Proceedings o f the
Danish Institute at Athens •
Edited by Seven D ie tz
T h e pub licatio n was sponsored by:
C o n su l G eneral G osta E nborns Foundation.
T h e D anish R esearch C o u n cil for th e H um anities.
K onsul G eorge Jo rck og H u stru E m m a J o rc k ’s Fond.
© C o p y rig h t T h e D anish Institute at A thens, A thens 1995
Proceedings o f th e D anish Institute at A thens G eneral E d ito r: Seren D ietz
G raphic design and P ro d u c tio n by: Freddy Pedersen P rin te d in D e n m ark on p e rm a n en t paper IS B N 87 7288 721 4
D istrib u te d by
A A R H U S U N IV E R S IT Y P R E S S U niversity o f Aarhus
D K -8 0 0 0 A rhus C Fax (+45) 8619 8433 73 Lim e W alk
H e ad in g to n , O x fo rd 0 X 3 7AD Fax (+44) 865 750 079 B ox 511
O akville, C o n n . 06779 Fax (+1) 203 945 94 9468
T h e cover illustration depicts a B ronze Statuette o f a H o rse fo u n d at th e Argive H eraion. N M 13943.
D raw in g by N iels Levinsen. See p. 55, Fig. 19.
Pagans in Late R om an Halikarnassos II
The voice o f the inscriptions
Signe Isager
N O T E 1
H in k s 1933, 125-143.
T h is article w ill provide a prelim in ary p resen tatio n an d discussion o f th e m osaic in scrip tio n s fo u n d in th e pavem ents o f a late R o m a n b u ild in g in H alikarnassos. It suggests th a t th e inscriptions testify to an o rp h ic /n e o p la to n ic m ilieu in 5 th ce n tu ry H alikarnassos and to a spiritual i f n o t p e r
sonal presence o f th e E g y p tian p o e t N o n - nos in th e to w n . T h e in scrip tio n s to be p resen ted w ere fo u n d d u rin g th e D anish excavations in B o d ru m in 1991 and 1992;
th ey w ere cleaned and restored in the cam paign o f 1993.
P art o f th e R o m a n b u ild in g was exca
vated in 1856 by C .T . N e w to n and the inscrip tio n s fo u n d th e n are already k n o w n .1 N evertheless it is n atural to dis
cuss th e m in c o n n e c tio n w ith th e n ew findings. In th e follow ing I shall fo r p ra c tical reasons m ake a d istin ctio n b e tw e e n th e parts o f th e R o m a n b u ild in g excavat
ed respectively by th e B ritish and the D anes.
A ll th e in scrip tio n s in th e D a n ish -e x cavated p a rt w ere set in th e tesselated flo o r o f th e apsidal ro o m F (Fig. 1). T h e
m o st ex citin g find was a nearly co m p lete in sc rip tio n in eig h t lines, an epigram set in a tabula ansata (Fig. 7). O p p o site this and ju s t in fro n t o f th e en tran ce was fo u n d the rig h t p a rt o f an in scrip tio n set in a m ed allio n (Fig. 2). T h e left p a rt - little m o re th a n h a lf - is com pletely missing.
T h e p erso n ified Seasons in th e fo u r c o r
ners o f th e ro o m w ere all id en tified n o t on ly by th e ir attributes b u t also by in scriptions (Figs. 3-6).
T h e p a rt o f th e b u ild in g excavated by N e w to n h ad som e inscrip tio n s w ith a p u rely id en tify in g fu n c tio n too: T h e Sea
sons w ere identified by the same nam es as in o u r p a rt o f th e building.
Several m ythical figures w ere identified by inscriptions, i.e. M eleag er and A talanta (Poulsen, Figs. 16-18), D id o an d Aeneas and finally D ionysos. T h re e m edallions each c o n ta in in g a fem ale head w ere id e n tified by inscriptions as H alikarnassos, A lexandria and B erytos respectively.
Finally th ere was a m edallion c o n ta in in g six w ords to be c o m m e n te d o n later.
T h e inscriptions o f the apsidal room F
1. T h e epigram Text:
B fjjia x edv <p£pe 5et>po poM>v 5’ i n i v e v e qxxeivoic;
a i m i c a v b v p L e tp a p o tq \jrq <p o0exov m p ^ x w
CHopa AiOtov TioXvpopqxov 07tep xexv Yj poveq av8peq
axopvupivo'D 8a7te8oD 7tavxo0ev fjyLddaav,
5 o<ppa Kev £ v 0 a K a i e v 0 a S o p o v 7 to k \)8 a t8 a k o v etS o q
i)\(/op6<pot) 7 ta x p r|i x f |i 5 ’ 6 v o p a o x 6 v o 7 tp tv a i K & t o v xek iO o v X a p iS q p oq
&y e ip e v
Etc y a v n q K a p A x o iq x p f j p a x a n X l m a Ttopcbv
Translation:
C o m e h ith e r and n o d y o u r approval w ith o u t delay w ith y o u r b rig h t shining
eyes. I p resen t a
m u ltifo rm b o d y o f stones laid in mosaics, a b o d y w h ic h skillful m en in spreading th e floor m ade shine all
over,
5 so th a t the richly w ro u g h t appearance o f th e h ig h -ro o fe d build in g
shall m ake this city re n o w n e d in m any places.
W h a t before was in a m iserable state C h arid em o s raised
from th e g ro u n d w ith toil and e n o r
m ous expense
R em ark s: T h e tex t was nearly co m p lete w h e n fo u n d in 1992, and the restorations carried o u t by B en n y B erg from the N a tional M u se u m o f C o p e n h a g e n in 1993 have co n firm ed the readings and elim inat
ed alternative possibilities.
T h e designer o f th e m osaic attained a balanced co m p o sitio n in th e tabula by fill
in g o u t gaps at th e e n d o f the line w ith a floral m o tif, resem bling a tw ig o f pine.
T h e same device is used in th e inscribed m ed allio n .2
T h e letters o f th e in scrip tio n are h a n d som e and regular, 5-6 cm in h eig h t, and are w ith o u t d o u b t “ late” , b u t they do n o t o f course in them selves provide evidence for a precise dating. Sigm a, epsilon and om ega (always divided) have square form s, w h ile th e o m ik ro n and th eta are circular a n d /o r oval respectively. It is obvious from th e beautifully ro u n d e d om ikrons and p h i’s th at it was n o t lack o f skill w h ic h m ade th e artist choose th e square fo rm for som e letters.3 It is difficult to generalize o n such m atters. F or instance a b u ild in g in T h eb es tentatively dated to th e late 5 th or early 6th c e n tu ry 4 has representations o f th e m o n th s in its tesselated p avem ent, tw o o f th e m w ith square sigmas in th e ir nam e, the rest b ein g oval. Likew ise th e m osaic o f th e N ereid es in A pam ea dated to 362-3 6 3 has th e nam e o f Krisis w ith square sigmas b u t an oval sigma in P o seid o n .5 I have so far n o t fo u n d close parallels for all letters o f th e epigram in o n e in sc rip tio n .6
T h e translation o f th e very first w ord, fifjjia, is a m atte r for discussion. It is very unlikely th at it had a neutral m ean in g to any c o n te m p o ra ry reader - n o r did it give
N O T E 2
C om pare the inscription in the bath o f P oim enios w h ere a n o th e r floral m o tive is used, B u d d e 1972, pi. 93.
N O T E 3
cf. A kerstro m -H au g en 1974, 50.
N O T E 4
Spiro 1978, 210-211.
N O T E 5
Baity 1981, fig. 242.
N O T E 6
M o re irregular b u t n o t w ith o u t sim ilarities are the letters o f an inscription from a baptisterium in the Basilica o f Saint J o h n at Kos, Pelekanidis 1974, pi- nax 44.
Fig. 2. Inscription in medallion seen from the west. Fig. 3. Personification o f Autum n.
*’ y \' r 7: S S fl - T ■*.' 4 r \ > (%,
I
W v ‘« w l ^ 3 9 £ m f e f ;
n < m
Fig. 4. Personification o f Spring. Fig. 5. Personification o f Summer.
' I \*i ••»*«<« * ■*
m f :
Fig. 6. Personification o f Winter. Fig. 7. Inscription in tabula ansata in front o f the apse.
th e same associations to every reader, since it was th e w o rd fo r step, fo r trib u n al and fo r th e area o f th e chu rch co n tain in g th e alter.
2. T h e in scrip tio n in the m edallion Text:
K IM E N O N E O N ITAP] A K O IT H N
E H N H 5 ,AA]OXOE
.EK JEO N Q N A EA A E IT IM H T ]E Y S H
A EK A
R em ark s: T h e m edallion m ust have b een a b o u t 115 cm in d iam eter w h e n c o m plete. Less th a n h a lf has b e en preserved and u n fo rtu n a te ly fo r th e in te rp re ta tio n o f th e in scrip tio n it was the rig h t-h a n d side w h ic h was preserved.
In th e course o f cleaning and restora
tio n d u rin g th e 19 9 3 -cam p aig n th e read
ings o f 1992 w ere co n firm ed , and a few loose le tte r fragm ents n o t given here, re covered.
T h e letters are 5 -6 .5 cm high. As in th e epigram , sigma, epsilon and divided o m ega are square, b u t in contrast to the e p ig ram som e o f the letters have a m o re cursive fo rm , delta, alpha, ksi an d m y (only once).
T h e excavators fo u n d n o trace o f re pair to the pavem ent. M an y o f the letters are very m u c h like those in the epigram ; the tesserae are th e same and, as already m e n tio n e d , th e way o f filling o u t th e e n d ings o f a line w ith a form alized tw ig o f p in e is identical w ith th e o ne used in the in sc rip tio n o f th e tabula ansata. T h erefo re w e assum e th a t th e in scrip tio n was m ade w ith th e rest o f the p av em en t and that th e use o f cursive letters was chosen to suit the circular fo rm o f the to ndo.
I shall n o t hazard any tentative transla
tio n o f this fragm entary in scrip tio n fo r the m o m e n t, b u t only p o in t to th e fact that m edallions o f this k in d w ith inscriptions have parallels in m any late R o m a n b u ild
ings, often in a C h ristian co n tex t. It seems fo r th e m o m e n t a reasonable guess th at a co m p lete te x t w o u ld have given the nam es o f a m a rrie d couple.
3. T h e Seasons a. a l a p - spring b. )(£i]}Kbv - w in te r c. (pG ivoraopov - a u tu m n d. 0[epoq] - sum m er
T h e Seasons are identified in the pavem ent partly w ith symbols, partly by inscriptions.
T h e restoration o f the te x t is su p p o rted by th e parallel inscriptions fo u n d in the p art o f th e b u ild in g excavated by N e w to n .
T h e letters are sim ilar to th e letters in th e epigram . N e w to n suggested th a t a l a p m ig h t be th e C aria n version o ffe a p , b u t the same spelling has since b e e n fo u n d in the Seasons panels from H agios Taxiarchis n ear A rgos in a b a th tentatively dated to the early 6 th century.8
Interpretation - the epigram
T h e epigram is th e in scrip tio n w h ic h has m ost to offer as regards in terp retatio n .
T h e general m ean in g o f th e tex t is n o t in d o u b t. T h e visitor is invited by the b u ild in g to approach, to acknow ledge and to adm ire w h a t skillful m e n have m anaged to do w ith th e floor. It will re d o u n d to th e city o f H alikarnassos, and th e m an w h o to o k u p o n h im se lf the trouble and the expense has deserved his em phatic place in th e verse.
W h ile the general m ean in g is plain, m any questions are raised by th e text.
W h o is th e visitor? W h o is th e d onor? Is he perhaps also the ho st and th e o w n e r o f th e building? Is it tru e th at th e bu ild in g was rebuilt from th e g ro u n d , o r is th at a form ulaic expression n o t to be taken lite r
ally?
As far as th e ex te n t o f th e rebuilding is co n c e rn e d , th e article by B irte Poulsen seems to su p p o rt a literal read in g o f the text. T h e nam e o f the d o n o r, C h a rid e - m os, is k n o w n from tw o H alikarnassian inscriptions on co lu m n drum s w h ic h we h o p e to be able to exam ine fu rth e r.9 It is
N O T E 7
For m y translation, cf. Lid
dell & S cott & Jones s.v.
(pepw A II. For the transla
tion tribunal, cf. R o b e rt 1948, especially A ddenda c o n ce rn in g an inscription from N icopolis ad Istrum , no. 30 in B ritsch k o ff 1923.
N O T E 8
N e w to n 1863, 310.
A k erstro m -H au g en 1974, 71 and 129.
N O T E 9 IG II 2656b.
N O T E 1 0
For N o n n o s ’ m etre see W ifstrand 1933, Vian 1976.
N O T E 1 1 Vian 1976, L.
N O T E 1 2 W ifstrand 1933, 23.
n o t e 1 3
C f. P P U A E S 3A, 167.
necessary to b ear in m in d th o u g h th at C h a rid e m o s m ig h t have lived elsew here and th erefo re have left n o fu rth e r traces in in scrip tio n s from H alikarnassos. In the fo llo w in g I w ill ex am ine in w h a t w ay the epigram and th e inscrip tio n s as a w h o le m ig h t th ro w som e lig h t o n th e id e n tity o f th e p e o p le m e a n t to use the building.
T he connection w ith N onnos - formal
It was the ch o ice o f w ords in th e p o e m w h ic h first p o in te d to th e E gyptian p o e t N o n n o s as a p erso n w h o m ig h t provide th e key to o u r u n d e rsta n d in g o f th e spirit th at reig n ed in th e b u ilding. T h e epigram - sh o rt as it is - has m o st o f its w ords in c o m m o n w ith N o n n o s ’ great p o e m D io- nysiaka in 48 songs. T h e y o ften o ccupy th e same p o sitio n in th e verse. T h e ep i
g ram is w ritte n in elegiacs an d consists o f fo u r disticha, w h ile N o n n o s ’ tw o k n o w n w orks are w ritte n in hexam eters. T h e re fore it is n o t possible to com p are th e m to o strictly, b u t still it is obvious th a t o u r p o e m follow s ra th e r slavishly th e rules used by K allim achos and revived by N o n nos. T h e G reek language was u n d e rg o in g a gradual process o f change w h e re b y the difference b e tw e e n lo n g an d sh o rt vowels te n d e d to disappear and tonal accen t to be replaced by dynam ic accent. O n e o f the effects on N o n n o s ’ verse was a preference for th e dactyl and a very restricted use o f th e sp o n d e e .10
Francis V ian has draw n up a list o f N o n n o s ’ p rio ritie s for th e first five feet in his h e x a m e te rs.11 T h e re are 9 different co m b in atio n s. In th e fo u r h e x a m e te r lines o f o u r p o e m lines 1, 5 and 7 follow N o n n o s ’ first p rio rity (ddddd - d = d acty l), line 3 follow s his th ird (dddsd - s= sp o n d - ee). Lines 1, 3 an d 5 have fem ale caesurae as p referred by N o n n o s and his follow ers w h ile line 7 has the m asculine caesura.
T h e latter sh o u ld n o t disturb us: line 7 is th e line w h ic h m e n tio n s th e n am e o f th e do n o r, C h a rid e m o s, and in fact p ro p e r nam es allow fo r certain prosodic devia
tions in N o n n o s ’ verses to o .12 To stress the p ro p e r nam e fu rth e r th e m osaicist left a
little extra space in fro n t o f it in th e m osa
ic. T h e fo u r lines consisting o f p e n ta m e ters are totally dactylic e x ce p t for the spondee constituting the first foot in line 8 .13
Special expressions
T h is is n o t the place to c o m p are every w o rd in th e epigram w ith N o n n o s ’ w ork.
I shall on ly p o in t to tw o c o m m o n d e n o m in ato rs. T h e rh y th m and c o lo u r o f N o n n o s ’ p o e try is ch aracterized inter alia by tw o expressions also fo u n d in o u r ep i
gram . T h e first is£v0(X Ktxi £v0(X, here an d th ere and everyw here, w h ic h occurs sixty tim es in N o n n o s , fifty -n in e o f w h ic h in th e sam e p o sitio n as in th e H alik arn as- sian epigram . It is n atu ral th a t only o n e o f these exam ples sh o u ld b e from his para
phrase o f the G ospel according to St. Jo h n . T h e expression is b e tte r su ited to th e rest
less, dram atic and som etim es slightly d ru n k e n atm o sp h ere in th e D ionysiaka.
T h e o th e r expression is 7toTl)8ou5ocXoV e i6 o q , rich ly w ro u g h t appearance, 1.5.
T h e translation “rich ly w r o u g h t” for 7toAu5cu8aAov is perhaps to o generic.
“ W ro u g h t w ith D aidalic skill” is th e m o re literal re n d erin g . T h e shield o f A chilleus is d escrib ed as polydaidalic in th e Iliad 11.32, and th e w o rd is very c o m m o n in N o n n o s. It occurs 23 tim es, o n e o f w h ic h in the paraphrase o f the Gospel according to S t.J o h n . T h e re are n o d o u b t several rea
sons for this. N o n n o s loves elab o ratin g on the fantastic an d m arvellous, be it in fo rm , in colour, o r in te c h n iq u e. H e prefers lo n g and graphic ep ith eta placed before th e (often short) n o u n to be qualified.
T h u s a typical constellation in his verse is precisely 7ToA,t)8(xl8(xA,OV EtSoq (six o c c u r
rences, all in th e D io n ysia ka ). A n o th e r rea
son fo r N o n n o s to use it m ig h t b e later philosophical in te rp re ta tio n s o f the d e scrip tio n o f th e shield in H o m e r as a d e scrip tio n o f cosm os (E ustathius ad locum - T a i r a x 8e p i p r i p a xoi) K<5apox> e lv o d q x x a iv o l Tcepi xov K p& xryta. A io 7to>a)8ou8aA(jc; xe e ip T y ra i 8 i d x q v xtbv K oapiK tbv eiStfjv T totK iX Jav K cd xfyv xtbv
%p6vtov Ktxi xtbv Ktxiptbv, kxA.)
T h e w o rd 7toA,X)8a(8aAx)q is especially
well adapted to dactylic poetry. In versi
fied inscrip tio n s it is nonetheless rarely fo u n d . T h e tw o inscrip tio n s c o n tain in g th e w o rd
7
toAu6
a i6
aAoc; (none o f th em c o m b in e d w ith ei5 o q ) are b o th from Late A n tiq u ity - o n e has a C h ristian co n tex t, the o th e r a pagan o n e.14
T h e latter is an in scrip tio n from A m yklai in S parta
.15
It was cut in a base m ean t to bear an h o n o ra ry statue. T h e verse is addressed to A pollo, it is in 4 dis- tichs and even i f it is less elegant and has its ep ith eta after th e n o u n unlike N o n n o s it has m u c h m o re th an the w o rd7
toAu5
a (5
aA oq in c o m m o n w ith o u r e p igram .
T h e o th e r in scrip tio n co n ta in in g the w ord
7
toAu5
a f5
aAoc; is a p o e m in h ex am e te r beautifully cu t in raised letters fo rm in g part o f the in te rio r as w ell as the ex te rio r d eco ratio n o f H agios Polyeuktos in B yzantion. It is thus dated to 5 2 4 -5 2 7 , i.e.th e p e rio d w h e n th e ch u rc h was built . T h e p o e m is n o t fully preserved in the ru in s o f th e ch u rch b u t all its 76 lines w ere already k n o w n from the Anthologia Palatlna (1.10). It was co m p o sed to cele
brate Juliana A nicia (462 - c. 528), w h o built th e c h u rch to the so ld ier-m arty r P o ly eu k to s
.16
T h e p o em has som e o f its w ords and its way o f using epitheta in c o m m o n w ith N o n n o s, and since it is w ritte n to praise th e d o n o r it is natural th at it should stress the splendour, the h o n o u r and the toil ju s t like the H alik ar- nassian epigram . T h e h e x a m e te r has to o m any sp o n d e e ’s by N o n n ia n standards.B u t th e general to n e, its interest in m y th o lo g y and in the celestial p h e n o m e n a w o u ld fit well in to his universe. T h e only in scrip tio n I k n o w o f till n o w w ith the typical N o n n ia n expression JtoAuficuoa- Aov e!5 o q is th e epigram in Halikarnassos.
T he connection w ith N onnos - content
W h ile the sim ilarities w ith o u r epigram in vocabulary and rh y th m can be fo u n d th ro u g h o u t the D ionysiaka o f N o n n o s, th ere are parallels in c o n te n t also. W e find in th e D ionysiaka tw o invitations e x ten d ed
to strangers, m ain characters in th e epos, to co m e and visit a palace (called inter alia a Sopioq). T h e first is K adm os w h o is led by P eith o in disguise to visit th e palace o f E le k tra
.17
K adm os is travelling all over the w orld in search for his sister E uropa, b u t th e jo u rn e y is in te rru p te d m o m e n ta rily at S am othrake so th at he can m eet his b rid e to be, H a rm o n ia . It has lo n g b ee n recognized th at the d escrip tio n o f K ad m o s’ visit to th e palace o f E lektra has its p recu rso r in th e d escription o f O dysseus’ visit to A lki- noos in the Odyssey. T h e palace o f E le k tra is described in detail in the D ionysiaka III 123-146.
T h e second invitation to visit a palace is ex te n d e d to D ionysos h im self by the Assyrian dynast Staphylos - G rape - w h o prostrates h im self before th e g o d and asks him to im itate his father Z eus the F riend o f G uests and e n te r his palace
.18
T h e palace is described in X V III 6 2 -9 2 . N e ith e r o f the palace visits in N o n n o s is “ c ite d ” in the sh o rt epigram from Halikarnassos, b u t in reading th em you ca n n o t help feeling that th ey w ere close to th e a u th o r’s m in d w h e n w ritin g . T h a t is n o t a dem onstrable fact. Still it seems safe to co n clu d e th at the epigram was w ritte n in the tim e o f o r lat
er th an N o n n o s ’ D ionysiaka. It m ay n o t be a co in cid en ce th a t an episode told in set
tin g the scene for K ad m o s’ visit is the one w h ere “ the Seasons ran to th e house (
8
o p o q ) o f E lektra b earin g the sceptre o f Z eus and th e robe o f K ronos and the staff o f O ly m p o s to prophesy th e indissoluble d o m in io n o f the A usonians” , th at is the R o m a n s, III 195-199.Parallels in stone
W h ile epigram s really sim ilar to th e H ali- karnassian o n e in m osaic pavem ents from Asia M in o r are n o t as yet k n o w n to us, there are som e c o n te m p o ra ry inscriptions in stone from A phrodisias w h ic h resem ble it in style as well as in purpose. All three inscriptions are cu t directly “ on the h ig h est rem ain in g course o f blocks
”19
in the restored so-called A gora Gate. T h e y celebrate the restoration w h ic h inter alia m ean t th at th e b u ild in g was adapted as a fo u n -
n o t e 1 4
See also Antlwlogia Palatina, 6.332 and 16.80.
N O T E 1 5
IG V,1,455, T sountas, E phA rch 1892, 21.
N O T E 1 6 H arrison 1989.
N O T E 1 7 III 83 sqq.
n o t e 18 X V III 39-40.
N O T E 1 9 R o u e c h e 1989, 68.
N O T E 20 R o u e c h e 1989, 71.
N O T E 21
Pelekanidis 1974 and Ase- n ak o p o u lo u -A tzak a 1987 are invaluable.
N O T E 22
A rch E p h 1917, 63 and 48, the last line o f w hich, line 5, does n o t seem to fo rm p art o f the poem . N O T E 2 3 Spiro 1978, 429.
N O T E 24 B u d d e 1969, 97.
tain, n o .38. T h e poetically best o f the th re e A phrodisian in scrip tio n s is a h e x a m e te r in six lines, R o u e c h e no .3 8 . It cele
brates th e d o n o r, A m pelios, titled “fath er o f his m o th e rla n d ” , and also th e p o e t h im se lf w h ic h is unusual. T h e p o e t is P y - th io d o ro s from Tralles and h e calls h im self a rh eto r. T h e tw o o th e r inscriptions, nos. 39 and 40, are in distichs like ours.
O n e o f th e m was am e n d e d - perhaps by th e city co u n cil - w h ic h did n o t add to th e elegance o f th e verse,20 b u t still the ch o ice o f w ords calls th e H alikarnassian in sc rip tio n to m ind.
Parallels in mosaic floors
F ro m w h a t I have b e e n able to discover till n o w it seems th a t m osaic inscriptions in floors are c o m m o n in Late A ntiquity, b u t th a t it is a com paratively rare o c c u r
rence to find such inscrip tio n s w ritte n in verse an d even m o re rare th a t this verse sh o u ld be o f such h ig h quality as th e o ne in H alikarnassos.
H elp fu l fo r research o n th e G reek m ainland is M arie Spiro, Critical Corpus o f the Mosaic Pavements on the G reek M ainland, Fourth /S ix th Centuries w ith Architectural Surveys, vol 1-2, N e w Y ork an d L o n d o n 1978.21
Several in scrip tio n s c o n ta in e d in this co rp u s m e n tio n th e s p o n s o r/d o n o r o f the m osaics o r perhaps o f th e w h o le building.
In these cases th e b u ild in g is always m ean t fo r th e public, a co n g reg atio n o r th e like - a fact to keep in m in d w h e n w e try to in te rp re t th e fu n c tio n o f o u r building.
N o rm a lly th e in scrip tio n s are in prose.
Versified inscriptions are, how ever, fo u n d in o n e o f th e five basilicas o f N i- kopolis, th e so-called basilica alpha. N o less th an th ree in scrip tio n s in its m osaic flo o r are in 4 -lin e h ex am eters an d have som e unm istakably N o n n ia n features.
T h u s nearly all th e w ords are k n o w n from N o n n o s and som e o f th e m o ften o ccupy th e same place in his verse. S om e o f th e m also o c c u r in o u r epigram . O n th e o th e r h an d , th e a u th o r seems, in contrast to N o n n o s, to prefer a sp o n d ee as th e fo u rth fo o t o f th e h e x a m e te r and th e re is only
occasional N o n n ia n h an d lin g o f epithets.
Tw o o f these in scrip tio n s are nearly id e n tical.22 S om etim es th e epsilon b efore th e iota is o m itte d as in o u r in sc rip tio n . O n th e basis o f th e c o n te n t o f th e inscrip tio n s th e first phase o f th e b u ild in g in N ik o p o lis has b e e n dated to th e seco n d q u a rte r o f th e 6 th c en tu ry .23
From Kilikia w e have an in scrip tio n in tw o distichs set in to th e m osaic flo o r o f a R o m a n bath. N o N o n n ia n echoes are de
tectable w h ic h is reassuring since th e m o saics are dated o n stylistic g ro u n d s as early as th e late 4 th cen tu ry .24
T he inscriptions in the part o f the building excavated by N ew to n
T h is is n o t th e place to devote an exhaus
tive e x am in atio n to th e in scrip tio n s re
covered d u rin g th e m id 19 th c e n tu ry B ritish excavations in H alikarnassos and to com p are th e m w ith the n ew D an ish fin d ings. It is obvious fo r instance th a t th e le t
te r form s o f the in scrip tio n s id en tify in g A talanta and M eleag er are n o t th e sam e as those fo u n d in th e p a rt o f th e b u ild in g ex
cavated by th e D anes. N evertheless the exam inations h ith e rto c o n d u c te d leave us in n o d o u b t th at th e b u ild in g fu n c tio n e d as o n e b u ilding. T h u s th e fo u r seasons in N e w to n ’s ro o m B w ere id en tified by the same w ords as in th e apsidal ro o m F exca
vated by th e D anes. T h e rest o f th e m y thological figures id en tified by inscriptions w ere all fo u n d in th e p a rt o f th e b u ild in g excavated by N e w to n . T h e y are nearly all k n o w n from N o n n o s ’ D ionysiaka, th e ex cep tio n s b e in g D id o and A eneas.
A square c o n ta in in g a laurel w re a th e n closing th e follow ing six w ords was fo u n d in th e p art o f th e b u ild in g excavated by N e w to n :
-byia - health
£6ll - life X a p d - j o y e i p f \ v n - peace
£\)9\>pUX- cheerfulness
£A,7ttq- h o p e
All these w ords are concepts w ith positive co n n o tatio n s. S om e o f th e m are c o m m o n in a C h ristian co n tex t, like Ttttmc; “ju o tk ;,
£A,jrfq, ay&Trrp o r e ip i^ v r j. In m osaic floors o f Late A n tiq u ity you o ften m ee t co ncepts personified and id e n tified by in scriptions25 o r y o u can m e e t th e m , th o u g h m o re rarely, as here, w ith o u t any figural representation. Still, I k n o w o f n o o th e r m osaic floor w ith so m any concepts spelled o u t in o n e frame. A n d since the im pression has already b e e n given th a t we are in th e m o re pagan w o rld o f N o n n o s ’ D ionysiaka, it is reasonable to lo o k fo r the appearance o f th e w ords in o rp h ic o r n e o - p latonic contexts. It seems to m e sugges
tive th a t five o f the six concepts are c o n tain ed in th e last fo u r lines o f th e so-called o rp h ic hym n to Zeus:
a X k h /r a p e i q A m P a i a i 8 ( S o d tppecriv a iC T tp a T tavxa ^tofjv x ’ dA-fhoODpov, o p o b 0 ’ h y t e i a v a v a o a a v e ip f jv riv xe 0 e6 v , KoDpoxp6<pov, ay X n to x ip o v , K a i p io v e tO tijio x c n v a e i Q a X X o v z a X o y i o i i o i q It is n o t im m ed iately clear w h y personifi
cations n o t only o f H alikarnassos b u t o f B erytos and A lexandria w ere given such a co nspicuous place in th e m osaic d eco ra
tio n o f th e floor. M any explanations are possible. T h e ow ners o f th e bu ild in g probably w a n te d to advertise th e ir c o n n ectio n s w ith p eo p le in th e coastal cities o f B ery to s and A lexandria. T h e c o n n e c tions m ig h t b e com m ercial. O r m aybe o n e should lo o k to N o n n o s o n ce m ore fo r an alternative explanation. A lexandria was the city w h ere he w ro te his Dionysiaka.
H e probably w e n t to B ery to s too, as d e m onstrated recen tly by P ierre C h u v in .26 B ery to s - m o d e rn B e iru t - plays a very im p o rta n t role in his D ionysiaka ju s t as it did in the intellectual h isto ry o f his o w n w orld. In his epic B erytos is dw elt u p o n as an im p o rta n t R o m a n co lo n y to b e and as th e city w h ic h is g o in g to h o ld th e “ reins o f L aw ” (XLI 389 -3 9 3 ). As fo r H alikar
nassos, th at is o f course th e city o f the building, th e 7tdipr| o f th e epigram . It is generally agreed, and it can be in ferred from th e D ionysiaka th at Asia M in o r at
tracted N o n n o s ’ lively atten tio n . It is o f course unnecessary to go so far as to th in k o f H alikarnassos as o ne o f th e places w h e re N o n n o s actually w en t, even i f o ne o u g h t n o t to exclude th e possibility that N o n n o s b ecam e o n e o f th e m any - m ost o f th em pagan - E gyptian w an d e rin g p o ets.T h ese are pu re speculations. It is a fact, th o u g h , th at A lexandria and especial
ly B erytos rem ained strongholds fo r in te l
lectual pagans fo r a lo n g tim e. T h e same mutatis mutandis m ig h t go fo r H alikarnas
sos.
C oncluding remarks
W h e re do these prelim in ary investigations lead us? As a m in im u m I th in k they give a relative datin g for the last phase o f the building. It was c o n te m p o ra ry w ith o r lat
er th an N o n n o s. F o rtu n ately th at fits well in to th e p ictu re given by the o th e r find
ings.28
A part from that, it is o f course a deli
cate q u estio n h o w far o n e o u g h t to go in the in te rp re ta tio n o f th e m osaics in a late R o m a n building. T h e re is m u ch to be said for R a e c k ’s m o d erate approach. O n e o f th e prem ises in R a e c k ’s illum inating b o o k is th at th e message o f a p icto rial rep resentation had to be im m ed iately u n d e r
standable. O n e could say w ith h im that o u r b u ild in g - the B ritish-excavated part o f w h ic h is well k n o w n to h im - fits well in to a p ictu re w h ere every th in g had to be spelt o u t in letters for the n o t so classically ed u cated g en eratio n o f th e n ew elite.
R a e c k actually points to the fact th at o u r M eleager was identified to th e users o f the b u ild in g by an in scrip tio n and th at the m yth as such m e a n t very little w h ile it is significant th at M eleag er is represented w e arin g th e c o n te m p o ra ry dress o f a m an o f the R o m a n elite. Perhaps the m osaics o f th e b u ild in g w ere only m e a n t to stress the felicitas temporum in a general way.29 O n the o th e r h an d it is te m p tin g to see the b u ild in g in a fram ew ork sim ilar to the one suggested for slightly earlier buildings in A pam ea in Syria30 and in N e a Paphos in C ypros31 an d to in te rp re t th e inscriptions as telling us th at this house was in te n d e d
N O T E 25
E.g. C am pbell, 1991 no.
IV A 1: k t i'o k; and no. IV A 36 : euKapTti'a and dycopa.
N O T E 26 C h u v in 1991, 198.
N O T E 2 7
Cf. C am eron 1965.
N O T E 2 8
Poulsen Supra 193-208.
N O T E 2 9
C f. also D resken-W eiland 1991.
N O T E 3 0
J. a n d j. C h . B aity 1974.
n o t e 31 Daszew ski 1986.
N O T E 32
Cf. the description o f the palace, alluding also to the Ausonians, D id o and A e
neas as the only strangers to the w orld o f N o n n o s ’ Dionysiaka, the personifica
tion o fB e ry to s , th e old R o m a n colony, and the expression i t a T p r j i T f j i S ’, w h ic h brings Halikarnassos to the fore am ongst im p o r
tan t cities in the w orld.
N O T E 3 3
A lfo ld i-R o sen b au m &
W ard-Perkins 1980. For th e Justinian renaissance, see especially K itzinger 1951.
to serve p eo p le w h o w ere conscious o f and p a rticip atin g in c o n te m p o ra ry sp iritu al life and w h o h ad chosen an alternative to C hristianity, th e o rp h ic /n e o p la to n ic ro a d .T h e y w ere citizens o f th e w o rld and th ey reco g n ized th e R o m a n s ’ rig h t - or destiny - as rulers and p ro tecto rs o f peace in th a t w o rld .32 T h e m osaics fo u n d in ch u rch es bu ilt o r reb u ilt d u rin g th e so- called Ju stin ian renaissance, e.g. in C y re - naica, should w arn us against to o clear-cu t c o n clu sio n s.33
It is possible th a t th e stress in th e in te r
p re ta tio n o f th e m osaics and th e ir in scrip tions sh o u ld be less on religious m atters an d m o re o n th e general a d m iratio n for classical cu ltu re w h ic h caracterized the p e rio d . I f so, th e th ree m edallions w ith th e p ersonifications o f H alikarnassos,
A lexandria an d B ery to s m ig h t sim ply an n o u n c e to w h o e v e r w e n t in to th e ro o m th a t h e w o u ld be received by an ed u cated p e rso n w h o le a rn e d his re to ric in A lex
an d ria and his law in B erytos. O r perhaps it a n n o u n c e d th a t th e visitor w o u ld find here a library o f in te rn a tio n a l standard.
A nyw ay it is im p o rta n t to keep in m in d th at th e m osaics w ith th e ir in scrip tions w ere laid in a p e rio d w h e n o n e and th e sam e p e rso n co u ld w rite a g reat epic o n D ionysiaka and a versified paraphrase o f th e G ospel acco rd in g to S t.J o h n .
* T h e E nglish was revised by P eter S pring. T h e p h o to g ra p h s, figs. 2 -7 , w ere m ade by Jacob Isager.
Bibliography
A fo ld i-R o sen b au m , E. & W ard-Perkins, J.1980
Justinianic M osaic Pavem ents in C yrenai- can C hurches (M onografie di A rcheologia Libica X IV ). R o m a.
A sem akopulu-A tzaka, P. 1987
C o rp u s M osaicorunr C h ristian o ru m Vetus- tio ru m P avinrentorum G raecorum II.
Thessaloniki.
Baity, J.C h . 1981
G uide d ’A pam ee. Bruxelles.
Baity, J. 1977
Mosai'ques antiques de Syrie. Bruxelles.
Baity, J. & Baity, J.C h . 1974
Julien et A pam ee. Aspects de la restauration de l’hellenism e et de la p o litique
a n tich retien n e de l’em pereur, Dialogues d ’histoire ancien n e I, 268-304.
B ow ersock, G.W. 1990
H ellenism in Late A ntiquity. C am bridge.
B ritschkoff, M . 1923
N e u e Inschriften aus N icopolis ad Istrunr, A M 48, 93-118.
B udde, L. 1969
A ntike M osaiken in Kilikien I:
Friihkristliche M osaiken in M isis-M opsu- hestia. R ecklingshausen.
B udde, L. 1972
A ntike M osaiken in Kilikien II: D ie h eid - nische M osaiken. R ecklingshausen.
C am ero n , A. 1965
W an d erin g poets: A literary m ovem ent in B yzantine E gypt, H istoria 14, 470-509.
C am pbell, S. 1991
T h e M osaics o f Aphrodisias in C aria. To
ronto.
C h u v in , P. 1986
N o n n o s de Panopolis entre paganism e et christianism e. Bulletin de 1’association G.
B ude 1986, 387-396.
C h u v in , P. 1990
A C hronicle o f the Last Pagans. Translated by B.A .A rcher. C am bridge, M assachu
setts/L o n d o n . C h u v in , P. 1991
M ythologie et geographie dionysiaques.
R ech e rch e s sur l’euvre de N o n n o s de Panopolis. Adosa.
Daszew ski, W.A. 1986
Cassiopeia in Paphos - a Levantine going West, in K arageorghis (ed.) 1986, 454-70 and Plate 51-52.
Daszew ski, W.A. et alii 1985
D ionysos der Erloser. G riechische M ythen im Spatantiken C y p e rn (T rier Beitriige zur A ltertum skunde. B.2). M ainz.
D resken-W eiland, J. 1991
R eliefierte T ischplatten aus theodosiani- scher Z eit (Studi di antichita Christiana 44).
R o m .
H arrison, M . 1989
A T em ple for B yzantium . T h e D iscovery and Excavation o f Anicia Juliana’s Palace- C h u rch in Istanbul. A ustin.
H inks, R.P. 1933
C atalogue o f the G reek, E truscan and R o m an M osaics and Paintings in the B ritish M useum . L ondon.
Karageorghis, V. (ed.) 1986
Acts o f the in ternational archaeological sym posium “ C yprus betw een the O rie n t and the O c cid en t, N icosia 8.-14. sept.
1984. N icosia.
K itzinger, E. 1951
M osaic Pavements in the G reek East and the Q u estio n o f a “R enaissance” u n d e r Justinian, in Actes du V l’e C o n g res In ter
national d ’E tudes B yzantines II. Paris, 209- 223.
Kitzinger, E. 1951
Studies o n Late A ntique and Early B yzan
tine F loor M osaics, D u m b a rto n O aks Pa
pers 6, 83-125.
K itzinger, E. 1965
E volution stylistique des pavem ent de m osaique dans l’O rie n t grec depuits l ’ep o q u e co nstantinienne ju s q u ’a celle de Justinien in Picard, M .G . & S tern, M .H .
1965, eds., 1,341-352 and figs.1-20.
Levi, D. 1971
A ntio ch M osaic Pavements. R o m e.
M ichaelides, D.1986
A new O rp h e u s mosaic in C yprus, in K a
rageorghis, V. 1986, ed., 473-89 and Pl.s 53-56.
N e w to n , C.T. 1863
A H istory o f Discoveries at Halicarnassus, C nidus and B ranchidte. L ondon.
Pelekanidis, S. 1974
C o rp u s M o saico ru m C h ristia n o ru m Vetus- tio ru m P avim entorum G raeco ru m I. T h es
saloniki.
Picard, M .G ./S te rn , M .H .(eds.) 1965, La M osaique G re co -R o m a in e . Actes du colloque a Paris, du 29 A o u t au 3 S eptem - bre 1963, 1-2. Paris.
R aeck , W. 1992
M o d ern isierte M y th e n . Z u m U m g an g der Spiitantike m it klassischen B ildthem en.
Stuttgart.
R o b e rt, L. 1948 H ellenics IV. Paris.
R o u e c h e , C h. 1989
A phrodisias in Late A n tiq u ity (Journal o f R o m a n Studies M o n o g ra p h s no.5). O x ford.
Spiro, M . 1978
C ritical C o rp u s o f the M osaic Pavements on the G reek M ainland, F o u rth -S ix th C en tu rie s, w ith A rchitectural Surveys.
N e w York.
V ian, F. 1976
N o n n o s de Panopolis, Les D ionysiaques, T om e I (C ollection B ude). Paris.
V ostchinina, A. 1965
M osaiques g reco-rom aines trouvees en U n io n Sovietique, in Picard, M .G . &
S tern, M .H . (eds.) 1965, 315 -3 2 3 and Figs.
l-1 2 + a ,b .
W ifstrand, A. 1933
Von K allim achos zu N o n n o s. Lund.
A k erstro m -H au g e n , G. 1974
T h e C alen d ar and H u n tin g M osaics o f the Villa o f the F alconer in A rgos (Skrifter u t- givna av Svenska In stitu tet i A then 4°, X X III). Lund.
