This book, a sequel to the author's treatment of the eighth-mode tracts, [1] proposes that the compositional principles underlying the tracts of the second mode are best understood in terms of the relationship between words and music, especially in the selective emphasis of particular theological and liturgical themes. This is a provocative approach. Although analytical studies carried out during the past thirty years have shown plainchant melodies to be acutely responsive to their texts at the level of prosody, grammar and syntax, and intelligent declamation and emphasis, most of these findings have been based on antiphonal genres of chant (introits, communion antiphons, Office antiphons, and, in certain respects, offertories) that make only sparing use of long melismas and in which there is little recycling of formulaic phrases. [2] Tracts, like graduals, alleluias and Office responsories, make use of extremely long melismas and of formulaic phrases, apparently recycled indiscriminately with unrelated texts, and this has been seen to preclude any meaningful relationship between words and music, except perhaps at the level of establishing an extended meditative space in which the words can be heard and digested. [3]
The book may be divided into two unequal parts. In the first (Chapters 1 to 5), Hornby argues, on textual and liturgical grounds, that four of the second-mode tracts comprise an originally Roman core repertory and that these possess a "musical grammar" according to which every verse of the tract will be expected to follow a standard melodic pattern. As is well known, second-mode tract verses are made up of four phrases with the target cadence pitches D, C, F and D (43), and there are various stock phrases that can be used in each of these positions. Hornby's central thesis, however, is that textual considerations have often influenced the choice of which stock phrase is used. Some of these, again, are familiar to scholarship, such as formulae chosen when the text ends in a dactyl rather than a spondee. In other cases, of which Hornby adduces some new examples, some phrases obviously became associated in cantors' minds with configurations of similar words (Hornby calls these "text cues"), like "verb plus pronoun." In still other instances, a phrase used to begin one verse of a tract will be repeated on a later occasion, or even in a different chant, when a phrase begins with the same word or words (31). Hornby's arguments are presented in constant dialogue with earlier writings on the second-mode tracts by Hans Schmidt, Helmut Hucke, Theodore Karp and Andreas Pfisterer, and she often has fresh insight to add to old observations.
A more intelligent aspect of the music's response to the text could be called punctuation, that is "articulating the text with reference both to the musical and to the textual grammar" (34). The point in a text where a division might be made is sometimes ambiguous--Hornby identifies a nice example where a tract divides a grammatically ambiguous sentence in the same way recommmended by St Jerome (33). Sometimes the music seems to contradict the natural grammatical division of the text, and this may sometimes be explained with reference to textual considerations; but as Hornby points out, musical factors are often more plausible (21).
Finally, the use of "emphatic formulaic phrases", which depart from the expected patterns of the "musical grammar", can sometimes be explained as a way of emphasizing or connecting certain texts, and "use of these phrases moves the second-mode tracts beyond a reading of the text which communicates its sense to a reading of the text which conveys an interpretation of its most important words and ideas" (64). This has nothing to do with any especially emphatic character of the melodies themselves. "Simply by virtue of transcending the usual formulaic shapes, they have the potential to emphasise text or to link texts within a chant where they are used" (67). The key to understanding the reasons for the use of these emphatic phrases, Hornby argues, is to be found in patristic and medieval exegesis of the texts. Hornby refers to the various series of Tituli Psalmorum, psalter collects, and the standard commentaries of Augustine, Jerome, Cassiodorus and others. So, for example, Augustine and Cassiodorus both remark on the rhetorical structure of Psalm 101 (Hebrew 102) as an intensifying series of petitions, and Hornby observes that several of the invocations in the tract text are set to emphatic phrases (103-4).
In Chapter 5, Hornby applies her "musical grammar" hypothesis to the question of genre. The gradual De necessitatibus is made up entirely of formulaic phrases from the second-mode tract repertory, so why is it not a tract? Responsorial performance (i.e. refrain-verse-refrain) is not a sufficient distinction, because it would appear from internal musical evidence and from a comparison of rubrication in a number of early manuscripts that two of the core-repertory tracts were also probably performed in this way, namely Domine exaudi and Domine audiui. Hornby proposes instead that the four tracts of the core repertory are distinguished from all other kinds of chant by "the treatment of text, presence of formulaic musical structures, and use of emphatic melodic material to promote particular theological messages" (135).
A second part of the book (Chapter 6 and 7) addresses the adoption and expansion of the second-mode tract repertory by the Franks. Chapter 6 is a fascinating study of the Frankish Good Friday tract Eripe me, composed perhaps as early as the 790s. Hornby notes that although the tract makes use of the same collection of formulaic phrases as the four tracts of the core repertory, they are not used "correctly" according to the tracts' underlying musical grammar (140-42). This could suggest that the Frankish cantors did not fully understand the compositional principles of the repertory they were imitating. Indeed, Hornby persuasively argues that the music for Eripe me (from Psalm 139, Hebrew 140) was probably a deliberate imitation of the tract Deus deus meus (Psalm 21, Hebrew 22), whose text, with its many parallels to the events of the Crucifixion, was seen by all early and medieval commentators as the Passion psalm par excellence. Noting that in patristic exegesis Psalm 139 "was not associated with the Passion, but was interpreted as the Church crying out for protection from evil, from the devil and particularly from heretics" (143), Hornby suggests that the insertion of this tract into the Good Friday liturgy was a response to the Adoptionist heresy (149-50) and that the use of melodic material from Deus deus meus promoted an interpetation of Psalm 139 against the backdrop of the Passion, since Christ's simultaneously perfect humanity and divinity--denied by Adoptionists--was seen as essential for the efficacy of the sacrifice of Calvary. Hornby's explanation of a musical link with the Nativity (first noticed by Theodore Karp) as a way of emphasizing the orthodox view of Christ's divine humanity is very striking (110). Chapter 7 surveys the appearance of four additional second-mode tracts in ninth- and tenth-century manuscripts, noting their connection with the liturgical calendar, their textual sources, and issues in establishing their melodies from the early notations. Hornby considers that "their separation from the core-repertory second-mode tracts is implicit in almost every characteristic of their notated form" (178) and concurs with Pfisterer's observation that the melodic transmission of these new chants, which seem to have entered the repertory around the same time that musical notation was coming into widespread use, was far less consistent than that of the core repertory, which was transmitted orally with great fidelity. Notation "may have damaged the understanding of the repertory" (179).
The issue of transmission, written and oral, is a constant theme throughout the book, and the disjunction between the supposed "core repertory" and later Frankish compositions is telling in this respect. The question of whether the Romano-Frankish or Old Roman tradition preserves the older form of the melody is frequently addressed, and Hornby finds many cases illustrating Pfisterer's model of "progressive stereotyping", in which non-standard phrases gradually gravitate, in the oral tradition, towards standard melodic shapes. In most cases, the Old Roman phrase choices are seen as a later divergence (49). Comparison of various manuscripts in the Romano-Frankish tradition reveals some interesting cases of what appears to be written-out ornamentation (61, Phrase 1c). Hornby's preference for musical "gesture" as the basic unit of oral transmission, instead of older authors' attempts to see variations of tonal skeletons, seems very sound (60 n.83, 71 n.117). Her comparison of formulaic phrases in the Romano-Frankish and Old Roman versions is illuminating: in some cases, apparently unrelated melodies are used to perform precisely the same "grammatical" function throughout the genre in each tradition, which lends considerable strength to the idea of a musical grammar as a governing principle not only in the tracts' composition, but in their transmission too (48, on Hornby's Phrase 1d).
Hornby's analysis of the second-mode tracts through the lens of the words-music relationship is very interesting and very promising. The book is marred, however, by inconveniences that make the argument hard to follow. The first three chapters are infuriating to read because the opening two chapters discussing the texts of the tracts are hopelessly cluttered with references to melodic material that has yet to be explained. Moreover, the book's copious appendices are poorly constructed as an aid to the reader. Frustration will be avoided by first photocopying Appendix 6, which gives transcriptions from representative manuscripts (she has consulted and cites variants from many more) of all the tracts in their Romano-Frankish and Old Roman versions, and annotating it with Hornby's phrase labels. These otherwise have to be ascertained by endlessly flipping back and forth between Appendix 1 (which gives the texts and phrase labels) and Appendix 4 (which presents in notated form every occurrence of every melodic phrase).
The reader's confidence is further shaken by errors and sloppiness. In the literal translations of the Latin texts of the tracts given in Appendix 1, to which the reader is constantly referred, Hornby inexplicably notes every instance in which her translation differs from "a sample of English-language Bibles with a reputation for being more-or-less literal translations", including every paraphrase of the New English Bible and every use of inclusive language in the New Revised Standard Version. Repeated warnings that certain words "are not usually translated literally" (e.g. 186 n.15) make no sense given that all these differences arise because these bibles are translated from the original Hebrew, while the Latin versions behind the tracts are all translated, unevenly, from the Septuagint. That Hornby is unaware of this is evinced by her treatment of Domine audiui, whose text is from a Vetus Latina version of Habakkuk. Hornby announces that since it is a translation from the Septuagint, "unlike most Bible translations" (191 n.1), she will compare her translation with Brenton's 1851 translation of the Septuagint (using an internet version--why not the more recent New English Translation of the Septuagint?). The grammatical discussion in Chapter 2 is far from reliable. For example, it is suggested (30) that Hornby's Phrase 1g is cued by the appearance of "a verb together with a possessive pronoun", when in fact she means a straightforward personal pronoun (an error repeated on pp. 50, 52, 56 and notes). It is in any case unclear what relevance the book's excessive and pedantic use of grammatical terminology has for the musical analysis. In the musical analysis itself, where Hornby is strongest, there are unsettling instances of carelessness. She notes, for example, that the typical second verse opening phrase in Domine audiui (her 1c) is here made an "open cadence", a term she elsewhere implies to mean a cadence on a note other than the final of the mode (45, 47; but here in Domine audiui it is on the final), and that this is "very unusual" (108). It is obvious from her comparative table in the appendices, however, that this is an insignificant adaptation to make the formula work with a dactyl instead of a spondee (243).
The treatment of patristic and medieval exegesis is perhaps most worrying, not least because it would appear that, with the exception of her effective use of the Tituli Psalmorum and the psalter collects, Hornby has limited her sources to those that can be searched in electronic databases, including an internet version of the English translations in the venerable Nicene and Post-Nicene Fathers collection, which she cites without reference to the Latin originals. These shortcuts inevitably lead to a few blunders. For instance, there is no such thing as "Bede's Calendar", although Bede did compile a martyrology, the edition of which in the Patrologia Latina must be used with great circumspection (158). Likewise, Bede did not write an In Psalmorum librum exegesis (143 n.20), although it is just possible that a small part of one of the two commentary traditions conflated under that name in PL may be from his pen. Some outdated secondary sources are also cited from the internet (158 n.22). In many cases, however, the textual emphases identified by Hornby are given no exegetical foundation at all, as in her discussion of the tract Qui habitat (Psalm 90), where she sees a musical link between a reference to God's sheltering wings and (winged) angels bearing the righteous man upon their hands (90). This may be absolutely correct, but no patristic quotations are adduced, nor is any iconographic evidence offered to support her assumption that the singers who put this chant into its extant form will have assumed that angels have wings.
We may heartily sympathize with the author's predicament: no musicologist can reasonably expect to master the whole field of Christian Latin or Patristics. There is, however, help available to those willing to visit an old-fashioned library. It is a pity that Hornby did not consult the wonderful four-volume Commentary on the Psalms from Primitive and Medieval Writers by J. M. Neale and R. F. Littledale (4th ed. 1884, recently reprinted by Lancelot Andrewes Press), which synthesizes and documents a wide array of patristic and medieval exegetical and liturgical material. Here we discover that there is in fact some patristic precedent for the interpretation of Psalm 139 (used in the Frankish tract Eripe me) as a "Passion psalm." Hilary of Poitiers identifies the peccator and uir linguosus of this psalm as Judas Iscariot, whom he elsewhere calls "the minister of the Lord's Passion" (in his commentary on Psalm 128). Neale and Littledale also translate two psalter collects from the Mozarabic liturgy that directly link this psalm to Christ's betrayal and Passion. [4] It is worth considering, therefore, that there may be a Gallican liturgical background to the Frankish use of Eripe me as a Passion psalm.
None of the flaws criticized here mean that Hornby's interesting analysis is incorrect. They simply mean that the reader is not equipped to return more than a Scottish jury's verdict of "not proven."
--------
Notes:
1. Gregorian and Old Roman Eighth-Mode Tracts: A Case Study in the Transmission of Western Plainchant (Aldershot: Ashgate, 2002).
2. The classic exposition is Ritva Jacobsson and Leo Treitler, "Medieval Music and Language: A Reconsideration of the Relationship", in Music and Language, vol. 1 of Studies in the History of Music (New York: Broude, 1983), 1-23. For a summary of more recent developments, and a consideration of the text-music relationship in the context of historical and contemporary "spiritualities" of liturgical chant, see Anthony Ruff, Sacred Music and Liturgical Reform: Treasures and Transformations (Chicago: Hillenbrand Books, 2007), 482-99.
3. See William Mahrt, "Gregorian Chant as a Paradigm of Sacred Music," Sacred Music 133 (2006), 5-14.
4. Both are attested in Verona, Biblioteca capitolare, MS 89, the late seventh- or early eighth-century Oracional Visigotico, ed. José Vives, Monumenta Hispaniae Sacra, Serie Liturgica 1 (Barcelona, 1946), 229, nos. 707 and 708.