Emperor Leo III the Isaurian, page 26
Leo also had numerous Armenians around him too, such as David Hypatos, future bishop Stephen Siwnik and possibly Artabasdos. There is some evidence of seventh-century ascetic Armenians following a doctrine that could be iconoclastic, while also venerating the cross,73 a combination that would come to reflect Roman iconoclasm. Another prominent Armenian sect of this time were the Paulicians, who reputedly rejected various aspects of Christian worship – baptism by water, eucharist, cross veneration and icons. Much like the Jews, it is difficult to attribute much or any Paulician influence on Leo III as he is recorded ordering the Paulician heresiarch Gegnesius to venerate the cross, with the very idea of a rejection of cross veneration being against Roman iconoclasm (not to mention the rejection of other vital aspects of orthodox Christianity). The later iconoclastic focus on the eucharist as the only true image may have stemmed in part from the Paulician belief in the words of the Logos being the only true icons.74 However, arguments like this, along with lack of Paulician persecution under Leo III and Constantine V and the presence of Paulician soldiers in the Isaurian army, ring hollow as evidence of a Paulician influence over the Isaurian emperors.
While the possible influences on Leo III focus on non-Christian and heretical sources, more general Christian influences should not be forgotten. It is difficult to find clear indications of Leo’s thoughts as the historical record has been purged of information claimed as iconoclastic. One source might be the introduction to the Isaurian law code, the Ekloga. It may present Leo’s sentiments not only on his law-giving mission, but in general. The Leo of the Ekloga is ‘serenely convinced of divine guidance in his task, who sees no need for an ecclesiastical hierarchy to interpret for him the divine directives.’75 There is almost some quasi-caesaropapism in such an outline.76 And such was the biblical simplicity that Leo seemed to follow that it could be that he took a straightforward view of icons as being idolatrous images.
Such straightforward thinking could also affect Leo’s treatment of icons should those icons be connected to any who might have thought to ‘usurp’ any of his prerogatives.77 This could be seen in the growing role of some icons as a focus of local patriotism as a substitute for the local ‘holy man’. The contraction of Roman imperial authority in its own lands78 saw the influence of these holy men grow, along with the icons that served as their stand-ins. Roman authorities felt the need to break this power. Perhaps this was ‘monachomachy’ rather than iconomachy;79 less a war on icons and more a war on monastic holy men, with icons a secondary target. However, the anti-monastic movement was more a policy of Constantine V, making it unlikely to be a driving force behind any iconoclasm under Leo III. It could also be that this anti-monasticism was separate from iconoclasm itself, particularly as there were iconoclastic monks.
In his iconoclasm, might Leo have been attempting to reach out to populations and groups in Asia Minor, specifically the bishops and the thematic armies?80 Evidence of contact between Leo and the Asian bishops is lacking, and said Asian bishops, even Monophysites, were not necessarily iconoclasts. This idea likely exaggerates the influence of iconoclastic bishops in Asia Minor,81 while the thematic armies of Leo III were lacking interest in the entire discussion of icons. It was not until the reign of Constantine V that iconodule/iconoclast became more of a dividing line amongst the soldiery.82 And even then, there would be a lack of unanimity regarding icons amongst the themes, with the example of Artabasdos relying on the Opsikon and Armeniac themes, both heavily iconoclastic, but then also restoring icons. This may add more regional rivalry to iconoclasm, and possibly even an aristocratic element to the dispute.83
The arguments regarding influences on Leo and his potential iconoclasm are lacking clarity on what the actual purpose of it was. There are plenty of potential answers, if not exactly consensus, on why iconoclasm happened – ‘an undercurrent of hostility towards religious images [had been] present for several centuries,’84 – while the answer to ‘why now’ seems slightly more settled. The correct observance of religious practice affecting military and political success had been part of Roman life for centuries, causing significant religious tension and even bouts of persecution. It was no different under Christian emperors. Defeat was a punishment, and while there had been some suggestions of the Muslim tide receding in the last quarter of the seventh century, the second Muslim wave of conquest shattered this illusion – what grave sin had caused this repeated ruin of imperial territory? Some decided that it was idolatrous veneration of icons. But while opposition to image-worship may have been galvanising in certain parts of the eastern Mediterranean, reverence for and worship of icons continued to spread across much of the Christian world, with images of Christ, Mary, saints, and Biblical scenes decorating churches in the form of mosaics, frescoes, carvings, and illuminated manuscripts. The belief that these were miraculous conduits to those depicted to act as intercessors with God was going to be hard to shake.
A Volcanic Spark?
Having looked at the possible reasons ‘why’ for iconoclasm, we must now look at ‘when’ Leo’s iconoclasm actually occurred, if at all. If Nikephoros is followed, which is tempting but difficult given his abbreviated style, the iconoclasm of Leo III had a literal volcanic spark. In the Aegean Sea, near the islands of Thera and Therasia, there was an eruption of ‘smoky steam’, followed by a fire burst that ejected so much pumice that it created a new island and saw pieces float as far as Abydos. Reputedly, the water became so hot that people could not touch it.85 There seems to have been a long precursor to such seismic and volatile activity, such as the ‘fiery hail’ and boiling sea that afflicted the Arab fleet in 718.86 Nikephoros would have it that Leo saw these volcanic eruptions as ‘signs of divine wrath’87 and responded to it by planning to remove holy icons, as he thought that God was telling him to stop their idolatrous adoration. Nikephoros does little to posit a year for these volcanic and iconoclastic eruptions, merely stating that they occurred ‘in those days’ and ‘during the summer season.’88 He has them take place after the coronation of Constantine V in Easter 720 and before the rebellion of the Hellenes and Cyclades (which he claims was due to iconoclasm), but his style of writing does not give any hint of how long after that coronation or before the rebellion the Theran eruption took place. Such chronological issues will be a consistent problem when looking at almost any part of the outbreak of iconoclasm.
Theophanes records this volcanic eruption in similar terms – ‘a vapour from a fiery furnace’, with pumice floating throughout the Aegean and a new island being formed89 – and places it in the summer of 726. He also connects this eruption to Leo’s iconoclasm, but does not have it as the specific spark in his chronological order. That is because by his entry of AM6218, Theophanes has already addressed the iconoclasm of Leo III on at least two occasions – AM6215 (722/723), in connecting it to the iconoclastic edict of Yazid II, where he also mentioned the influence of ‘boorish’ Besr, and in AM6217 (724/725) where Theophanes has Leo making ‘pronouncements’ against the holy and venerable icons. Such was the extent of these pronouncements a year before the supposed Nikephoran spark of iconoclasm, that Theophanes has Pope Gregory II withholding the taxes of ‘Italy and of Rome’ and sending the emperor a doctrinal letter stating that ‘it was not proper for the Emperor to issue a command concerning the faith or to make innovations in the ancient doctrines of the church, which had been established by the holy fathers.’90 Theophanes would have the Theran eruption see Leo, encouraged by Besr, double down on his iconoclastic policies rather than see the eruption as a warning to cease his ‘ruthless war on the holy and venerable icons.’91
Due to his epitomising, Nikephoros does give the more straightforward account, while Theophanes is a little more muddled; however, it is unlikely that Leo III suddenly jumped into a policy of iconoclasm due to a single volcanic event as the former suggests, and we have already laid out a list of potential influences to make Leo an iconoclast. It could even be that the reason Theophanes has information regarding Leo’s iconoclastic ideas from before the Theran eruption is because Leo had been trying to persuade his imperial subjects through peaceful discussion in the years prior to his turning to edicts, a notion that might explain why he tolerated the iconodule Germanos as patriarch for so long.92 However, this may be giving too much credence to Theophanes’ chronology. He was not above making errors in his own records. Regarding the two pre-Theran claims of Leonid iconoclasm, connecting Leo to Yazid’s iconoclasm is fraught with problems, with the AM6217 entry containing substantial issues, not necessarily of Theophanes’ making, but still damaging to the account. Firstly, and more importantly, the word used by Theophanes for Leo’s ‘pronouncements’ regarding the removal of icons – λόγον – is problematic, with ‘a bewildering multiplicity of possible meanings.’93 This can be seen in the Oxford Classical Greek Dictionary, which for λόγος lists…
‘saying, speaking, speech, mode of speaking; eloquence, discourse; conversation, talk; word, expression; assertion; principle, maxim; proverb; oracle; promise; order; command; proposal; condition, agreement; stipulation, decision; pretext; fable, news, story, report, legend; prose-writing, history, book, essay, oration; affair, incident; thought, reason, reckoning, computation, reflection, deliberation, account, consideration, opinion; cause, end; argument, demonstration; meaning, value; proportion; New Testament the Word.’
There is a colossal difference between an authoritative, imperial ‘order’ to remove icons and the initiating of a discussion regarding them, but both have received support.94 Is Theophanes claiming that Leo initiated iconoclasm at this point in 724/725? Nikephoros would seem to have it after the summer of 726 (without actually telling us), while the Vita Stephani iunioris,95 the most detailed hagiographic account of iconoclasm, places the outbreak of iconoclasm in the tenth year of Leo’s reign – 726 in inclusive counting. It is also important to note that if there was an imperial ‘order’, no text has survived.
There are also factual issues highlighted in Theophanes’ AM6217 entry. He conflates popes Gregory II (715–731) and Gregory III (731–741) into a single pope of that name with a reign of 725–734. And the Liber Pontificalis indicates that the taxation issue between the papacy and the empire was already happening before any ‘order/discussion’ on icons was initiated by Leo III in 724/725. Indeed, it does not state that iconoclasm was partly a result of the economic falling out between Gregory II and Leo III – instead, imperial authorities reputedly responded to Gregory’s withholding of tax revenue by trying to remove the pope and even kill him.96 There is also significant discussion about the existence of Gregory’s doctrinal letter to Leo. It may even be in reference to two late-eighth/early-ninth century letters of dubious authenticity. And even if they are genuine or record some actual contemporary papal contact with Constantinople, their content gives allusions to the Lombard capture of Ravenna, which was not until 732 or definitively in 751, undermining the chronological position of this Gregorian ‘doctrinal’ letter.97 This gives little concrete in chronological terms, only to further prove that we face considerable trouble in determining when, or even if, Leo’s iconoclasm was initiated.
Instances of Icon Breaking: The Chalke Gate and Walls of Nicaea
On this already shaking historiographical ground, we move on to what is considered the first open act of Leonid iconoclasm: the removal/destruction of the icon of the Chalke Gate. This image – depicting the Lord above an entrance to the imperial palace – would seem to be of considerable prominence; however, even in iconodule writings, it was not accorded ‘any outstanding significance.’98 This could demonstrate that at least the Chalke Gate icon and maybe icons in general were not as widely revered as iconodules would later suggest; at least not until its removal. There are several versions of the story of this initiative iconoclastic act, each with their own additional/alternative aspects. Theophanes has the ‘Lord’s image’ that hung above the Bronze/Chalke Gate taken down by men of Leo in the immediate aftermath of the Theran eruption.99 Already roused in opposition to the ‘new-fangled doctrines’100 of the emperor regarding icons, the populace of Constantinople attacked and killed some of these men. Leo responded with ‘mutilation, lashes, exile and fines’, creating the first iconodule martyrs. Theophanes also suggests that in his iconoclastic anger, Leo targeted many of noble birth and high learning, destroying many of the schools and education systems that had been around since the Christianising of the empire.
Such a pronouncement of a Leonid destruction of Roman education at this point is specific to Theophanes. The iconoclastic age was not a great one for art, but some traditions continued, and some influences were absorbed from abroad, giving the era its own distinct flavour. There was a parallel decline in some expertise, but this was not necessarily due to iconoclasm itself, as many of the richer cities of the empire were already showing decline by the late-sixth century, followed by the upheavals of the seventh. This decline was long claimed to be in marked contrast to the artistic boom that encapsulated the ‘Carolingian Renaissance’ in the Frankish Empire, but this is incorrect due to iconodule sources presenting the ‘Byzantine Renaissance’ as beginning much later than it did. The green shoots of Roman recovery were Isaurian in date and origin. That said, the iconoclastic period did see a significant change in the basic education programme. While philosophy, rhetoric, mathematics and their many constituents remained part of Roman education by the late-eighth century, the core changed from classical elements like Homer to biblical texts.
A more detailed story of the Chalke Gate incident comes in the first letter of Gregory II to Leo III, although it does not appear to be a contemporary text, either a later interpolation or a complete fabrication.101 It records Leo sending a spatharocandidatus called Julian (some manuscripts call him Jovinus) to the Chalke Gate to destroy the image of the Saviour, which had been the site of many miracles. When he arrived there, Julian was met by a group of zealous women, who begged him not to carry out his task. The spatharocandidatus did not listen, climbing a ladder and starting to hack away at the icon with an axe. Upon seeing this, the women pulled down the ladder and beat Julian to death. Gregory bemoans how Leo sent his soldiers to punish and kill many of these women. ‘The specific details of this account … appear at first sight to be due to an eye-witness,’102 but it is instead problematic. Not only does it entail exaggeration, the text misidentifies the location and therefore the actual icon targeted – it speaks of the Chalkoprateia, the name of a quarter of Constantinople and a church. This could be a mistake based on the similar names, but other comments suggest that the identification of the Chalkoprateia was intended and understood, detailing the account from a Constantinopolitan origin, who would not have made such a mistake. Such is the problem that even those who argue in favour of the authenticity of these Gregorian letters find the whole ‘Chalkoprateia’ incident to be so problematic that they posit the whole section as a later interpolation into an authentic contemporary source.
The Vita Stephani iunioris has Leo III order the Chalke icon be taken down and burned. It repeats the pious women pulling down the ladder and killing the spatharius charged with the deed, but adds that they made for the patriarchal palace where they threw stones at the patriarch. The Vita also provides an extra piece of information regarding the nature of the Chalke icon. There has been argument over what exactly the icon above the Chalke Gate was made from. It could have been a stone statue, more easily connected to pagan practice and therefore explains why Germanos did not strenuously attack Leo’s actions in 726/727. In such a notion, the dispute between the emperor and patriarch only came later when iconoclasm was escalated in 730 with paintings being targets. However, that Leo would target the Chalke icon for destruction through burning increases the likelihood that it was a wooden painted panel rather than a stone statue.
Other hagiographic sources provide slightly different forms of this Chalke incident, specifically what are considered the ‘Marian’ and ‘Theodosian’ versions. The ‘Marian’ version stems from an anonymous Passio, also called ‘The Acts of the Ten Martyrs’, which belongs to the second half of the ninth century and relies heavily on Theophanes and the Vita Stephani iunioris, following the latter in terms of date. It also places Besr at the head of the group of soldiers undertaking the destruction. A group, led by a patrician woman of imperial lineage called Mary, but also containing ten others – Gregory the protospatharius, Julian, Marcian, John, James, Alexius, Demetrius, Leontius, Photius and Peter – attacked and killed a spatharius. The incensed Leo sent 500 soldiers against the people, killing many. The group were thrown in prison for eight months, where they were punished daily. After that, they were brought before the emperor, who tried to make them renounce their support for the icons. When they refused, Leo had their faces branded with hot irons, before they were beheaded at the Kynegion, their bodies thrown into the ta Pelagiou, the burial place for criminals on 9 August. Their bodies would later be moved to the Church of St Demetrius at the Aninas monastery. The ‘Theodosian’ version first appeared in the Menologium of Basil II of c.1000 and replaces Mary with Theodosia the nun. Theodosia was executed with a ram’s horn, a peculiar method. Showing some naivety, the Menologium has Theodosia live during the reign of Constantine V, despite having been martyred under Leo III. This is not completely incorrect as Constantine was co-ruler with his father from 720.
Another aside from the Chalke incident is the suggestion that Leo replaced the Christ icon with a plain cross, using an epigram of Theodore the Studite. This epigram could refer to Leo V and his son Symbatios-Constantine,103 although it would not be surprising for Isaurian emperors to mount a cross above the Chalke Gate due to the prominence they gave to the cross in their iconoclastic policies. It was also claimed that the Chalke Gate image was not destroyed, making its way to the papal collection in the Lateran. While this has been refuted as a myth,104 it shows that regardless of whether it happened or not, the incident at the Chalke Gate and the icon itself had taken on significant importance for iconodule opposition to Isaurian icon policies.
