Continuity and change in medieval Iberian processional practices

ABSTRACT In this article, we explore the Palm Sunday palms procession in León across the Middle Ages. How might the experience of a tenth-century citizen of León compare with that of his/her descendant 400 years later? Did the palms procession still have the same devotional goals, reached in similar ways? We focus on questions of continuity and change, with the palms procession as our focus. Some processional elements continued without change after the Old Hispanic rite was replaced by the Roman rite. Some elements were still present, but took a different form in the Roman rite. Other elements were lost entirely. This case study introduces the present critical cluster, which provides multiple examples of how scholars can interrogate the evidence - often preserved piecemeal across sources, or providing only partial information - in order to provide a rich picture of medieval ritual practice and its contemporary meanings.

branches. 2 The people crowd forward, approaching the bishop as he hands out olive and willow stems. 3 Direct physical contact with such a godly man is a rare privilege, and it affects each person in different ways. One seamstress recognises her own handiwork in his fine embroidered robes; a cobbler notices that the bishop is wearing the sandals he mended last Advent; an elderly widow reaches eagerly towards the holy touch of the bishop's hands; a mother comes forward, babe in arms, carrying him towards the next stage of his initiation into the Christian community. As the last willow branch is handed out, the bishop cries out "Thanks be to God! Thanks be to God! Thanks be to God!" The townsfolk start to respond by rote, three times "Thanks be to God! Thanks be to … " and a clear voice -the archdeacon's -rings out above their chorus, soon joined by the cathedral choir, singing "When the crowd that was to come to the festival day heard that Jesus Christ was coming, they took palm branches, and went out to meet him and cried: 'Hosanna, blessed is he who comes in the name of the Lord'." A frisson runs through the townsfolk. Standing ready with their freshly-cut foliage, seamstress, cobbler, widow, mother, babe and all have become that biblical crowd and, just as the song instructs, they go out into the streets, with their bishop and all his clerics, to meet their Lord. As they walk through the narrow streets, the singing continues, echoing off the ancient city walls and silencing those few who stand on the edge watching rather than processing. The choir sings of the donkey on which Jesus Christ entered Jerusalem, and again and again they sing "Hosanna! Blessed is he who comes in the name of the Lord." The townsfolk reach the top of a small rise, seeing the towering mass of the cathedral across the plaza. This is the goal of their procession, their Holy Jerusalem. Now the mood changes as the choir begins a chant about Mary Magdalene anointing Jesus's body for his burial. There is sorrow here, amid the joy: Christ's entry to Jerusalem will lead to triumph only through his humiliation and death. The following chant uses a psalm text to articulate the solemnity of the day on which foliage is carried to the altar, an Old Testament text that foreshadows the Palm Sunday narrative of the New Testament and of their own present day. The singers complete the chant by articulating again "blessed is he who comes." Now all are massing by the cathedral door, and the singers burst into a new song, sounding just like the angel choir at Christmastime: "Glory to God in the highest, and on earth peace to men of goodwill." As everyone files inside, a baby bawls at the new smells and the change of temperature, but is swiftly hushed by its mother (an old hand, she is nursing on the move). The choir's angelic voices turn to the now-familiar text: "Blessed is he who comes in the name of the Lord" before the routine recitation of praise "Glory and honour to the Father, Son and Holy Spirit for ever" and, as the singers solemnly move up into the choir, they repeat this whole chant again "Glory to God in the highest … " The bishop says a prayer whose sense is lost in the ecclesiastical shuffling, and a deacon instructs everyone to arise. Mass has begun. 2 The prayers are not given in L8, which is a chant book; however, six prayers for this moment are preserved in the lateninth century orational BL52, eight prayers in the eleventh-century liber ordinum S4 (two are shared with BL52, but not in the same sequence), and the first two S4 prayers are also present (but no other prayers for this moment) in the late thirteenth-century liber misticus, T5. (For all manuscript sigla, see Appendix 1.) T5 preserves a distinct branch of the Old Hispanic liturgical tradition, labelled Tradition B by modern scholars, which differs from L8 (Tradition A) in liturgical genres and placement, the texts used on particular occasions, and their melodies. On the relationship between the traditions among certain Lenten genres, see Hornby and Maloy,Music and Meaning. 3 Only S4 specifies that the bishop handed out the palms.
This imagined description is based on the Palm Sunday rubrics of one manuscript, 4 the mid tenth-century antiphoner (hereafter L8) that has been owned by the cathedral of León since the tenth century, and whose layers of annotation and correction attest to its authority through a long period of use. 5 The number of manuscripts preserving Old Hispanic liturgy is so small, and their placing and dating often so contested, that we have tended to focus on each liturgical rubric individually, and explore its implications in the cultural milieu implied by its manuscript witness(es). 6 At the same time, there is value in extrapolating to speculate about liturgical practice across a longer time or wider space, when the evidence permits it. In what follows we explore the possibilities for taking just such a longer view in our understanding of the Palm Sunday palms procession in medieval León. How might the experience of a tenth-century citizen of León compare with, say, that of his/her descendant 400 years later? Some elements of the Palm Sunday palms procession are, and have been, universal in Christian practice. However, specific aspects of this procession changed at different times and in different places. These changes can sometimes give the procession a different theological emphasis. In this article we focus on questions of continuity and change, with the palms procession as our focus. Some processional elements continued without change after the Old Hispanic rite was replaced by the Roman rite. Some elements were still present, but took a different form in the Roman rite. Other elements were lost entirely. 7 There are few Spanish institutions for which one can compare liturgical practice between the early and late Middle Ages. León Cathedral is one of the exceptions, preserving several manuscripts from before 1080, and many more from the later period. L8 contains the most detailed extant information about Old Hispanic chant in its liturgical context, with both extensive rubrics and notated chants, and scholars agree that it was used at the cathedral of León. 8 Furthermore, the León Cathedral archive preserves several liturgical books with processional information, including two customaries from the fourteenth and fifteenth centuries. 9 The processional repertory for Palm Sunday, including the blessing of the palms, is transmitted in a late fourteenthcentury epistolary (L23, 62r-77r), and in a fifteenth-century ritual (L67, 15r-27r). A mid fifteenth-century missal from the cathedral also attests to the same repertory (L44, 21ff) (see Table 1 for a summary comparison of the Palm Sunday processional 4 On Palm Sunday in the Old Hispanic rite, see Pinell, Liturgia hispánica, 290-99, with the palms procession described on 293-95. In making our phenomenological description, we have been inspired by the archaeological perspective that Dianne Scullin has brought to our research network. 5 On the dating of L8, see most recently Gutiérrez González, "«Librum de auratum conspice pinctum»." In a forthcoming publication, Thomas Deswarte also places the León antiphoner in the mid tenth century. See Deswarte, Les folios introductifs. On the corrections, see Boudeau and De Luca, "Erreur, variante et correction." 6 There are several examples of this approach in the present critical cluster. Individual rubrics are explored in detail in Ihnat, "Singing to the Tomb of Leocadia;" Carrero Santamaría, Hornby, and Maloy, "Processional Liturgy;" Hornby and Maloy, "Old Hispanic Pre-baptism Initiation." The wider Visigothic context for ceremonial movement in urban spaces, drawn from literary sources, is explored in Wood, "Narrating Processions." 7 In this three-fold approach, we were inspired by Nettl, Western Impact. 8 See Appendix 2 for comparisons with the less complete information preserved in the other Old Hispanic manuscripts. 9 These are L1 and L21, "libros de constituciones" from the fourteenth century, with information about processions, including Palm Sunday (L1, 15v), and the use of capes and the obligation of attendance for the parish church rectors in León (L21,29r,13r,respectively materials in L67 and L44). L44 and L67 have the same prayers, clearly reflecting the same tradition. L23 is a fourteenth-century notated epistolary with readings for several occasions. For the purposes of the present article, we are concerned only with a late fourteenth-or early fifteenth-century booklet inserted after f. 62v. This booklet contains the two principal Holy Week versus chants sung in the Roman liturgy: the Palm Sunday versus Gloria laus, followed by the versus O redemptor summe carmen, sung during the triduum paschale. L77 is a sixteenth-century processional book from León Cathedral. It may be a copy of a previous processional belonging to the cathedral, since L67 has a rubric referring to a processional book that was in use when L67 was written. 10 L77 is unfortunately incomplete and many folios are disordered. 11 L67 mostly has incipits rather than complete chants. L44 has complete Atollite portas (incomplete) Entering the church R. Ingrediente domino* as in the processional book * = the chant is presented in the MS as an incipit Boldface is used to show concordances across the three manuscripts. chant texts without musical notation. L77 has therefore been useful to us, despite its late date and incomplete status, because most of the Palm Sunday chants it preserves are both complete and notated. The preservation of such a rich body of material in the León Cathedral archive permits us to draw direct comparisons between this cathedral's tenth-century Old Hispanic liturgy and the Roman rite liturgical practices at the same institution later in the Middle Ages.

Continuity
Some aspects of the Palm Sunday procession of the palms did not change at all when the Old Hispanic rite was replaced by the Roman rite. 12 There was no immediate change to the urban and architectural surroundings, 13 and we imagine that the same population processed through the same streets, from and to the same churches, carrying branches (olives in Iberia, presumably, rather than palms). The procession itself still involved people singing, and probably much the same people -or at least the same classes of people -as before. At this level, the experience was (and is) fundamentally the same across hundreds of years.
While the physical details of the processional experience are unique to each urban context, the palms procession always places the laity symbolically in the position of the crowds who welcomed Jesus to Jerusalem. As early as Egeria's fourth-century description of Palm Sunday, the Jerusalem procession involved palm and olive branches, and articulation of the biblical acclamation "blessed is he who comes in the name of the Lord." 14 As O.B. Hardison notes, this biblical event continued to be commemorated one week before Easter when the Palm Sunday procession was adopted in the west, "but the space necessarily became symbolic." 15 Structurally, too, the palms procession had the same liturgical structure right across western Europe. 16 As articulated in the post-950 Ordo Romanus L, the ceremony comprised blessing, distribution, and procession of the palms. 17 The ritual blessing of palms or olive branches took place in the morning after Terce, usually outside the main church; palms were distributed among the people; and then they processed to the main church door, where (in the León version of the Roman rite) the [A]tollite portas dialogue was performed. After the liturgical participants entered the church, Mass began. In its fundamental structure, this ceremony parallels that of the Old Hispanic ceremony described above.
These elements of the Palm Sunday palms procession remained stable through the Middle Ages, regardless of the shift from the Old Hispanic to the Roman rite in Iberia. 18 If scholars such as Gisèle Clément, Alejandro Olivar, and Michel Huglo are correct that the western versions of this procession originated in Visigothic Iberia, such points of continuity are hardly surprising. 19 Indeed, at this level, the palms procession continues unchanged into present Roman Catholic practice, integrating the laity as participants in the biblical narrative as it is brought into the liturgical present. 20

Adaptation
As noted above, in the Old Hispanic annual re-enactment of the biblical palms procession, the cathedral stands symbolically for Jerusalem. This was particularly pointed in early medieval Iberia since, as argued elsewhere in this critical cluster, Visigothic cathedrals were often conceptualised as "Holy Jerusalem." 21 If, at this time, the cathedral was always considered "Holy Jerusalem," we cannot be sure when the idea faded and its identification as "Holy Jerusalem" became a characteristic of Palm Sunday alone. A further shift is evident in the Roman rite procession as preserved in L44. While the triumphant entry of Jesus into Jerusalem is implied in many of the processional chant texts, it is spelled out explicitly only in the responsory Ingrediente domino: "Ingrediente Domino in sancta civitatem hebręorum" ('The Lord entered the holy city of the Hebrews'). In L44's palms procession, this responsory was sung during the entry into the city. This emphasised the whole intra muros city of León as Jerusalem, where previously the cathedral had had that role. In this way, the symbolic presence of Jerusalem within the León palms procession was adapted over time.
An attentive processional participant before and after the transition to the Roman rite would notice that while several of the chant topics remained familiar, 22 there were both textual and musical changes. 23 In both rites, the procession celebrates Jesus riding the 18 We should acknowledge here that the ceremony was not standardised in the Roman rite. There is much diversity in the precise directions for the palms ceremony in tenth-century Roman rite manuscripts, and also in the choice of chant texts and melodies. Young, Drama of the Medieval Church, I:90-98. 19 Clément, Le Processionnal en Aquitaine, 129-31, following Olivar, "Survivances wisigothiques," 162-63; Huglo, "Source hagiopolite." Palms were not present in Roman liturgical sources until the eighth-century Gelasian and Gregorian sacramentaries. There are rubrics in the sacramentaries in Mohlberg 23 In what follows, we base our discussion on the repertoire preserved in L44, acknowledging that this is an individual snapshot of the repertorial possibilities within the Roman rite.
donkey into Jerusalem, using a combination of free biblical paraphrases and texts that are closer to their biblical sources. 24 Three Old Hispanic chants and three Romano-Frankish chants are not directly related textually. 25 At first sight, we might think that Cum audisset populus and Quum audisset turba are cognates across the two rites. On close inspection, however, they are not closely related. 26 The similarity of the first part of the text is misleading; only thirty-six syllables are in fact related between the two chants; that is, over half of the sixty-three Old Hispanic syllables, but only 19% of the Romano-Frankish syllables ( Figure 1). In the parts with similar text, there is no discernible relationship between the text pacing; different syllables are lengthened in the two traditions ( Figure 2). The detail of the melody itself is also worth comparing between the two versions. Here, we draw on a quantitative method developed for comparing different versions of Old Hispanic chant, which results in a "relationship ratio" between two melodies. Because the Old Hispanic melody is preserved without pitch, we cannot securely define how closely related it is to the Romano-Frankish melody; however, we can say, for each syllable, how many of the Romano-Frankish notes are compatible in their melodic contour with the Old Hispanic notes for the same syllable, and thus we derive the relationship ratio from this. 27 For Quum audisset turba and Cum audisset populus, the relationship ratio between the two melodies (only including the syllables in bold in Figure 1) is 0.55; 28 this can be observed more informally by comparing the melodic contours in Figure 3. In other words, these melodies are no more related than one would expect to occur by chance. Thus, four processional antiphons in the Old Hispanic rite and three in the Roman rite shared the topic of Christ's triumphant entry into Jerusalem, but were not directly related across the traditions. The Romano-Frankish Palm Sunday processional antiphon, Quum adpropinquaret, does share most of its text with the first Old Hispanic processional antiphon. 29 This text 24 The episode of Jesus entering Jerusalem on a donkey is described in Matthew 21:7-9; Mark 11:7-10; John 12:12-13 and, with a different text, in Luke 19:36-38. It is anticipated in Psalm 117:26-27 (Vulgate numbering). 25 Compare the Old Hispanic Osanna benedictus and Constituite (close to the Bible) and Quum introires (free paraphrase from John 12, Luke 19, Matt 21) with the Romano-Frankish Palme fuerunt (freely paraphrased from Revelation 7:10, 7:9 and 5:13), Occurrunt turbe, and Ingrediente domino (freely paraphrased from John 12:12-13). The Old Hispanic Quum introires (Ingrediente te domine in T5) has previously been compared in detail with the Aquitanian antiphon Introeunte/Ingrediente te Domine. Huglo, "Source hagiopolite," 367-74, argued that the text was composed in seventhcentury Jerusalem, was translated into Latin for the Old Hispanic liturgy ca. 700, and was later transmitted from Iberia to Aquitaine; he identified different melodies in each of these traditions (and two melodies in Aquitaine). The Aquitanian chant was also studied by Brockett, "Osanna! New Light," 122-27, who spotted some similarities in the L8 musical outline for the cognate text.   would have been directly familiar to listeners straight after the transition to the Roman rite.
In fact, aspects of the melodic delivery of the text would also have been familiar. 30 As can be seen in Appendix 3, Quum adpropinquaret has almost the same text length in the two traditions, and the texts were broken up in very similar ways, with most phrases comprising two to four words. We cannot directly compare the beginnings of the chants, since the Romano-Frankish León version in L77 has a lacuna at the beginning, and the first clause of the preserved text does not parallel the Old Hispanic text. We begin our comparison at "Solventes," from which point the two traditions have the same text. The proportions of the chant are similar in the two traditions as well, with the Franco-Roman chant (186 notes in the comparison) being somewhat less prolix than the Old Hispanic one (229 notes in the comparison). The pacing of text delivery is also similar (Figure 4). Some syllables have a melisma in one version and not the other (e.g., on the final "[Da]vid" in L8 but not in L77; "[imposuerunt] illi" in L77 but not in L8), but in the main, each syllable has a comparable melodic density across the two chants. The listener experience of the textual flow was extremely similar in the two different versions. At a detailed level, the Old Hispanic and Romano-Frankish versions of Quum adpropinquaret are distant cousins. We can recognise some kinship between their contours ( Figure 5), but they could not be described as closely related: their relationship ratio is 0.70. 31 Attentive listeners around the transition will certainly have noticed the difference between the Old Hispanic version and the Romano-Frankish version that replaced it.
One chant in each procession anticipates the darker events of Holy Week, but in very different ways. The Old Hispanic antiphon Amen dico uses the story of Mary Magdalene anointing Jesus with oil "for his burial." The Roman rite León ceremony instead has the Atollite portas dialogue. This dialogue was understood in medieval times as representing Christ's descent to hell. 32 The precise narrative choice was different in the Roman and Old Hispanic palms processions, but both had the effect of pointing forward towards Christ's death. 33 The last Old Hispanic processional chant was sung at the church door. This shares textual components with the Romano-Frankish chant sung on the city wall, during the procession (ad portam civitatis … super muram). 34 As noted above, the Old Hispanic antiphon combines the Christmas angel song "Gloria in excelsis deo" with the text "Benedictus qui venit in nomine domini" that permeates the Palm Sunday procession. The closest textual equivalent in the León version of the Roman rite is instead an abbreviated 30 We concentrate exclusively on L8 in our analysis here, maintaining our focus on León. One other manuscript contains melodies for the Palm Sunday procession, the Toledan Tradition B manuscript T5. Analytical comparisons of the palms procession antiphons between T5 and L8 may be found at http://plainsong.org.uk/publications/hornby-andresgutierrez-and-scullin-processional-melodies-in-the-old-hispanic-rite-appendices/ 31 145 compatible notes; 229 notes in L8; 186 notes in L77. 2×145/(229+186). For a point of comparison, we calculate the relationship between the Old Hispanic Quum introires and its cognate Aquitanian antiphon Introeunte/Ingrediente te Domine at 0.85, based on the transcriptions by Brockett, "Osanna! New Light," 123-25. This will have been much more recognisable as "the same" melody in a different version. 32 There is a very helpful summary of the text sources and narrative implications of the Atollite portas dialogue in Brockett, "Scenarios.   version of a hymn by Theodulf of Orleans (d. 821), Gloria, laus et honor. 35 This non-biblical text alludes to the palms carried by the Hebrews and -in the third person -to Christ being the blessed one who comes in the name of the Lord ("et inclyta proles nomine qui in domini rex benedicte venis"). 36 The size of the procession increased after the transition to the Roman rite, in several ways. Across the Old Hispanic chants for the palms procession, there are 591 syllables and 1439 notes. It is not possible to calculate the size of the Romano-Frankish chants with such exactness. The León manuscripts do not preserve complete melodies for Cum approprinquaret or Cum audisset. We have supplied those notes, speculatively, from E-Mn 1361, a fourteenthcentury manuscript whose melodies are very closely related to those of L77. In the sixteenthcentury L77, Palme fuerunt, Occurrent turbe and Cum audisset all end with a differentia: the word "seculorum," with six notes above, to provide the melodic formula for the end of the doxology "seculorum amen." This implies that in these three chants, and possibly in Cum appropinquaret as well (where the end of the antiphon is lost in a lacuna), there was a psalm and doxology, with antiphon repeat. There are no differentia indications at the ends of the antiphons in the unnotated fifteenth-century missal L44. We cannot tell whether there were in fact psalms, doxologies and repeats in the performance practice of these antiphons in late-medieval León, but there is at least a possibility of it. 37 Without the possible psalms, doxologies and antiphon repeats, the Romano-Frankish chants have 1264 syllables and 2003 notes. For participants in the Roman rite palms procession in León, there was thus more than twice the amount of text for singers to learn and listeners to digest than there had been in the Old Hispanic procession. Assuming a roughly equivalent performance speed, in terms of numbers of notes per minute the palms procession in the León Roman rite lasted more than one-third again as long as the earlier Old Hispanic procession; if there were psalms, doxologies and antiphon repeats within the procession, the Roman rite procession was even longer than this estimate.
For non-singers, a musical kinship of the Romano-Frankish and Old Hispanic Quum adpropinquaret might have been recognised by an attentive listener in the years immediately after the transition. For other palms procession antiphons, the familiarity was textual, or more generally topical. At this level, and for these liturgical participants, there was a significant level of continuity in this aspect of the rite. More text was delivered in the Roman rite version of the palms procession in León, and many more musical notes, probably reflecting an increased length of the procession. 38

Rupture
In the Old Hispanic rite, Palm Sunday primarily had an initiation function. The Palm Sunday ceremonies associated with initiation have been explored by several scholars, sometimes from an archaeological perspective, 39 sometimes in dialogue with Visigothic theology, 40 and, in the present critical cluster, from a musicological perspective. 41 These ceremonies included processions, many initiation-specific texts, and an elaborate exorcism ceremony. This aspect of Palm Sunday was not present in the Roman rite. Such a significant shift of emphasis on Palm Sunday is best described as a "rupture." Other aspects of the Roman palms procession in comparison to the Old Hispanic procession that preceded it will similarly have been perceived as a rupture in the practice. Some of these aspects are topographical, some are textual, and some are musical. We introduce each in turn. The L8 rubrics outline the tenth-century processional topography. They instruct that five of the antiphons should be sung en route to the cathedral, and the final antiphon, Gloria in exceslsis, should be sung at the church door. Because L8's Palm Sunday rubrics are very detailed, we consider that other stations on this procession would have been signalled if they were present. In the León Roman rite (as reflected in the fifteenth-century Missal, L44) the series of processional antiphons was interrupted by a station, at which there was a reading and sermon. At a second station, the hymn Gloria laus was sung not at the church door (like the closest Old Hispanic equivalent), but at the city gate, on the wall. 42 In the missal, this is followed by the responsory Ingrediente domino, sung while entering the city. According to the fifteenth-century ritual, L67, the dialogue Atollite portas was recited at the church door (see above, Table 1). 43 Topographically, the stations for the reading and at the city gate will have had a significant impact on the experience of all participants, punctuating the processional movement with periods of standing still, without direct precedent in the Old Hispanic procession.
The Roman procession had new textual elements. Although the themes of Gloria laus, Atollite portas, and Ingrediente domino were anticipated in different ways in the Old Hispanic procession, the text structures have no cognates in the Old Hispanic procession of the palms, where almost all the antiphons comprise a single section without verses or repeats. 44 Gloria laus is a strophic hymn with a repeat of the first strophe after each subsequent strophe in L44 (28r-v). 45 This was probably performed as an alternation between soloist(s) (most likely boys as attested in L44), and the whole choir, with the latter singing the refrain strophe. 46 Atollite portas has a dramatic repeat structure. The door was struck three times and the presbyter intoned three times (in an increasingly loud voice) "Lift up the gates … " Each time he was answered from within: "Who is this king of Glory?" He (once) or a "sacerdos" (twice) responded, identifying Christ, the first time as "the strong and powerful Lord," the second time as "the Lord, mighty in battle" and finally as "the Lord of hosts: he is the king of glory." Finally, the doors of the cathedral were opened and the procession entered. There is no equivalent to this 39 Quevedo Chigas, "Early Medieval Iberian Architecture." 40 Ramis Miquel, La iniciación cristiana; Akeley, Christian Initiation in Spain; Hormaeche Basauri, La pastoral de la iniciación cristiana; Pijuan, La liturgia bautismal; McConnell, "Baptism in Visigothic Spain." 41 Hornby and Maloy, "Old Hispanic Pre-baptism Initiation." 42 See note 35, above. On the performance of Gloria laus from within the west wall of the cathedral in various medieval English cathedrals, see Malone, "Architecture as Evidence." 43 In this ritual, the responsory Ingrediente domino was sung while entering the church. The sixteenth-century processional L77 does not confirm which of these was in fact practised, since it has Ingrediente domino as a later addition, without a rubric; the usual liturgical position of this responsory is therefore unclear. 44 The one exception is Gloria in excelsis deo, where the structure is antiphon-doxology-antiphon repeat. 45 The late fourteenth-century epistolary L23, 64v, has a repeat of the first strophe signalled only after the last strophe. 46 Haggh-Huglo, "Review," 182. dramatic dialogue and entrance in the Old Hispanic palms procession. The responsorial form of Ingrediente domino is not anticipated in the Old Hispanic procession either. In this chant, an abbreviated version of the last part of the text ("cum ramis palmarum osana in excelsis") is repeated after the verse. 47 This emphasises one of the key Palm Sunday texts.
For the singers in the choir, the change from the Old Hispanic to the Roman palms procession was significant and immediate: they had a whole new repertory to learn. As we have shown above, the Romano-Frankish antiphon Cum appropinquaret may be distantly related to its Old Hispanic counterpart, but this would hardly have been helpful to the singers charged with learning and performing the chant. For these specialist liturgical performers, there was an undoubted rupture when the practice changed from the Old Hispanic to the Roman rite.
On processions in medieval Iberia: the present critical cluster and next steps As we have shown, the procession of the palms on Palm Sunday in León conserved some elements across hundreds of years while other aspects changed, reflecting different devotional, ceremonial and musical priorities. Both before and after the transition to the Roman rite, the triumphant entry of Jesus into Jerusalem was brought into the liturgical present by the laity carrying foliage and processing through the streets while the choir sang chants about this biblical event. The precise expression of this fundamental concept changed over time, both musically and ritually. A ritually active observer will have been acutely aware of such shifts, since texts, melodies and ritual actions changed in their details, albeit not in their general purpose. Even in the one case where the chant text is the same -or very similar -we have shown that the melodies changed considerably.
This critical cluster provides multiple examples of how scholars can interrogate the evidence -often preserved piecemeal across sources, or with only partial informationin order to provide a rich picture of medieval ritual practice and its contemporary meanings. In her article on the veneration of Saint Leocadia in León, Kati Ihnat takes as her starting point the two-word processional rubric in L8: "Ad sepulcrum." 48 Present near the end of the Mass on the feast of Leocadia, the patron saint of Toledo, this rubric relates to the saint's tomb. What does this mean in the ritual context of León, a city that did not host the saint's remains? Ihnat draws out the implications of this processional rubric in the light of what is known about the history of Leocadia's relics and of stational liturgies. In contrast to the two-word Leocadia rubric, the L8 Mid-Lent Sunday and Palm Sunday rubrics provide considerably more information about Old Hispanic initiation ceremonies in tenth-century León, and these are discussed by Emma Hornby and Rebecca Maloy. 49 Such ceremonies occurred daily for the fortnight before Palm Sunday, with further initiation ceremonies on Palm Sunday itself. At the same time as being an intensive period of Lenten penitence and fasting, these two weeks 47 Repetition "per latera." On this phenomenon, usually with the word "presa" in the Aquitanian and Spanish repertories, see Huglo, "D'Hélisachar à Abbon de Fleury," Appendice I. The closest structural equivalent in the Old Hispanic palms procession is the antiphon Gloria in excelsis deo, where the entire antiphon is repeated after the doxology. 48 Ihnat, "Singing to the Tomb of Leocadia." 49 Hornby and Maloy, were also punctuated by daily exorcisms of those to be baptised at the Easter Vigil. These were surrounded by processions. Indeed, on Palm Sunday itself, despite the headline palms procession we have discussed in the present article, the main focus of the feast was an explicit, day-long, welcoming of the "new shoots" of the next generation to the sheepfold of the Christian church, where they were to be taught the love of God. At matutinum, inside the cathedral, this welcome included ritual movement whose complexity and solemnity rivals that of the Palm Sunday palms procession.
L8 is a comprehensive manuscript, containing chants for almost the whole calendar year together with multiple explanatory rubrics, some detailed (like those for Mid-Lent Sunday and Palm Sunday discussed by Hornby and Maloy) and some gnomic (like the Leocadia rubric discussed by Ihnat). L8 has been the focus of the vast majority of scholarship on Old Hispanic chant because of its early date, its high production values, its excellent state of preservation, and the amount of material it contains. We offer a balance to that concentration in this critical cluster, with a close look at Visigothic Tarragona in the contribution by Eduardo Carrero Santamaría,Emma Hornby,and Rebecca Maloy. 50 This was a major urban centre in the Visigothic period, and is thought to have been the origin of the earliest extant Old Hispanic liturgical book, the Verona Orational (dated before 732). 51 In OV, there are rubrics on Carnes Tollendas Sunday, six weeks before Easter, that mention different churches. The implications of the rubrics are most clearly understood -together with their limitations for modern scholars attempting to reconstruct the Visigothic urban topography -when we consider the wider liturgical context and the likely practicalities of the processions to which they refer. As these articles illustrate, each Old Hispanic processional rubric invites close and detailed study, integrating the possible architectural and urban context into the discussion. 52 There is evidence pertaining to early medieval Iberian processions beyond the liturgical books. In his contribution, Jamie Wood maps out the wider Visigothic context for ceremonial movement in and beyond urban spaces. 53 He draws primarily on hagiographical sources, showing different ways in which collective movement -"religious," "secular" or a mixture of both -could display, underline and undermine contemporary power structures. Among the sources, a saint's passion was usually read at matutinum in the Old Hispanic rite on his/her feast day. As Wood shows, several passions articulate the role and potential of processions for constructing sainthood. For liturgical participants, the public liturgical reading of each passion helped to develop and sustain their understanding of processions, which they then enacted practically in the liturgical processions in which they participated on days such as Carnes Tollendas Sunday, Mid-Lent Sunday, Palm Sunday, or on special occasions such as a king's departure to war.
The research collaboration represented in this cluster has brought together scholars from multiple countries and disciplines. We shared data, offered each other iterative feedback, and each contributed our disciplinary-specific insights to a growing understanding of the role and characteristics of processions in the early Middle Ages. In fluid co-writing combinations, we brought together a range of evidence bases and different disciplinary perspectives. The result, in the current issue, is a substantial body of work that provides multiple snapshots of medieval Iberian culture and religious life. In future, it will be possible for scholars to draw on this body of work when thinking about the functional topography of late antique and early medieval cities. Some Old Hispanic liturgical genres are potentially processional but lack explicit rubrics. Our body of work provides a departure point for exploration of such genres. Finally, as we have illustrated in this introductory article, we are now in a position to draw comparisons between the Old Hispanic processions and their cousins in other liturgical traditions, identifying points of continuity, adaptation and rupture.

Disclosure statement
No potential conflict of interest was reported by the author(s).  Text length in syllables (verse(s)) n/a n/a n/a n/a n/a n/a n/a Text length in syllables (antiphon repeat) n/a n/a n/a n/a n/a n/a n/a

Notes on contributors
Text length in syllables (doxology) n/a n/a n/a n/a n/a n/a 27 Text length in syllables (final repeat) n/a n/a n/a n/a n/a n/a 45 n/a n/a n/a n/a n/a n/a (ca. 27 in doxology) Number of phrases 9 phrases 27 12 6 7 4 6 (antiphon) Number of words per phrase 2-4 words x 8 1 word x 3 2-4 words x 11 2-4 words x 5 2-4 words x 7 2-4 words x 4 1 word x 1 5 words x 1 2-4 words x 21 5 words x 1 5 words x 1 1-4 words x 4 5 words x 1 6 words x 1 6 words x 1 7 words x 1