Images of Byzantine Adolescents

Byzantine teen-agers are virtually invisible. Their bones are noted – though rarely discussed – in archaeological excavations of burial grounds, but they are hard to track in the written sources,1 and they have been ignored in discussions of Byzantine imagery: unlike women and children, no one has ever even looked for the portrayal of Byzantine adolescents in the visual evidence.2 Portraits of Byzantine teen-agers are, however, important to find and to evaluate, not just because no one has ever done this, but because they illuminate an area of Byzantine social history that has been neglected and is in itself significant. How Byzantine authors represented the transition and transformation of the Byzantine male or female across his or her lifetime is beginning to be studied, but how that transition and transformation was represented visually is even more important to understand, because images provide a data set produced by a different community than the mostly urban and mostly elite males who provide the bulk of textual evidence preserved from the Byzantine empire, and for a much broader constituency. We mostly hear the voices of a small and select group of Byzantines; images, in contrast, allow us to see with everyone in Byzantium who had not lost their sight. While we have not learned (and can never learn) to see in the same way as a Byzantine did, looking at portrayals of adolescents nonetheless adds valuable new layers of information, from different social and cultural perspectives, to the evidence about the Byzantine life course – how people, from ordinary peasants to the imperial family, experienced growing up – provided by words. The Byzantine life course itself has only recently received scholarly attention. In a PhD thesis awarded in 2012, Eve Davies looked at how the Byzantines described and valued infancy, youth, maturity and old age; and she compared and contrasted

the ways thatm ale and female experiences of childhood, adolescence, adulthood and olda ge were discussed.³ Her focus waso nt he archaeological and textual evidence for family roles across the life courses of both men and women from late antiquity until the thirteenth century.
Davies' key findingsa bout Byzantine adolescents mayb es ummarised as follows.⁴ Byzantine authors sawadolescencesomewhat differentlyt han we do, though then as now the potential emotional and physical eruptions of male adolescents weref requentlyn oted, decried, and (often) later forgivena sy outhful indiscretions. The idea of the 'teenager',h owever,i sn ot found; the Byzantines seem to have viewed 'youth' as beginning roughlya round the ageo f1 2o r1 3( for girls) or 14 or 15 (for boys) and ending,a tl east for males, around the ageo f25. As this indicates, the ways that Byzantine authorsdiscussed 'youth' weregendered. Normally, they described femaleyouth in terms of beauty,with marriagethe signal marker of transition from youth to maturity.M ale youth, in contrast, was described in terms of activity, physical strength, and socialisation. When the ageo fa dolescents is specificallyr ecorded (whichi sn ot common), there are no years that are especiallyp rominent for males; for females, however,virtually the onlya ge mentioned is 12,which was apparentlyt he symbolic turning point for females exual awakening: Mary of Egypt,f or example, was 12 when she rejected her familyt or un riot in Alexandria.⁵ In short,as eries of distinctive textual conventions cluster around descriptions of both male and femalea dolescence.
Davies was not able to look at how the visual evidence from Byzantium intersects with the writtena nd archaeologicalr ecord, and, as it happens, the distinctive nature of adolescencenoted in Byzantine written sources is onlypartially seen in Byzantine visual sources. How and whythis is the case, and what this tells us about the significanceofthe Byzantine experience of, and social response to,adolescence, are the subjects of this chapter.
The first thing to be said is thatt he Byzantine painters could and did clearlydistinguish between various stages of the life course trajectory,sometimes quite intentionally. The scene of Gregory'sburial in the copyofthe Homilies of Gregory of Nazianzos in Paris (ca 880), for example, surrounds the dead Gregory with figures representing the three ages of, in this case, man: abeardless youth lowers the saint'sfeetinto the sarcophagus; abearded middle-ageman lowers his head; and an elderlypriest,with grey hair and beard, presides ( fig.1).⁶ Though they are normallyl ess schematically  Paris,B NF,g r. 452r.F or discussion of this image,s ee L. Brubaker,Vision and meaningi nn inthcentury Byzantium: Image as exegesis in the Homilies of Gregory of Nazianzusi nP aris.C ambridge displayed, almost all other facets of the Byzantine life course also find visual expression. Infantsa ppear in representations of the Nativity,a sw ell as in other examples such as the votive images at Hagios Demetrios;⁷ small children are shown in many representations of the Crossingo ft he Red Sea,t he toddler Benjamin is depicted in some pictorial sequences of Joseph'slife, and ayoungboy (perhaps intended to represent the brother of Jesus) alsoo ccurs in the Kokkinobaphos manuscripts.⁸ Young men -for example Joseph before his sale to Potiphar -and youngw omen such as Jairos'sd aughter appear regularly,⁹ though less frequentlyt hanm ature adults of both sexes, who are visualisedw ith specific attributes:t he males bearded to signify maturity and the females veiled to signal matronlymodesty¹⁰.Old men are distinguishedb yt heirg rey hair and beards;o nlyo ld women do not appeari nt he Byzantine visual record.¹¹ But although painters could clearlyd istinguish between the different stages of the life course, they did not always do so: in an inth-century miniature illustrating the life of St Basil ( fig.2 ), for example, the child St Basil hiding in ac avew ith his parents is shown as an adult, though elsewherei nt his manuscript the painter portraysc hildren, and elsewhere on the same paget he miniaturist has been happyt o portray Basil as ay outh, an adult, and as an old man.¹² There are two keyp oints to takef rom this example. First,what we might see as an issue of verisimilitudethis is the images elected to represent Basil'sc hildhood and family, so he should 1999,136 -37, fig.46; on the 'threea ge system' moreg enerally, see E. Sears,The Ages of Man. Medieval Interpretations of the Life Cycle. Princeton NJ 1986.  Imageso ft he Nativity,which invariably show Christ as ab aby, often onced ressed in swaddling clothes and oncen aked, in the bath, areu biquitous;f or Hagios Demetrios, see, e. g., C. Hennessy, Iconic imageso fc hildren in the church of St Demetrios, Thessaloniki, in: Icon &w ord. The power of images in Byzantium, eds.A .E astmond -L. James.A ldershot 2003,1 56 -72, though Id isagree with some of her conclusions (L. Brubaker,E lites and patronagei ne arlyB yzantium: the evidence from Hagios Demetrios at Thessalonike, in: Elites old and new in the Byzantine and earlyI slamic Near East [TheB yzantine and early Islamic Near East 6], eds.J .H aldon -L. Conrad. Princeton NJ 2004,6 3-90).  Ford iscussion, see Meyer,A no bscurep ortrait,102 -5. Fort he Kokkinobaphos child see C. Hennessy,The stepmum and the servant: the stepson and the sacredvessel, in: Wonderful things:Byzantium through its art,e ds. A. Eastmond -L. James.F arnham 2013,7 9-98.  Both found, for example, in the Paris Gregory (Paris.gr.510,f f.69v, 143r): Brubaker,V ision and meaning, figs1 2, 19.  Forb eards, see I. Kalavrezou-Maxeiner,The Portraits of Basil Ii nP aris gr.5 10. JÖB 27 (1978) 19-24;for veiling,see Meyer,Anobscureportrait,120: 'According to Byzantine protocol, older or married womena re depicted with head coverings and maidens are not'.  See, e. g., representationsofthe Old Testament prophets Isaiah and Ezekiel: for ninth-century examples,B rubaker,Vision and meaning, figs11, 44.With the occasional exception of Mary of Egypt, old women weren ot represented until the lateB yzantine period, but older women were sometimes distinguished fromy ounger ones, as in exampleso ft he Nativity where the older midwife is helped by ayoungerassistant: see Meyer,Anobscureportrait,119 -20,218. Ithank Tony Eastmond for discussion on this point.  Paris,B NF,g r.510, f.104r:B rubaker,Vision and meaning1 37-41,f ig.1 7( as in n. 6).

Images of Byzantine Adolescents
be represented as achildwas not necessarilyrelevant to the Byzantines. The painter or patron chose whether or not to depict someone as achild, an adult, or aged; so (and this is the second keypoint)when this decision resulted in aportrait that seems to us to contradict the actual agethatweknow the person portrayed was, we should not assume, as scholars sometimes do, that the painter was inept,orthat the Byzantines sawchildren as miniature adults. We need instead to ask whythe painter made this choice. In this case, the child-adult Basil is probablyavisual reference to his mature behaviour even as achild (the puer senex trope), which is carefullyelucidated in the accompanying text,Gregory of Nazianzos' funeral oration to Basil.¹³ The important issue for us here, however,isthat depictions of individuals at anygiven stageof their life cycle represents ac hoice based on criteria other than 'naturalism':t hat there are images of old men but not of old women does not mean that no Byzantine women grew old,¹⁴ but that old men wereimportant to depict while old women were not.That social conventions, conscious or subconscious,underpinned this decision needs to be recognised, and we will return to this issue at alater stageinthis chapter.
As econd qualification that we need to consider is the impact of status on representations of Byzantine men and women. The presenceo ra bsenceo fabeard is akey distinguishing mark between adolescent and adultmales in Byzantine imagery, but it is also asign of status.¹⁵ In imperial portraits after the mid-seventh century,for example, the junior emperor is portrayed as beardless no matterwhat his age, in contrast to the bearded senior emperor.¹⁶ Similarly,the presenceorabsence of head covering can distinguish adolescent and adultfemales in Byzantine representations, but head covering tooisasign of status. Prostitutes and, often, female servants,whatever their actual age, are often depicted without veils.¹⁷ Because most generic depic-  PG 36,5 09,5 12 (Eng.trans. in NPNF 7,399). On the puer senex trope in Byzantium, see Davies, From wombt ot he tomb, 81-82.  See note 11 above.  Lack of facial hair maya lso, of course, signal that the male portrayedi saeunuch, but context usuallya llows us to aget he man appropriately: see, e. g., the eunuch patriarch Ignatios portrayed in aninth-century mosaic at Hagia Sophia, and the eunuch Leoshown gifting his bible to the Virgin  See Kalavrezou,P ortraits of Basil I; and notet hat on the coins minted during the period when the empress Eirene was in control of the empireh er son Constantine VI, though an adult,was portrayedwithout abeard(see note 48 below), as had been Heraklonas Constantine when Martina was evidentlyt he power behind the throne (see note 37 below). Similarly, very young co-emperors were often depicted as youths rather than the small children they actuallyw ere,aswith Constantine, toddler son of Michael VII Doukas,onthe Royal CrownofH ungary:E. tions of figures who appear to be Byzantine youths -men without beards;w omen with uncovered headsalso represent workers in the field or servants,m ale or female,¹⁸ it is hard to determine whether the beardless males shown (say) pruning vines, holding washing implements or running errands, and the femaleservants portrayed without veils, are actuallym eant to represent young men and women or are beardless and veil-less simplytoindicate theirlesser status. Iwill thereforefocus on portraits of identifiable adolescents for the remainder of this chapter,which, sadly, effectively obliges us to look at saints (like St Basil), aristocrats, and the imperial family. This means thatthe run of portraits we are about to surveyoverlaps to aconsiderable extent with the evidence of the written sources in thatthe images are urban productions associated, in manycases, with the court or with acourtlymilieu. Iwill, however,open out the data set at the end of the chapter,and draw some conclusions applicable to al argera nd less restricted viewing audience.
The following several paragraphs present agallery of portraits, arranged chronologicallybythe ageofthe youth portrayed. Even this brief survey illuminates the issues that portraits of Byzantine youth requireu st oc onfront.¹⁹ The frontispiece of the Paris Gregory (Paris, BNF gr.510,f .Br),d ated to ca 880, portrayst he (eventual) emperor Alexander at around age1 0( fig.3 ),²⁰ in ap ortrait quite distinct from his mosaic portrait in the north gallerya tH agia Sophiaw here, at age2 3o r2 4, he wears the beard of maturity as well as imperial vestments.²¹ A slightlyo lder male youth appears in am iniature in the copy of Dionysios the Areopagitei nt he Louvre, which depicts Manuel II Palaiologosa nd his wife Helena with their three children, the eldest of whom, John, was born in 1392, and was therefore11, 12 or 13 when the familyportrait miniature was painted between 1403 and 1405 (fig.  Iknow of no studydedicated to depictionsofByzantine male servants,but arepresentative samplingo fi mages mayb ef ound in: Everydayl ife in Byzantium, ed. D. Papanikola-Bakirtzi.A thens 2002.S ervants also often attend rulers in later Byzantine images (e. g. as ervant mayh old ab asin jugi ni mages of Pilate washingh is hands after the condemnation of Christ), and these men arev irtuallya lways shown without beards.  Ihavemade no attempt to cluster images in one century,orevenone period: there is not enough preserved material to do this effectively,a nd Ia mn ot suret hat portrayals of adolescents changed significantlybetween the 4 th and the 14 th century.Wewill return to this issue at the close of the chapter. Note toothat manyofm yexamples come from manuscripts.This is due simplytothe fact that in Byzantium manuscripts had agreaterrangeofimages than other media. Because of arguments about its date, Ih aveo mitted the Vatican epithalamion (Vat.gr.1851) from this overview,though we will return to its miniatures at the end of this chapter: on the manuscript,see most recentlyC.Hennessy,A child bride and her representation in the Vatican Epithalamion, cod.g r. Images of Byzantine Adolescents 4).²² Fora nother 12-year-old, we need onlylook at images of Christ in the temple,as for example in the Paris Gregory again (f.165r), which presents an extended version of this episode that portrays Christ three times(fig.5): once hurrying to Jerusalem, once preachinginthe synagogue, and finally reunited with his anxious parents.²³ Fora12year-old female, we maylook to images of the Virgin Mary when she is betrothed to Joseph in the Kokkinobaphos manuscripts( fig.6 ).²⁴ A1 3-(or possibly1 6 -)y ear-old Alexios is portrayedi namosaic in the south-west gallery of Hagia Sophia to the side of his parents John II Komnenos and Eirene ( fig.5in Hennessy article); Alexios was born in 1106,and wasnamed co-emperor in 1119,aged13;²⁵ my belief is thatthe mosaic commemoratesh is elevation, but the mosaic is usuallyd ated to 1122, in which caseh ei s1 6i nt his image.²⁶ The Paris Gregory -on the same pagea sw e have justs een Alexander at 10 -also portraysh is elder brother LeoV I, born in 866,ataround age14( fig.3). All of these youths in theirearlyteens are distinguished from both younger children and adults by their height (with children shortera nd adults tallert han the youths portrayed) and theirs light builds, which usually contrast with the slightlyc hubbier children and the somewhat stockier adults.
Aside from the Virgin, the examples we have seen thus far have all been male, but the Lincoln Collegetypikon (Oxford, Bodleian Library,Lincoln College, f. 11r), in aminiature dated around the year 1300,introduces another female, with Theodoule, the foundress of the monastery for which this is the typikon, leading her daughter Euphrosyne (born 1285/6), who took over the monasteryather mother'sdeath.²⁷ Euphrosyne is aged about 16 in this portrait ( fig.7). Forthe end of the teen-agedecade, the Paris Gregory,s ourceo fm anyofo ur examples, depicts St Basil as ay outh aged 18 to 20 with his good friend Gregory studying in Athens ( fig.2). The faces are abraded, but the figure on the left has clearlygrown abeard to indicate his increased maturity;i tc ontrasts with the grey beard of the wise old philosopher who teaches the pair.The significanceo ft he beard as as ign of mature authority mayb eu nderlined by the Trier ivory,which has recentlyb een interpreted as ad epiction of the translation of relics to, and dedicationof, Hagia Euphemia in 796 by Constantine VI and his mother Eirene:²⁸ if this is correct,i tp ortraysC onstantine VI (born 771) beardless at 25.T his would be unusual,b ut givent he relationship between Constantine and his mother,a nd her apparent attempts to share the throne -which she would succeed in taking ayear later after Constantine wasblinded -it is perhapsn ot surprising.Certainlywesee the samesuppression of Constantine'sauthority on the coin issues of the same year,which suggests thatE irene and her supporters had control of the mints.²⁹ Coins are, in fact,i nteresting for our purposes, for three reasons.F irst,c oinsand particularlyl ow denomination coins -often circulated widely. They weret hus viewed, or had the potential to be viewed, by aw ide swath of the Byzantine populace, and had legal standing, so anyr epresentations on them had specific requirements: images of the emperor(s) validated the worth of the coinage.Thought he Byzantines never tell us this, the function of coinage requires their ornamentation to be sufficientlyc onservative thatw hen (and if)p eople actuallyl ookedc loselya tt heir money they would not distrust the coins.Weare probablysafe to assume,therefore, that the images we find on coins would be broadlyacceptable to the Byzantine populace, and represent conventional images. In our case, thatmeans that when we see adolescents represented on coins,they are likelyt oshow us portraits of youths that the Byzantines accepted as suitablea nd appropriate.
Coins are also interesting to us here because they could be used to conveyp articular messagesf rom the emperoro re mpress to his or her people. Whether or not those people paid anyattention is amootpoint;all the same, we mayregard images of adolescents on Byzantine coins as representing what was thoughtt ob eagood public face of imperial youth.
The third reason coins provide interesting evidence for understanding Byzantine teenagers is perhaps the most important one: duringt he seventh century,a nd (although less vividly) into the eighth, some coin sequencesvisualise the agingprocess across adolescence.  .9);³¹ at age18to 19,with as hort but more fullydevelopedb eard (fig.10);³² at age20t o21, with -on manyb ut not all coins -as lightlyl onger beard ( fig.11);³³ and from age21o nward, with af ull beard and moustache (and oftenwith, from 654, his son Constantine IV, proclaimed co-emperorinthat year,aged4 : fig.12).³⁴ The most notable development is, clearly,C onstans' beard, which becomesp rogressively more prominent until, at the ageo f2 1-four years before the 'legal' year of male adulthood in Roman law -it becomes the dominant feature of his portrait,t hus earning him the nickname pogonatos (the bearded one). ³⁵ Other examples can be adduced from the numismatic evidence, though they rarelyv isualise the stages of the life course as graphicallya st he coins of Constans II.³⁶ But the coins of Constans II'sgrandfather,H erakleios (610 -641), anticipate the sequence,with Herakleios Constantine (son of Herakleios and father of Constans II, born in 612a nd crowned in 613) shown small on the earlycoins,then graduallyb ecomingl argera nd,f inally,bearded; his younger brother,H eraklonas( borno fH erakleios' second wife, Martina, in 626and named augustus at age12),r emained small and beardless,both because he was the junior co-emperorand because he was only 15 when Herakleios died and he and his mother were, shortlyt hereafter,e xiled.³⁷ And coins minted towardthe end of Constans II'sreign show Constantine IV growing older,a ccompanied, between 659 -668, by his younger brothers Herakleios and Tiberios, with the ages of the threes ons clearlyd ifferentiated by their size.³⁸ During  DOC 3, 422, pl. XXIV,5 c( minted 647).  DOC 3, 423, pl. XXIV,13c (minted 648/9).  DOC 3, 424, pl. XXIV,16a, 17b( minted 650/1).  DOC 3, 425 -27,p l. XXIV,19a-30a (minted 651-659).  In fact,his grandfather,Herakleios,was often depicted on coins with an equallyimpressive beard and moustache: see, e. g., DOC 2, pl. VIII, 26eand 27 (both minted 629 -631inConstantinople, when Herakleios was about 55).  In addition to the coins citedinthe followingparagraph,Maurice'sson Theodosios -born in 583 or 585, named augustus in 590and killed in 602(PLRE 3, 1293 -94) -appears on coins in 591/2, when he was not yeta na dolescent (ibid.,1294), and on as eries of coins that are, sadly, of indeterminate date(DOC 1, 373 -76), but the coins are toobadlyabraded to allow us to speculateabout his appearance.  On this extremelycomplicated sequenceofcoins, see DOC 2, 216 -383, 389 -401.Duringthe few months that Heraklonas remained on the throne after his father'sd eath, he remained beardless on the coins, perhaps because of his age, or perhapsbecause of his mother's( apparentlyunofficial) regency. In this respect,these coinsanticipatesome of those minted during the reign of Constantine VI and Eirene (on which see notes 16 abovea nd 48 below).  See, e. g., DOC 2, 434 -35,p l. XXV, 40c, 41d, 42e, 43b( minted 663 -668). The three brothers remained on the coinageafter the death of Constans II in 668, until Constantine IV deposed his brothers in 681 to clear the wayfor his own son, Justinian II; because it is uncertain when Herakleios and Tiberios were born, however,their ages on the coinagea re equallyu nclear and Ih avet herefore not consideredthem here: see PmbZ I/2, Nr.2556 and I/5, Nr.8484;for the coins, see e. g. DOC 2, 525 -29, pls XXXII-XXXIII, 1c-10g ,2 3.9 -27.1. his first reign, the coinage of Constantine IV'ss on, Justinian II (685 -695), followed the pattern established by his grandfather (Constans II): at 16,heappears beardless; at 17 or 18, his beard is formedofsmall dots; at 18 or 19 his beard, though still short, is more fullyd eveloped; and by the lasty ears of his first reign -famouslya lso signalled by the introduction of Christ on the gold coinage -he wears af ull beard.³⁹ After the demise of the Herakleian dynasty,t he importance of the beard stops being primarily as ignifier of maturity and becomesi nstead an indication of rank: under the Isaurians, beginning with LeoI II (717-741), junior emperors on coins are shown without beards,nomatter what their age, and the senior emperor is bearded, as are the Isaurian ancestors who appear on the reverse of coins from the reign of Leo III'sson Constantine V(741 -775) onward.⁴⁰ On LeoIII'sgold coins,Constantine Va ppears, aged 2, as soon as he was nameda ugustus in 720, and Philip Grierson tracks his increasing maturity on coins minted from 720( ' headv ery small, his hair not covering his ears':f ig.13);⁴¹ to the years 720 -725, when Constantine Vwas between 2a nd 7( ' heads lightlyl argera nd older,with longer hair':f ig.1 4);⁴² to 725 -732, when he was7to 14 ('head still larger'),⁴³ through the rest of Leo III'sr eign, 732 -741, by which time Constantine Vwas between 14 and 22 ('bust more mature': fig.1 5).⁴⁴ The key developments here are, as Grierson noted, the size of the head, which becomes more or less the same size as LeoI II'so nlyi nt he coins minted when Constantine was between 14 and 22;h air length; and, though Grierson does not mention this,f acial shape: Constantine Va cquirest he almost pointed chin of his father onlyinthe latest coins of this sequence.A tthis point,perhapstoindicate dynastic unity,the portraits of LeoI II and Constantine Vare virtuallyi ndistinguishable except for the small beard worn by Leo.
Coin portraits of Constantine V'ss on, Leo IV (born 749a nd namedc o-ruleri n 751), follow as imilar pattern: his round face with short hair graduallye longates and the hair lengthens until he closelyr esembles his father,a nd, on the reverse, his grandfather.⁴⁵ This changed, however,w hen LeoI Vb ecame senior emperor in 775, and namedh is own son, Constantine VI, augustus in 776. Constantine VI was then 5, and -following the precedent of the coins of LeoI II and Constantine Vone would expect to see the child'sf ace and hair lengthen as he grew.T his does not happen. Constantine VI appears on his father'sg old coinage immediatelya fter his elevation, but is distinguishedf rom him onlyb yh is lack of facial hair ( fig.16

Images of Byzantine Adolescents
we never see Constantine VI as achild on the coins.⁴⁶ It is possiblethat this was due to Leo IV'sill health, or to his apparent wish to ensure the succession of his son rather than of his brothers,⁴⁷ either of which mayh aveu rgedt hat dynastic succession was stressed aboveany indication of his heir-designate'sy outh. In fact,Constantine VI'sp ortrait changed remarkablyl ittle across the 20 years he appearedo nc oins: after his father'sd eath in 780, when Constantine VI was 9, he appeared, beardless, with his mother,t he regent empress Eirene, and he remained beardless on the gold coinage until his deposition in 797( fig.1 7), when he was 26.⁴⁸ Political circumstances mayh avep rompted Constantine VI'se xtended numismatic youth, but the absence of aportrait sequence visualising his adolescent growth and increasing maturitycontinued under Nikephoros I, whose son Staurakios joined him on coins after he was namedaugustus in 803,aged10or12, and remained there until Nikephoros' death in battle in 811, but whose features never varied from the beardless but mature facial type that had previouslyt ypified rulersi nt heir late teens.⁴⁹ The same is true of Theophylact,s on of the emperor Michael I, who ruled with his father from 811 to 813,a ged8to 10;⁵⁰ and it is also true of Smbat/Constantine, bornsometime between 800 and 810,who ruled with his father LeoVfrom 813 -820.⁵¹ What seems to be happening here is that the age-specificityw ef ound under the Herakleians is fading:t he beard becomes the dominant signifier of status, and it wasa pparentlym ore important to signal that therew as an heir-designatew ho looked the part thant ot rack that heir'sg rowth spurts. The insecurity of the period seems to have suggested that dynastic stability was more important to display than the growingm aturity of the co-ruler.
Coins minted under Michael II the Amorianwhich show Theophilos, born in 812/ 3and elevated to the throne by his father in 821, appear to continue this pattern, but are problematic. They have not been dated more preciselyt han '821-829' so that while some examples show the junior emperor with as lightlym orer ounded face than others, it is impossible to establish whether or not these werem inted earlier than coins thats how him with as lightlym ore elongatedf ace -whether,i ns hort, the variation (which is in anyc ase slight) represents an attempt to indicate aging or whether it is simply due to the use of different dies.⁵² Theophilos' first son, Constantine (who was born in the 820s but who died young), appears on the coinage as a child, but (like Staurakios, Theophylact and Smbat/Constantinebefore him) is never shown aging;⁵³ his daughters appear on the rare class IV nomismata of the late 830s with their relative ages indicated by scale but not physiognomic details (which would, in anycase, be difficult giventhe size of the coins);⁵⁴ and Michael III appears virtuallyi mmediatelya fter his birth, portrayed -despite his infancy -with the beardless but mature facial type that characterised images of the young Constantine VI, Staurakios and Smbat/Constantine earlier in the century⁵⁵.O nce again, the appearance of am ale heir,p resented in the appropriate and approved fashion sanctified by the last five generations of rulers, was more important to signal than Michael III'si ncreasingm aturity.
After his father'sdeath in 842, however,Michael III appears as achild (with his older sister Thekla) on the reverse of the coin; as regent,h is mother,t he empress Theodora, is shown on the obverse.⁵⁶ Soon thereafter,M ichael III -now once more with the beardless but maturef acial type he displayedd uringhis father'sl ifetime -is joined by his mother,and an imageofChrist occupies the obverse.⁵⁷ In 856, when Michael III was 16,h is mother disappears,a nd the emperori ss hown as a bearded adult⁵⁸.This tells us two things, neither of which is surprising.F irst,a si s clear from the coins of co-rulers from Constantine VI onwards, coin imagery shows what is believed to be appropriate;i ti sn ot visual reportage.T heodora's mint-master followed the precedent that had been established most recentlyw hen Eirene became regent in association with the young Constantine VI, and visualised this relationship on the coinage by presentingt he pair as an easilyr ecognisable mother and youngs on. Second, the beardless but mature facial type has become the normative wayt or epresent imperial youth, and this continues until the end of the Empire. Henceforth, adolescents on coins follow the pattern established by Constantine VI, Staurakios, Smbat/Constantine and, except on the coins minted immediatelya fter his father'sd eath, Michael III: they mayo rm ay not be smaller in scale thant he senior emperor, and they are beardless,b ut they do not displaya ny other signs of aging.⁵⁹ Never again will theirp rogressive maturation be portrayed as vividlya su nder the Herakleians and, though to al esser extent,t he Isaurians.  (DOC 3,; Constantine VII during his mother'sregency (DOC 3, 541-42); Romanos I'ssons Christoher,Stephen and Constantine Lekapenos,sometimeswith ConstantineV II, whofrom9 21 is sometimes presented as abearded adultand sometimes not,a pparentlyd epending on the political tide (DOC 3, 544-57); Constantine VII'ss on Romanos II (DOC 3, 551-69); Nikephoros II'ss tepson Basil II (DOC 3, 582-83); Eudokia'ss ons Michael VII and Constantios (DOC 3, 783 -84); Romanos IV'ss tepsons (DOC 3, 789 -91); Alexios I'ss on John II (DOC 4,

Images of Byzantine Adolescents
The run of portraits we have just examined allows us to draw an umber of conclusions, one of which is that the visualisation of adolescencei sh eavilyg endered. Formales, from the mid-seventh century on, the keyindicator is the presenceorabsence of abeard, which is, as we have seen, asignal of both maturity and of ashift in status. Forf emales, the key indicator is less blatant.A lthough we do not have af ull run of femaleportraits, the distinction between the just-turned-12-year-oldMary and the 16-year-old Euphrosyne is, however,s triking.W em ay add to it the imageo fa princess in the Vatican epithalamion (Vat.gr.1851), for,whomever the girl represents, she is presented as ap re-pubescent female ( fig.1 8).⁶⁰ The epithalamion child is slight,r aises her hands in supplication, and is bareheaded, allowing us to admire the pearls in her blond hair.T he 16-year-old Euphrosyne is far more solid, stands nearlyf rontal, and is portrayed with covered head. The child-princess,e venw hen she is shown dressed in Byzantine imperial regalia and enthroned amidst her court,displays the top of her head, and her hair remains visible, unlike the presumablyo lder girls and women who attend to her.E uphrosyne is solemn and displays due decorum; the princess appears to smile.I np art,t he differences between the child and Euphrosyne have to do with status: the younger girl is ap rincess,t he elder is anun. But if we lookatother young but married females in the Lincoln Colleget ypikon some of the distinctions continue. As equence of familyp ortraits was added to the typikon around 1335,d isplaying eight married couples in the family, some of whom are mentioned in the typikon.⁶¹ The older married couples - fig.1 9 portraysE uphrosyne'se lder brother Theodore Komnenos Doukas Palaiologos Synadenos and his wife Eudokia Doukaina Komnene Synadene Palaiologina -are blessed by the Virgin of Sure Hope, to whom the monastery was dedicated. The youngermarried couples - fig. 20 depicts the protosebastos Constantine Komnenos Raul Palaiologos with his wife, and Euphrosyne-the-nun'sn iece, Euphrosyne Doukaina Palaiologina -are blessed by the youthfulC hrist.W ed on ot have accurate dates for the second Euphrosyne, but she is probablys lightlyy ounger thanE uphrosyne-thenun, so in terms of chronological ages he is well within our category of Byzantine femalea dolescents, though she is clearlym arried.L ike the epithalamion princess, the typikon youngw omen are dressed in courtlyf inery,but they all differ distinctly from the princess in their headcovering and solemn demeanor.The princess,though arriving in Constantinople for her wedding, is not yetm arried age; the typikon women are, either to elite menortoChrist.The defining difference between how adolescent females wered epicted had little to do with actual age-in-years, but was in-224-26); Michael VIII'sson AndronikosII (DOC 5,nos 36,44,; Andronikos II'ssons Michael IX and Andronikos III (DOC 5,560,; John V'ss ons Andronikos IV and Manuel II (DOC 5, 200a nd no. 1196).  The problematic datingofthis manuscript and its illustrations arediscussed in Hennessy,Achild bride.  See the references in note 27 above. stead dependent on whether or not they were married (to at errestrial man or,a si n Eudokia'sc ase, to Christ).
Twoa dditional examples drive this point home. EudokiaD oukaina Komnene Synadene Palaiologina (wife of Euphrosyne-the-nun'se lder brother Theodore), whom as we have alreadynoted was pictured in the Lincoln Collegetypikon around 1335 ( fig.19), was,apparentlybefore her marriage, also portrayed in an unpublished prayer scroll made for her personal and privateu se, wheres he is identified by inscription ( fig.2 1). The young Eudokiai ss trikingly similar to the Vatican princess, down to the pearls interwoven into her long and looselyflowing hair.She is, equally, strikingly different from her laterp ortrait.A lthough we cannot see the top of her head, her hair is unbound and her facial features appear much more relaxed than they do in the lateri mage. We do not know how old Eudokia was when the prayer scroll was made (whichi sw hy Id id not includeh er in our rogues' gallerye arlier), but her size relative to John Chrysostom, before whom she anachronisticallys tands in prayer,a nd her bulk indicatest hat she is not as mall child. She is evidentlyn ot, however,yet married,and so appears as af emaley outh. Marriage, not age-in-years, and -as the next example will demonstrate -not even puberty itself, determines whether ay oungw oman of status is presented as an adolescent or am atron.
This convention of femalea dolescent portraiture is brought out sharplyi nt he two twelfth-century manuscripts illustrating the life of the Virgin Mary attributed to the monk James of the Kokkinobaphos monastery.⁶² Mary appears in about three-quarters of the miniatures, which depict her life from her birth until her virginal purity,despite being pregnant,isvindicatedbytrial by water. She is shown as an infant in the bath at her Nativity,⁶³ but as soon as she is accorded anyagency she is presented as ac hild, and her appearance changes little as she grows up in the temple. Accordingt ot he Protevangelion of James, after her precocious seven first steps at six months, her mother Anna vowed that Mary'sf eetw ould never touch the ground until she reachedthe ageof3 ,when she would be dedicated to the temple.⁶⁴ The first seven steps are not illustrated in the Kokkinobaphos manuscripts, but she appears in several miniatures that picture episodes thattook place before she was 3, and the miniaturist adheres to the prescription of the Protevangelion: Mary is always   Vat.gr.1162, f.29r;P aris.gr.1208, f. 38v:reproductions in the facsimiles cited above, and in Linardou,R eadingt wo Byzantine illustrated books 42 -43,p ls 18 -19.  E. Hennecke -W. S chneemelcher,N ew Testament Apocrypha 1. Philadelphia PA 1965, 377. shown either in bed or in the arms of her mother.⁶⁵ Between the ages of 3and 12 she is shown severalt imes, always as as light youngw oman considerablys horter in height than the adultf igures in the images ( fig.2 2),b ut with no attempt to portray the various stages of her physical development as she grew older.⁶⁶ When Mary turned 12,t he highp riestZ acharias begant ow orry about what to do with her,a s she could not stayi nt he temple once she reachedp uberty. He prayed; God told him to summonw idowers;a nd from among the widowers Joseph was selected to be Mary'sh usband and guardian.⁶⁷ Seven miniatures accompanyt his narrative, and Mary appears in all but the two scenes that depict Zacharias in prayer and the summoning of the widowers.⁶⁸ In all of them, including the miniature that pictures her presentation to Joseph ( fig.6), she is presented in the child-like format familiar from earlier episodesi nt he manuscript.M inutes later, Mary bidsf arewell to the priests, and leavest he temple with Joseph to travel to his house: she is here, and henceforth, portrayedasanadultwoman ( fig.23).⁶⁹ This is agraphic demonstration of the role of marriageindetermininghow young women wereportrayedinByzantium:what defines Mary as ayoung woman rather than an adolescent girl is neither puberty nor her betrothal to Joseph, but her actual partnership with him. As visualised in Middle and Late Byzantium, growingup, for women, was not agradual process, but an instant transformation effected by marriage.
Indeed, after Iconoclasm, young (pre-married) females look remarkablyt he same. The epithalamion princess ( fig.18) and the young Eudokia ( fig.21) are almost interchangeable; and Mary looks the samea t3( fig.2 2) as she does at 12 ( fig.6). As the coin evidence demonstrated, duringa nd afterI conoclasm, the same is true of adolescent (pre-beard) males. This is not because Byzantine painters (or coin die strikers,f or that matter)w erei ncapable of depicting different facial types: adult males are oftens harplyd istinguishedo ne from another,a nd even when they were clearlyr eproducingt ypes rather than painting portraits, painters wereu sually careful to make adultm en in the sames cene look different.⁷⁰ In contrast,a fter Iconoclasm, even portraits of actual adolescents look exactlyl ike all other male youth: the 10-year-old Alexander and 14-year-old Leo VI from the Paris Gregory ( fig.3 ) are distinguishedf rom one another onlybyt heirh eight and attributes,a nd look remarkablys imilar to the 12-year-old Christ ( fig.5 )a sw ell.
Though Byzantine texts present adolescents with aseries of distinctivelydefined characteristics,t hese characteristics are themselves topoi (e. g. impetuous youth, saintlyyouth). In stories -hagiographyand the twelfth-century (and later)Byzantine novels -adolescents playw ell-defined roles,⁷¹ but they are otherwise rarelyd escribed as differentiated actors:once Byzantine males become active participants in the political or social narrative,t hey are treated as adults. Clearly, the Byzantines were well aware of the wayt hat 'real' males agea cross adolescence; we could assume it,b ut the coins of the seventh century confirm it for emperors. Nevertheless, in terms of representations, in all other contexts, the overwhelmingly dominantw ay youths are portrayed is the generic. The visual evidence corroborates the suspicion that adolescent males weresimply not importantenough as characters for Byzantine painters to want to distinguish them. This givess ome context to textual representations of unrulyorsaintlyy outh; therewereofcourse clichés, but also clichés which, thanks to the visual evidence, we can see were the work of writers who did not care how accurate they were, for in asense all adolescents weremuch the same.⁷² (In this sense, on suspects thatadolescents,likewomen, weregood to think with). This was, definitely, not ac ulturew hich at anyp oint across its millennium of history paid much attention to the experience of living through adolescence.
My final point is related to this. Women, like youthful males, all look the same. There are young, unmarried women, and there are older,m arried women (brides of men, brides of Christ,a nd occasionallyw idows), but none of them have anyd istinguishing physiognomic characteristics except for the colour of theirhair. After Iconoclasm, onlymen weres ignificant enough players to be individualised,a nd this fits with my earlier observation that onlym en grow old in Byzantine imagery.S o: Females appear as children (rarely), unmarried adolescents (rarely), servants (ageu ncertain) and generic adults, but undifferentiated inside each group. Males appear as generic children, generic adolescents,g eneric servants (ageuncertain), but then differentiated adults, includinge unuchs, and often (even if not always) differentiated old men. The conventions of gendered social status and interpersonal relations in Byzantium could hardlyb ec learer.