r26

JOURNAL, f{..\.s. (( Fl'r.O-\)

(Vor. XXXVI.

t'{o.9q--1945)

GNDIGE

127

GEDIGE
By S. PanRuavtrlna
i

A number of ancient buitdings in Ceylorr, belongiug to ciilTerent periods, and situated in various parts of the island, are known tc;<lav bv the name ' Gedig6'. A ruinecl brick structure within tire citadel of Anurddhapura and the remains of a shrine of stone construction in the village of Ndla.nda in the Md.tal District, are known b-v this name, while a Gedig Viirdr:r is situa':ed near Kandy. The Galmaduva and the Larikd.-tilaka temples,
present, these tr,vo shrines are not usualll. referrecl to, in writings, by that appellation. In none of these instances is there anv evidence to show that ' Gedig6 ' r'vas the name giver:r to it by its builclers and the interpretation of this name has given rise to considerable speculation: In his detailed description o{ the Gedig6 at N6lancla, Bell has expressed his opinion that " the origin of the na.me is doubtless found in the Tamil vernacnlar gaiS,itai, or in the Sanskrit ghntika". He has also quotecl

and Krishn:r :l Sastri' rvlrich is based purer.r'on tlre ii-it"rity of sorincl, to me as prausible. on tne other hancr, th".'" h;;iil;;; ao"= nrt appear ff,h""l,,ncie't riteraturc of the island, which, i,r *i-trin_ton, concrusiverv establisires ,rtrached r{, tt,; \vn;; si[ipa br. the p;;p[;f i.;rton in :ij.:t-t'jr:l.rrce

mo'uments now known, or at one time known,- by the name of Gedig does' trrerefo'e' appear to have "u"rpoir"r.ef lhe character of the class of estabtishments known as staatikds, i" so"tt i;g1;."ir?i trr""itri"nry ,"_ sarding the origin ,]l: name GeQigi, sug.gested by Bell

bpth in the Kandy District, have been knorvn as Geq1ig6s, though, at

-rI(1ijri;u-"n:i[ayn 1n.'r.s. cai,r""l n. :o5) is cluotecr as arr iilust'atiorr. itr," si"'t"i"se explanato.v s",i,r" of tlre Rtipasiddhi, a w.rk w.hich has b..";;;l;; as an aurrority bv tire cet"brarert schr-rlar- Sri Ri.lr'lir ot tr," nli*J,lL i.ntorj:;"'ni, p"i;riia ))ra,tipn, a.'d which, from its language, seems to nave the ccrmposition of tni napaiiiii; ;;'th; beJn w;rtt;;,;;;;;;g after seconcr half of centurv,r comments o' the word Giijai;;orrii;-^;'i;it"*, ,1e twerfth ,,i:
ir"yona of doubt tlrat the word.qt(ti*i-*e.,,,, tn" Rnpasiddhi'sawte' i'e.. i,i p";il' rr.tru.",r rhe rzth ,nJ-it" ,5tn centuries, a building of i qirtjcu-:4;1p" "i "*Utecture anct tir:rt the w'rd wars considered to be eqriivalent i" J.*iis to the pali

In the Pali grammar_Rd,pasiddkr. of tsuddhappiya, tlie sentence ,, Ekait. sLunct'ya'it Bltagautr :\'rldihc uiltar,rti ciiiinai,iritr,.,j''#rlri"*,1,, rhe cfrtagosinga sutta o[ the

the opinion of Rao Bahadur Krishna Sastri, the late Government Epigraphist for India, which is as follows :* " Thr word gaiS,igai is not nolv used among the 'lamils to denote any part of a. temple. Ce{i-ge may, however, be n derivative of the Tamil ge(.i or gaSinai, the {ormer of wtrich denotes'halting place' 'rest-house' and the latter ' the alms-disli of :r religious mendicant'. i\tiy strong conviction is that the rvord has to be connected with Sanskrrt gh,atikd,, corrupted in the vemaculilrs into geQige or gaiQ,igai. 'l'he term occurs in man5r inscriptions, and alwal's in connection rvith impr:rtant centres of learning-.el1ntiha,-stlt&na being a high attribute conferred on a town. " There is also reason to believe that theseg/za.tikds were attachecl to temples. 'Ihe town Kairchi, for instance, was a glttttika,-sth,ar'ta. Sholingur, a station on S. M. and M. Railwarr, is still referred to in Sanskrit worl<s as Ghatihdchalo", evidently meaning a place llt which, on a hili, was once locatecl an institution of the kind in question. I have mentionecl in m1' Repc-,rt for rgro-rr (page 5t, pu.og.iph 7) 'the ghatihci o{ IJrahmanas'.

je(igetti., (rneans) in thc buircling which ;;s ."'*l*.r"a entireh of brick, th*t is "r'his.refe;;; to s..v in rhe Gedigd ". 'r""^pilrrrririry ;;r"brist"r

Ciiijahduasalhe. sivnltt ttlu-nttru|i !elti

ltetal

,.Gifijak,iasollrc

ift

,iL" tl;;'Ji;;";.itll,"J.i

" 'lhe terni occLLrs ir:r the'lalguncl inscription of abciut the sixth centur-\, A.p. Frofessor- Kielhorn (E.pi,graph.ia Indiaa,p. 35)' is of opinion that it denotes an establishment (probablv founded in most cases bv a king) for holy and leamed men. There catn, therefore, be little doubt tiiat the

ruined' gaid,igai' at Ndlanda does represent an institution of the kind "." The Gedige at Nalanda, though it is buiit in the Pallava st'n'le of architecture, is, in l.rll probability, a Buddhist shrine. 'Ihe ruins of a strlpn by its side prove this fact. The Gedig at Anur5.dhapura is a part of the Sinhalese royal estzrblishment and the Gedige at Kanclv is a Buddhist shrine built in the lifteenth centur_v'. I-anka-tilaka is a Bucldl'rist shrine built in the {ourteenth century while Galmaduva Vihdrir dates, according to tradition, from the reign of I{irtti Sri Rljasiri-rlia. None oi the

Tlre explanation of.^tbe_ ylord Gifija.kytas_atln b-v the author of the Riipasiddhi-san*e is in agreement ivith that of Budclhaghosa, who irr his cc-,mnrentar\. toJh!-Viijii""'"ithiya_,,*tates: Gifr,jukduasathe. ti irthahamaj'e ai'asa!rte. m"t*ii''nir)"'o,',,rri'st,og^ii,ill,air'r'llrlircot,n,;, liaronlo vaj.jilq.tlhc carihan't, car*ntaito""'Nadirrn,n,)tltth^nttn \r t,',tsitto wanuisd Eitosuuttt.o ntahu,t,Trtni joi,i aiii'i*'ln,i,iil!i,li,',',?o h;ur;'!;:,';_ ltaduyi Sarthu i,trsuttattrtdnarit harissiii,T:i" t,mtttctra i*/tnhdrtira bhitti_ . sofidnn-rltambha-rirantpartrtti dassritii iiraiii iri)u'';',,ii'ii,,'r"'ri*prtr,, tnaldhantma-la!akarrndtrt,i y;y:li,ijrira ''"iii,,,,*nttt,nvatttr-tnctii'ctr_ftttltddltti pr,r'iiapetvti Sattlw niyt,qtesurix.i ,;.'i;;;(i;r"a'sathe built of bricks'. onie upon a time irre-"g'i;ecl nteans, in the i-esidence one, to.ring the vajji courrtr\- in trrc c"lll'se of l,ris ptttrt;c f iirr peopre of Naclika gave.arms i" ti;; ,,,iniitr-rtrrrn. arriveti at \adir<ri. R;dha, rirt"n.Iio iiJaur"*5,, trr the l)lrammrr . rn.i b"ing,,rtf,"i,""iii'i,"rrr. t,,rk corrrrslJ irrnong tlremsel'cs (and res,,r'ccr)..th-at ttre-i'lrlrrrt,t.t' const'rct a prircc of resi_ ttence I,r the Te;rcher ii-; "iiin.. -n",ui,ig "*r)rr," nigr,,, -i.;a ," r'I steps, pillars anrl fisures of )-,,,,/,i.'c,,,i-,.r.t",r of brick rrlop1.. plxsf6spll tt witlr lime, comoret"x it= o..oi"iiu,''r'riin'i*t;e"r rf wirh carpets, r,.ar,' ln"t;:;, ";;.., andflowers ",,i[r""p"r., cteclicated it to the +:::;::+,t,

Gifi,jakavnsathct,.

:;,r"o:!:;1!Xl:.i"".T:lT*,:I#Ty,f -"}if ::-to-tlrethirteenthccntur-y.butinarl 'ii:'1".i;t',:fii3t It'a""srro"n.irri;d;;#;?i::rT;';^:'{iX";r3!:k1i;, ,';*.,i;j.1,;';E i{l:;i^T)i:,iijiii;.Y;:;;;'\,;:'{t',J',',': {l

t. Vol. VIIf ? -fhe reference to the Volnme does not occur in Beil's report z. Annual Reporr o.f tlre Arch.eeological Suraey oJ Ceylolr, rgro-rr, p. -5o.

p,

164.

Eclition. part z. p. zj-5.

--a

JOLrR\rAr-, R.A.S. (CEYLON)

(Vor. XXXVI'.

(io. 99---.rqa5)
nneans ' residence

GEDXGE

1-29

ancient iomrnentaries which contained exegetical materia.l going bach Sinha.lese io a stiil etrriier age. The tradition of the architectural character of the Gifiiakavasatha wliich he has embodiecl in his commentary rnay, therefore, Ou,j". which rvas bn,rught to Cc1'l,rn bv the ealliest Rrtddhist inissionaries rpnro can-le to tiris isiancn frc'rn Nr'rrth Inclia. Frclm the clescriptierns of p:rlaces and other important edifices occurring irn the Fifakus,,,rie geis tG impression that the architecture of ancient l^al" a*furg the tirie of tl.re tsuciclha rvas essentially wooden. This is ,fu,t tfr" coiclusion ttrat has beei.r arrived at bV an examination of the

tsudclhagtrrosa rvrote in the fifth century, but he utilised

the

earlies'} monumerltal remains in India. But the Gifljakdvasatha seerns t r have belongecl tr.t, a cliff erent type of architecture. As it is said to harte U."m .otrutr1ritecl entirei,v of brick, it rnust have he"d a vaulted roof. The hrick vault, as an archltecturatr feature, seems therefore to harte been knor.vn in India in the time of tl-re Buddha. No actual example of a Uo;.L i"u"ft clating from such an early period has been.prese-rr.ed in India, L,"i ilr" ren'ra.ins at }lohenio-claro sl]oiv that the principle of the corbelled' lu". knorvr-r to tlie huilclers of the hidus Valley in the third milleniurn

oy rlie jnfluence of an v in the pneceding or'following .viri,n]". ft,1, ln the present instance. ]rrstalrces, Iiowever, are not atrtogether wanting in which a d has been cerebralisecx withoiJ socn ar, influence. compare_for instance sintraiese Qah,a .[.or p. and s{t-. aana Jnr the compound daha-diyo nrcdern (.dd{ta' sweat ' ;
reature_rs wantlng Tn rnodern sinhalese,thg word gediga is arso applied to an . orntrmenta,l pcng*la or kiosk ernbellished by-flowers ana fiuits'. Tiresc temp,,rary 1]rlictgre;, put up for festive occasions rvith bamboos, areca_Dal'ms or tright tim-ber, are usually hung with bunches of coconuts and ottei rruit.

gifi,jak_d.aasatlatz, the se,i:ond nremtrer d:uasatha or ' house ' and has the sanre *g"inc""." ,, l;"'i, ttu" sinhalese ge!.i'ge- Pali giii.iaka,. tl'rerefore, corr.sp.rnd, ,o.cratl" si""halese. ovr_the analogy o{ th9 Sanskrit kifijarka, p. h;iijahkha,'rreco*;ne r.r?a; in sjntralese, P. gifija-ka can assurne thc fornr .i:ia;. tn'ts-ir*i".in gi"" rise to the^form sed! by the cereirralisation of d, rvhich u.""rtu i, ca,rsea

ln the Fali.cornpound \rrord

'

"uuti B.C.

clf the Ge<iig6-r knr,rwr-l ttr erist in cer'lon, the monument of that narne to be ,.eeru at-Auuradha.pura corresponds rtery rvell to thc definition of
'.irl.-rr3o*i-frnn

in rlle Iliitltosiddhi-sottne. .\pari from the door-1rames, es and steps., rvhich are of stone, the building is .entireXy *f Url.fo construction, woocl having hacl no place in the architectural ."fr.*., even for tire roof. The sprlnging of the dome which roofed the erlifice can be noticed ire tXre portion of the building still preserve.d- The i-mfia-tifnf," is aXso oi'brick*construction, save foi the base r'vhich is of ;;;;;. Of the Gaknaduva tempie the r:ault is of brick construction wtiitre
pcdiuE'gir-en

the

brase

is

r..rf stone.

t-lruiliur" appears that tUe terin. teSigA r,vas -first app]ied to brick-built but, at a. later <late, the distinguisl'ring characteristic

C*+rgC at

i['he other

tu,'er Gecliges are built of stoire, but in con1mo_ll rvith tire AnrLrradh*1riotu, thel'' are.provided with vaulted roofs' It

"r"fi"A =ttlitot"r to be the vaulted constmction of the roof and not the a{ i geSige was ta}ieit rnateriai c,f wlaich the edif,ce rvas built. The application of tlie term rvas in*r.it""a"d to vaultecl edifices of stone, as rvell as of brick construction' were

(]ed'i in Sinhalese means 'fmit' or 'nut' a'd gertigE n ;Itoii-to"..'. llhe terrn rrray therefore be a modern one inventecl to suit the nature of tlhese ornarnental structures and have nothjng to do with the tural ter-rn geQ,igE- loand in the R$,pasiddhi-siiute, the irJentitv^r.Ht".in forrn "between the.trvo lreing punely accidental. On the other hand, tLe re.rense i"s a,lso possible. llvery s$ige is not necessarily hung witrr'buncrles of fruit. sorne are decorated with tender reaves orf th" irrto various onirarnerrtatr designs; briglrt colorrred tissue papei oiid.lotr, "i.on"t fnr*l*"au lra,,'e entered rno.e and nrore irtn rhJdecorarir.e scherne of'thesc qe(igts rn recent tirnes. llsselitiaily, a motlern gcrligi is o* orno**niol of a ternporary nature. It is possib,le that ^n.tr arches of this type, 'nialneiltal a. perrnanent character, exilted in ancient days, ir"l front of t"*01". ;;a "rf palaces.-,The ter*r *sed to bnildings ot the same type as the Adunadha'pura ge{'igE- could have been extended to thern. W}len tenaporuru o**r,'rrental arches were constructed tlaey were also givell the surri* ,rnire ,ryhen the original me-aning of the woid trad "rd been"forgotrera, loilr-etiiiotogy interpreted the word to rnea' 'fru.it-house' ar;c1 f,he o"itor' ilr;^-rrlgi::aterl of decorating such structures with tnrr:ches ot' fruit, ira or,ler to lustrfy the narne as it was populanly r,indenstorid.

.ancient edifices in cel'I"n, rvhether tirerr '*vere of brick or of stone, ..u:"*lfu pla.stered ove; i1lld tbre nraterial of their construction vrould not lon"o U.;" eviCent tt.r tire people rvho saw tirem when thev were in a- good it^te of n-laintenallce. The eiterior forrn of an edifice, on the other hand, *ii"ianot'" been the characteristic to be r-roticed more than anvthing else as the ;rnttr it rnust have had as much to clo in tlre applicartion of narnes .lvhich it r,vas [tui]i. we p-lav, therefore, conclude that a materlat of C"eGe at first signihectr a brick-built eriifice rvith a vaulted roof and later ,,-r, ilriilcline of t-'trat architectural forna, rvhether built of stone or brick. 'ilr'* rn*" therelore trlas nottrring t,l clci *'ith the purpnse rvhich such a

bulldlnq sel-YeG. had tlrc sittrte siStrific;ince ll I i- llrus tleor :1,-i lhc:inl,al.:.e tt rlrqcrlige it * e.ll g;Ai*,taussalltt:r.. it is, hor,vever', nc,t so certrtin whether ^, -(i;r11,xl;..'lvrirci is etynroJiosiciiXl-r' connectecn rvith its Pali cr-runterpart' li-1"