Gullo


 

————— 
Greek, Roman, and Byzantine Studies 62 (2022) 328–340 

Article copyright held by the author(s) and made available under the 
Creative Commons Attribution License 

 CC-BY  https://creativecommons.org/licenses/by/4.0/ 
 
 
 

The Artistic Object as µάρτυς and the 
Formula µάρτυρές εἰσι πόνων in Late-antique 

Epigrams for Public Monuments 
Arianna Gullo 

µαρτυρίην γὰρ Ἔρως µόνος οὐκ ἐφίλησεν  
ἔµπνουν  Phld. Anth.Pal. 5.4.3–4 

 
T IS A RATHER COMMON literary convention in laudatory 
poetry, particularly in inscriptions, that people, places, and 
things are “witnesses” to the deeds or the work of the 

honorand. The pattern is pervasive, especially in epigrams from 
the Garland of Philip.1 

This paper falls into three sections. In the first two I study late-
antique laudatory epigrams for illustrious citizens that feature 
the motif of artwork (mostly statues) testifying to the honorand’s 

 
1 In Parmen. Anth.Plan. 222.4 = Gow-Page, GP XV 2627, the statue of 

Nemesis at Rhamnous bears witness to Attica of victory and of art; in Crin. 
Anth.Pal. 9.283.3 = GP XXVI 1919 = 26.3 Ypsilanti, mountains and rivers 
are depicted as witnesses of Germanicus’ victories over the Celts; in another 
epigram by Crinagoras, Anth.Pal. 9.419.4 = GP XXIX 1938 = 29.4 Ypsilanti, 
Pyrenean waters attest the fame of Augustus; in Antip. Thess. Anth.Pal. 
9.238.1–2 = GP LXXXIII 535–536, the bronze statue of Apollo by Onatas 
bears witness to Zeus and Leto of their beauty; in Alph. Anth.Pal. 7.237.2 = 
GP VI 3543 Apollo is represented on Themistocles’ funerary monument as 
µάρτυς of his deeds. Also in Antip. Sid. Anth.Pal. 7.427.3–4 = Gow-Page, HE 
XXXII 398–399, the tombstone features four dice which represent (µαρ-
τυρέουσι) the throw called “Alexander”; in Greg. Naz. Anth.Pal. 8.62.2, a table 
bears witness to the passing of Nonna, Gregory’s mother, while she was 
praying, while in Anon. Anth.Pal. 15.10.1, the sea and the waves should give 
testimony to a shipwreck. For the Greek Anthology I follow the text of H. Beckby, 
Griechische Anthologie I–IV (Munich 1967–68). 

I 



 ARIANNA GULLO 329 

————— 
Greek, Roman, and Byzantine Studies 62 (2022) 328–340 

 
 
 
 

deeds. In particular, I focus on the iunctura µάρτυρές εἰσι πόνων, 
which marks a precise subcategory of public honorific inscrip-
tions in which the artistic object acts as µάρτυς. Based on this 
review of the late-antique evidence, the third section shows that 
it is possible to trace back the original form and model of the 
conventional iunctura to µάρτυρές εἰσι in the classical age, when 
this expression is already attested in inscriptional poetry. I argue 
that the widespread motif of artwork, particularly statues, as 
“witnesses” in epigrams probably derives from the practice of 
calling upon gods and goddesses as witnesses. Late-antique in-
scriptions naturally ascribed to statues as the nearest representa-
tives of the gods their own function as witnesses. They thus 
referred to them as µάρτυρες. This theme of the artistic object 
that bears witness likely in turn stimulated in late-antique 
honorific inscriptions the expansion of the original formula 
µάρτυρές εἰσι into µάρτυρές εἰσι πόνων. 
1. The role of artwork as “witness” in late-antique honorific inscriptions 

The ekphrastic theme of the statue bearing witness to the 
achievements of the honorand was very popular, especially in 
late-antique epigrams. In this section I examine occurrences of 
this motif in inscriptions of the fourth to sixth centuries. A few 
examples were already collected by Federica Giommoni.2 To 
these I add two. First, an inscription for the general Theodosius 
the Elder, whose στήλη in Ephesus acts as µάρτυς to his virtues 
(Merkelbach-Stauber, SGO 03/02/24 = Late Statues of Antiquity-

 
2 F. Giommoni, “La testimonianza catalogica dei barbari: tra retorica e 

iconografia,” in D. Gigli Piccardi et al. (eds.), Studi di poesia greca tardoantica 
(Florence 2013) 135–147, at 138–143. The chapter focuses on the rhetorical 
topos of a list of subdued barbarians—a well-established convention to 
celebrate the emperor’s victories in both imperial panegyric and late-antique 
eulogistic epigraphy—as it appears in an epigram from the Cycle of Agathias 
(Arab. Anth.Plan. 39 = II Giommoni = LSA-476), which I discuss below. The 
“witness motif” is briefly touched upon at the beginning, solely in connection 
with barbarians as witnesses, and viewed through the lens of its visual impact 
in contemporary works of art dedicated to the emperor. 



330 THE ARTISTIC OBJECT AS ΜΑΡΤΥΣ 

————— 
Greek, Roman, and Byzantine Studies 62 (2022) 328–340 

 
 
 
 

722; A.D. 390): 
Εὐδικίης, µώλοιο, σαωφροσύνης, ἀρετάων 
   µάρτυς ἐγὼ στήλη Θεοδοσίῳ τελέθω.  
To justice, to toil, to temperance, to the virtues, I, the statue 
dedicated to Theodosius, stand as witness. 

Second, an honorific inscription for the proconsul Eustathius on 
his statue base, in which the statue conventionally speaks in the 
first person and declares that it testifies to the good and noble 
things he did, as well as to those that the assembly did for him in 
return (SGO 05/01/09 = LSA-516; Smyrna, fourth/fifth cent.): 

εἰκὼν Εὐσταθίοιο πέλω, φίλος· εἰµὶ δὲ µάρτυς  
ἀµφοτέρων κραδίης πα̣[ν]ετήτυµος, ὅσσα µὲν αὐτὸ[ς]  
βουλὴν ἐσθλὰ ἔοργε̣ πονεύµενος, ὅσ⟨σ⟩α δ̣ὲ̣ βουλή 
ἀνθυπάτων τὸν ἄριστον ἀµείψατο κυδαίνουσα. 
I am the statue of Eustathius, friend. I bear true witness to the 
purpose of both sides, to all the noble things that he has done 
working hard for the council, and to all the things that the council, 
holding him in honor as the best of proconsuls, has done in return. 
In Agath. Anth.Pal. 1.36.5–6 = 17 Viansino the picture of 

Theodorus, the illustrious magister officiorum and proconsul Asiae in 
the mid-sixth century, portrayed while receiving the insignia of 
office from the archangel, testifies to his gratitude, for he faithfully 
had the archangel’s grace toward him painted in colors: 

τῆς δ᾽ εὐγνωµοσύνης µάρτυς γραφίς· ὑµετέρην γὰρ 
   χρώµασι µιµηλὴν ἀντετύπωσε χάριν. 
This picture testifies to his gratitude, for he faithfully depicted 
your grace towards him in colors. 

An epigram by Michael the grammarian on an icon of the poet 
and historian Agathias Scholasticus claims (lines 3–4) that his 
hometown, the city of Myrina, gave him this portrait as a testi-
mony of his love and his own literary skill (Anth.Plan. 316 = SGO 
05/04/01 = LSA-663 and LSA-2494): 

… καὶ πόρε τήνδε 
εἰκόνα, καὶ στοργῆς µάρτυρα καὶ σοφίης· 
(The city) gave him this portrait as a witness to his love and his 
literary skill. 



 ARIANNA GULLO 331 

————— 
Greek, Roman, and Byzantine Studies 62 (2022) 328–340 

 
 
 
 

The “witness” motif endures down to the Byzantine era. For 
example, it is expressed by the verb µαρτυρέω in votive epigrams 
or inscriptions for small objects (Rhoby, Byz.Epigramme II no. 
Me28; France, before 1204): 

καὶ πρὶν ὑπούργει τὸ τρύβλιον Δεσπότῃ 
κεῖνῳ µαθητὰς ἐστιῶντι τοὺς φίλους 
καὶ νῦν ὑπουργεῖ τοῖς µειλιγµοῖς Δεσπότου· 
µαρτυρεῖ τοῦτο δῶρον εἰσειργασµένον. 
For previously the bowl served that Lord when he entertained his 
beloved pupils, and now it is used for the offerings to the Lord. 
This is witnessed by the gift made for this purpose. 

No. Te6.2 Rhoby (Meteora, fourteenth cent.): 
ὡς µαρτυρεῖ τὸ λύθρον ἐν τῇ φιάλῃ.  
As the blood in the bowl testifies. 
Especially in Byzantine epigrams for sacred objects, like the 

two just quoted, the offering “testifies” or “bears witness” to the 
religious devotion of the donor. The motif is also featured, again 
with µαρτυρέω, in the epitaph for Basil II (d. 1025), buried in the 
church of St John the Evangelist near the Ἕβδοµον in Con-
stantinople (lines 15–16):3  

καὶ µαρτυροῦσι τοῦτο Πέρσαι καὶ Σκύθαι,  
σὺν οἷς Ἀβασγός, Ἰσµαήλ, Ἄραψ, Ἴβηρ. 
And this is witnessed by the Persians and the Scythians, [and] 
with them the Abasgians, the Ishmaelites, the Arabs, and the 
Iberians. 

The different peoples listed as witnesses could have been actually 
carved in Basil’s funerary monument or, by exploiting the power 
of ekphrasis through allusion, they might have been only evoked 
in the mind of the reader. 

From the cases examined above, we conclude that the motif 
of the statue bearing witness to the achievements of the 
honorand is a key poetic convention of inscriptions on artifacts, 

 
3 S. G. Mercati, “Sull’epitafio di Basilio II Bulgaroctonos,” Bessarione 25 

(1921)137–142 = Mercati, Collectanea Byzantina II (Bari 1970) 226–231. 



332 THE ARTISTIC OBJECT AS ΜΑΡΤΥΣ 

————— 
Greek, Roman, and Byzantine Studies 62 (2022) 328–340 

 
 
 
 

monuments, and buildings, a convention sanctioned by imperial 
power or, more generally, bearing the official authority of a 
formal literary code. The use of µάρτυς peculiar to Christians4 
(“martyr” or “witness to the truth”)5 likely shaped the “witness 
theme” in late-antique honorific epigrams with ekphrastic impli-
cations: the task of bearing witness is performed by mute statues, 
which bear real and living testimony though fixed in marble to 
the eyes of the viewer (or visualizing reader). 
2. The iunctura µάρτυρές εἰσι πόνων 

A group of epigrams for honorific statues that give expression 
to the “witness theme” feature the iunctura µάρτυρές εἰσι πόνων. 
Alan Cameron studied it in his work on Porphyrius and other 
charioteers from fifth/sixth-century Constantinople but limited 
his analysis to very few examples.6 Τhis formula appears to have 
been a favorite of late-antique epigrammatists. In this section I 
review all its occurrences and expand previous studies to include 
its variants and adaptations. I show that the expression was not 
only good literary phrasing: its political significance suited the 
formal language of inscriptions in late antiquity. My approach 
allows me to reconstruct the original form of the iunctura against 
the strong conventional background on which it rests.  

The formula µάρτυρές εἰσι πόνων occurs in the clausula of the 
pentameter in Anon. Anth.Plan. 353.2 = LSA-361, one of the 
epigrams carved on the decorated bases of the bronze statues of 
Porphyrius in the Hippodrome:   

 
4 In late-antique honorific inscriptions for eminent dedicatees the word 

µάρτυς is never employed in the Christian sense of “martyr.” This use, 
however, is common e.g. in Book 1 (Christian epigrams: 5 occurrences) and 
Book 8 (epigrams of Gregory of Nazianzus: 17 occurrences) of the Greek 
Anthology. For a comprehensive survey of the word family of µάρτυς in Nonnus 
see F. Vian, “Μάρτυς chez Nonnos de Panopolis: étude de sémantique et de 
chronologie,” REG 110 (1997) 143–160. 

5 E.g. Greg. Naz. Anth.Pal. 8.118.6 µάρτυρες ἀτρεκίης. Cf. also the use of 
µαρτύροµαι in Anth.Pal. 8.169.1. 

6 A. Cameron, Porphyrius the Charioteer (Oxford 1973) 91. 



 ARIANNA GULLO 333 

————— 
Greek, Roman, and Byzantine Studies 62 (2022) 328–340 

 
 
 
 

Εἰ φθόνος ἠρεµέοι, κρίνειν δ’ ἐθέλοιεν ἀέθλους, 
   πάντες Πορφυρίου µάρτυρές εἰσι πόνων. 
If envy would keep quiet and they should be willing to judge the 
contests, all bear witness to Porphyrius’ labors. 

Everyone is a witness to the glorious deeds of Porphyrius. The 
iunctura is found adapted in other poems of roughly the same age. 
But before we look at such adaptations, consider the epigram in 
Anth.Plan. 39.4 = II Giommoni = LSA-476, by Arabius Scho-
lasticus (one of the Cycle poets), which is dedicated to an icon of 
Longinus, hyparchos in A.D. 537–539 and 542 and magister militum 
in Constantinople in 551: 

Νεῖλος, Περσίς, Ἴβηρ, Σόλυµοι, Δύσις, Ἀρµενίς, Ἰνδοί 
   καὶ Κόλχοι σκοπέλων ἐγγύθι Καυκασίων 
καὶ πεδία ζείοντα πολυσπερέων Ἀγαρηνῶν 
   Λογγίνου ταχινῶν µάρτυρές εἰσι πόνων· 
The Nile, Persia, the Iberian, the Solymoi, the West, Armenia, 
the Indians, the Colchians near the peaks of Caucasus, and the 
burning plains of the widespread Hagarenes are witnesses to the 
swift labors of Longinus. 

A list of peoples and places perhaps fixed in marble (including 
the river Nile) act as witnesses to Longinus’ deeds. An epigram 
by Leontius Scholasticus (Anth.Plan. 37.1–2 = XV Giommoni = 
LSA-477) offers another instance. It celebrates Peter Barsymes, 
comes sacrarum largitionum (A.D. 542 and 557/8–560), patricius 
(542), praefectus praetorio Orientis (543–546 and 555–559), and 
honorary consul: 

Πέτρον ὁρᾷς χρυσέοισιν ἐν εἵµασιν· αἱ δὲ παρ’ αὐτὸν 
   ἀρχαὶ ἀµοιβαίων µάρτυρές εἰσι πόνων· 
You see Peter in his golden robes, and the offices beside him bear 
witness to his alternating labors. 

The representation of Peter’s offices (ἀρχαί) next to his portrait 
are witnesses to his achievements as honorary consul and as 
praefectus praetorio Orientis. Perhaps Peter’s offices were personified 
and painted standing by him—the first praefectura, then the 
consulship, and once again the praefectura (a second time), the 



334 THE ARTISTIC OBJECT AS ΜΑΡΤΥΣ 

————— 
Greek, Roman, and Byzantine Studies 62 (2022) 328–340 

 
 
 
 

genitive ἀµοιβαίων alluding to the fact that the two prefectures 
alternated with the consulship. 

As to adaptations of the formula µάρτυρές εἰσι πόνων, the 
closest parallel identified by Cameron dates back to the second 
or third century A.D. It is an inscriptional poem from Rome (GVI 
1937.1–2 = IGUR 1328): 

ἀθάνατ[ος] µερόπων οὐδεὶς ἔφυ· τοῦδε, Σεβήρα, 
   Θησεύς, Αἰακίδαι, µάρτυρές εἰσι λόγου· 
No man is immortal: to this statement, Severa, Theseus and the 
sons of Aeacus testify. 

A further adaptation is echoed by an epigram (Anth.Plan. 38 = 
VIII Giommoni = SGO 20/11/03 = LSA-484) by John Barbu-
callus (another Cycle poet) on an εἰκών of Synesius Scholasticus 
“erected in Berytus on the occasion of his victory in battle” (εἰς 
εἰκόνα Συνεσίου σχολαστικοῦ ἐπὶ νίκῃ µάχης ἀνατεθεῖσαν ἐν Βη-
ρυτῷ): 

Οὐχὶ παρ’ Εὐρώτᾳ µόνον ἀνέρες εἰσὶ µαχηταί, 
   οὐδὲ παρ’ Ἰλισῷ µνάµονές εἰσι δίκας· 
ὡς ἀπὸ τᾶς Σπάρτας, ὡς αὐτᾶς ἀστὸν Ἀθάνας 
   Συνέσιον Νίκα καὶ Θέµις ἠγάσατο. 
Not only by Eurotas are there warriors, and not only by Ilissus are 
there men mindful of justice. Victory and Themis reverenced 
Synesius as if he were from Sparta, as if he were a citizen of 
Athens itself.7 

The Athenians dwelling by the Ilyssus are not the only ones who 
remember (µνάµονές εἰσι) to act and live according to justice. In 
this epigram the Eurotas and the Ilyssus, the respective rivers of 
Sparta and Athens, to our surprise do not bear witness to the 
glorious deeds of their compatriots (as rivers often do in 
laudatory poems). Here they are replaced by men who take up 
the role of µάρτυρες. It is important that, from a formal and 
stylistic point of view, the words µνάµονές εἰσι δίκας are modeled 
precisely after µάρτυρές εἰσι πόνων. This attests that, at least 
 

7 Cf. also Anon. Anth.Plan. 351.1 = LSA-361 κήρυκες ἀµεµφέες εἰσὶν ἀγώνων, 
where the structural parallel is close to µάρτυρές εἰσι πόνων, but the meaning 
differs. 



 ARIANNA GULLO 335 

————— 
Greek, Roman, and Byzantine Studies 62 (2022) 328–340 

 
 
 
 

within this genre, the latter was recognized as an established 
poetic formula which, like all formulas, could be modified and 
adapted to suit new contexts and expressive needs. It is uncertain 
whether the two rivers were depicted in Synesius’ monument. 
From what we know about typical ekphrastic play, we can 
assume that the elements and figures denoted as witnesses in the 
epigrams for Porphyrius and Longinus, and in the second-
century Roman inscription and the epitaph of Basil II, were 
actually represented in the monuments. It also seems quite 
certain that the offices mentioned in Leontius’ epigram for Peter 
Barsymes were actually depicted (and probably personified).  

In inscriptional contexts like those analyzed above, both the 
image and the accompanying text combine to bear joint witness: 
the text refers to, clarifies, and amplifies the representation, en-
hancing its authority and expressive power. The interpretation 
of such epigrams must always bear in mind the inextricable bond 
between text and image. This facilitates a suitable appreciation 
of the “witness theme” and prevents traditional labels from 
dulling its original sense. Even if the “witnesses” were not 
represented in the artwork, the evocative power of the written 
word could call to the reader’s mind relevant images from 
suitable contexts, such as those seen, for example, during the 
triumphs celebrated in the Hippodrome of Constantinople. 

The examples reviewed above, including the variants, show 
that the expression µάρτυρές εἰσι πόνων was an established poetic 
formula in late antiquity. This conventional unit belongs to a 
language code that falls within the so-called “stile moderno,” 
that is, the language and meter codified in late-antique literary 
poetry, of which the supreme exponent is Nonnus of Panopolis.8 

 
8 For this reason it is called “Nonnian,” though it was not originally de-

veloped by him. On the definition of the “stile moderno” see, e.g., M. Whitby, 
“From Moschus to Nonnus: The Evolution of the Nonnian Style,” in N. 
Hopkinson (ed.), Studies in the Dionysiaca of Nonnus (Cambridge 1994) 99–155; 
G. Agosti and F. Gonnelli, “Materiali per la storia dell’esametro dei poeti 
 



336 THE ARTISTIC OBJECT AS ΜΑΡΤΥΣ 

————— 
Greek, Roman, and Byzantine Studies 62 (2022) 328–340 

 
 
 
 

Strong support for this statement is provided by Nonn. Dion. 
46.29 ἐπιµάρτυρές εἰσι κεραυνοί, which may have played a de-
cisive role in licensing its adoption as an approved formula. It 
was certainly entrenched in the poetic language of artwork 
carved, hammered, or painted for public display, and endorsed 
by secular power. Its use in inscribed verses marked the message 
as official. The depiction of monuments as witnesses (or of 
elements represented on them) was a common practice in 
laudatory inscriptions for statues exhibited in public spaces. This 
was perceived by the audience as a conventional motif of public 
poetry focused on self-representation.  
3. The origin of the expression µάρτυρές εἰσι πόνων 

The use of µάρτυς, or of cognate compounds and derivatives 
of it, is often accompanied by a genitive. This is abundantly at-
tested in the Greek Anthology.9 The expression µάρτυρές εἰσι πόνων 
 
cristiani greci,” in M. Fantuzzi et al. (eds.), Struttura e storia dell’esametro greco I 
(Rome 1995) 289–434. 

9 Antig. Anth.Pal. 13.28.5 = Page, FGE I 37 µάρτυρα Βακχίων ἀέθλων; Call. 
Anth.Pal. 6.311.2 = Gow-Page, HE XXVII 1172 νίκης µάρτυρα τοῦ Ῥοδίου; 
Hedyl. Anth.Pal. 5.199.6 = HE II 1836 = 2.6 Floridi ὕπνου καὶ σκυλµῶν τῶν 
τότε µαρτύρια; Anon. Anth.Plan. 268.2 ἠὲ σὺ τῆς κείνου µάρτυς ἀκεστορίης; 
Anon. Anth.Plan. 279.6 χερµάδι τοῦ κόµπου µαρτυρίην κοµίσαι; Alph. Anth.Pal. 
7.237.2 = Gow-Page, GP VI 3543 ἀµφοτέρων µάρτυρα; Euen. Anth.Pal. 
11.166.2 = GP XI 2339 πλούτου µάρτυς; Euen. Anth.Plan. 166.4 = GP XI 2341 
µάρτυρα τῆς τέχνης; Parmen. Anth.Plan. 222.4 = GP XV 2627 νίκης καὶ σοφίης 
Ἀτθίδι µαρτύριον; Antip. Thess. Anth.Pal. 7.369.2 = GP XLIX 338 Πανελ-
λήνων πεύθεο µαρτυρίης; Phil. Anth.Pal. 7.234.4 = GP XXXI 2842 ἐς προτέρην 
ἔργων ἄρσενα µαρτυρίην; Phil. Anth.Pal. 6.236.2 = GP II 2643 Ἀκτιακοῦ πολέ-
µου κείµεθα µαρτύρια; Anon. Anth.Pal. 3.15.6 µύθων ἐσθλῶν µάρτυς; Anon. 
Anth.Pal. 7.331.4 πιστὸν ἐµοῦ βιότου µάρτυρα σωφροσύνης; Greg. Naz. Anth. 
Pal. 8.118.6 µάρτυρες ἀτρεκίης; Pall. Anth.Pal. 10.54.3–4 τοῦδ’ ὁ τυραννήσας 
Διονύσιος Ἡρακλείας / τῆς ἐν τῷ Πόντῳ µάρτυς; Jul. Aegypt. Anth.Pal. 6.18.2 
γηραλέων ... µαρτυρίην ῥυτίδων; Jul. Aegypt. Anth.Pal. 6.19.4 δώρου ... 
µαρτυρίην; Jul. Aegypt. Anth.Pal. 7.583.3 µάρτυρα µόχθων; Maced. Anth.Pal. 
9.645.3 = 26 Madden µάρτυς ἐγὼ πρώτη γενόµην Διός; Anon. Anth.Plan. 344.5 
ἐπιµάρτυρα νίκης; Leon Anth.Pal. 9.578.8 τοῦτων δὲ µάρτυς; SGO 04/02/06.2–
4 = LSA-654 (Sardis, fourth cent.) Ἀχόλιος, / ᾧ βουλὴ µεγάλων ἀγαθῶν χάριν 
 



 ARIANNA GULLO 337 

————— 
Greek, Roman, and Byzantine Studies 62 (2022) 328–340 

 
 
 
 

follows this same syntax. In this final section I propose a possible 
origin for this formula. From my discussion, deities in epigrams 
containing the expression µάρτυρές εἰσι emerge as having played 
a key role in the development of the µάρτυς theme. 

As Cameron notes, “it would be superfluous to cite examples 
of so common a theme.”10 But the pervasiveness of the theme 
does not prevent the search for the original model of this iunctura. 
In a funerary verse inscription (GVI 1457 = CEG 82 = 12 Tentori 
Montalto) from Kokkina Chomata (Lemnos), dating to the 
second half of the fifth century B.C. and dedicated to an ἄωρος, 
familiar wording appears in the clausula of line 4:11 

   [ἄκ]ρως µὲν σοφίας µέτρο[ν ἐπι]στάµενος 
   καὶ ψυχὴν ἀγαθός· τούτω̣[ν µάρ]τυρές εἰσιν ἐµ[οί].   
Knowing the measure of wisdom at the farthest point and noble 
in soul: I have witnesses to this. 

The inscription consists of an elegiac distich, followed by a pen-
tameter and a final, irregular pentameter with an extra foot. The 
deceased fell in battle for his fatherland. The anonymous 
speaker states that he has witnesses (µάρ]τυρές εἰσιν ἐµ[οί]) that 
testify to the wisdom of the dead and of his brave soul. The for-
mula µάρτυρές εἰσι occurs in a Hellenistic literary epigram too. 
In Diosc. Anth.Pal. 5.56.7–8 = Gow-Page, HE I 1469–1470 = 
1.7–8 Galán Vioque, an erotic poem on the charm of the poet’s 
mistress, we read: 

ἀλλὰ τί µηνύω κυσὶν ὀστέα; µάρτυρές εἰσι 
  τῆς ἀθυροστοµίης οἱ Μίδεω κάλαµοι. 

 
εἰκόνα βαιήν / στήσαµεν, εὐνοµίης µάρτυρα πιστοτάτην; Kaibel 905.5 = I.Cret. 
IV 323 = L. Robert, Hellenica IV (1948) 89–94 = LSA-785 (ca. fourth cent.) 
Δίκης ἐπιµαρτυρα θεσµῶν; SGO 02/09/09.6 = ala2004 41 = IAph2007 8.608 
= LSA-225 (Aphrodisias, mid-fourth cent.) µάρτυς σῶν καµάτων. 

10 Cameron, Porphyrius 91. 
11 Perhaps under the influence of Il. 3.280, ὑµεῖς µάρτυροι ἔστε. For “they 

bear witness to …” and further parallels see also A. Harder, Callimachus. Aetia 
II (Oxford 2012) 629–630 on Callim. Aet. fr. 75.48 Pf. = 174.48 Massimilla 
ψήφου δ̣’ ἂν ἐµῆς ἐπιµάρτυρες ε̣ἶ̣ε̣ν̣. 



338 THE ARTISTIC OBJECT AS ΜΑΡΤΥΣ 

————— 
Greek, Roman, and Byzantine Studies 62 (2022) 328–340 

 
 
 
 

But why do I point out bones to dogs? Midas’ reeds bear witness 
to impertinent loquacity. 

This is a very close stylistic parallel, which proves that the 
phrasing was codified within the epigrammatic genre as for-
mulaic. Finally, let us consider the following, from an epitaph 
(IG II2 13165.1–2 = GVI 1058): 

Εἰµὶ µὲν ἐκ Πιρέως, ἱερὰ δέ µε ἐδέξατο Ἐλευσείν 
   ζήσασαν σεµνῶς· µάρτυρές εἰσι θεαί. 
I come from Piraeus, but the sacred land of Eleusis received me 
after I lived in an honorable way: the goddesses bear witness to 
this. 

It is a funerary inscription from Eleusis for a woman who had 
lived σεµνῶς, as the goddesses testify. It dates to the third/fourth 
century A.D.  

From the evidence thus far we may conclude that µάρτυρές εἰσι 
πόνων likely rests on an original formula µάρτυρές εἰσι, which the 
three instances reviewed above prove to be a codified 
epigrammatic expression. Of these, the epigram from Lemnos 
might be its earliest attestation. Moreover, I believe that the 
Eleusinian epitaph serves as a link that clarifies, best exemplifies, 
and marks as completed the transition from the conventional 
phrasing µάρτυρές εἰσι to µάρτυρές εἰσι πόνων, the standard form 
in late-antique honorific epigrams. It also illuminates the general 
development of the theme of the statue as witness.  

Of course one cannot rule out accidents of transmission: there 
may have been earlier attestations or other contemporary epi-
grams, now lost, that feature deities as “witnesses” with the same 
or similar language. Nonetheless, given its parallel structure, 
meaning, and metrical placement, it seems certain that µάρτυρές 
εἰσι is the Ur-formula and model for the late-antique µάρτυρές 
εἰσι πόνων. At some point along the way, probably before the 
late-antique period or even the high imperial age,12 the asso-
 

12 Since the parallel of GVI 1937 = IGUR 1328 (second or third cent.) 
provided by Cameron already shows a variation of the formula µάρτυρές εἰσι 
πόνων. 



 ARIANNA GULLO 339 

————— 
Greek, Roman, and Byzantine Studies 62 (2022) 328–340 

 
 
 
 

ciation of “gods as witnesses” with the formula µάρτυρές εἰσι was 
expanded into the iunctura µάρτυρές εἰσι πόνων, driven by the 
growth and expansion of the theme of the work of art as µάρτυς 
(especially statues).  

It is hardly surprising that a god should be conventionally 
called upon as a universal witness.13 And the depiction of gods 
as µάρτυρες in the context of µάρτυρές εἰσι πόνων appears fully 
developed in the second or third century, as shown by the first 
couplet of the Roman inscription singled out by Cameron and 
quoted above (GVI 1937 = IGUR 1328), which offers a slight 
variant with demigods as witnesses (Theseus and the Aeacidae). 
Artistic embodiments of deities, whether statues, reliefs, or 
paintings, may play the same role in a representative mode. The 
shift of meaning is easily explained: in votive epigrams gods are 
witnesses to the dedicatee’s gesture, and their statues are most 
intimately proximate to them. Thus, statues can readily bear 
witness to the accomplishments of the honorand in laudatory 
epigrams.  

To conclude, the formula µάρτυρές εἰσι was already attested in 
the classical age, specifically in connection with deities as wit-
 

13 The Sun, whether as Helios or Apollo, had been portrayed as the uni-
versal witness since Homer: Il. 3.277–280; Aesch. PV 91, Eum. 576, 594, 609; 
Soph. Aj. 845–849; Nonnus Dion. 17.283–284; Antip. Thess. Anth.Pal. 7.367.4 
= Gow-Page, GP LXIII 416 Ἥλιε. See also Aesch. Ag. 632–633, where the 
herald believes that the Sun is the only one who knows whether Menelaus is 
still alive or dead; Alph. Anth.Pal. 7.237.2 = GP VI 3543, where Apollo is 
presented as µάρτυς; GVI 1498.1 = CEG 623, epitaph for Pantaleon (Piraeus, 
fourth cent. B.C.): the Sun and the parents of the deceased are called upon as 
witnesses to the kindness of the untimely dead. The guilty flee the Sun: Soph. 
El. 1493–1494; Eur. HF 1231–1232. In general, the sun and the stars observe 
human events: Anon. Anth.Pal. 7.357.2 on Dike’s all-seeing eye, and Crin. 
Anth.Pal. 7.633.3 = GP XVIII 1869 = 18.3 Ypsilanti, for the moon which 
“sees” (cf. M. Ypsilanti, The Epigrams of Crinagoras of Mytilene [Oxford 2018] 
203). For the Hellenistic commonplace that stars observe mortal loves see J. 
D. Reed, Bion of Smyrna (Cambridge 1997) 175. Gods are also called upon in 
oaths as witnesses to ensure that the oath-taker will do as he promises. The 
inscriptional µάρτυρες, on the other hand, are witnesses for deeds done. 



340 THE ARTISTIC OBJECT AS ΜΑΡΤΥΣ 

————— 
Greek, Roman, and Byzantine Studies 62 (2022) 328–340 

 
 
 
 

nesses. Since the “witness theme” was so common in epigrams, 
one may reasonably infer that µάρτυρές εἰσι, in association with 
a noun in the genitive case, developed into the standard official 
poetic iunctura µάρτυρές εἰσι πόνων. This formula marks a well-
defined subcategory of late-antique honorific inscriptions for 
public monuments which often feature artistic elements (possibly 
illustrated) acting as µάρτυρες. This process must have been 
facilitated by the theme of the artwork as µάρτυς, based on the 
belief that statues, in particular, as visible embodiments of the 
gods, could easily bear witness on their behalf—especially if they 
should be of a deity who, like Eros, does “not desire a living 
witness.”14 

 
May, 2022 School of History, Classics  

    and Archaeology 
 Newcastle University 
 arianna.gullo@newcastle.ac.uk 

 
 

 
14 Cf. Phld. Anth.Pal. 5.4.3–4 = Gow-Page, GP I 3162–3163 = 7.3–4 Sider 

µαρτυρίην γὰρ Ἔρως µόνος οὐκ ἐφίλησεν / ἔµπνουν.