Skip to the story
BibleMapThe human story
Illustrated journeySources

Antioch, Roman Syria · around 303 CE

Romanus

The voice they could not silence

Romanus was a Christian deacon from the church around Caesarea. In Antioch, during the persecution associated with Diocletian, he challenged people sacrificing to the gods. He was arrested. His tongue was cut out. He died in prison.

We know him through the writers who remembered him. They do not all tell the same story about what happened to his voice.

Begin with the person
About this encounter and its title

AI-assisted editorial draft, awaiting human historical and theological review. The title names a conviction in expanded traditions; it does not settle the difference between them. Original summaries follow identified sources. This is a historical Christian account, not Scripture. Eusebius 2.1–5; Prudentius 10.1–15.

A man from one city, remembered in another.

We meet Romanus in someone else's writing. Eusebius, a Christian historian who lived through the persecution, calls him a deacon and exorcist of the church at Caesarea: a member of its ministry. The longer account places that service in a village around the city. Neither account gives us a childhood or a private life.

Caesarea lay on the eastern Mediterranean coast. Antioch lay farther north, near the coast in Roman Syria. Romanus was in Antioch when the events began. The shorter account does not explain why he was there. The distance between the cities is part of his story; an itinerary is not something we possess.

Around 303, imperial measures against Christians were changing ordinary life. Eusebius describes church buildings being demolished and sacred books destroyed, followed by orders to imprison church leaders and compel sacrifice. His story of Romanus begins among people going to sacrifice: men, women and children.

Sources for this movement

Original editorial summaries. Reported actions and theological interpretations remain attributable to the texts named here.

A city larger than this story.

Before Romanus, someone near Antioch lived with this image underfoot. The floor comes from Daphne, a retreat used by wealthy people from the city. At its center is a figure of abundance or good living.

The object cannot tell us who joined the crowd Romanus confronted. It reminds us that his account belongs to a world of homes, wealth and images as well as persecution.

The Met’s object record and interpretation

He raises his voice in public.

Romanus rebukes the people he sees sacrificing. Eusebius describes his religious zeal; he does not preserve the words of this first rebuke in the shorter account. For Romanus, as Eusebius presents him, worship of these gods is incompatible with his Christian allegiance.

Sacrifice was a public act as well as a religious one. Under persecution, it could become the act by which authorities measured compliance. Eusebius elsewhere describes people being counted as having sacrificed even while they resisted. Yet his shorter Romanus account does not say that every person in this crowd had been ordered to sacrifice. The longer Syriac version explicitly describes people being compelled.

His intervention brings arrest. The judge sentences him to burn. He is bound to a stake, with wood around him, while those attending wait for the emperor before lighting the fire. At this point the historian gives Romanus a question.

Sources for this movement

Original editorial summaries. Reported actions and theological interpretations remain attributable to the texts named here.

“Where is the fire for me?”
Words attributed to Romanus by Eusebius, 2.2.
McGiffert translation, 1890. A reported question, not a surviving recording.

The fire is not lit.

Romanus is brought before the emperor again. The punishment changes: his tongue is cut out. The shorter account describes him offering it willingly and interprets his endurance as the presence of divine power. That is Eusebius's Christian reading of bodily suffering, not access to everything Romanus felt.

Then he is put in prison. The text does not tell us that he speaks again. It also does not state that renewed speech was impossible. The absence is precise: this account gives us no further words from him.

Sources for this movement

Original editorial summaries. Reported actions and theological interpretations remain attributable to the texts named here.

Another text gives him speech.

If we turn to the longer account preserved in Syriac, the story after the wound changes. It says Romanus speaks through God's gift, greeting people, exhorting them and praying. The difference is inside the textual tradition of Eusebius itself.

Other writers will make that voice central to his memory. A preacher addresses a Christian congregation; a poet asks the martyr for the ability to write. They are doing different things with the same remembered life. Opening their texts lets us hear their claims without turning them into one continuous eyewitness account.

Sources for this movement

Original editorial summaries. Reported actions and theological interpretations remain attributable to the texts named here.

Eusebius · shorter Greek account

After the wound

“After this punishment he was thrown into prison”

2.4 · McGiffert translation, 1890

No further speech is reported.

The narrative continues to his death. This omission neither narrates a miracle nor disproves one.

Read this text’s provenance

Keep that account in view. Open another.

Open the longer Syriac account

Eusebius, in Syriac transmission

Longer work c. 311; surviving manuscript dated 411
Martyr narrative · translated textual witness

Speech continues.

After the tongue is removed, Romanus speaks through God's gift. He greets acquaintances, encourages worship of God, and prays. The account continues with his death by strangulation.

This is not an independent witness added to Eusebius. The long work likely preceded the short reworking, but the surviving Syriac translation may contain expansions. We cannot securely assign this passage to the lost Greek original simply because it appears here.

Editorial summary · Cureton p. 8; Minov, E00295

Open the preacher’s remembrance

John Chrysostom

Late fourth / early fifth century, 386–407
Festival homily · Christian theological interpretation

The voice becomes a sermon.

In the accepted homily, a doctor cuts out the tongue, yet a stronger spiritual voice sounds. Chrysostom presents the attempted silencing as defeat for the Devil. He begins with love: martyrdom without love has no effect.

This summary follows Efthymios Rizos's catalog account, not a directly inspected complete translation. The accepted homily is distinct from a second, generally inauthentic sermon under Chrysostom's name. Neither is Romanus speaking to us without a preacher between.

Editorial summary · CPG 4354, PG 50.605–612; Rizos §§1–4; Tsartsidis pp. 30–31

Open the poet’s invocation

Prudentius

Around 400
Latin Christian poetry · polemic and invocation

A poet asks for a voice.

The poet addresses Romanus and asks for eloquence. A martyr who spoke without a tongue can, in the poem's logic, help a faltering poet speak. Later, Prudentius gives Romanus a long argument against traditional cults, even after the mutilation.

uox ueritatis testis extingui nequit

The poem's speeches are literary composition, not a transcript. Its attacks on other religions belong to Prudentius's polemical Christian perspective. His opening declares the conviction behind the title of this encounter.

Editorial summary · 10.1–25; 931–1010

Why a longer text is not necessarily a later story

Eusebius’s longer work is usually dated around 311; the shorter Greek recension reworks it. The full longer text survives in Syriac, including a manuscript dated 411. Some scholars have proposed that the renewed speech was an expansion in translation or transmission. We cannot resolve that question by putting the two texts in date order on a screen.

Minov, CSLA E00294 and E00295; Tsartsidis, pp. 29–30. Manuscript date, composition date and event date are different things.

His life ends in prison.

The shorter account returns to a prisoner held for a long time. An imperial anniversary approaches, with a release of prisoners. Romanus is instead stretched in the stocks and strangled. The longer Syriac account also ends in strangulation. Continued speech, in that telling, does not mean escape from death.

Eusebius counts him among the Palestinian martyrs even though he died in Antioch. For this Christian historian, his death belongs to a community's remembrance. But the person reaches us through that act of remembrance: selected events, transmitted words, a writer's judgment.

Prudentius ends his poem by imagining two records. The official's papers decay; an angel's record before God endures. That is the poet's theological vision. What we can read is the poem itself, alongside other texts that preserve Romanus differently. We have their writing. We do not have his voice apart from it.

Sources for this movement

Original editorial summaries. Reported actions and theological interpretations remain attributable to the texts named here.

Romanus

Remembered by Eusebius.
Preached by Chrysostom.
Given words by Prudentius.
Read by us.

When we keep someone's voice alive, how much of what survives is theirs—and how much is ours?

A question from this encounter, not a quotation from the sources.

Read the sources, uncertainties and image record

The writing that reaches us.

This is a historical Christian story, not a passage of Scripture. Its surviving accounts speak from particular Christian perspectives. Multiple versions and later retellings do not amount to multiple independent eyewitnesses.

AI-assisted editorial draft for human review. Summaries, transitions, diagram and typographic composition are modern work for BibleMap. Historical, theological and literary review is pending. No original words or private thoughts have been invented for Romanus.

Why “around 303”? Which emperor?

The shorter account connects the death to Alphaeus and Zacchaeus and an approaching imperial twentieth anniversary. It separates the arrest from death by a long imprisonment. Precise dates and the emperor’s identity cannot be flattened into one certain sequence.

The short Greek text leaves the emperor unnamed. Cureton’s Syriac translation names Diocletian; Prudentius names Galerius in his opening setting. McGiffert argues for Galerius; Tsartsidis offers a different chronology. We use c. 303 for the death and leave the encounter’s emperor unnamed.

McGiffert, notes 2627–2632; Tsartsidis p. 29 n. 57; Prudentius 10.31–35.

Eusebius of Emesa and the other Chrysostom sermon

Tsartsidis identifies a passage from Eusebius of Emesa’s De resurrectione (PG 24.1097–1100), preserved in Latin and once attributed to Eusebius of Caesarea. It reports both rain extinguishing the fire and renewed speech. We have inspected the scholarly discussion, not that underlying sermon, and do not quote it.

The accepted Chrysostom homily (CPG 4354, PG 50.605–612) must be distinguished from a second sermon (PG 50.611–618), generally judged inauthentic, which includes a child martyr. These are distinct literary traditions, not extra episodes silently added to this journey.

Tsartsidis, introduction §5, pp. 30–31.

Eusebius · Martyrs of Palestine, shorter Greek recension

Open the source record or edition

Introduction; 1.4–5; 2.1–5; 3.1

Arthur Cushman McGiffert, Nicene and Post-Nicene Fathers, second series, vol. 1 (1890), transcribed by New Advent. Early fourth-century Christian commemoration; the shorter recension is a reworking of a longer work.

Eusebius lived through the persecution, but this chapter does not say he witnessed the events at Antioch. It reports no speech after the mutilation. Its theological judgments remain the author's. Neither a court transcript nor Romanus's own surviving writing.

Historical public-domain translation. Two brief credited excerpts; all other narrative is original editorial summary. Checked 2026-11-04.

Eusebius · longer account preserved in Syriac

Open the source record or edition

Cureton's printed pp. 6–8, especially p. 8, lines 1–23

William Cureton, History of the Martyrs in Palestine (1861), English translation of Syriac. Relevant translation inspected through digitized OCR; the Syriac manuscript was not inspected. No OCR wording quoted.

The longer Greek recension is generally dated around 311; its complete text survives in Syriac translation, including a manuscript dated 411. This is a textual witness to Eusebius, not a separate eyewitness. Translator/redactor expansion is a scholarly possibility; shorter does not automatically mean earlier.

Public-domain historical edition; original summaries only. Checked 2026-11-04.

Sergey Minov · Eusebius and the Syriac transmission

Open the source record or edition

CSLA E00295, overview and discussion; companion E00294, doi:10.25446/oxford.13797248

The Cult of Saints in Late Antiquity, University of Oxford. Complete catalog discussions inspected; includes BL Add. 12150 and its November 411 colophon, plus bibliography.

A scholarly synthesis, not direct examination of the manuscript here. References to Barnes's proposal about the miracle are mediated through this entry; Barnes's full book was not read.

Catalog metadata CC0. Original editorial summary; no manuscript reproduction. Checked 2026-11-04.

John Chrysostom · On Romanos

Open the source record or edition

CPG 4354; BHG 1601; PG 50.605–612; CSLA E02538, summary §§1–4

Accessed through Efthymios Rizos's full scholarly catalog summary. Critical edition: Rambault and Allen, Sources Chrétiennes 595 (2018); English translation: Wendy Mayer (2006), pp. 227–237. These complete editions were not inspected.

Accepted Chrysostom homily, distinct from the second sermon transmitted under his name (PG 50.611–618), generally judged inauthentic. CSLA dates this to 386–407, at Antioch or Constantinople; a narrower Antioch dating is also argued. Christian festival preaching, not eyewitness corroboration.

CSLA record CC0. Original paraphrase of the identified summary; no translated homily quotation. Checked 2026-11-04.

Prudentius · Peristephanon, poem 10

Open the source record or edition

1–25 (invocation); 31–85 (setting); 891–1010 (mutilation and speech); 1096–1140 (death and records)

Latin text, The Latin Library, inspected alongside CSLA E00947 (doi:10.25446/oxford.13802798), which identifies Cunningham's 1966 edition and dates the poem around 400. Original summaries; one short Latin quotation.

A Latin Christian poem with extended anti-pagan argument, invocation and literary speeches. Written long after Romanus's death. Its depiction of a heavenly record is theological poetry, not a recovered document. Relationships to earlier passions are debated.

Ancient public-domain Latin text; no modern English translation reproduced. Checked 2026-11-04.

Thomas Tsartsidis · the sources for Romanus

Open the source record or edition

Commentary on Prudentius' Hymn to Romanus 1–650 (Edinburgh PhD, 2017), introduction §5, printed pp. 29–38

Repository PDF downloaded; the introduction's source survey inspected (PDF pp. 39–48), not the full thesis. Distinguishes the accepted and pseudo-Chrysostom sermons and identifies Eusebius of Emesa, De resurrectione, PG 24.1097–1100.

Eusebius of Emesa's underlying sermon was not directly read. Its attribution and reports of rain and renewed speech are presented here through Tsartsidis, not as newly verified independent testimony. Proposed source relationships and dates are scholarly arguments.

Copyright retained by author/university. Linked and briefly paraphrased; no scanned page reproduced. Checked 2026-11-04.

McGiffert · notes on date, sacrifice and emperor

Open the source record or edition

NPNF notes 2627–2632 to Martyrs of Palestine 2

McGiffert's 1890 editorial notes, read separately from Eusebius's text and compared with Tsartsidis (2017), p. 29 n. 57.

An older scholarly interpretation, not consensus. Short Greek leaves the emperor unnamed; Syriac names Diocletian; McGiffert argues for Galerius. Arrest and execution were separated by imprisonment. Exact sequencing and reconciliation remain uncertain; the experience uses c. 303 rather than a precise-day reconstruction.

Historical public-domain commentary, linked and summarized. Checked 2026-11-04.

The Metropolitan Museum of Art · a floor from Daphne

Open the source record or edition

Mosaic floor panel, 38.11.12; object description and Open Access API object 253565

Roman, second century CE, stone, tile and glass. Found at Daphne-Harbiye near Antioch in 1937. Purchase, Joseph Pulitzer Bequest, 1938. Object record and image-level public-domain API flag inspected.

The central figure is a personification, traditionally Spring and now described as probably abundance/good living. It is not Romanus, a portrait of someone in his story, or evidence of his arrest. It predates the events.

The Met Open Access / CC0. Local optimized derivatives of DP140137.jpg; full composition retained. Checked 2026-11-04.

Cornell Archaeology · Caesarea Maritima

Open the source record or edition

Site introduction; compare Eusebius 2.1,5 and the Met's Daphne object record for Antioch

Cornell University Archaeology Program, site description. The diagram uses broad coastal orientation; no ancient itinerary is reconstructed.

Modern site orientation cannot establish Romanus's exact village, journey, trial site or prison. Why and when he came to Antioch remain unknown in the short account.

Linked site description; original schematic drawing, no institutional map copied. Checked 2026-11-04.

About the title and the modern artwork

“The voice they could not silence” names a conviction in the expanded traditions, especially Prudentius’s opening. It is not a historical verdict about the disputed speech. The question at the ending is a modern editorial reflection.

The map is an original AI-assisted schematic, not an archaeological map or a recovered route. The blank interval after the punishment and the opening textual leaves are modern typographic interpretations, not ancient manuscripts. No sound or simulated voice is used.

The Met’s second-century mosaic is shown intact. Its female personification is not Romanus. It supplies material context for the broader world around Antioch; it cannot establish what happened to him. Open Access API object 253565 confirms the image is public domain under The Met’s CC0 program.

This encounter remains on the collection’s no-index review surface. Technical verification does not approve historical or theological interpretation. Return to Scripture: read Jonah.