTESY accidental \u2014 the\n%% note is diatonic in A-flat and only wants marking because the natural stood in\n%% the bar before \u2014 so LilyPond does not reprint it, and it is not forced here.\n%%\n%% SLURS.  The plate slurs several note-pairs that stanza 1 sings SYLLABICALLY \u2014\n%% the two pickup notes, the first pair of measure 1, the last pair of measure 5 \u2014\n%% because its slurs are drawn for the union of all five stanzas' melismas.  In\n%% this corpus a drawn slur means the REFERENCE stanza's melisma, so only stanza\n%% 1's twelve are drawn here (continuations at soprano notes 9, 16, 21, 23, 32, 37,\n%% 41, 46, 56, 59, 67, 70); the other stanzas' held notes live as '_' in the\n%% setting's stanza_lines.  verse_variable_melisma, because Sankey composed this\n%% extempore and the five stanzas do NOT scan alike \u2014 stanza 3's third line has\n%% eleven syllables where stanza 1's has nine.\n\\version \"2.24.0\"\n\nglobal  = { \\key aes \\major \\time 6/8 \\partial 8 }\n\nsoprano = {\n  ees'16 ees'16 |\n  aes'8. aes'16 aes'8 aes'4 aes'8 |\n  g'8( g'8) g'8 aes'4 aes'16 aes'16 |\n  c''4 c''16( c''16) aes'4 aes'8 |\n  bes'4. r4 ees'16( ees'16) |\n  aes'8( aes'8) aes'8 aes'4 aes'16 aes'16 |\n  c''4 c''8 aes'4 aes'16( aes'16) |\n  bes'4 bes'16 bes'16 g'8( g'8) g'8 |\n  aes'4. r4 aes'16( aes'16) |\n  c''8. c''16 c''8 aes'8( aes'8) aes'8 |\n  c''4 c''8 aes'4 ees''8 |\n  ees''8. ees''16 ees''8 ees''8( c''8) aes'8 |\n  bes'8( aes'8) bes'8 c''4 ees''8 |\n  ees''8. ees''16 f''8 ees''8( c''8) aes'8 |\n  bes'8( c''8) bes'8 aes'4\n}\n\nalto = {\n  c'16 c'16 |\n  ees'8. ees'16 ees'8 ees'4 ees'8 |\n  ees'8( ees'8) ees'8 ees'4 ees'16 ees'16 |\n  aes'4 aes'16( aes'16) ees'4 ees'8 |\n  g'4. r4 c'16( c'16) |\n  ees'8( ees'8) ees'8 ees'4 ees'16 ees'16 |\n  aes'4 aes'8 ees'4 ees'16( ees'16) |\n  f'4 f'16 f'16 ees'8( ees'8) ees'8 |\n  c'4. r4 c'16( c'16) |\n  e'8. e'16 e'8 f'8( f'8) f'8 |\n  e'4 e'8 f'4 ees'8 |\n  aes'8. aes'16 aes'8 aes'8( aes'8) ees'8 |\n  g'8( f'8) g'8 aes'4 aes'8 |\n  aes'8. aes'16 aes'8 aes'8( ees'8) d'8 |\n  des'8( ees'8) des'8 c'4\n}\n\ntenor = {\n  aes16 aes16 |\n  c'8. c'16 c'8 c'4 c'8 |\n  bes8( bes8) bes8 c'4 c'16 c'16 |\n  ees'4 ees'16( ees'16) c'4 c'8 |\n  ees'4. r4 ees'16( ees'16) |\n  c'8( c'8) c'8 c'4 c'16 c'16 |\n  ees'4 ees'8 c'4 aes16( aes16) |\n  aes4 bes16 bes16 bes8( bes8) bes8 |\n  aes4. r4 aes16( aes16) |\n  g8. g16 c'8 c'8( c'8) c'8 |\n  c'4 c'8 c'4 bes8 |\n  c'8. c'16 c'8 c'8( ees'8) aes8 |\n  ees'8( ees'8) ees'8 ees'4 c'8 |\n  c'8. c'16 des'8 c'8( aes8) bes8 |\n  g8( g8) g8 aes4\n}\n\nbass = {\n  aes,16 aes,16 |\n  aes8. aes16 aes8 aes4 aes8 |\n  ees8( ees8) ees8 aes4 aes16 aes16 |\n  aes4 aes16( aes16) aes4 aes8 |\n  ees4. r4 aes16( aes16) |\n  aes8( aes8) aes8 aes4 aes16 aes16 |\n  aes4 aes8 aes4 c16( c16) |\n  des4 des16 des16 ees8( ees8) ees8 |\n  aes,4. r4 f16( f16) |\n  c8. c16 c8 f8( f8) f8 |\n  c4 c8 f4 g8 |\n  aes8. aes16 aes8 aes8( aes8) c8 |\n  ees8( ees8) ees8 aes4 aes8 |\n  aes8. aes16 aes8 aes8( aes8) f8 |\n  ees8( ees8) ees8 aes,4\n}\n"}, "harmonizations": [{"slug": "ninety-and-nine-gospel-hymns-1883", "tune": "ninety-and-nine", "music_file": "ninety-and-nine-gospel-hymns-1883.ly", "canonical_key": {"tonic": "aes", "mode": "major"}, "time": "6/8", "pickup": "8", "system_breaks": [11, 19, 30, 39, 50, 61], "breaks_manual": true, "final_barline": "|.", "amen": false, "verse_variable_melisma": true, "provenance": {"layer": "harmonization", "source": "gospel-hymns-consolidated-1883", "source_kind": "hand_encoded", "pd_basis": "pub_pre_1931", "pd_evidence": "Ira D. Sankey's own four-part setting, read off the page: Gospel Hymns and Sacred Songs, Consolidated Nos. 1-4 (P. P. Bliss & Ira D. Sankey; New York: Biglow & Main / Cincinnati: John Church & Co., 1883), No. 6, printed page 8 \u2014 archive.org gospelhymnsconso00blis_202606, leaf n9 \u2014 headed 'The Ninety and Nine' over the epigraph 'Rejoice with me, for I have found my sheep that was lost.' \u2014 Luke 15: 6, credited 'Elizabeth C. Clephane, 1868.' and 'Ira D. Sankey, by per.', with stanzas 1 and 2 underlaid and 3-5 as text below. It is a four-part SHORT SCORE (soprano and alto on the treble staff, tenor and bass on the bass staff): A-flat major, 6/8, a two-sixteenth pickup, fourteen measures, 72 soprano notes. US public domain \u2014 published 1883, well before 1931 \u2014 and no modern arranger stands anywhere in the chain.\n\nWHY THIS PRINTING AND NOT THE FIRST. The same setting is in the first printing, Gospel Hymns and Sacred Songs, 1875, No. 6, page 8 (archive.org gospelhymnssacre00blis_0, leaf n11), which was read bar by bar first and agrees with the 1883 plate in every pitch. (NOT in every rhythm, and this record used to claim otherwise: at mm. 12 and 14 both plates print a held quarter in the lower staff under the soprano's slurred pair, where this encoding restrikes. That is the corpus's universal convention \u2014 278 of 278 other harmonizations are homorhythmic \u2014 but it is a deviation from the page and is now stated as one, in the .ly header and here.) But this setting's whole character is four chromatic accidentals, and the 1875 scan does not resolve them: the alto's E natural at the head of measures 9 and 10 ('way on the' / 'wild and bare'), which with the soprano's c'' and a bass c lifts a plain C-major triad into the middle of A-flat, and the alto's D natural on the last chord of measure 13 against the tenor's des', answered by the alto's D flat at the head of measure 14 ('from the ten-der Shep-'). All four are printed, each with its own accidental before the LOWER notehead, on the 1883 leaf and can be read there; on the 1875 leaf they are not legible. So the 1883 plate is the witness of record \u2014 the page these chords were actually taken from. A third dated printing, Christ in Song (rev. and enl., 1908), No. 61, page 61 (archive.org songfora00beld, leaf n76), carries the same notes with four verses underlaid.\n\nWHAT THE PLATE PRINTS AND THIS RECORD DOES NOT. (a) Fermatas, omitted per house style as the Bach chorales in this corpus omit theirs: 1883 prints them over the final chord and at two interior cadences, 1875 prints only the final one. (b) The 'rit.' over measure 9, which 1883 and 1908 print and 1875 does not. (c) The plate slurs several note-pairs that stanza 1 sings SYLLABICALLY \u2014 the two pickup notes, the first pair of measure 1, the last pair of measure 5 \u2014 because the slurs are drawn for the union of all five stanzas' melismas; in this corpus a drawn slur means the reference stanza's melisma, so only stanza 1's twelve are drawn here and the others live as '_' in the setting. \n\nNOT the Cyber Hymnal. Its NoteWorthy file (TheNinetyAndNine.nwc) was decompressed and read as a transcription scaffold only, and its |SongInfo| Comments names the edition it was copied from \u2014 'Trinity Hymnal (Philadelphia, Pennsylvania: Orthodox Presbyterian Church, 1961), # 137', a POST-1930 book, so it can be no part of the public-domain basis. It differs from the dated plates in exactly two places, and in both the plate was followed: (1) measures 4 and 8 ('fold,' / 'gold,') print a dotted quarter, a QUARTER REST and then the two sixteenths that pick up the next line, in all four voices \u2014 the Cyber Hymnal has no rest and ties a quarter on instead; (2) the plate ends on the quarter of measure 14 under a fermata, five eighths long, exactly what the one-eighth pickup owes back \u2014 the Cyber Hymnal ties that quarter across a further barline into a half plus a quarter rest, a fifteenth measure that appears in no dated printing.", "pd_jurisdictions": {"US": true}, "encoding_origin": "Hand-encoded from the page images of archive.org gospelhymnsconso00blis_202606 leaf n9 (Gospel Hymns and Sacred Songs, Consolidated Nos. 1-4, 1883, No. 6, p. 8), read staff-position by staff-position and render-diffed against that leaf and against archive.org gospelhymnssacre00blis_0 leaf n11 (the 1875 first printing); the Cyber Hymnal NoteWorthy file TheNinetyAndNine.nwc (decompressed from byte 6) served only as a scaffold and was corrected against the plate where it differs.", "encoding_license": "PD-dedication"}, "music_ly": "%% THE NINETY AND NINE (Irregular) \u2014 Ira D. Sankey's own setting, 1874.\n%% Four SATB voice streams, canonical key A-flat major, 6/8, \\partial 8 (the\n%% pickup is TWO SIXTEENTHS, not one eighth); 14 measures, 72 soprano notes.\n%% RULES (CI-enforced): exactly the variables soprano/alto/tenor/bass/global;\n%% absolute octaves; each voice rhythmically complete; NO chords <>, NO <<>>,\n%% NO \\secondShift/\\unisonWholeShift/tightSpacing, NO \\break, NO \\bar, NO lyrics.\n%%\n%% SOURCE OF THE NOTES.  The witness of record is \"Gospel Hymns and Sacred Songs,\n%% Consolidated Nos. 1-4\" (P. P. Bliss & Ira D. Sankey; New York: Biglow & Main /\n%% Cincinnati: John Church & Co., 1883), No. 6, printed page 8 \u2014 archive.org\n%% gospelhymnsconso00blis_202606, leaf n9.  The page is a four-part SHORT SCORE:\n%% soprano/alto share the treble staff, tenor/bass the bass staff, one rhythm per\n%% staff.  The FIRST\n%% printing, \"Gospel Hymns and Sacred Songs\" (1875), No. 6, page 8 (archive.org\n%% gospelhymnssacre00blis_0, leaf n11) was read bar by bar first and agrees in\n%% every PITCH \u2014 but its scan does not resolve the four accidentals\n%% below, which the 1883 leaf prints legibly, so the 1883 leaf is what these chords\n%% were taken from.  A third dated printing, \"Christ in Song\" (rev. & enl., 1908),\n%% No. 61, page 61 (archive.org songfora00beld, leaf n76) carries the tune with four\n%% verses underlaid, but as a DIFFERENT arrangement bearing \"Copyright, 1904, by\n%% Ira D. Sankey\" (still public domain) \u2014 it is a corroborating witness to the tune,\n%% not to these chords.\n%%\n%% ONE DELIBERATE DEVIATION FROM THE PLATE, recorded because the header used to\n%% claim there was none.  At mm. 12 and 14 the lower staff prints a HELD QUARTER\n%% across the melody's slurred pair \u2014 m.12 has five treble noteheads over four bass\n%% onsets (tenor E-flat/bass E-flat sustained while the soprano moves bes'->aes'),\n%% m.14 four over three (tenor G/bass E-flat held under bes'->c'').  Both the 1875\n%% and the 1883 leaf agree, so it is not a scan ambiguity.  This encoding RESTRIKES\n%% instead (ees'8( ees'8) / g8( g8)), because that is this corpus's universal\n%% convention: all 278 other harmonizations are strictly homorhythmic across the\n%% four voices, and at all 816 moving-soprano slurs corpus-wide the lower voices\n%% restrike and never hold.  Encoding the holds would be the corpus's first\n%% voice-independent rhythm and needs an engine check and an owner decision \u2014 so\n%% the deviation is stated here rather than hidden behind a \"note for note\" claim.\n%%\n%% NOT the Cyber Hymnal.  Its NoteWorthy file for this hymn (TheNinetyAndNine.nwc)\n%% names its own edition in |SongInfo| Comments \u2014 \"Trinity Hymnal (Philadelphia:\n%% Orthodox Presbyterian Church, 1961), # 137\" \u2014 i.e. a POST-1930 book, so it can\n%% be no part of the public-domain basis.  It was decompressed and read only as a\n%% transcription scaffold, and it differs from the dated plates in exactly the two\n%% places recorded below; in both, the plate wins.\n%%\n%% THE TWO BREATH BARS (measures 4 and 8, \"fold,\" and \"gold,\").  The plate prints,\n%% in ALL FOUR voices, a dotted quarter, then a QUARTER REST, then the two\n%% sixteenths that pick up the next line \u2014 3 + 2 + 1 = 6 eighths.  The Cyber\n%% Hymnal file has no rest there: it ties the dotted quarter to a quarter instead.\n%% The rest is what both dated printings show (the old-style quarter rest, drawn\n%% as a reversed eighth rest, is plain on leaf n9 and on leaf n11), and it is the\n%% reason the singer breathes before \"But\" and before \"A-way\".  Kept as printed\n%% rather than collapsed: the house collapse convention is documented for 4/4 only,\n%% and here the rest carries no lyric slot at all.\n%%\n%% THE LAST BAR.  Measure 14 is FIVE eighths (three eighths + a quarter), which is\n%% exactly what the one-eighth pickup owes back.  The plate ends on that quarter,\n%% under a fermata, with no rest after it and no further bar.  The Cyber Hymnal\n%% file instead ties that quarter across a bar into a half plus a quarter rest \u2014 a\n%% fifteenth measure that appears in no dated printing; it is dropped.\n%%\n%% FERMATAS, omitted per house style (as the Bach chorales in this corpus do), and\n%% recorded here instead: 1883 prints one over the final chord (soprano note 72) in\n%% both staves, and two more at interior cadences; 1875 prints only the final one.\n%% 1883 and 1908 also print \"rit.\" over measure 9; 1875 does not.\n%%\n%% CHROMATICS \u2014 the fingerprint of this setting, all four printed on leaf n9 with\n%% their own accidental before the LOWER notehead.  (a) The ALTO's E natural at the\n%% head of measures 9 and 10 (\"way on the\" / \"wild and bare\"), against the\n%% soprano's c'' and a bass c: the plain C-major triad Sankey lifts into the middle\n%% of A-flat.  (b) The alto's D natural on the last chord of measure 13, against\n%% the tenor's des', answered by the alto's D flat at the head of measure 14\n%% (\"from the ten-der Shep-\").  The plate's D flat is a COURTESY accidental \u2014 the\n%% note is diatonic in A-flat and only wants marking because the natural stood in\n%% the bar before \u2014 so LilyPond does not reprint it, and it is not forced here.\n%%\n%% SLURS.  The plate slurs several note-pairs that stanza 1 sings SYLLABICALLY \u2014\n%% the two pickup notes, the first pair of measure 1, the last pair of measure 5 \u2014\n%% because its slurs are drawn for the union of all five stanzas' melismas.  In\n%% this corpus a drawn slur means the REFERENCE stanza's melisma, so only stanza\n%% 1's twelve are drawn here (continuations at soprano notes 9, 16, 21, 23, 32, 37,\n%% 41, 46, 56, 59, 67, 70); the other stanzas' held notes live as '_' in the\n%% setting's stanza_lines.  verse_variable_melisma, because Sankey composed this\n%% extempore and the five stanzas do NOT scan alike \u2014 stanza 3's third line has\n%% eleven syllables where stanza 1's has nine.\n\\version \"2.24.0\"\n\nglobal  = { \\key aes \\major \\time 6/8 \\partial 8 }\n\nsoprano = {\n  ees'16 ees'16 |\n  aes'8. aes'16 aes'8 aes'4 aes'8 |\n  g'8( g'8) g'8 aes'4 aes'16 aes'16 |\n  c''4 c''16( c''16) aes'4 aes'8 |\n  bes'4. r4 ees'16( ees'16) |\n  aes'8( aes'8) aes'8 aes'4 aes'16 aes'16 |\n  c''4 c''8 aes'4 aes'16( aes'16) |\n  bes'4 bes'16 bes'16 g'8( g'8) g'8 |\n  aes'4. r4 aes'16( aes'16) |\n  c''8. c''16 c''8 aes'8( aes'8) aes'8 |\n  c''4 c''8 aes'4 ees''8 |\n  ees''8. ees''16 ees''8 ees''8( c''8) aes'8 |\n  bes'8( aes'8) bes'8 c''4 ees''8 |\n  ees''8. ees''16 f''8 ees''8( c''8) aes'8 |\n  bes'8( c''8) bes'8 aes'4\n}\n\nalto = {\n  c'16 c'16 |\n  ees'8. ees'16 ees'8 ees'4 ees'8 |\n  ees'8( ees'8) ees'8 ees'4 ees'16 ees'16 |\n  aes'4 aes'16( aes'16) ees'4 ees'8 |\n  g'4. r4 c'16( c'16) |\n  ees'8( ees'8) ees'8 ees'4 ees'16 ees'16 |\n  aes'4 aes'8 ees'4 ees'16( ees'16) |\n  f'4 f'16 f'16 ees'8( ees'8) ees'8 |\n  c'4. r4 c'16( c'16) |\n  e'8. e'16 e'8 f'8( f'8) f'8 |\n  e'4 e'8 f'4 ees'8 |\n  aes'8. aes'16 aes'8 aes'8( aes'8) ees'8 |\n  g'8( f'8) g'8 aes'4 aes'8 |\n  aes'8. aes'16 aes'8 aes'8( ees'8) d'8 |\n  des'8( ees'8) des'8 c'4\n}\n\ntenor = {\n  aes16 aes16 |\n  c'8. c'16 c'8 c'4 c'8 |\n  bes8( bes8) bes8 c'4 c'16 c'16 |\n  ees'4 ees'16( ees'16) c'4 c'8 |\n  ees'4. r4 ees'16( ees'16) |\n  c'8( c'8) c'8 c'4 c'16 c'16 |\n  ees'4 ees'8 c'4 aes16( aes16) |\n  aes4 bes16 bes16 bes8( bes8) bes8 |\n  aes4. r4 aes16( aes16) |\n  g8. g16 c'8 c'8( c'8) c'8 |\n  c'4 c'8 c'4 bes8 |\n  c'8. c'16 c'8 c'8( ees'8) aes8 |\n  ees'8( ees'8) ees'8 ees'4 c'8 |\n  c'8. c'16 des'8 c'8( aes8) bes8 |\n  g8( g8) g8 aes4\n}\n\nbass = {\n  aes,16 aes,16 |\n  aes8. aes16 aes8 aes4 aes8 |\n  ees8( ees8) ees8 aes4 aes16 aes16 |\n  aes4 aes16( aes16) aes4 aes8 |\n  ees4. r4 aes16( aes16) |\n  aes8( aes8) aes8 aes4 aes16 aes16 |\n  aes4 aes8 aes4 c16( c16) |\n  des4 des16 des16 ees8( ees8) ees8 |\n  aes,4. r4 f16( f16) |\n  c8. c16 c8 f8( f8) f8 |\n  c4 c8 f4 g8 |\n  aes8. aes16 aes8 aes8( aes8) c8 |\n  ees8( ees8) ees8 aes4 aes8 |\n  aes8. aes16 aes8 aes8( aes8) f8 |\n  ees8( ees8) ees8 aes,4\n}\n", "preferred": true}], "license": "The hymn texts and tunes are in the public domain; per-layer provenance is recorded on each record."}