Theodosius Arrives at Ephesus

I recently found this old paper from my senior year of college sitting in my drafts folder — a very cut-and-dry analysis of a particular Medieval stained glass window at the Cloisters Museum in New York.  I like it not because it’s particularly interesting, but because I recall writing it in one night, feeling as though I’d mastered the academic essay form: boy did I have a long way to go. (ELP, 2015)

As with many objects surviving from medieval times, the Theodosius Arriving at Ephesus stained glass panel (Fig. 1, Cloisters Museum:1980.263.4) has a history that is nearly as interesting as the artifact itself. With so many pieces lost over the centuries, and so many pieces added, there is an imaginative gap that must be leapt if we are to understand this artifact within the setting and context of the world and people that created it. Beginning with an analysis of the Theodosius panel in its current state, I will move back in time to imagine what this piece was like in its original condition as part of a greater stained glass window, and discuss art historian Michael Cothren’s hypothesis regarding the reasoning behind the choice of this Christian myth for the Cathedral of Rouen.

Fig. 1: Theodosius Arriving at Ephesus, stained glass, 1200-1204, 25 x 28¼ inches, Cloisters Museum (1980.263.4).

The Theodosius panel is 25 x 28¼ inches, and has been dated with unusual precision to 1200-1204.1 In it, three figures are depicted riding horses towards a portal or door of some sort. The central figure, wearing a golden yellow crown with a brownish purple costume rides a that is somewhere between tan and pink in color. A man to the right of the crowned figure points to the architectural space, represented by a wall of alternating blue and yellow bricks alongside a door formed by thin, light green supports and a white threshold. He points at this door, and his horse has already pushed his nose through. The horse, the gesture of this man, and the gaze of the crowned figure all create a sense of movement that leads our eye rightward, towards the portal and beyond. If we step back from the Theodosius panel, and imagine it as part of a greater composition, we can see how this rightward motion moves our eye across to the adjacent panel of stained glass, in which the next scene of the story would be depicted. The three figures are layered atop each other very closely in a way that does not provide much illusion of depth, however the illusion of depth is not something the medieval master artisan was striving for, nor does stained glass as a medium lend itself to the representation of three-dimensional spaces.

For the purpose of depth, the nose of the hose poking through the portal is enough to indicate the narrative motion of the story. Rather than depth, the strength of stained glass lies in color and radiance. The majority of this panel is occupied by vibrant blues, deep reds, and golden yellows. A number of secondary colors illuminate smaller areas: green (on cloak of the leftmost rider, on the vertical elements of the portal, and on the trappings of the king’s horse), light blue (for the middle horse, and the brickwork); white (on the backmost horse, and several of the architectural elements; and flesh tones, which vary slightly in the case of each figure.

Perhaps the most baffling aspect of the composition of this image is the surrounding border, which seems at first glance a bit of a hodgepodge. On either side of the panel there is a vertical line of white dots, like a string of pearls; inside of that is a framing line of blue and red rectangles, regularly interrupted by golden yellow images of castles and fleurs-de-lys.

Fig. 2: Image showing the different parts of the Theodosius panel that are not original. Image scanned from the Metropolitan Corpus Vitrearum, vol.1, p. 101.
The bottom framing element is strangely curved, and it is here that we must transition into a discussion of the provenance, as large portions of the Theodosius have been replaced, touched up, or downright fabricated (Fig. 2). The original stained glass window was circular, (consistent with the shape of the bottom framing element) and was part of a much larger stained glass window with numerous panels depicting scenes from the story of the Seven Sleepers. In 1270, a mere sixty odd years after the window was finished, reconstruction of the Rouen Cathedral called for new altars to be built in place of this stained glass. Rather than trash the window, prudence and economy compelled the church to carefully dismantle and refinish the circular panels into rectangular panels so they could fill the lancets of the new chapels. This explains the abruptness of the fleurs-de-lys and castle border. According to Cothren, the pearls on the outmost edge are entirely a fabrication. They were added by some 20th century dealer in an effort to better balance out the somewhat forced and disjointed border.

Fig. 3: Cothren’s hypothesis as to the full composition of the Theodosius narrative within the stained glass window. 17 is where the Theodosius in question was probably located.

The pearls are thus mees the story further alonant to make the stained glass panel more appealing to the sensibilities of 20th century collectors, making the artifact both easier to handle and easier to sell. Having removed these later additions, it is possible to imagine the Theodosius panel as it existed as one-fourth a medallion within a tall Gothic window (Fig. 3). Each individual panel thus progressg from beginning to conclusion. In order to place the Theodosius panel properly within the framework of this dismantled window, we must first look at the story it is depicting.

The Golden Legend tells us the story of the seven sleepers, who lived in Ephesus during the rule of the Roman Emperor Decius. Seven brothers (named Maximianus, Malchus, Marcianus, Dionysius, Johannes, Serapion, and Constantinus, so the story tells us) fled to a cave in nearby Mount Celion, where they hid to avoid the persecution Decius ordered against the Christians. Decius found their cave, but could not find them within, and instead decided that a proper punishment would be to wall up the entrance, sealing their martyrdom. Unbeknownst to the pagan King, God put the seven Christians into a trance that lasted 372 years, well into the reign of the Christian King Theodosius. Theodosius was having political trouble as theologians and scholars debated the veracity of the miracle of resurrection, a tenant of Christianity that, if proven false, would significantly diminish the power of Jesus in the minds of believers, and undermine the religion. God in his wisdom compelled a sheepherder to open the cave, dispelling their divine hibernation. The seven Rip van Winkles awake at last, feeling as though they have only slept a night, and decide to send one of their number (Malchus) out to buy food in the city. Malchus cannot understand what has happened to Ephesus, how it has grown, why there is a cross over the gate of the city, why indeed, no one he knows is around, and his money is suspiciously refused. The Ephesian citizens capture Malchus, thinking he has found some hoard of ancient treasure, and take the pleading man on a leash to the bishop and consul, who in turn send word to King Theodosius. Malchus recalls his six friends to explain their situation to the King, and they present a scroll that they had written when the cave was initially sealed, detailing their martyrdom. The miracle proven, the seven stand before Theodosius as living testaments to the veracity of the resurrection. The seven then return to sleep, or death, proving that a Christian have no need to fear death, as all Christians will be resurrected at the Second Coming just as the seven were before the eyes of Theodosius.
Narratively, the panel illustrates a moment near the end of the story, in which Theodosius is led to witness the miracle of the seven sleepers. The good king is accompanied by two unspecified men. A good guess is that the two represented are the bishop and consul who first vetted Malchus’ story. Cothren hypothesizes what the exact makeup of the whole window was, and assumes that, like many other medieval stained glass window narratives, the story was meant to be read from the bottom up. By piecing together the fragmented remainders of the window, Cothren was able to show that the window probably consisted of five medallions, each comprised of four parts (Fig. 3). Being at the end of the story, the Theodosius was likely part of the topmost medallion, and, given the shape of the curve at the bottom of the panel, must therefore have constituted its bottom left quadrant. The rest of the medallion is entirely lost, and can only be left to speculation. The motion rightward draws our eye to one such missing panel that would probably have depicted King Theodosius meeting the seven sleepers and witnessing the miracle. The top half of the medallion would have concluded the narrative, perhaps depicting the seven sleepers returned to their hibernation by the will of God.
The problem that has baffled art historians to this day is the question of why this particular Christian myth was chosen for illustration in the Cathedral of Rouen. It seems an odd myth particularly because of its relative obscurity. According to Cothren, the tale was popular in Byzantine works, however, no such recreation of the Seven Sleepers story has ever compared to this stained glass window in scale or prominence. Art historian Jean Lafond proposed that the subject was probably chosen simply as a byproduct of the fact that Byzantine art was popular in Normandy. However a close reading of the history of Rouen around the turn of the 13th century indicates that the window might very well have had a political inspiration.
It is not common that anything from medieval times can be dated so precisely as this window is (1202-1204), but a number of historical contingencies make such a specific attribution possible. Edward the Confessor ruled as King of England from 1042-1066. While his career was a turbulent one, the most relevant aspect of reign was a prophetic dream he had, in which he saw the seven sleepers turning over in their graves. After his canonization a century later (1161), Edward’s vision became associated with his sainthood, and a symbol of the divine right of the English royal family. When Richard the Lionhearted died in 1199, his brother King John (ruled 1199-1216) took over as King of England and Duke of Normandy. This rule would last a mere five years, at which point the French conquered Normandy; however, five years was time enough to commission the stained glass window at Rouen Cathedral. After a fire destroyed most of Rouen in 1200, King John pledged to finance the reconstruction and repair of the Cathedral, which held many of his friends and relatives. For John, the choice of the Seven Sleepers legend fit perfectly. Personally, he understood the tombs of his relatives and friends within Rouen as modern day sleepers, waiting for the Second Coming. Politically, it allowed him to legitimize his authority as the new King by installing a program of artwork that associated the prime cathedral in Normandy with the English monarchy through the symbols of Edward the Confessor. Further, by funding such a program, he corroborated his own connection to the royal bloodline by paying homage to his royal predecessors. Unfortunately for John, the French successfully drove the English out of Normandy in 1204, by which date the Theodosius must have already been completed.
Perhaps the French overlooked the symbolic significance of the Theodosius windows in Rouen, and therefore didn’t feel a need to destroy or replace them. Or perhaps the French felt that the story was still fitting. If we associate Theodosius with the French monarchy and Decius with the English monarchy, then Normandy could be understood as a modern Ephesus, plagued by the rule of the British. Having slept through the rule of the British, the French peasantry (the true believers) awoke again in 1204 to find the city free from English control and restored to the power of the true French king—a miracle! The unpleasant period during which the English ruled is thus subverted when the story of the Seven Sleepers is reimagined in this way, and becomes merely an unfortunate period of hibernation during which the English (barbarians, pagans) ran amuck.


The Theodosius panel stands by itself as a masterpiece of stained glass. The vibrant colors, the composition, the control of line and movement, all carefully planned and executed, is beautiful and valuable in and of itself. Yet it is the history that brings it to life: the chance fires and shifts in political power that led to its creation; the luck that allowed it to survive political shift to French rule despite its English agenda; the reworking and reformation of it into new windows; and finally, the years of being passed among different owners and dealers. All this provenance is obscured from the casual viewer sauntering through the Cloisters Museum, yet the evidence of it is present. It is this history that adds dimension to what otherwise would simply be a well worked piece of artisanship severed from its original body, removed from its setting and divorced from its significant political and historical import.


1 Michael W. Cothren, “The Seven Sleepers and the Seven Kneelers: Prolegomena to a Study of the “Belles Verrieres” of the Cathedral of Rouen,” Gesta, Vol. 25, No. 2, p. 209. (International Center of Medieval Art, 1986) .
The Metropolitan Corpus Vitrearum more liberally dates the Theodosius between 1200-1210.

Jane Hayward, Mary B. Shepard, and Cynthia Clark, English and French Medieval Stained Glass in the Collection of the Metropolitan Museum of Art. Corpus vitrearum, vol. I, (New York, N.Y: Harvey Miller Publishers and the Metropolitan Museum of Art for American Corpus Vitrearum, Inc, 2003) p. 101.

1 Eugene-Emmanuel Violett-le-Duc, trans. Francis P. Smith, Mediaeval Stained Glass, (Atlanta: Lullwater Press, 1946) p. 13.
“An opaque painting rightfully should define various planes, and even when it consists of a single figure against a flat background, the artist attempts to indicate three dimensions in his figure. If the early painters did not always attain this result, it was nevertheless their goal. To transpose this quality of the opaque painting to a translucent medium, is wrong. The true ideal of translucent painting is that the design should express as energetically as possible a beautiful harmony of colors, which, when realized, is all that can be desired.”

1 Cothren, 203.
2 Cothren, 205.
3 I recognize that The Golden Legend was written well after the creation of this stained glass panel. Nonetheless, it serves as the closest authority on early Christian mythology for the purposes of this paper.
4 Paraphrased from The Golden Legend.
5 Cothren, 208.
6 To this day, Edward is the official Saint of royalty.
7 Qtd in Cothren, 222: “quod fratrum et amicorum nostrorum sepultura nobis venerabilem in perpetuum
commendat.” trans., “the Cathedral holds in venerable perpetuity the graves of our brothers and our friends.”
8 We see a similar tactic employed in the 18th centruy by Cathrine II in Saint Petersburg, whose enormous artistic propaganda campaigns aimed to prove her bloodline was directly Peter the Great, though not a drop of it flowed through her German veins. The Bronze Horseman is perhaps the best example.
9 Cothren, 209-212.


Cothren, Michael W. “The Seven Sleepers and the Seven Kneelers: Prolegomena to a Study of the “Belles Verrieres” of the Cathedral of Rouen,” Gesta, Vol. 25, No. 2, p.203-226. International Center of Medieval Art, 1986: .
Hayward, Jane, Mary B. Shepard, and Cynthia Clark. English and French Medieval Stained Glass in the Collection of the Metropolitan Museum of Art. Corpus vitrearum, 1. New York, N.Y: Harvey Miller Publishers and the Metropolitan Museum of Art for American Corpus Vitrearum, Inc, 2003.
Hayward, Jane, and Walter Cahn. Radiance and Reflection: Medieval Art from the Raymond Pitcairn Collection. New York: Metropolitan Museum of Art, 1982.
Jacobus de Voragine, trans. William G. Ryan. The Golden Legend: Readings on the Saints. Princeton, N.J: Princeton University Press, 1993.
Loisel, Armand, and Jean Lafond. La Cathédrale De Rouen. Paris: H. Laurens, 1927.
Viollet-le-Duc, Eugène-Emmanuel, trans. Francis P. Smith. Mediaeval Stained Glass. Atlanta: Printed at the Lullwater press, 1946.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s