Performance, Relic, Poem: Ephrem’s Hymns on Saints

A Byzantine Syriac poem recounts, in first-person voice, the dying reflections of the Syriac poet Ephrem. [1] In this pseudonymous autobiographical poem, Ephrem, who died in 373 CE in the East Roman city of Edessa, tells his gathered disciples that when he was a child, resting in his mother’s lap, he had a vision. [2] In that vision, a vine sprouted from his infant tongue and began to grow. As it grew, it put forth leaves and bore fruit in abundance—more than he could count. It grew up and grew out, covering the whole of the earth horizontally, and the space between heaven and earth vertically. People took its leaves and plucked its fruit, but it was not diminished. For every piece of fruit that was taken, another grew miraculously in its place. The leaves, he tells those gathered around his death bed, were memre, homiletic poems. The fruits were madrashe, hymns. God was the vinedresser, giving rise, through the poet, to this harvest of musical, metered poems.

How do words become things? How does a poet sing and find her words as material as vines, as concrete as leaves, as tangible as fruit? From where do the words come before they take over and spring forth from the poet’s unsuspecting mouth? Once the words have left the poet’s mouth, how do they take on a life of their own, growing of their own accord, producing new poems—poems somehow derived from that single mouth, yet still distinct from it?

The fecundity of this image and the questions that arise from it direct us, in a contemporary context, toward the realm of the performative. The word “performative” can signify a range of ideas about what language does, what it is, and how its presence, meaning, and creative capacity relate to the spaces in which it arises, and the audiences before whom it is uttered. For the tradition that arises from J. L. Austin, “performative” denotes the now seemingly obvious idea that words do not merely describe some thing out in the world, but in varying degrees shape that world—bring it into being. [3] “Performative” also emphasizes the spaces of performance in which words are uttered, and argues that these performative spaces and the people gathered within them are as crucial to a work’s production of meaning as the utterances of the work itself. [4] These two notions of performativity—that language is creative; that the spaces in which words are uttered play a role in constructing the meanings of those words—have taken yet further shape in what has come to be known as Oral Formulaic Theory, and in its reception by historians of the senses. [5]

Working initially as textualists, oral formulaic theorists have found ways through close attention to the repetitive use of small bits of language to argue that literary works that appear to us stable (such as the Homeric epics) in fact capture merely a single moment in the life of a work that existed originally and primarily in an oral, performative framework. [6] This move from written text to imagined oral context has profoundly expanded our notions of textuality, and—especially important for this chapter—the materiality of nonwritten literature. Scholars of the history of the senses have especially emphasized the durability of oral language, and the unique physicality of the sensoria that such orality afforded. [7] Moving through the spaces of the imagination (spaces no less real for having been imagined), the language of oral literature—its formulaic phrases, set scenes, narrative patterns—possessed a concrete, spatially oriented existence, even when free of the moorings of written texts and single, identifiable authors. [8]

For the anonymous Syriac poet who sang the dying vision of the fourth-century poet Ephrem, words clearly were things. Poets clearly could be possessed by the words they sung and the divine presences from which those words came. Taking as inspiration the anecdote with which I began, but rerouting it through contemporary theories of the oral and the performative, I turn in this chapter to a series of fourth- or fifth-century hagiographical poems not about Ephrem, but attributed to him: the hymns “On Abraham Qidunaya” and “On Confessors.” [9] The chapter builds from a basic observation: certain ephremic hymns sung to Christ share linguistic formulae with certain ephremic hymns sung to the saints. I trace these formulae in detail to articulate two ways that this attention to Ephrem’s language can reshape our perception of the poetry attributed to him, as well as the performative contexts within which it arose. First, the formulaic language of Ephrem’s poems suggests their origin as a kind of oral literature. This observation challenges our preoccupations with the authorship of these poems and emphasizes the role that performance played in their composition. Second, the unique blending of fragmentation and concreteness of the poem’s language can be connected to the fragmentary concreteness of the relics around which the poems were sung.

But first, who was Ephrem, and how do these hagiographical poems come to be attached to his name?

The poet known to us as Ephrem the Syrian is cloaked in uncertainty. Having reposed at the end of the fourth century, he appeared on the historical, textual stage in both third-person biographical sources and first-person musical poems. [10] Though Ephrem appears to have composed only in Syriac, the earliest biographical sources for his life arose not in Syriac, but in Latin and Greek, and only took canonical Syriac shape in the sixth century. Alongside this multilingual web of biographical sources stands the hymnographic corpus for which he is primarily known. The hymnody attached to the name Ephrem is first attested in a Syriac florilegium in the late fifth century, before appearing in a variety of manuscripts in the sixth and seventh centuries. [11] The biographical sources for his life appear to have developed mostly without a close connection to the hymns he wrote in the fourth century. Similarly, these hymns offer little detail of a biographical sort at all, much less one that can be mapped onto the later biographical tradition. The complicated relationship between the biographical sources and the early hymns is made only more difficult by the extensive corpus of literature in Greek, Syriac, and Armenian that would attach itself pseudonymously to the name Ephrem. While some of this pseudonymous tradition can easily be identified as inauthentic, much of it appears to have been composed relatively early. [12]

Because of the volume of Syriac hymnody attributed to Ephrem, and because the actual life of the saint is mostly unknown, the modern critical study of Ephrem has worked towards two primary goals: on the one hand, to uncover the biography of what we might call the “historical Ephrem,” and, on the other hand, to distinguish the authentic from the inauthentic hymnody of this Ephrem. [13] It is mostly only in this context—in the context of the search for the authentic hymnody of Ephrem, and the elimination of works deemed spurious—that the hagiographical hymns attributed to him have been treated. [14] The earliest evidence for this ephremic hagiographical hymnody appears in a manuscript from the late sixth century. [15] In this manuscript, we find, alongside hymn collections by Isaac of Antioch (fl. fifth c.) and Jacob of Serugh (d. 521), fifty hymns—madrashe—attributed to Ephrem the Syrian. These fifty hymns are organized into six collections: five hymns “On Confessors” (a word that, in early Syriac Christianity, denotes a martyr), fifteen hymns “On Abraham Qidunaya,” four hymns “On Barsauma,” twenty-four hymns “On Julian Saba,” one “On the Sons of Shmuni” (the mother of the Maccabees), and one “On the Chorepiscopos of Nisbis.”

The authorship and authenticity of these hymn collections is uncertain. Generally, the five hymns “On Confessors” and the single hymn “On the Sons of Shmuni” have been taken as early, but inauthentic, while the hymns “On Abraham Qidunaya” and “On Julian Saba” are thought to represent a mixture of authentic and inauthentic elements. [16] The hymns “On Barsauma” and “On the Chorepiscopos of Nisibis” have not, as far as I can tell, received any attention at all. My interest in this chapter is in two of these collections—the hymns “On Abraham Qidunaya” and “On Confessors.” I am not immediately concerned with the questions of authenticity that these hymns have provoked, but with habits of speech they develop around the saints, linguistic habits that have, however, only been registered in connection with these questions of authenticity.

This early hagiographical hymnody crafts a tone of awe and transcendence around the saints. The hymns’ “I” routinely doubts its ability to speak about the saints and laments the frailty of its words compared to the resplendent glory of the saint. This oratorical hand-wringing, moreover, has a decidedly formulaic character, one that operates stably within the hagiographical hymns, but also migrates between hymns on saints and hymns on Christ. For example, in the second of fifteen hymns written in honor of the Syriac ascetic Abraham Qidunaya, who died in 367 CE, the poet sings directly to St. Abraham:

Hymn “On Abraham Qidunaya” 2:7 [17]

When Edmund Beck critically edited these lines in 1972, he thought it unlikely that Ephrem actually wrote them. While he admitted that they sounded similar to authentic Ephrem, he thought that Ephrem could only apply such language, in which the object of the poet’s speech seemed so clearly to transcend that speech, to the divine. [18] Beck did not further unpack or explain his aversion, but it likely arose because of the resemblance of this stanza to another stanza, one from a hymn that Beck had edited ten years earlier, and thought undoubtedly authentic. This latter stanza comes from a set of four hymns that go by the title of “On the Oil and the Olive and the Symbols of our Lord.” [19] At the conclusion of this short hymn series, the singer finds himself overwhelmed by all the symbols of Christ that he has perceived in an olive, and cries out in the concluding stanza:

Hymns on the Olive and its Oil and the Symbols of our Lord 7:15

These two passages barely overlap in terms of specific vocabulary. But in terms of grammatical structure, their syntactic relationship is much closer. Both start the first sentence with the preposition men (“from”) and end it with a passive verbal form. Both use a purpose clause, joined to a first-person imperfect verb, in the second line. These parallel grammatical structures house parallel conceptual structures. In the hymn “On Faith” 10, the poet is passively carried to the Lord’s body, as his christological speech emerges organically from his encounter with that object. In the hymn “On Confessors,” the passive movement is the same, but the referents have changed: the poet encounters not the Eucharist, but the martyrs’ bones, from which the power to narrate them emerges.

As we zoom out from this single hymn “On Confessors” to catch sight of its place within the broader landscape of ephremic hymnody, we witness a shared formulaic lexicon and a consistent recourse to story patterns. These literary tools are likewise put to a shared end—to mark the poet’s posture before the holy. We find a further parallel—both grammatical and conceptual—in the second stanza of hymn “On Faith” 16. In the initial stanza of hymn “On Faith” 16, the poet reflects hymnically on his own task as poet. He asks, “Lord, how shall your servant stop singing your hymn? / How shall my tongue cease from thanksgiving?” He then, in stanza 2, echoes what will become a standard liturgical idea, namely, that what humans give in liturgy is merely what they have first received from the Lord. Here, in hymn “On Faith” 16, the gift refers specifically to the poet’s song. And so he declares:

Hymn “On Faith” 16:2

We can compare this with the line we have already seen in Hymn on Confessors 8:

Hymn “On Confessors” 8:21

As before, these two passages link through a shared formulaic repertoire and a shared theme. Their grammatical structures mirror one another, as do their conceptual senses. Both express the idea that the content of the song they sing is merely what the object of the song has given. The objects, however, differ: in the hymn “On Faith,” it is the Lord to whom belongs the poet’s song, whereas in the hymn “On Confessors,” the song belongs to the martyrs.

Joining these passages on Christ and the saints reveals the poets’ use of a range of formulaic phrases, narrative patterns, and themes. These formulaicisms are united conceptually in the way that they place the poet before God or the saint as the poet asks for inspiration. Just as Christ and the saints stand in mimetic relationship to one another, so do the concrete objects—Eucharist and relic—through which their presence is mediated to the poet. This idea that in the economy of poetic inspiration, Eucharist and relic are conceptualized in relationship to one another will be important as I move to the final parts of this chapter and argue for a relationship between the fragments of poetry and the fragments of relics. But, more basically, this analysis of formulaic speech raises questions around the hymns’ authorship. When in the 1970s Edmund Beck edited these hagiographical hymns, he assumed a fixed author: Ephrem had written certain hymns and had not written others; some hymns had been ascribed to him even though someone else had written them. But looking at these shared lines through the lens of oral formulaic theory suggests a different paradigm for relating these texts to one another. Rather than assuming an author who created final, written versions, and an imitator who did the same, only of a second-rate quality, these formulaicisms suggest a different model of composition. They suggest that, like the movement of relics throughout the landscape, poems grew in a fugal-like way, expanding, compressing, and morphing in the process. It is within this context that we can situate the hagiographical hymns and their striking speech to the saints.

So what kinds of texts are these? I have tried to show that the hymns on Christ and the hymns on the saints have a relationship to one another. What Christ is in the christological hymns, the saints are in the hagiographical hymns—both objects of unknowability nevertheless made known in concrete things (Eucharist and relic), in response to which the poet’s song emerges. What I want to suggest now is that these poems developed out of a certain kind of practice, in which poets did not so much pen single, complete poems, but learned, through a process that was at least partially, if not primarily, oral, the building blocks of the poems they would compose in the context of particular performances.

Though the hymns ascribed to Ephrem come to us as written texts, there is virtually no doubt that performance played a crucial role in their realization. In the textual versions that have come down to us, they bear on every page the marks of these performative origins—what Walter Ong and Albert Lord have called an oral residue. [26] From a certain perspective, the performative setting of the hymns could not be more obvious. Though we no longer know the tunes, the hymns are marked with melodies, indicating that they were sung. [27] Each stanza carries a refrain, suggesting that the hymns were sung with multiple people—specifically, by choirs of women. [28] The first-person voice within the hymnody routinely exhorts some unseen audience to listen, sing, cry out, or glorify. [29] Many of the early hymns were written for liturgical feasts and seasons—Easter, Christmas, Epiphany, Lent, Holy Week, Easter, and, as we have seen in this chapter, the feasts of the saints. [30] Our later evidence, too, suggests liturgical performance as key to the hymn’s setting. While I have argued elsewhere that some of the hymns emerged in settings of study, my point in making that argument was to suggest the musical, performative character even of educational settings. [31] These hymns were composed not to be read alone in a library or a monastic cell, but to be sung aloud with others—often, perhaps, many others.

This evidence for the performative origins of the hymnody can be combined with formulaic aspects of the hymns’ language, aspects that suggest an oral environment. The hymns we looked at above, for example, share a formulaic use of the narrative of the Wedding at Cana, one which is signaled through key vocabulary, a minimalist sense of narrative shape, and a shared rhetorical end toward which the narrative is put in different performative contexts. [32] We can think, too, of the stylized scenes, in which poet stands before God or saint, as formulaic—stock, micro-narrative images that poets learned as stylized ways to depict themselves before the gates of heavenly inspiration. [33] These scenes were connected to certain set, introductory phrases that triggered poem, audience, and the God whose divine gift inspired the hymn that followed. [34] To these formulaic aspects of the hymnody, we can join the various mnemonic devices they employ. Melody and meter rightly strike us for their sonic capacities, but they also surely functioned—for ancient composers just as for us—as aids to memory. The hymns feature other devices that can be understood mnemonically, such as acrostics. [35] Alongside these mnemonics of form stands a remarkably small lexicon shared among the ephremic hymns. While much of the hymnody ascribed to Ephrem is notorious for the complexity of its syntax, the hymns articulated this twisting syntax via a micro-lexicon that lent the whole a repetitive, formulaic feel.

All of this evidence taken together suggests that we can think of the hymns—both christological and hagiographical—as arising, to some degree, in a shared context of oral composition. As recent scholarship on orality in late antiquity and the Byzantine period has argued, the relationship between orality and literacy in these later periods was not a simple binary, but one that emerged through a complex interchange between oral and written elements. [36] The hymnody ascribed to Ephrem arose in a robustly literary culture, and it is thus certainly likely that its oral components existed alongside written elements. But scholarship on the transmission of Ephrem’s hymnody has focused almost entirely on their process of textualization, despite all these cues that suggest orality as an equally crucial aspect of their composition and transmission. [37] In reflecting on these hymns as oral compositions, I distinguish the authorial model from one primarily and fundamentally literary, in which a poet would write a script that he, and perhaps other poets, would later perform. Instead, by oral compositions, I indicate texts that emerged, even if with written prompts, through a process of singing and listening, of learning the building blocks of meter and melody, narrative and thematic pattern, specific words, phrases, and modes of argumentation, to which were attached certain theological themes and ideas. Out of the pieces of these building blocks, poems came together intermittently in the context of liturgical performance. Contexts of liturgical performance, moreover, changed: while a poem emerged from the same cache of stories, rhetorical topoi, and set ways of positioning self and audience vis-a-vis God and other, the end to which these pieces were put differed depending upon the occasion. In this model, the changing referent between the christological poems and the hagiographical poems, resting underneath a shared poetic veneer, is not necessarily one of chronology and development. Rather, the similarities may instead suggest participation in a shared culture of poetic learning and production, one in which easy chronological boundaries between early and late remain difficult and unhelpful to draw. What we witness between the christological and hagiographical hymns is a shared style of singing about the holy, which migrates between Christ and the saints as “other christs,” between the Eucharist and the relics.

The idea that words moved between poets and poems in the context of performance suggests a compelling analogue to the movement of relics across the late antique landscape. The movement of the holy in the early Christian world was rarely one of sudden, contextless appearing, but one that pressed outward from certain charged locales, always looking back to the places from which it came. [38] Like the poetry of Ephrem as imagined by the later pseudonymous portrait with which we began, God’s manifestation in the things of the world moved outward like a vine, channeled through holy sites and holy people, gradually covering the expanse of the earth. In a late antique Christian context, the symbol par excellence for this locative movement attaches to the liturgy of Jerusalem, where the sites of the life and death of Christ, and the holy men and women of the Bible, were linked in the practice of pilgrimage. [39] The pilgrims, their journeys mapped by the movement of Bible and liturgy, traversed the earth, binding it as they walked, sang, and prayed in a web of divine signification. [40] This structural logic is repeated throughout the Mediterranean in the cult of the saints. As scholars have noted, the spatial movement of saints and their relics throughout the late antique world represented a process both historical and conceptual. [41] Historically, saints’ bodies were divided and distributed throughout the world. Conceptually, the holiness of saints’ bodies radiated out from central locations, spreading that localized sanctity universally. The local became ecumenical, and the space of the world sanctified through a network of cult sites that always harkened back to their place of origin, while simultaneously emphasizing the part as capable of manifesting the whole. [42] The cult was built through the movement of pieces that were both mundane and holy, fractured and whole, dead and living.

Late antique poetry broadly, like the cult of relics, has come to be recognized for its cultivation of a certain aesthetics of fragmentation. In the context of Latin poetry, Michael Roberts has argued that, unlike a classical aesthetic that privileged “the unity of the whole, the proportion of the parts, and the careful articulation of an apparently seamless composition,” late antique poetry not only let the seams show, but blatantly advertised them. [43] Though the Syriac culture out of which ephremic hymnody emerged was different in important ways from Latin poetic culture, both Syriac and Latin poets share this display of the fragmentary. Like relics, Syriac poems emerged from the pieces of words, phrases, and narratives. These pieces passed through the mouths of poets and consecrated the ritual spaces in which they were sung, even as they were consecrated by those spaces. The spatial movement of relics mirrors what we find in these hagiographical hymns, but in the hymns this process of sacred movement works on the level of language, moving through the spaces of the imagination, spaces that are both real and collective.

I suggest the linking of the movement of oral words and physical relics not simply as an imaginative exercise, but to underscore the materiality of the oral words of Syriac poems and to reconceive the relationship between artistic production, human agency, and ritual in early Byzantine Syriac poetry. Theories of performativity and orality have helped us to appreciate the tangible, quasi-material character of uttered language. Building on oral formulaic theory and the history of the senses, Reyhan Durmaz helpfully portrays orality as a kind of space out of which hagiographical texts emerge. The space, while elusive, was nevertheless concrete: “When we read hagiographical texts, we capture a moment in the life of a living, moving, changing story, one that mostly stays in the space shared by speech and hearing.” [44] So, like relics, stories—or, in the case of this chapter, poetic fragments—lived, moved, changed, and shaped the poets who uttered them. In the examples from the hymnody at which we have looked, formulaic ways of speaking migrated between different poems and different contexts of poetry. Like the words in this chapter’s opening vision, they acted upon poets, as much as poets acted upon them. If we take seriously the link between poetry and relic, it has consequences for the way we think about the agency of early Byzantine poetry. Just as pilgrims were shaped by the relics they touched, tasted, and looked upon, so poets were affected by the words that came upon them. [45] If oral words can move through space just like written words, then they, too, can be heard, tasted, touched, kept as protection, and redirected yet again.

These observations can help us think finally about the role that ritual context played in early Byzantine artistic production. In these Syriac hagiographical hymns, literary form, theological content, and ritual practice developed in a relationship of reciprocity. The fracturing of Syriac hymns, and the composition of wholes through attention to parts, developed alongside veneration of martyrs’ relics and the Eucharistic bread. Hymnody came to the poet in parts, but those parts emerged as pieces of a whole that nevertheless existed beyond the poet’s grasp. The consecration of relics that these hymns worked to achieve proceeded as a musical weaving of presence into presence—Eucharist into relic into poem and back again. All of these pieces played together. Rather than emphasizing a model of religious authority in which artistic production emerged as a human, agentival response to the stimuli of religious ritual, we can think of poetry as itself carrying out the early Byzantine relic’s unique negotiation of part and whole, movement and stasis, material and immaterial.