349Arheološki vestnik 68, 2017, str. 349–386 Začetki zgodnjeslovanske poselitve Prekmurja Beginnings of the Early Slavic settlement in the Prekmurje region, Slovenia Daša PAVLOVIČ Izvleček Na najdišču Nova tabla pri Murski Soboti sta bili iz vzorcev oglja v jamah, poimenovanih SZ 6 in SO 149A, pridobljeni 14C-dataciji, ki nakazujeta, da je bilo najdišče poseljeno že v prvi polovici 6. stoletja. Lončenina iz polnil omenjenih jam je primerljiva s t. i. lončenino praškega tipa s hrvaških, slovaških, čeških, poljskih, nemških in ukrajinskih najdišč. Na Slovaškem in v Nemčiji so jame in grobovi s primerljivo lončenino datirani tudi s pomočjo 14C-datacij, na ukrajinskih najdiščih pa s tipokronologijo kovinskih predmetov in bizantinskimi novci. Zgodnja naselitev nosilcev praške kulture, najverjetneje zgodnjih Slovanov, na zahodne obronke Panonske nižine je utemeljena tudi s prostorsko razporejenostjo poznoantične in langobardske poselitve v 6. stoletju, z jezikovnimi analizami in s prvimi omembami Slovanov v Panoniji v pisnih virih. Ključne besede: Slovenija, Panonska nižina, Prekmurje, Nova tabla, zgodnji srednji vek, zgodnji Slovani, lončenina, datiranje 14C Abstract The site of Nova tabla near Murska Sobota revealed the remains of an early medieval settlement. These remains in- cluded the sunken features SZ 6 and SO 149A, charcoal samples from which have yielded 14C dates that indicate the site was already inhabited in the first half of the 6th century. The associated pottery is comparable with the so-called Prague type pottery unearthed in Croatia, Slovakia, the Czech Republic, Germany and Ukraine. In Slovakia and Germany, sunken features and graves with such pottery have also been dated with the help of 14C analyses, while the sites in Ukraine are dated on the basis of typochronological analyses of the metal artefacts and finds of Byzantine coins. Such early settle- ment of the western fringes of the Pannonian Plain on the part of Prague culture groups, most likely corresponding to the Early Slavs, is supported by the spatial distribution of the Late Antique and Langobard settlements in the 6th century, by the results of linguistic analyses and by the first written records mentioning the presence of Slavs in Pannonia. Keywords: Slovenia, Pannonian Plain, Prekmurje, Nova tabla, Early Middle Ages, Early Slavs, pottery, 14C dating 350 Daša PAVLOVIČ UVOD* Sodobna arheološka izkopavanja na trasi avto- ceste med Mariborom in Lendavo so med letoma 1998 in 2008 odkrila kopico novih najdišč zgo- dnjesrednjeveške poselitve med 6. in 9. stoletjem. Sistemsko urejeno financiranje je omogočilo, da je bilo precejšnje število vzorcev oglja iz polnil odkri- tih jam datirano z analizo radioaktivnega izotopa 14C, ki je med najbolj razširjenimi naravoslovnimi metodami za datiranje arheoloških kontekstov (od tu 14C-datacija ali radiokarbonsko datiranje). Največ 14C-datacij analiziranih vzorcev oglja izhaja z najdišča Nova tabla pri Murski Soboti.1 Vsi prikazani razponi 2σ kalibriranih datacij se gibljejo med letoma 414 in 941. Datacije se v grobem delijo na starejše, z razponi med letom 414 in sedemdesetimi leti 7. st., ter mlajše, z raz- poni od sedemdesetih let 7. st. do leta 941. Tak časovni okvir ustreza zgodovinsko-arheološkemu razlagalnemu modelu, da so se nosilci materialne kulture, za katere menimo, da so zgodnji Slova- ni, na območje današnjega slovenskega prostora priselili po iz pisnih virov izpričanem odhodu Langobardov iz Panonske nižine v severno Italijo leta 568, najverjetneje v sedemdesetih ali osem- desetih letih 6. st. V uvodu bi bilo morda smiselno opozoriti tudi na (pre)nekatere probleme, ki spremljajo vsakega raziskovalca zgodnjesrednjeveške poselitve, pred- vsem tiste, ki jo povezujemo s Slovani. V prispevku v tem smislu kar mrgoli terminov, formulacij in tem, ob katere se lahko tisti z drugačnim pogledom obregnejo ali celo diskreditirajo tu predstavljeno argumentacijo. Če se ognem sami problematičnosti opredeljevanja začetka zgodnjega srednjega veka, nam ostaneta dva temeljna sklopa problemov. Pri poskusih razreševanja teh sta se oblikovali dve vidnejši struji, ki še zdaleč nista edini. Ena temelji na pogledih Florina Curte, druga na modificiranih in dopolnjenih pogledih Luborja Niederleja, ki jih * Doseženi raziskovalni rezultati so nastali v okviru nalog, ki so bile (so)financirane iz državnega proračuna po pogodbi med Javno agencijo za raziskovalno dejavnost Republike Slovenije in Narodnim muzejem Slovenije (raziskovalni program P6-0283). 1 Najdišče Nova tabla pri Murski Soboti je bilo raziskano v treh izkopavalnih sezonah (1999–2001, 2002–2003, 2007–2008). Odkritih je bilo 193 jam in 13 grobov, ki pripadajo zgodnjesrednjeveški poselitvi prostora južno in vzhodno od umetnega jezera Kamešnica pri Murski Soboti (Guštin, Tiefengraber 2002; Guštin 2003; Pavlovič 2008, 2013, 2015). v večji ali manjši meri upošteva večina vidnejših raziskovalcev zgodnjega srednjega veka v Sloveniji, na Češkem, Slovaškem, Poljskem, Hrvaškem, v Ukrajini, Rusiji in Belorusiji.2 Prvi sklop problemov se osredinja na predstave o tem, kdo so bili zgodnji Slovani, in s tem povezano vprašanje, od kod so prišli. Curta meni, da Slovani niso etnična skupnost, ki se je priselila z območij današnje zahodne Ukrajine in južne Belorusije (Curta 2001a, 337), temveč je nastala kot proces urejanja socialnih razlik znotraj skupnosti. Ta proces formiranja elit ter politična in vojaška mobilizacija pa naj bi bili odgovor na zgodovinske pogoje, ki jih je ustvarila utrditev meje na Donavi (ib., 343–344). Samo ime etnične skupnosti “Sclavene” naj bi bil čisto bizantinski konstrukt, oblikovan kot potreba po ureditvi zapletene situacije različnih etnij na drugi strani severne meje imperija. Slovanska etnija naj bi bila po njegovem mnenju iznajdba Bizantincev (ib., 118–119). Raziskovalci širše sprejete struje menijo, da Slovani izvirajo z razmeroma majhnega območja v zgornjem porečju Dnepra, da je njihovo samorazu- mevanje temeljilo na jeziku ter da so se razselili in do konca 8. st. poselili območje vse od Hamburga v današnji Nemčiji do Peloponeza v današnji Grčiji (Pleterski 2013b in tam citirana literatura). Drugi sklop problemov je povezan z vprašanji, ali so Slovani pustili materialne sledi, so te sledi arheološko prepoznane ter ali lahko specifično materialno kulturo povezujemo z zgodnjimi Slovani (o tem tudi Pleterski 2013a, 207). Poleg perečih, vendar splošnih, teoretskih arheoloških problemov, kot so povezovanje arheoloških kultur s posame- znimi ljudstvi in razumevanje materialne kulture kot izraz identitete, ki se jih tukaj ne bom lotevala, se ta sklop osredinja predvsem na vprašanje, ali je t. i. kultura lončenine praškega tipa (tudi praška kultura), ki jo je leta 1940 opredelil Ivan Borko- vský, res arheološka sled zgodnjih Slovanov ter kaj točno sploh predstavlja to kulturo. Kot pove že ime kulture, je bila kot njen naj- značilnejši element določena lončenina, predvsem prostoročno izdelani lonci “vazastih” ali “sodčastih” oblik brez izrazitih ramen ter s kratkimi pokončnimi ustji. Druge karakteristike te kulture so še izbira 2 Poglobljene nove študije o izvoru in poselitvi zgodnjih Slovanov od Rusije do Sredozemlja, temelječe tudi na naravoslovnih analizah ter posvečene Luborju Niederleju, so zbrane v posebni izdaji revije Stratum Plus (Rabinovich, Gavritukhin 2015). 351Začetki zgodnjeslovanske poselitve Prekmurja prostora naselitve ob rekah, nezavarovana naselja s polkrožno organiziranimi vkopanimi hišami s kamnitimi pečmi, grobišča z majhnim številom žganih grobov v žarah ali preprostih jamah. Te lastnosti se le na nekaterih najdiščih pojavljajo vse skupaj. Navadno prepoznamo ali odkrijemo le dve ali tri na najdišču (Profantová 2012, 255–256). Na Novi tabli je, na primer, prisotna praška lončenina, naselbina je nižinska in nezaščitena ter v bližini več manjših vodotokov, hiše so bile najverjetneje razporejene v vrstah, nekatere skupine pa polkro- žno s skupnim dvoriščem. Odsotne so značilne kvadratno oblikovane zemljanke z obokanimi kamnitimi pečmi (glej Pavlovič 2013, 135–139). Praška kultura naj bi se oblikovala v 5. st. ozi- roma po vdoru Hunov v Evropo leta 375. Huni so s pritiskom na germanska ljudstva ob obalah Črnega morja ustvarili razmere, da so se na obmo- čjih kijevske in černjahivske kulture formirale tri zgodnjeslovanske kulture, med njimi tudi praška (Pleterski 2013b, 21–22). Po zgoraj navedenih karakteristikah so si sorodna najdišča v Ukrajini, Belorusiji, Moldaviji, Romuniji, Poljski, Slovaški, Češki, Sloveniji, Hrvaški ter Nemčiji, in ker so tista v Ukrajini in Belorusiji datirana najzgodnejše (konec 4. st.), predvidevamo, da so od tam nosilci te kulture migrirali proti jugu in zahodu konec 5. in v 6. st. (npr. Profantová 2012, 256; Gavrituhin 2009, 11). V 6. in 7. st. je v večjem delu centralne Evrope in tudi Panonske nižine mogoče arheološko slediti poselitveni diskontinuiteti in kompleksni transformaciji. Na območjih te diskontinuitete ter prvih pisnih omemb Slovanov se pojavi pra- ška kultura (Profantová 2012, 255). Preprostost te kulture je težko pojasniti zgolj kot tehnološko nazadovanje preživelih poznoantičnih skupnosti, verjetneje je, da gre za priliv novih skupin ljudi s tehnološkim znanjem na tem nivoju (Pleterski 2013a, 207). Analize pisnih virov (Fusek 2008; Bratož 2014, 481–503), primerljivost gradiva med najdišči Ukrajine in centralne Evrope ter kontinu- iteta od praške lončenine do keramike gradišč 8. in 9. st. na Češkem, ki jo historično zanesljiveje povezujemo s Slovani, utemeljujejo tezo, da je praška kultura arheološka sled zgodnjih Slovanov (Profantová 2012, 260). V zadnjih dveh desetletjih se je pojavil izrazito drugačen pogled tako na obstoj praške kulture kot na njeno povezovanje z zgodnjimi Slovani. Očitkov je veliko, od metodološke neustreznosti pri kla- sifikaciji praške lončenine, intuitivno določenih tipov, selektivnega izbiranja območij raziskovanj do nerazumevanja in nekritičnega branja pisnih virov. Zaključki Curte, na katere je za ta prispevek smiselno opozoriti, so, da je (relevantno) arhe- ološko gradivo s slovenskih in hrvaških najdišč mogoče datirati šele v drugo polovico 7. st. ter da ni pregledne in čiste kronološke razlike med pra- ško lončenino in lončenino, okrašeno z valovnico, katere pojav naj bi bil kasnejšega datuma (Curta 2010, 34–35). Poglobljene in utemeljene kritike tez Florina Curte podajajo tudi drugi raziskovalci, tako arheologi kot jezikoslovci in zgodovinarji (med drugimi Biermann 2009; Profantová 2009; Pleterski 2009; id. 2013b, 22–23; Snoj, Greenberg 2012, 283; Bratož 2014, 484–487). V tem prispevku izhajam iz dveh 14C-datacij z Nove table, katerih razpon 2σ se zaključi leta 543 oziroma 547. Predstavljeno in ovrednoteno bo arheološko gradivo, ki je bilo odkrito skupaj z datiranima vzorcema. To primerjam z gradivom najdišč zgodnjega srednjega veka s čim širšega območja oziroma s tistimi, katerih objave so mi bile dostopne. V zadnjem delu članka sta v razpra- vi podani interpretacija in utemeljitev možnosti obstoja skupin zgodnjih Slovanov v Prekmurju že v prvi polovici 6. st., tako na podlagi arheoloških primerjav, razprostranjenosti gradiva v prostoru, zgodovinskih okoliščin, pisnih virov in jezikovnih analiz. Kot bo pokazano v nadaljevanju, podrobna analiza gradiva z najdišč v okolici Murske Sobote3 in številne 14C-datacije ter pisni viri potrjujejo, da so nekatere skupine, ki jih historično in lingvistično povezujemo s Slovani, migrirale z območij Ukra- jine, posledice teh migracij pa lahko zasledimo na obronkih Panonske nižine že v prvi polovici 6. st. NAJDIŠČE NOVA TABLA PRI MURSKI SOBOTI O materialni kulturi prvih priseljencev, ki jih povezujemo s Slovani, je bilo na ozemlju današnje Slovenije znanega sorazmerno malo. Takšna slika se je spremenila ob koncu 20. in prvih letih 21. stoletja, ko so na območju severovzhodne Slovenije potekala obsežna izkopavanja na trasi avtocestnega križa. Južno od mesta Murska Sobota je bil odkrit cel niz arheoloških najdišč, ki so razkrivala do takrat slabo poznano nižinsko zgodnjesrednjeve- ško poselitev. Največje raziskano območje, kar 40 hektarjev, je najdišče Nova tabla pri Murski Soboti 3 Pregled arheoloških raziskav zgodnjesrednjeveških najdišč Prekmurja: Guštin (ur.) 2002 in id. (ur.) 2008; Kerman 2011. 352 Daša PAVLOVIČ (Guštin (ur.) 2008 in tam citirana literatura; Pavlovič 2008; 2012; 2013). Poleg ostalin od eneolitika do novega veka je bilo iz obdobja zgodnjega srednjega veka odkritih 193 jam, ki jih pripisujemo naselbini (sl. 3), in 13 skeletnih grobov. Jame so bile razporejene v tri večje skupine. Največja je ležala v centralnem delu izkopnega polja, manjši dve zahodno od nje. Jame v manjših dveh skupinah so bile razporejene ob nekdanji (arheološko potrjeni) strugi potoka Dobel. Grobovi so ležali ob eni od rimskodobnih grobnih parcel, obdani z jarkom, nad parcelo je bila morda nekoč nasuta zemljena gomila. Natančnejša opredelitev jam kot bivalne hiše, odpadne jame ali gospodarskih objektov je težavna zaradi neznačilnih oblik jam ter majhnega števila ognjišč in peči, ki so najbolj zanesljivi pokazatelji funkcionalnosti jame (o tem podrobno v Pavlovič 2017). Glavno prepoznavno znamenje za umestitev v čas od zadnje tretjine 6. do 9. st. so bili odlomki lončenine z značilno luknjičavo površino ali z vrezanim okrasom snopov valovitih in ravnih linij. Kronološki okvir je dopolnil velik nabor 14C-datacij (Guštin, Tiefengraber 2002; Guštin, Pavlovič 2013, 151; Pavlovič 2015, 61). V zgodnjesrednjeveških naselbinskih jamah Nove table je lončenina najštevilnejše gradivo in tudi edino, ki dopušča nadaljnjo kulturno in časovno opredelitev, četudi povsem grobo. Na Novi tabli je bilo odkritih 6038 odlomkov (95 kg) zgodnjesrednjeveške lončenine. Vsa pripada t. i. grobemu kuhinjskemu in shranjevalnemu posodju, med katerim prevladujejo lonci, v veliko manjšem številu se pojavljajo pekači. Prostoročna lončenina je sorodna in primerljiva s t. i. lončenino praškega tipa oz. praško kulturo, ki jo povezujemo s priho- dom prvih slovanskih rodov v srednje Podonavje (Fusek 2008, 645; Fusek, Zábojník 2003; Kuna, Profantová 2011, 415). Naselje na Novi tabli ter večina naselij, ki jih povezujemo z zgodnjeslovansko poselitvijo, tako v zahodni Panonski nižini kot na preostalih območjih centralne in vzhodne Evrope, so nižinska, s spe- cifičnimi ostalinami vkopanih delov naselbinskih stavb in jam. Objekti oz. jame so navadno razpo- rejene vsaka sebi in se razen v redkih primerih ne prekrivajo. Najdišča zgodnjeslovanskih naselbin nimajo kompleksnejše stratigrafije,4 ki bi vzpostavila 4 Na problem neprekrivanja in s tem na pomanjkanje stratigrafije pri nižinskih zgodnjesrednjeveških naselbinah so opozorili tudi raziskovalci velike naselbine v Roztokách pri Pragi, kjer sicer poznajo nekaj primerov prekrivanja relativne kronološke odnose med objekti in s tem tudi gradivom. Jasno se je na Novi tabli pokazala le t. i. horizontalna stratigrafija. Objekti s starejšo lončenino so bili v centralnem delu izkopnega polja, z mlajšo pa na zahodnem delu. Sklepamo lahko, da se je poselitev premaknila proti zahodu (Guštin, Tiefengraber 2002, 62; Pavlovič 2012, 322–327). Časovno lahko prepoznamo dve fazi naselbine (starejšo in mlajšo), med katerima se je morda zgodila krajša prekinitev poselitve na tem obmo- čju. Fazi sta bili prepoznani na podlagi razlik v lončenini (Pavlovič 2015; 61, sl. 2–4), vidna pa je tudi razlika izbire prostora poselitve (Pavlovič 2012, 325). V starejšo fazo so umeščeni predvsem objekti v največji skupini jam, s posameznimi bolj od- daljenimi, satelitskimi objekti. V mlajšo fazo pa predvsem objekti v približno enako velikih dveh zahodnejših skupinah (Pavlovič 2015, sl. 5). Dve časovni fazi sta jasno izraženi tudi v razponih 14C-datacij (Pavlovič 2015, sl. 6). Nova tabla: 14C-datacije Na najdišču Nova tabla so iz 34 (19 %) objek- tov pridobili 36 vzorcev oglja (sl. 1) za datiranje po metodi 14C.5 Vseh 36 datacij izhaja iz vzorcev oglja iz polnil jam, ki so bile med izkopavanji na podlagi primerjave lončenine s tisto s čeških, slovaških in poljskih najdišč začasno opredeljene v zgodnji srednji vek. Vzorčeno oglje je zelo verjetno pripadalo posa- meznim stavbam ali lesu za kurjavo, vendar način njegovega nastanka v večini primerov ni jasen.6 Zavedam se, da je kvaliteta 14C-datacije odvisna od zanesljivosti arheološkega konteksta, iz katerega je bil vzorec vzet, “čistosti” (nekontaminiranosti) dveh ali celo treh zemljank: 8 primerov, kar je glede na intenzivnost poselitve prostora še vedno zelo malo (Kuna, Profantová 2005, 208). V prekrivajočih se objektih je bilo gradiva malo in razlike med posameznimi skupki majhne. Le v enem primeru je v mlajšem objektu večje število odlomkov, ki pa pripadajo le enemu rekonstruiranemu okrašenemu lončku in enemu “klasičnemu” lončku praškega tipa (Kuna, Profantová 2005, 347), tako da ne ponuja možnosti natančnejše relativno kronološke razvrstitve. 5 Vzorci so bili analizirani v nemškem laboratoriju (Leibniz Labor für Altersbestimmung und Isotopenforschung, Christian-Albrechts-Universität Kiel). 6 Pomembnost tafonomije konteksta, v katerem leži vzorec, je v literaturi, vezani na 14C-datacije, čedalje bolj izpostavljena (npr. Bayliss 2009, 129; Pleterski 2010a, 86–87). 353Začetki zgodnjeslovanske poselitve Prekmurja ohranjenih hodnih površin s tlaki ali plasti fito- litov, je bilo vzorčeno le oglje, odkrito v polnilih jam. Vzorci so bili vzeti povsod tam, kjer je bila količina oglja dovolj velika. Čeprav je oglje manj zanesljiv vzorec, ker je navadno razpršen v plasti z veliko možnosti intruzivnosti, pa lahko ob pre- gledu mikro in makro konteksta zagotovi relativno zanesljiv vzorec in s tem datacijo (Boaretto 2009, 278). Makro kontekst predstavlja plast, v kateri leži oglje. Razumevanje nastanka te plasti od- kriva tudi kvaliteto in zanesljivost vzorca. Mikro kontekst predstavlja lokalno okolje vzorca znotraj plasti. Izključiti je treba vzorce, ki bi lahko bili kontaminirani zaradi bioturbacije, predvsem ob rovih glodavcev, ali človeške intervencije (ib., 278). Na Novi tabli je v zvezi s tem nujno opozoriti na nekaj problemov. Ključen je ta, da je bilo najdišče izpostavljeno intenzivni poljedelski obdelavi. Tako so vsi zgornji deli jam poškodovani z oranjem. Edini deli, ki v tem smislu, po končnem zasutju jam, niso bili kontaminirani, so polnila čisto na dnu jam. Drug problem je, da so redko ohranje- ne večje konstrukcije, ki bi omejile čas nastanka oglja, kot so strešne konstrukcije objekta, peči ali kurišča z ohranjenim zoglenelim kurivom ali plasti podrte zgorele hiše, skupki zoglenelega žita v silosu in podobno. Tretji problem je, da natančna lega odvzetih vzorcev oglja v polnilih ni zabeležena. Vzorci z Nove table so oglje znotraj polnil, ki ne pripadajo strjeni plasti ali večji konstrukciji, zato je očitek, da gre za star les, vedno mogoč. Na podlagi poznavanja izkopavalcev in nji- hovega znanja, predvsem pa iz izkušenj, ki sem jih pridobila kot namestnica vodje izkopavanj v sezoni 2007/2008 ter med osebnim sodelovanjem pri vzorčenju jame SO 149A, predpostavljam, da je bilo oglje vzorčeno tam, kjer je bila njegova ko- ličina največja, in nikoli iz zgornjega dela polnila, kjer je velika možnost bioturbacije ali premešča- nja zaradi poljedelske obdelave. Kljub naštetim problemom mi številčnost7 in, kot bom pokazala v nadaljevanju, homogenost pridobljenih datacij vlivata zaupanje, da je bilo vzorčenje dobro in so iz njega pridobljene informacije verodostojne.8 7 “Ker celoten postopek gradi na množici podatkov, s številom vzorcev narašča zanesljivost opredelitev.” (Pleterski 2010a, 86, poglavje: Datiranje skupin ustij z metodo radioaktivnega ogljika 14C). 8 Zavedam se, da je oglje, pridobljeno v tafonomsko nejasnih okoliščinah, težko opora za gradnjo sistematičnih kronologij (prim. npr. Boaretto 2009, 276). Sl. 1: 14C-datacije (razponi 2σ) iz zgodnjesrednjeveških objektov najdišča Nova tabla pri Murski Soboti (prilagojeno po Pavlovič 2013, sl. 76). Fig. 1: 14C dates (2σ ranges) from the early medieval sunken-featured buildings at Nova tabla near Murska Sobota (adapted from Pavlovič 2013, Fig. 76). vzorca ter od ustreznosti in natančnosti analitične metode (Boaretto 2009, 275). Zanesljive arheološke kontekste za datiranje pomenijo npr. tlaki iz malte ali plasti ometov, plasti fitolitov in prižgani ostanki hrane znotraj celih posod ter kosti. V skladu s prakso in vedenjem ob izkopavanju najdišča Nova tabla med letoma 1999 in 2008 ter glede na specifičnost najdišča, ki ni imelo 354 Daša PAVLOVIČ Nove raziskave in morebitne drugačne analize bodo to potrdile ali ovrgle. 14C-datacije z Nove table sem kalibrirala po enotni kalibracijski krivulji.9 Pri njihovi obravnavi sem upoštevala razpon 95 % verjetnosti (2σ). Upo- raba ožjih datacij je namreč strokovno sporna in lahko vodi v velike napake in napačne zaključke (Michczyński 2007, 401). Za en vzorec se je izkazalo, da pripada objektu rimske dobe.10 Ostale 14C-datacije so prikazane na sl. 1. Razponi datacij segajo od začetka 5. do začetka 10. st. Datacije se delijo v dve skupini (sl. 1). Dvajset datacij ima starejše razpone, 15 pa mlajše; nekaj razponov datacij se le malenkostno prekriva. Mejno obdobje so sedemdeseta leta 7. st. Pri prvi skupini lahko opazimo dve podskupini. Prva zajema 8 datacij, pri katerih se razponi zač- nejo v prvi polovici 5. st. (SO 149A, SZ 6, SO 75, SO 161, SO 105, SO 149, SZ 10 in SO 11). Druga podskupina vsebuje datacije s krajšimi razponi, med drugo polovico 6. st. in sedemdesetimi leti 7. st., ter zajema 12 vzorcev (SO 169, SZ 2, SZ 11, SO 73, SZ 1, SZ 9, SO 47, SZ 3, SZ 4, SO 60 × 2 in SO 58). Razponi podskupin se pri večini datacij deloma prekrivajo, zato je sočasnost objektov, iz katerih izvirajo vzorci, možna. V drugi skupini so datacije z mlajšimi razponi. Pripada ji 15 datacij (SO 164, SO 34, SO 44, SO 39, SO 156, SO 144, SO 33, SO 43, SO 155, SO 141, SO 166, SO 168, SO 40, SE 189 in SO 46) z razponi od sedemdesetih let 7. do začetka 10. stoletja. 14C-datacije torej nakazujejo vsaj dva časovna horizonta najdišča. Kartiranje objektov, v katerih 9 Calib Rev7.1.0, RADIOCARBON CALIBRATION PROGRAM*, Copyright 1986-2016 M Stuiver and PJ Reimer. *To be used in conjunction with: Stuiver, M., and Reimer, P.J., 1993, Radiocarbon, 35, 215–230 in P.J. Reimer et al. 2009, Radiocarbon 51, 1111–1150. Zadnja kalibracija 4. 2. 2017. 10 Vzorec iz objekta SO 8, 14C 2 sigma od sredine 2. do konca 4. st. Na terenu so objekt SO 8 opredelili kot zgodnjesrednjeveški na podlagi luknjičave površine odlomka dna z ostenjem lončene posode. Ponoven pregled gradiva je pokazal, da ima to dno sledove izdelave na hitro vrtečem se vretenu, kar jasno kaže na rimsko dobo; poleg tega je dno v primerjavi z zgodnjesrednjeveškimi dni izrazito tanko. Razen tega so v istem objektu našli ustje rimske posode. Objekt SO 8 torej ni zgodnjesrednjeveški, temveč sodi v rimsko dobo. Rimskodobno poselitev na Novi tabli jasno kažejo drobne najdbe iz naselbinskih objektov in dobro ohranjeni inventarji sočasnega grobišča (Guštin 2003; id. 2004a; Pavlovič 2013, 58–62). so bili vzeti vzorci za analizo radioaktivnega ogljika 14C (Pavlovič 2013, 199, sl. 79), kaže, da so bili objekti z datacijami s starejšimi razponi razpo- rejeni predvsem na centralnem delu najdišča, pri čemer so bili objekti z mlajšimi razponi predvsem na zahodnem delu najdišča. Ujemajočo se dvojnost je pokazala tudi analiza lončenine. Ločijo se namreč jame, ki so vsebovale prostoročno izdelano in neokrašeno lončenino, od jam z lončenino, ki je bila dodelana na počasnem lončarskem vretenu in je praviloma okrašena. Prostorsko se jame s prvo oziroma drugo vrsto lončenine izključujejo.11 Objekti s prostoročno izdelano lončenino so se pojavljali predvsem v večji grupaciji objektov na osrednjem delu najdišča, pri čemer so se objekti z obvrteno in okrašeno lonče- nino pojavljali skoraj izključno na zahodnem delu najdišča (Pavlovič 2013, 186, sl. 74). Torej, dvojnosti, ki smo jih prepoznali tako pri lončenini, v razponih 2σ 14C-datacij, kot tudi v razprostranjenosti objektov na najdišču, nam dajejo dovolj trdno oporo in samozavest za trditev, da sta na najdišču Nova tabla v času zgodnjega srednjega veka obstajali vsaj dve izrazitejši kronološki fazi, starejša in mlajša. Na podlagi zgoraj zapisanega je, na tej stopnji raziskav, težko pojasniti, ali je razvoj iz ene v drugo kontinuiran ali ne. V tem trenutku se bolj nagibamo k predpostavki, da vzrok za spremembe v arheološkem gradivu ni neposreden razvoj starejše v mlajšo fazo, temveč da je med fazama obstajala časovna vrzel in lahko spremembe v drugi fazi poselitve najdišča pripišemo zunanjim dejavnikom. Najstarejši zgodnjesrednjeveški 14C-dataciji z Nove table Najstarejši zgodnjesrednjeveški 14C-dataciji ne presegata sredine 6. st. (sl. 1; 2). Verjetnost je, da sta polnili jam SO 149A in SZ 6 (sl. 3), iz katerih izvirata, nastali v istem obdobju. Jama SO 149A, pripadajoče gradivo in 14C-datacija so že objavljeni (Pavlovič 2008), jama SZ 6 pa bo v celoti v tem prispevku predstavljena prvič.12 11 Dva časovna horizonta sta prepoznala Mitja Guštin in Georg Tiefengraber, ki sta po prvi izkopavalni sezoni tudi uvidela, da se naselbina najverjetneje “širi” od vzhoda proti zahodu (Guštin, Tiefengraber 2002, 62). Podrobneje in nekoliko drugače o tem Pavlovič (2012, 322–327; 2013, 203–228; 2015). 12 Odlomek lončenine iz jame SZ 6: Pavlovič 2015, sl. 3: 12. 14C-datacija: Guštin, Pavlovič 2013, sl. 3. 355Začetki zgodnjeslovanske poselitve Prekmurja Jama / Pit Koda vzorca / Sample Code Radiokarbonska starost / 14C Age Kalibrirana datacija / Calibrated Age (Calib 7.10) SZ 6 KIA 19417 1576 ±28 415−547 SO 149A KIA 35477 1582 ±27 414−543 Sl. 2: Nova tabla. 14C-dataciji vzorcev oglja iz jam SO 149A in SZ 6. Fig. 2: Nova tabla. 14C dates of the charcoal samples from the sunken features of SO 149A and SZ 6. Sl. 3: Območje najdišča Nova tabla pri Murski Soboti, skupine jam zgodnjesrednjeveške poselitve in lega jam SO 149A in SZ 9 (za natančno lokacijo najdišča glej Pavlovič 2013, sl. 1, 2, 3). Fig. 3: Excavated area at Nova tabla near Murska Sobota with clusters of sunken features as traces of the early medieval habitation and the locations of SO 149A and SZ 9 (for the exact location of the site see Pavlovič 2013, Figs. 1, 2, 3). Dataciji odpirata nove možnosti razmišljanja o času prihoda prvih zgodnjeslovanskih skupin v Prekmurje in navajata k ponovnem pretresu dolgo zasidranih tez, ki temeljijo na pisnih virih. Jami sta bili podobne velikosti in oblike (sl. 4; 5). Oblikovani sta bili nepravilno ovalno z relativno položnimi stenami in ravnim dnom, SO 149A (dolž. 2,4 m; šir. 2 m; glob. 0,25 m) in SZ 6 (dolž. 2,25 m; šir. 2,2 m; glob. 0,45 m). V jamo SO 149A je bila vkopana mlajša jama SO 14913 (sl. 1) (Pavlovič 2008, 50–51). Ta je vsebovala veliko več oglja in lončenine (55 odlomkov oz. 928 g) kot SO 13 Oglje iz jame SO 149 je bilo prav tako datirano po metodi 14C, in sicer 91,6 % verjetnost je med letoma 533 in 635 n. št. (KIA 35476). 356 Daša PAVLOVIČ 149A (tri odlomke oz. 134 g). V jami SZ 6 je bilo 167 odlomkov lončenine (2206 g) in kos žlindre. Primerjava gradiva jam SO 149A in SZ 6 je pokazala očitne skupne lastnosti. Lončenina iz obeh jam je izdelana prostoročno, ima luknjičavo površino, ki je posledica izgorelega organskega pustila ali izluženega apnenčastega peska ter je neokrašena. Lonci so oblikovani sodčasto z naj- širšim obodom v zgornji tretjini. Rame posode je neizrazito. V jami SO 149A je bil en odlomek ustja lonca, v SZ 6 pa pet odlomkov (sl. 6). Spadajo v skupino neizrazitih kratkih ustij, ki so pogosto navpična, redko pa blago izvihana in zaobljena. Višina ustja je kronološko pomembna. Kratka ustja so starejša, daljša in bolj izvihana ustja pa mlajša (npr. Kuna, Profantová 2005, 340 in tam citirana literatura). Tovrstne lonce pripisujemo lončenini praškega tipa, ki je razširjena na prostoru od Rusije do Slovenije in je opredeljena kot najzgodnejša lon- čenina zgodnjih Slovanov.14 Konteksti, v katerih je samo lončenina praškega tipa s t. i. arhaičnimi ustji, med katero sodijo odlomki lončenine iz jam SO 149A in SZ 6, so opredeljeni kot najstarejša faza slovanske poselitve (Parczewski 1993, 56–58; Fusek 1994, 101; Kuna, Profantová 2005, 212, 213, sl. 84; Pleinerova 2000, 147–149). Od tega odstopa le odlomek iz jame SZ 6 (sl. 6: 11), od ostalih se razlikuje v tem, da ima vrezan okras, ki je na najdišču edinstven. Opis jam in katalog predmetov Jama SO 149A (sl. 4): Dobro vidna temnosiva lisa nepravilne oblike. Jama je sestavljena iz dveh delov, in sicer prvega objekta SO 149A, velikosti 2,4 × 2 × 0,25 m,15 iz peščene ilovice rjavo-sive barve, z ogljem in nekaj odlomki lončenine. SO 149A je presekal mlajši in manjši vkop SO 149, velik 1,4 × 1 × 0,25 m, ki je imel polnilo iz peščene ilovice temnosive barve. Vseboval je večjo količino oglja in veliko odlomkov lončenine. Oba vkopa sta imela položne stene in ravno dno. Drobne najdbe: – 3 odlomki lončenine (134 g), – pol prozorne, modrikaste steklene jagode, – 2 odlomka rimske lončenine (5 g).16 14 Z zgodnjo slovansko lončenino se je ukvarjalo mnogo avtorjev (glej npr. Gavrituhin 1997; Terpilovskij 2005; Buko 1990) predvsem iz Ukrajine, Poljske, Češke in Slovaške. Zgodovina raziskav in metod za klasifikacijo lončenine (npr. Fusek 1994, 18–34 in tam citirana literatura; Kuna, Profantová 2005, 154–159) ter geneza lončenine praškega tipa (npr. Parczewski 1993, 62–65) sta bili večkrat obravnavani. Kritično k opredeljevanju lončenine praškega tipa Curta (2001b). 15 Mere jame so v besedilu povzete s terenskih opisnih obrazcev. Risba jame pa je izdelana po digitalni dokumentaciji narejeni v programu ACad. Mere z obrazcev in v digitalni obliki se nekoliko razlikujejo. Razhajanja ni mogoče pojasniti. 16 V večini zgodnjesrednjeveških jam so bili odkriti predmeti, ki jih pripisujemo antičnemu ali prazgodovinskemu obdobju, od celih ali razlomljenih antičnih opek, odlomkov stekla, lončenine, prazgodovinske kamnite sekire itd. Ker za nekatere od teh predmetov vemo, da so bili v času zgodnjega srednjega veka ponovno uporabljeni, sicer v drug namen, menim, da ti predmeti v jame niso prišli naključno, temveč so jih prebivalci pobrali v okolici in jih ponovno uporabili. Najočitnejši primer je uporaba celih antičnih opek ali njihovih odlomkov kot podlag za ognjišča. Odkriti so tudi SO 149 SO 149 A A B A B 0 1 m SO 149 SO 149A ornica / topsoil geološka osnova / sterile layer Sl. 4: Nova tabla. Jama SO 149A. M. = 1:50 (Pavlovič 2008, sl. 2, 3). Fig. 4: Nova tabla. Sunken feature SO 149A. Scale = 1:50. (Pavlovič 2008, Figs. 2, 3). 357Začetki zgodnjeslovanske poselitve Prekmurja Opis izbranih drobnih najdb (sl. 6): 1. Odlomek pekača. Lončarska glina: finozrnata. Površina: obe površini luknjičavi. Barva: svetlorjava do oranžnorjava. Velikost: viš. 2,2 cm; šir. 7,2 cm; deb. do 1,3 cm; rekonstr. premer 23,6 cm. – Začasna inv. št. SO 149A/2. 2. Odlomek ustja in ostenja lonca. Lončarska glina: finozrnata. Površina: obe površini luknjičavi, gladki. Barva: obe površini lisasti, od svetlorjave do rjave. Izdelava: z lepljenjem in neobvrteno. Ve- likost: viš. 8 cm; šir. 9,6 cm; deb. 0,9 cm; rekonstr. premer 15,2 cm. – Začasna inv. št. SO 149A/1. 3. Del steklene jagode. Barva: modra, prosojna. Velikost: viš. 0,4 cm; deb. 0,16 cm; rekonstr. premer 0,38 cm. – Začasna inv. št. SO 149A/3. Jama SZ 6 (sl. 5): Jama nepravilne okrogle oblike, velikosti 2,25 × 2,20 m. V geološko osnovo je bila vkopana 0,46 m. Polnilo je rumenorjava peščena zemlja z ogljem, prodniki, odlomki lončenine in semeni. Drobne najdbe: – 167 odlomkov lončenine (2206 g), – 6 odlomkov prazgodovinske lončenine (103 g), odlomki antičnega stekla, npr. votli robovi čaš ali ustja žar, ki so lahko uporabljeni kot obeski na verižici. Predmeti so lahko imeli tudi apotropejski pomen, npr. prazgodovinska kamnita sekira. – 8 odlomkov prodnikov (232 g), – 1 odlomek žlindre (70 g). Opis izbranih drobnih najdb (sl. 6): 4. Odlomek ustja in ostenja lonca. Lončarska glina: drobnozrnata; posamezni grobi vključki; luknjičava. Površina: gladka. Barva: zunaj lisasta, sivorjava in svetlorjava, slednja mestoma z rdeč- kastim odtenkom; znotraj rjavordeča in temno rjavordeča. Izdelava: prostoročna. Velikost: premer ustja 13,4 cm; viš. 7,6 cm. – Začasna inv. št. SZ 6/3x. 5. Odlomek ustja in ostenja lončka. Lončarska glina: grobozrnata; luknjičava. Površina: gladka. Barva: zunaj rjava; znotraj temno rjavosiva. Izde- lava: prostoročna. Velikost: premer ustja 7,6 cm; viš. 6,5 cm. – Začasna inv. št. SZ 6/10. 6. Odlomki ustja in ostenja lonca. Lončarska glina: drobnozrnata; posamezni grobi vključki; luknjičava. Površina: hrapava, prvotna ni ohranjena. Barva: zunaj lisasta, oranžnorjava, temnorjava in temno rjavosiva; znotraj rjava. Izdelava: prosto- ročna. Velikost: premer ustja 15,1 cm; viš. 10,9 cm. – Začasna inv. št. SZ 6/6x. 7. Odlomek ustja in ostenja lonca. Lončarska glina: grobozrnata; luknjičava. Površina: hrapava. Barva: oranžnorjava. Izdelava: prostoročna. Veli- kost: premer ustja 14,6 cm; viš. 16,9 cm. – Začasna inv. št. SZ 6/7x. 8. Odlomki ostenja in dna lonca. Lončarska glina: drobnozrnata (posamezni grobi vključki; luknjičava). Površina: hrapava. Barva: zunaj te- mnorjava do temno rjavosiva; znotraj temnosiva. Izdelava: prostoročna. Velikost: premer najširšega oboda 16,2 cm; premer dna 9,4 cm; viš. 18,3 cm. – Začasna inv. št. SZ 6/8x. 9. Odlomek ustja in ostenja lonca. Lončarska glina: grobozrnata; rahlo luknjičava. Površina: hrapava, prvotna ni ohranjena. Barva: zunaj sve- tlorjava s posamezno liso svetlo sivorjave; znotraj temno rjavosiva in rjava. Izdelava: prostoročna. Velikost: 4,5 × 4,4 cm. – Začasna inv. št. SZ 6/3. 10. Odlomek ostenja in dna lonca. Lončarska glina: finozrnata; posamezni drobni vključki. Po- vršina: gladka. Barva: zunaj lisasta, rjavooranžna, oranžna, rjavordeča, oker in temnorjava; znotraj rjavordeča in temno rjavordeča. Izdelava: prosto- ročna. Velikost: premer dna 5,0 cm; viš. 4,3 cm. – Začasna inv. št. SZ 6/1x. 11. Odlomek ostenja lonca. Lončarska glina: grobozrnata; luknjičava. Površina: gladka, prvotna površina v glavnem ni ohranjena. Barva: svetlorjava. Izdelava: prostoročna. Velikost: 7,0 × 8,2 cm. Okras: vrezovanje. – Začasna inv. št. SZ 6/2x. 0 1 m A B A B Sl. 5: Nova tabla. Jama SZ 6. M. = 1:50. Fig. 5: Nova tabla. Sunken feature SZ 6. Scale = 1:50. 358 Daša PAVLOVIČ Sl. 6: Nova tabla. Izbrane najdbe iz jam SO 149A (1–3) in SZ 6 (4–11). 3 steklo; ostalo keramika. M. 1–2,4–11 = 1:3; 3 = 1:1. Fig. 6: Nova tabla. Select finds from the sunken features SO 149A (1–3) and SZ 6 (4–11). 3 glass; others pottery. Scale 1–2,4–11 = 1:3; 3 = 1:1. (Risba / Drawing: J. Tratnik Šumi) 359Začetki zgodnjeslovanske poselitve Prekmurja NAJSTAREJŠE ZGODNJESREDNJEVEŠKE 14C-DATACIJE Z DRUGIH NAJDIŠČ V PREKMURJU Pod Kotom – jug pri Krogu Na najdišču Pod Kotom – jug so raziskali jamo (SE 095/096) s podobno 14C-datacijo,17 kot sta najstarejši z Nove table (Šavel 2009, 157–159; Šavel 2002, 11–16). Datacija je bila prvič objavljena leta 2010 (Pleterski 2010a, 88; Pleterski 2010b, 47, tab. 1). Končna letnica razpona, leto 535, se je takrat zdela “veliko prestara”, zato ni bila upoštevana v nadaljnji analizi (Pleterski 2010a, 127). Polnilo jame je bila temnosiva ilovnata zemlja, ki je vse- bovala kamnite oblice, oglje, hišni lep, kosti in lončenino ter železen predmet (Šavel 2009, 157). Datirano je bilo oglje iz polnila, vendar točna lega vzorca iz objav ni znana. Lončenina se deli v dve skupini: a – prostoročno narejeno in brez okrasa; b – dodelano (obvrteno) na počasnem lončarskem vretenu in z vrezanim okrasom. Med prostoročno izdelano lončenino so odlomki kratkih navpičnih oziroma blago izvihanih ustij, ki jih pripisujemo zgodnjemu praškemu tipu lončenine (npr. Šavel 2002, t. 13: 3,6). Popava pri Lipovcih 1 Z najdišča Popava 1 (objekt št. 17, SE 390) iz- vira vzorec oglja, za katerega je analiza 14C dala datacijo (2σ) med letoma 420 in 540. 14C-datacija se časovno ne ujema z najdbami v objektu (Karo 2012, 53). Lončenina iz iste stratigrafske enote ni izdelana prostoročno in je ne moremo pove- zati z lončenino praškega tipa, ampak je mlajša, dodelana na lončarskem vretenu ter okrašena z vrezanimi linijami ali valovnicami. Diskrepanco med 14C-datacijo in oceno starosti najdb v polnilu iz objavljenih podatkov ni mogoče pojasniti. Ker je nastanke jam in njihovih polnil z zgodnje- srednjeveških najdišč v Prekmurju težko razumeti, predvsem dolžino časa zapolnjevanja jam, so poja- snjevanja takih diskrepanc nemogoča. Primernosti vzorca lahko na hitro oporekamo z razlago, da gre za star les, ki je z neznanimi procesi ali kasnejšo poljedelsko obdelavo zašel v polnilo, ali da gre za sredico starega debla, ki je veliko starejša od časa, ko je bilo deblo dejansko uporabljeno. Lahko pa nam tovrsten vzorec ponuja indic, da je bilo najdišče aktivno tudi v času razpona tega vzorca. 17 Datacija: Pleterski 2013a, 211, 212, sl. 1; id. 2010b, 47, tab. 1; id. 2010a, 88. RAZPRAVA O DATIRANJU LONČENINE Zgodnjeslovanski lončenini praškega tipa s t. i. arhaičnimi ustji iz Prekmurja najdemo najbližje primerjave na Slovaškem. Lončenino s slovaških najdišč, ki jo pripisujemo zgodnji slovanski poseli- tvi, je zbral, klasificiral in opredelil Gabriel Fusek (1994).18 Določil je tri kronološke faze, I, II in III. Fazo I je razdelili na Ia in Ib (ib., 65–76). Faza Ia je opredeljena s konteksti, v katerih se pojavlja izključno prostoročno izdelana lončenina brez okrasa z “arhaičnimi” robovi ustij (sl. 7). Taka je tudi lončenina iz jam SO 149A in SZ 6 z Nove table (sl. 6). Fusek je fazo Ia umestil v čas od prehoda 5. v 6. st. do začetka zadnje tretjine 6. st. oziroma do odhoda Langobardov iz Panonije v Italijo. Zgodnjo naselitev Slovanov na prostor zahodne Slovaške je z arheološkega gledišča utemeljil z odsotnostjo germanskih najdišč ter s primerlji- vostjo gradiva na izvornem ozemlju Slovanov, tj. Zakarpatje (Fusek 1994, 93, 101), s historičnega pa je ponovno ovrednotil relevantne pisne vire, za katere meni, da podpirajo zgodnjo naselitev (Fusek 2008, 645–656). Z novejšimi izkopavanji na najdišču Suhograd na skrajnem zahodu Slovaške je bil odkrit objekt s pečjo (objekt 1), v katerem je bila prostoročna lončenina praškega tipa arhaičnih potez. Iz odlomkov oglja, najdenih v polnilu tega objekta, sta bili pridobljeni tudi dve 14C-dataciji, ki objekt umeščata v prvo polovico 6. st. oziroma gotovo pred leto 550 ali 570 (Fusek, Zábojník 2010, 155–180).19 Torej, tako lončenina (ib., sl. 4; 5) kot razpona 14C-datacij (ib., sl. 7; 8) so primerljivi z gradivom iz jam SO 149A in SZ 6 z Nove table. Primerjave lončenine zgodnjega praškega tipa so tudi na najdiščih Nižná Myšľa-Alamenev in Ždaňa (vzhodna Slovaška), ki sodita v skupino zgodnje- slovanskih najdišč v zgornjem Potisju (na stičišču današnjih držav Slovaške, Madžarske, Romunije in Ukrajine), na katerih je prva faza slovanske poselitve datirana po letu 470 in pred prihodom Avarov v letih 567/568 (Fusek, Olexa, Zábojník 2010, 354, sl. 7, 16; Béreš 2013, 31, t. I; II). Podobno kot Fusek za Slovaško je Darinka Jelínková naredila relativno kronologijo za Češko (Jelínková 1990, 273, sl. 20) in Moravsko (ib., sl. 19). Ugotavlja, da se lončenina prve faze na Moravskem 18 Seveda, večino do takrat odkritega gradiva. 19 Problematiko najzgodnejše slovanske poselitve na območju današnje Slovaške obravnava G. Fusek tudi v članku Fusek 2015. 360 Daša PAVLOVIČ Sl. 7: Lončenina faze Ia s Slovaške (po Fusek 1994, 102, sl. 68, 69). Fig. 7: Phase Ia pottery from Slovakia (from Fusek 1994, 102; Figs. 68, 69). dobro ujema z najzgodnejšo lončenino z ukrajinskih najdišč, vendar meni, da direkten prenos ukrajin- skih absolutnih datacij na Moravsko, ki bi kazal na začetek slovanske poselitve Moravske ob koncu 5. ali v prvi polovici 6. st., zaradi drugačne historič- ne in kulturne situacije ni upravičen. Slovansko poselitev Moravske datira v drugo polovico 6. st. (ib., 277). To datiranje potrjujejo tudi nove 14C-da- tacije z najdišča Pavlov – Gorní Pole (objekt 953) z razponom med letoma 551 in 648 (Jelínková 2015, 144). Razmah praške kulture na južno Moravsko šele v drugi polovici 6. st. povezujejo z dejstvom, da so tam pred tem bivali prebivalci merovinškega kulturnega kroga oziroma do odselitve v Italijo historično izpričani Langobardi (Jelínková 2015, 135–143; Fusek 1994, 93; Fusek, 2008, 648; Fusek, Zábojník 2010, 170, sl. 10). Najnovejšo relativno kronologijo za Češko je naredila Nad'a Profantová (Kuna, Profantová 2005, 211–219, sl. 83, 90). Najboljše primerjave za kera- miko iz jam SO 149A in SZ 6 na Novi tabli so v najzgodnejši fazi RI (sl. 8). Profantová meni, da 361Začetki zgodnjeslovanske poselitve Prekmurja primerjava lončenine faze RI z lončenino drugih regij, pojavljanje germanskih najdb v kontekstih faz I in II na najdišču Roztoky in historične interpre- tacije nakazujejo slovansko poselitev Češke v prvi polovici 6. st. (ib., 219–225), vendar 22 14C-datacij predvsem z najdišč v okolici Prage, Roztoky, Liboc in Běchovice tega eksplicitno ne potrjuje (Profan- tová 2015, t. 1, 101–102; Profantová, Bureš 2013). Geografsko bolj oddaljeno – a nič manj podob- no – je najstarejše slovansko keramično gradivo na Poljskem, ki je datirano v drugo polovico 5. in prvo polovico 6. st. (Parczewsky 1993, 56, 58, sl. 13). Najstarejša je skupina najdišč na jugu Poljske, kjer imajo jame tudi nepravilno ovalno obliko20 (ib., 101–102, sl. 26), tako kot v Prekmurju (Pa- vlovič 2017). Ne nazadnje je treba opozoriti še na primerljivost praške lončenine Prekmurja z najdišč današnje Ukrajine. Tipologijo praške lončenine ter kronologijo praške kulture za prostor Ukrajine je izdelal Igor O. Gavrituhin21 (1997). Praško kulturo je razdelil na tri faze (I, II, III) ter predfazo (0). Predfaza se zaključi do prehoda v 5. st., faza I je opredeljena v čas od sredine 5. do prve polovice 6. st., faza II od sredine 6. do začetka 7. st. ter faza III v čas od leta 650 do konca 7. st. Po Gavrituhinovi tipokronologiji 20 Podobno tudi zgodnja najdišča na vzhodnem Slovaškem, tj. zgornjem Potisju (Fusek, Olexa, Zábojník 2010; Béreš 2013). 21 Več različnih transkripcij: Gavrituhin in tudi Gavrituchin, Gavritukhin. lahko lončenino iz jam SZ 6 in SO 149A umestimo med slabo oziroma rahlo profilirane lonce z nav- pičnimi ali rahlo izvihanimi ustji (tip II, podtipa Б in Д z robovi ustij oblik б, в in г; Gavrituhin 1997, sl. 1). Tovrstne oblike so značilne za njegovo fazo I, torej za čas od sredine 5. do začetka 6. st. (ib., 43). Raziskovalci praške kulture Ukrajine in Belorusije ločijo regionalne različice, ki jih pripisujejo različ- nim kulturnim podlagam ali integraciji različnih sosednjih kultur (Baran 1981, 86), in znotraj njih lokalne posebnosti praške lončenine, vendar je njena analiza pokazala, da znotraj različnih regij odseva podobno pravilo spreminjanja. Razvoj in raznolikosti znotraj praške kulture obravnavajo kot časovno in prostorsko dinamičen sistem (Gavritu- hin 1998, 200). Kljub temu je mogoče lončenino tu obravnavanih jam primerjati s praško lončenino ukrajinskih najdišč, npr. z najdišči Zymne (objekti 19a, 21, 23 – Gavrituhin 1998, t. 1; 2; 3), Zelenyj Gaj (objekt 4 – Baran 1987, sl. 2), Raškov III (objekta 12 in 38 – Baran 1988, t. 31; 39), Kodyn I (objekti 17, 21 in 26 – Baran 1987, sl. 5), Kodyn II (objekti 12, 21 in 30 – Baran 1987, sl. 6) in drugimi. Pomembnost primerjav z lončenino Ukrajine je v dobro utemeljeni dataciji v drugo polovico 5. in začetek 6. st., ki temelji na kovinskih predmetih bizantinsko-germanskih kultur, predvsem fibulah in zapestnicah, ter bizantinskih novcih, odkritih v zemljankah praške kulture (Baran 1981, 67 in sl. 2; Baran 1987, sl. 13; Gavrituhin 1997, 43, sl. 3: 44–46; Gavrituhin 2005). Sl. 8: Lončenina najstarejše slovanske faze na Češkem (Kuna, Profantová 2005, sl. 83). Fig. 8: Pottery of the earliest Slavic phase from Bohemia (from Kuna, Profantová 2005, Fig. 83). 362 Daša PAVLOVIČ Prve tipološke razvrstitve loncev iz Prekmurja so se precej navezovale na tipologije, vzpostavljene na Slovaškem in Poljskem, vendar se je kronologija razlikovala. Tipologijo lončenine z Nove table in kronologijo, pri kateri sta uporabila 14C-datacije, sta izdelala Mitja Guštin in Georg Tiefengraber. Uporabila sta gradivo iz izkopavalne sezone 1999–2001. Razdelila sta ga na dva časovna horizonta, Murska Sobota 1 in Murska Sobota 2 (MS 1 in MS 2). Starejši ho- rizont (MS 1), v katerega sodi lončenina praškega tipa, sta na podlagi 14C-datacij umestila v drugo polovico oziroma na konec 6. in v 7. st. (Guštin, Tiefengraber 2002, 58–60). Andrej Pleterski je leta 2010 izdelal kronologijo zgodnjesrednjeveškega gradiva za celotni vzhodni alpski prostor (Pleterski 2010a, 85–160). Pri njeni izdelavi je upošteval lončenino, razvrščeno glede na skupine ustij (S1-S7, V1-V2), ki so določene na podlagi klasifikacije ustij lončenine s Pristave na Bledu (Pleterski 2010a, 63–84), ter 14C-datacije (ib., sl. 4.63) z relevantnih in objavljenih (dostopnih) najdišč Slovenije, Hrvaške, Madžarske, Italije, Av- strije in Nemčije (ib., sl. 4.2). Izdelal je referenčno tabelo (ib., sl. 4.95), v kateri so podana obdobja pojavljanja posamezne skupine ustij (ib., 158–160). Natančne letnice, ki omejujejo ta obdobja, so bile po posebnem postopku (ib., 126) pridobljene iz razponov kalibracij 14C-datacij. Kot najstarejša Pleterski opredeljuje ustja loncev skupine S1. Kri- terij pripadnosti skupini S1 je neobvrtenost. Po tej kronologiji se prostoročno izdelana ustja pojavljajo med letoma 584 in 674, vendar zgodnejši začetek ni izključen (ib., 127–129). Zadnji je poskušal lončenino iz Prekmurja ter predvsem hrvaških, avstrijskih in madžarskih najdišč tipološko, kulturno in kronološko ovredno- titi Luka Bekić (2011, 37; 2016, 85). Prekmurska najdišča je umestil v skupino, ki jo je poimenoval Drava-Mura-Sava. Meni, da ta skupina predstavlja sledove najzgodnejšega slovanskega naseljevanja ob koncu 6. in v prvi polovici 7. st. na zahodnem delu Panonske nižine. Kronologijo je izpeljal iz tipologije ustij loncev ter 14C-datacij vzorcev oglja iz polnil jam z najdišč Stara ves, Pod lipom, Bli- zna, Brezje, Šarnjak in Brekinjova kosa z območja severne in severozahodne Hrvaške (Bekić 2016, 85–99, sl. 51 in 52).22 Stefan Eichert je podal podoben predlog razvoja zgodnje slovanske lončenine na podlagi najdišč 22 O lončenini praškega tipa na Hrvaškem: Bekić 2012, 21–35. avstrijske Koroške oziroma vzhodnoalpskega prostora. Meni, da je na območje Koroške, kjer je bila v uporabi lokalna poznoantična lončenina, na prehodu iz 6. v 7. st. z novim prebivalstvom prišla lončenina praškega tipa. Domneva, da sta se dve lončarski tradiciji v tem času združili v prehodno vrsto lončenine,23 ki se od keramike praškega tipa razlikuje po fakturi, je pa prostoročno izdelana in brez okrasa (Eichert 2010, 133, sl. 33). Eichertov pogled na razvoj lončenine na Koroškem je še toliko bolj zanimiv, ko ga primerjamo z njegovo kronologijo (Eichert 2010, 155, sl. 43). Za območje avstrijske Koroške je opredelil štiri faze, ki se lahko deloma med seboj časovno prekrivajo (Eichert 2013b, sl. 6). Opredelil jih je na podlagi pisnih virov in historične analize ter predvsem na podlagi temeljite analize arheoloških najdb. Domneva, da se takoj po prihodu Slovanov v vzhodnoalpski prostor okoli leta 590 začne preho- dna faza, ki se zaključi okoli leta 660, ko nastopi skupina grobov A, za katero ne poznamo lončenine, saj je definirana z grobovi elitne skupine (grobovi tipa Grabalja vas). Moški te skupine so pokopani z avarsko-bizantinskimi pasnimi garniturami, lahko tudi s frankovskim orožjem, redke ženske pa s sre- dozemskim nakitom. Na grobiščih je bil velikokrat odkrit samo po en grob tega tipa, vedno obdan s skupino grobov brez pridatkov. Taka razporeditev naj bi nakazovala specifično socialno in ekonom- sko razslojeno prebivalstvo (Eichert 2013a). Šele okoli leta 740 naj bi se pojavila grobišča občega prebivalstva s pridatki (začetek skupine grobov B), v katerih je lončenina, okrašena z valovnicami in vodoravnimi linijami ter dodelana na lončarskem vretenu. Absolutne časovne meje faz so določene predvsem z analizo in primerjavo grobnih inven- tarjev ter preverjene s posameznimi 14C-datacijami, pridobljenimi iz grobov z reprezentativnim gradi- vom (Eichert 2013b). Zaradi pomanjkanja najdb v času med koncem 6. in zadnjo tretjino 7. st. je začetek prehodne faze okoli leta 590 definiran s prenehanjem pojavljanja poznoantičnih najdb (ib., 419), konec pa z začetkom pojava grobov skupine A (ib., 421). 23 Stik med staroselci in nosilci praške kulture je “trhlo” utemeljen na arheoloških najdbah z najdišča Sv. Hema pri Globasnici, kjer so bili v prostoru zakristije četrte cerkve, nad žganinsko plastjo, odkriti prostoročno izdelani lonci, domnevno praškega tipa, in lonci poznorimske tradicije skupaj (Ladstätter 2000, 159–160). 363Začetki zgodnjeslovanske poselitve Prekmurja RAZPRAVA O ZAČETKU ZGODNJESREDNJEVEŠKE POSELITVE Bogo Grafenauer (1970–1971, 20) je bil pre- pričan, da so prvi Slovani v Panonsko nižino prišli s severa v prvi polovici 6. st. in se ustavili na Donavi. Na podlagi omemb v pisnih virih se je utrdilo mnenje, da se je val naselitve Slovanov okoli leta 550 obrnil proti jugu in pred letom 577 zajel severovzhodno Slovenijo, ker se škofa iz Petovione (danes Ptuj) ne omenja več. Drugi obsežen val Slovanov na slovensko ozemlje naj bi pljusknil ob prodoru Avarov z juga in vzhoda po odselitvi Langobardov v Italijo leta 568, najverje- tneje v zadnjih dveh desetletjih 6. st. (Grafenauer 1970–1971, 23; Štih 2008, 20; nekoliko drugače Žužek 2007, 266–268, 275–278). Grafenauer je pri raziskovanju slovanskega naseljevanja današnjega slovenskega ozemlja izhajal iz ohranjenih sinodalnih zapisnikov oglejskega patriarhata, ki po njegovem mnenju kažejo postopen propad antičnih škofij. Izostanki podpisov škofov naj bi pričali o propadu njihovih škofij, ta propad pa pripisuje prodirajo- čim Slovanom (Grafenauer 1970–1971, 24–25). Propadanje škofij je podrobneje obravnaval Rajko Bratož (2014, 505–547),24 ki meni, da se škofje iz ogroženih škofijskih sedežev sinod niso udeleževali iz neznanih razlogov, nevarnost slovanskega vdora omeni le kot notico v oklepaju (ib., 535).25 Glede problematike slovanskega prodora proti vzhodnim Alpam pa meni, da se “rezultati analize literarnih virov v veliki meri ujemajo z dognanji arheoloških raziskav zgodnjih slovanskih selišč v Prekmurju in Podravju”, in povzema datiranje v pozno 6. in zgodnje 7. st. (ib., 493). Tu je zaznati krožno dokazovanje, saj so bile prve kronologije arheološkega gradiva, kljub 14C-datacijam, prilagojene zgodovinskim interpretacijam pisnih virov, Bratož pa analizo literarnih virov potrjuje s tistimi arheološkimi interpretacijami, ki upoštevajo zgodovinske. V prejšnjih poglavjih sem pokazala, da naj- zgodnejše 14C-datacije in primerjava keramike z najdišči Češke, Slovaške, Poljske in Ukrajine nakazujejo, da so se najstarejše skupine nosilcev 24 Tu tudi celoten historiat raziskav naseljevanja Slovanov v jugovzhodni alpski prostor – predvsem zgodovinski pogled, dragocen, ker upošteva tudi arheološke raziskave. 25 Nekoliko drugačen pogled na razloge za propadanje škofij poda Hrvoje Gračanin. Meni, da bi bil lahko izostanek omembe škofa povezan s formalno ukinitvijo civitas, zaradi nefunkcioniranja mestnega središča in majhnega števila prebivalstva v diecezi (Gračanin 2008, 25). praške kulture lahko v Prekmurje naselile že v prvi polovici 6. st. Tako datiranje se ne sklada z dosedanjim dati- ranjem prekmurskih zgodnjesrednjeveških najdišč niti s kronologijami bližnjih regij, kot sta Koroška in Hrvaška, niti z uveljavljenimi zgodovinskimi pogledi na naseljevanje zgodnjih Slovanov na za- hodne obronke Panonske nižine. Drugačen pogled sem predstavila v svoji disertaciji (Pavlovič 2013, 203–210) in na možnosti, ki jih odpira, je bilo že opozorjeno (Pleterski 2015, 241). Zdi se, da je bil nižinski del severovzhodne Slovenije v 5. in deloma v 6. st. precej prazen, na kar kaže odsotnost poznoantičnih najdb (Cigle- nečki 2000a, sl. 2). Zaradi negotovih političnih in ekonomskih, morda tudi klimatskih (Oblomskij, Petrauskas, Terpilovskij 1999, 84) okoliščin ter predvsem za- radi nenehnih vdorov in premikanj prebivalstva so nekatera – lahko tudi precej obširna – območja med Karpatsko nižino in srednjo Evropo opustela. O takšnih opustelih območjih nam poročajo tudi pisni viri. Prokopij namreč omenja, da so Heruli, ki so živeli na severnem bregu Donave, po tem, ko so jih porazili Langobardi leta 509, prešli ozemlje vseh plemen Sclavenoi, prečkali precejšna območja praznega podeželja, preden so dosegli pleme Var- nov na severu Nemčije. Na malo kasnejši čas se nanaša poročilo Gregorja iz Toursa, da so Avari v pohodu proti Frankom trpeli pomanjkanje živeža, kar implicira, da so prečkali opustela območja (Kobyliński 2005, 531; Steinacher 2010, 353, op. 132). Komes Marcelin, kronist na dvoru Vzhodnega rimskega cesarstva do leta 534, je zapisal, da so barbare med drugim naselili na opustošena obmo- čja in v mesta Romanov. To pojasni s primerom Herulov, katerih del so po porazu z Langobardi leta 509 sprva naselili na območju Gepidov, leta 512 pa znotraj imperija (Croke 2001, 74, 131),26 najverjetneje na opustelo območje v okolici in severno od Singidunuma (Beograd) (Steinacher 2010, 350, 351). Pisni viri torej nakazujejo obsežna neposeljena območja tudi v Panonski nižini severno in južno od Donave (sl. 9). Podobno, kot je bil izpraznjen prostor v okolici Singidunuma, lahko glede na odsotnost arheoloških najdb domnevamo, da je bil opustel tudi prostor južnih in zahodnih obronkov Panonske nižine, npr. Podravina (Sekelj Ivančan 2010, 165) in Prekmurje (Ciglenečki 2000, sl. 2). 26 To obdobje zgodovine Herulov je podrobno predstavil Alexander Sarantis (2010). 364 Daša PAVLOVIČ Prazna območja so skupinam zgodnjih Slovanov omogočala enostavno napredovanje iz Zakarpatja v Panonsko nižino, prek Donave in vse do Prek- murja, brez vojaških spopadov in zato brez omemb v bizantinskih pisnih virih. V prid domnevi o naselitvi skupin nosilcev kul- ture praškega tipa, ki jih povezujemo z zgodnjimi Slovani, na območje današnjega Prekmurja in severne Hrvaške v prvi polovici 6. st. je tudi prva epigrafska omemba Slovanov (v ednini Sclavus) v Panoniji. Leta 558 je Martin iz Brage v epigrafskem posvetilu tega leta dograjene katedrale v samostanu Dumio pri Bragi (Bracara Augusta) na Portugalskem napisal hvalnico Martinu iz Toursa. V njej našteva ljudstva, med drugimi Slovane, ki so živela v Panoniji in so po zaslugi Martina iz Toursa spoznala Kristusa (Šašel 1976, 151, 152). Martin iz Toursa se je rodil v Savariji (današnji Szombathely na Madžarskem), Sl. 9: Panonija v prvi polovici 6. stoletja. Označeni so pomembnejši antični kraji in v besedilu omenjena poselitvena območja (1–7) ter zgodnjesrednjeveški naselbini Nova tabla pri Murski Soboti (A) in Cerklje ob Krki (B). Obseg poznoantične poselitve: 1 – območje Pólis Norikón (hipotetična meja po Ciglenečki 2017, sl. 2); 2 – poselitev Langobardov (po Vida 2008, sl. 1); 3 – poselitev Gepidov in 4 –Herulov (oboje po Sarantis 2009, sl. 2). Zgodnjeslovanska poselitev: 5 – severno od Donave in 6 – v zgornjem Potisju (po Fusek, Zabojník 2010, sl. 10; Fusek 2008, sl. 4); 7 – poselitev severovzhodne Slovenije (po Guštin, Pavlovič 2009, sl. 1). Fig. 9: Pannonia in the first half of the 6th century. Significant Antique towns and settlement areas (1–7), as well as two early medieval settlements – Nova tabla (A) near Murska Sobota and Cerklje ob Krki (B) – mentioned in the text are marked. Extent of the Late Antique settlement area: 1 – Pólis Norikón (hypothetical boundary after Ciglenečki 2017, Fig. 2); 2 – Langobard settlement area (after Vida 2008, Fig. 1); 3 – settlement of the Gepids and 4 – the Herules (both after Sarantis 2009, Fig. 2). Early Slavic settlement area: 5 – north of the Danube and 6 – in the upper reaches of the River Tisa (after Fusek, Zabo- jník 2010, Fig. 10; Fusek 2008, Fig. 4); 7 – settlement of north-eastern Slovenia (after Guštin, Pavlovič 2009, Fig. 1). 365Začetki zgodnjeslovanske poselitve Prekmurja Martin iz Brage pa najverjetneje na območju sotočja Drave in Save z Donavo (Bassianensis Pannonia). Jaroslav Šašel je pokazal, da Martin iz Brage ni našteval plemen šablonsko, temveč spontano kot spomin iz svojih mladostnih dni, najverjetneje iz dvajsetih let 6. st. Upravičeno se je vprašal, katere Slovane je imel Martin iz Brage v mislih. Glede na takrat znane podatke je domneval, da je imel v mislih Slovane na območjih Gepidov v današnji Romuniji ali severno od Donave na današnjem Slovaškem (Šašel 1976, 156, tudi 155 na karti). Nova odkritja, ki sem jih navedla, me vodijo k domnevi, da je lahko imel v mislih Slovane, ki so živeli v bližini rojstnega kraja Martina iz Toursa, to je v Prekmurju in v severni Hrvaški.27 Pomembna podpora domnevi o najstarejši na- selitvi Slovanov v Prekmurje v prvi polovici 6. st. izhaja iz proučevanja sledov naselitve Langobardov v Panonsko nižino in na območje jugovzhodnih Alp.28 Arheološki viri kažejo, da so poselili območje med črto Szombathely–Keszthely–Pécz in kolenom Donave, živeli so tudi na višinskih naselbinah zahodno od linije Ptuj–Sisak. Na prostoru med tema linijama ni arheoloških sledov langobardske ali poznoantične poselitve. Domnevam, da je bil prostor med Blatnim jezerom in Ptujem v prvi polovici 6. st. že poseljen s Slovani. Podrobneje bom to pojasnila v naslednjih odstavkih. Langobardi so se z območja ob spodnji Labi v današnji Nemčiji ob koncu 5. st. preselili na prostor Spodnje Avstrije in Moravske (Pohl 2008b, 26).29 Na konec 5. ali začetek 6. st. je na podlagi lončenine (Fusek 1994, 102) datiran prihod zgodnjih Slovanov iz Zakarpatja na območje Slovaške. Iz Prokopijeve pripovedi o langobardskem pretendentu za prestol Hildigisu, ki je med drugim iskal pribežališče pri Slovanih (Fusek 2008, 646), lahko sklepamo, da so bili Slovani južno od Karpatov in v bližini Donave nastanjeni najkasneje v prvi polovici 6. st. Skupnosti Langobardov in Slovanov sta bili torej v prvi po- lovici 6. st. v Panonski nižini severno od Donave sočasni in sta mejili. To dokazujejo predmeti, ki 27 O Martinu iz Brage in omembi Slovanov od kasnejših avtorjev tudi H. Gračanin (2008, 23–24) ter tam citirana literatura (ib., op. 74). 28 Ta teza ni povsem nova. Zgodnejši prodor posameznih slovanskih skupin že v času, ko so tam živeli Langobardi na obrobju njihove panonske države, so omenili že drugi raziskovalci (Pleterski 1990, 51; id. 2015, 241; Gavritukhin 2015, 20–21). 29 Pregled selitev Langobardov med Elbo in Italijo v luči pisnih in arheoloških virov poda npr. W. Pohl (2008a, 1–12; id. 2008b, 23–33). jih sicer pripisujemo merovinškemu kulturnemu krogu oz. Langobardom, vendar so bili odkriti v jamah z lončenino praškega tipa (Profantová 2008, 623–631; Fusek, Zábojník 2010, 172). Skupine nosilcev prodirajoče kulture z lončeni- no praškega tipa so ob stiku z Langobardi lahko krenile proti jugu vse do današnjega Prekmurja. Tej domnevi v prid je lončenina praškega tipa z Nove table, ki je sorodna najzgodnejši slovanski lončenini v Ukrajini s konca 5. ali začetka 6. st. (glej zgoraj). Čeprav so langobardske migracije dobro znane iz pisnih virov (glej npr. pri Pohl 2008a), so arheološki viri in 14C-datacije osteološkega gradiva ključni za razumevanje naselitve Langobardov v Panonijo. Ta je bila postopna in se je začela na začetku 6. st. Grobišča kažejo, da so sprva, v prvih dveh tretjinah 6. st., poselili severni del Panonije, pas južno od Donave. Tu se pojavljajo dolgotrajnejša, po številu grobov obsežna grobišča tipa Szentendre od leta 510 pa do preselitve v Italijo leta 568. Med letoma 535 in 568 so datirana manjša, kratkotrajnejša grobišča tipa Vörs-Kajdacs na območju med (jugo)vzhodno Panonijo in zahodnim delom Blatnega jezera (Vida 2008, 76, sl. 1; Stadler et al. 2003, 265–269, t. 2).30 Zadnji v pisnih virih izpričan premik v lan- gobardski panonski fazi je bil kmalu po letu 546 (najkasneje do leta 548), ko jim je bizantinski cesar Justinijan, v senci gotskih vojn in v strahu pred prodirajočimi Franki, z darovnico podaril pólis Norikón, panonske utrdbe in številne druge kraje ter veliko vsoto denarja (Bratož 2014, 454–458). Ne glede na to, ali se strinjamo s trditvijo, da je ime pólis Norikón označevalo mesto z upravnim območjem, najverjetneje nekdanje Petovione ali Celeje (Šašel Kos 1994, 294) ali območje med Petoviono in Celejo, panonske utrdbe pa višinske utrjene naselbine med Siskom (Hrvaška) in Ajdno (SZ Slovenija) (Ciglenečki 2005, 272, 274–275, 30 Poskus absolutnega datiranja z metodo analize radioaktivnega izotopa 14C langobardske poselitve severno od Donave in v Panonski nižini je bil zasnovan na 63 vzorcih z najdišč Moravske, Spodnje Avstrije in zahodne Madžarske, pridobljenih večinoma iz kolagena ostankov človeških kosti. Prav tako je bil za statistično obdelavo pridobljenih 14C-datacij uporabljen program OxCal, v katerega so vnesli tudi historično ugotovljene faze naseljevanja Langobardov v Panonsko nižino. Avtorji tega projekta zaključujejo, da je absolutni časovni okvir naseljevanja Langobardov na obravnavano območje, temelječ na 14C,-datacijah zelo verjetno primerljiv s časovnim okvirom, pridobljenim z raziskavami langobardskih grobišč, opisanih zgoraj (Stadler et al. 2003). 366 Daša PAVLOVIČ sl.  5; Ciglenečki 2001, 187, 188, sl. 5), pri obeh interpretacijah so Bizantinci Langobardom podarili območje zahodno od črte Ptuj–Sisak. Arheološke najdbe kažejo na majhne langobardske vojaške posadke z družinami, ki so zahodno od linije Ptuj–Sisak bivale na višinskih utrjenih nase- ljih skupaj z romaniziranimi staroselci in varovale prehode v Italijo (Ciglenečki 2000a, 122–123; id. 2005, 271, 272). Najzgodnejša lončenina s pečatnim in glajenim okrasom na slovenskem ozemlju, ki jo povezujemo z Langobardi in je bila najdena na višinskih naselbinah (Ajdovski gradec nad Vra- njem, Rifnik, Tinje nad Loko pri Žusmu, Puštal nad Trnjem ter iz Kranja; – Knific 1994, 222, sl. 15; Modrijan, Novšak 2015, 30), je datirana v prvo tretjino 6. st. (Knific 1994, 219). Podobno data- cijo je pokazala tipološka analiza za Langobarde značilnih S-fibul s slovenskih najdišč, ki kaže tudi na povezave z langobardsko predpanonsko fazo, tj. pred letom 510 (Milavec 2007, 333). Presenetljivo je, da so Langobardi poselili velik del Panonije, celo malo zahodneje od Blatnega je- zera, ter osrednji in zahodni del Slovenije, območje Prekmurja in širše okolice pa ne. Langobardskih najdb skorajda ni niti na avstrijskem Štajerskem, kjer so našli le eno langobardsko fibulo in malo pečatne lončenine, vse brez jasnih najdiščnih okoliščin (Gutjahr 2002, 147). Tivadar Vida meni, da so bili vzrok za odso- tnost langobardske poselitve zahodno od linije Szombathely–Keszthely–Pécz provincialni rimski staroselci, ki naj bi to območje poseljevali in naj bi pomenili oviro pri nadaljnjem širjenju Lango- bardov proti zahodu (Vida 2008, 76). To se mi zdi malo verjetno. Geografsko najbližji, z arheo- loškimi viri izpričani staroselci v prvi polovici 6. st. izvirajo namreč s hribovitih predelov Štajerske (Ciglenečki 2000a, sl. 2). Obsežne arheološke raziskave nižinskih predelov Dravskega polja, Prekmurja in zahodne Madžarske niso odkrile sledov poznoantične poselitve, ki pa je dobro izpričana na zahodneje ležečih višinskih naseljih. Poleg tega najdbe, ki jih pripisujemo merovin- škemu kulturnemu krogu oziroma historično izpričanim Langobardom na višinskih naselbinah v Sloveniji, nakazujejo sobivanje Langobardov s staroselci (Ciglenečki 2001; 2005). Domnevam, da so bili langobardski ekspanziji ovira nosilci praške kulture, ki so v prvi polovici 6. st. že živeli v Prekmurju in v severnih delih Hrvaške. Menim, da so Langobardi v Prekmurju spoštovali meje naselitve zgodnjih Slovanov. Tako je bilo namreč severno od Donave. Za območje reke Morave sta Gabriel Fusek in Jozef Zábojník z analizo najdišča Suchohrad in s kartiranjem sočasnih langobardskih najdišč in najdišč nosilcev praške kulture (faze Ia, prva polovica 6. st.) pokazala jasno upoštevanje naselitvenih mej med tema skupnostma (Fusek, Zábojník 2010, 172, sl. 10; 11). Sožitje med nosilci kulture praškega tipa, tj. Slo- vani, in Langobardi na območju današnje Slovenije morda nakazuje odlomek lončenine, ki ga po tehniki izdelave lahko pripišemo lončenini praškega tipa, vendar ima žigosan okras (sl. 10), ki je značilen za langobardsko lončenino. Izvira z najdišča Cerklje ob Krki,31 jugovzhodna Slovenija. Najden je bil v jami SE 1224/1225 z več polnili. V vseh treh polnilih so bili odlomki prostoročno izdelane lončenine s kratkimi, navpičnimi ali blago izvihanimi ustji, ki jo lahko tipološko opredelimo kot praški tip.32 Žigi na omenjenem odlomku so preprosti krožci, ki se zdijo na ohranjenem delu razporejeni neurejeno. Arhaičnost praške lončenine ter časovno možnost sobivanja potrjuje tudi 14C-datacija, ki je bila pri- dobljena iz vzorca oglja iz polnila SE 1225. Vrhnja meja razpona 2σ je leto 560 (poročilo: Vojaković et al. 2016, 120). Okras žigov na lončenini praškega tipa je na čeških in moravskih najdiščih razmeroma redek, malo pogostejši je na zahodnem Slovaškem. Pripisujejo ga vplivu germanskih ljudstev, predvsem Langobardov (Fusek 1994, 62; Kuna, Profantová 2005, 341, sl. 69: A). Profantová meni, da je večja pogostnost tega okrasa na zahodnem Slovaškem povezana z daljšim sobivanjem zgodnjih Slovanov z Langobardi in Gepidi. Na najdišču Roztoky na Češkem lončenino s tem okrasom uvršča v najsta- rejšo fazo najdišča (Kuna, Profantová 2005, 174). Tovrstni krožci niso pogosti na langobardski lončenini. Nekaj primerov lahko najdemo na Moravskem, in sicer na prostoročno izdelani lončenini – npr. najdišče Šaratice (Tejral 2008, sl. 7: 5) – iz grobov, ki jih pripisujejo Langobardom, vendar še z izrazito tradicijo polabskih Germanov, in so datirani v drugo polovico 5. st. (ib., 64–66). Domnevam, da odlomek lončenine s pečatnim okrasom z najdišča Cerklje ob Krki odseva stike 31 Neobjavljeno poročilo: Vojaković et al. 2016. Raziskave: Arhej, d. o. o., vodja izkopavanj Matjaž Novšak, namestnica vodje Petra Vojaković. Ekipi Arhej, d. o. o., se najlepše zahvaljujem za prijazno obvestilo o novih odkritjih. 32 Gre za preliminarno opredelitev brez temeljitejše analize. Nadejamo se, da bo ta sledila v kratkem. Informacijo mi je posredovala P. Vojaković, za kar se ji najlepše zahvaljujem. 367Začetki zgodnjeslovanske poselitve Prekmurja med germanskimi ljudstvi, domnevno Langobardi, in nosilci praške kulture, zelo verjetno zgodnjimi Slovani, na območju jugovzhodne Slovenije. Taka interpretacija kaže na datacijo pred langobardsko selitvijo v Italijo leta 568. Podoben kontakt med nosilci praške kulture in merovinškega kulturnega kroga južno od Donave je viden na grobišču An den Klostergründen, Großprüfuning, Regensburg, kjer se pojavljajo žgani grobovi v žarah, ki so opredeljene kot lonci praškega tipa (Losert 2011, 482–483, sl. 7). Ti gro- bovi so na podlagi kovinskih pridatkov datirani v drugo polovico 6. in začetek 7. st. Na možnost, da sta skupnosti sobivali že v prvi polovici 6. st., pa nakazujeta dve 14C-dataciji, in sicer iz groba 11 z zgornjo mejo 2σ razpona v letu 546 in iz groba 20 v letu 569 (ib., 489; Pleterski 2013b, 23–25). V kontekstu predstavljene nove hipoteze so zanimive tudi ugotovitve jezikoslovcev o jeziku slovanskih prebivalcev med Donavo in Jadranom v srednjem veku. Za širitev slovanščine na to ozemlje obstajata dva glavna pogleda, migracijski in difuzijski (Snoj, Greenberg 2012, 282). Marko Snoj in Marc L. Greenberg sta trdno prepričana, da migracijski model jezikovnega razvoja slovanščine na ozemlju današnje Slovenije pojasnjuje sledove razlik iz zgodnejšega obdobja, tj. obdobja pred letom 500. Kot enega od poučnih primerov navajata ravno analizo razvrstitve leksema *gъlčěti 'govoriti', ki ga najdemo v prekmurščini in okoliških panonskih govorih, ne pa v ostali slovenščini, pojavi se spet v določenih narečjih v Bolgariji. “Pomenski razvoj glagola *gъlčěti s pomenom 'govoriti' iz 'hrumeti, brneti, grmeti' je bila inovacija v pomensko-obli- kovnem ujemanju z antonimom *mъlčěti 'molčati'. V tem primeru lahko najdemo isti novi pomen v osrednjeruskih govorih, kar kaže na povezavo med vzhodnoslovanskim področjem in prebivalstvom, ki se je izselilo na področji kasnejših prekmurskih in bolgarskih govorov” (Snoj, Greenberg 2012, 283).33 Z gledišča jezikoslovcev je torej prihod slovansko govorečih skupin v Prekmurje možen pred letom 500 ali okoli tega leta. 33 Podrobna analiza glagola *gъlčěti je v članku Josepha Schallerta in Marca L. Greenberga (2007, 9–76). Sl. 10: Odlomek lončenine praškega tipa z žigosanim okrasom z najdišča Cerklje ob Krki. M. = 1:2 (A) in 1:1 (B). Predmet hrani Posavski muzej Brežice. Fig. 10: Jar fragment of the Prague-type pottery with stamped decoration from Cerklje ob Krki. Scale = 1:2 (A) and 1:1 (B). Kept in the Posavski muzej Brežice. (Risba / Drawing: T. Leskovar; foto / photo: © Narodni muzej Slovenije, T. Lauko) A B 368 Daša PAVLOVIČ ZAKLJUČEK V zadnjih treh desetletjih se je močno obogatilo vedenje o zgodnjih Slovanih na ozemlju današnje Slovenije. Lončenina praškega tipa s primerjavami med najstarejšo lončenino tega tipa na najdiščih Ukraji- ne, Poljske, Češke in Slovaške ter 14C-dataciji oglja iz jam SO 149A in SZ 6 na Novi tabli ter iz jame SE 095/096 z najdišča Pod Kotom – jug kažejo na njuno datacijo pred sredino 6. st. To nakazuje, da so se prve skupine nosilcev kulture praškega tipa na zahodne obronke Panonske nižine priselile že v prvi polovici 6. st. To območje je bilo morda pred zgodnjeslovansko naselitvijo nenaseljeno, na kar kaže odsotnost poznoantičnih najdišč v Prekmurju in nižinah severne Hrvaške. Slovanski poselitvi Prekmurja v prvi polovici 6. st. v prid govorijo jezikovne analize pa tudi dejstvo, da so Langobardi pri poselitvi Panonske nižine izpustili prostor med Blatnim jezerom in Dravskim poljem, kar lahko nakazuje, da so spoštovali meje na tem mestu že naseljene skupnosti, tj. Slovanov. V luči domnevne poselitve Slovanov v Prekmurju v prvi polovici 6. st. je prva epigrafska omemba Slova- nov v Panoniji pri Martinu iz Brage povezana z dejansko demografsko sliko Panonije v tem času in ni le konstantinopelski vpliv in retorični efekt ali mehanizem pretiravanja in poudarjanja hitre širitve krščanstva.34 Zato zaključujem z vse bolj prepričljivo hipotezo, da so se manjše skupine nosilcev praške kulture, ki jih povezujemo z zgodnjimi Slovani, na zahodne obronke Panonske nižine začele naseljevati že v prvi polovici 6. stoletja. 34 Zadnje je mnenje F. Curte, ki izključuje relevantnost podatka Martina iz Brage kot dokaz za obstoj Slovanov v Panoniji že tako zgodaj (Curta 2001a, 46). Drugačno mnenje o relevantnosti zapisa pri Martinu iz Brage podaja R. Bratož (2014, 285–286). BARAN, V. D. 1981, The question of Origin of the Slavic Culture in the Early Middle Ages. – Acta Archaeologica Carpathica 21, 67–88. BARAN, V. D. 1987, Istoki pražskoj kultury. – Studia nad etnogeneza Slowian i kultura Europy wczesnosredniowiecznej 1, 35–51, Wroclaw, Warszawa, Kraków, Gdansk, Lodz. BARAN, V. D. 1988, Pražskaja kul´tura Podnestrovja (po materialam poselenij u s. Raškov). – Kijev. BAYLISS, A. 2009, Rolling out Revolution: Using Radiocarbon Dating in Archaeology. – Radiocarbon 51/1, 123–147. BEKIĆ, L. 2011, Rani srednji vijek između Panonije i Jadrana: ranoslavenski arheološki nalazi od 6. do 8. stoljeća. – Neobjavljena disertacija / Unpublished PhD dissertation, Univerza na Primorskem, Koper. BEKIĆ, L. 2012, Keramika praškog tipa u Hrvatskoj. – Zbornik Stjepana Gunjaće i hrvatska srednjevjekovna povjestno-arheološka baština 2, 21–35, Split. BEKIĆ, L. 2016, Rani srednji vijek između Panonije i Jadrana. Ranoslavenski keramički i ostali arheološki nalazi od 6. do 8. stoljeća. / The Early Medieval bet- ween Pannonia and the Adriatic. Early Slavic ceramic and other archaeological finds from the sixth to eight century. – Monografije i katalozi 27. BEMMANN, J., M. SCHMAUDER (ur. / eds.) 2008, Kul- turwandel in Mitteleuropa. Langobarden – Awaren  – Slaven. Akten der Internationalen Tagung in Bonn vom 25. bis 28. Februar 2008. – Kolloquien zur Vor- und Frühgeschichte 11. BÉREŠ, J. 2013, Včasnoslovanský sídliskový objekt zo Ždane.  – Študijné zvesti archeologického ústavu SAV 53, 23–37. BIERMANN, F. 2009, Kommentar zum Aufsatz von Florin Curta: Utváření Slovanů (se zvláštním zřetelem k Čechám a Moravě). – The Making of the Slavs (with a special emphassis on Bohemia and Moravia), Archeologické rozhledy 60. – Archeologické rozhledy 61, 337–349. BOARETTO, E. 2009, Dating Materials in Good Archaeo- logical Contexts: The Next Challenge for Radiocarbon Analysis. – Radiocarbon 51/1, 275–281. BRATOŽ, R. 2014, Med Italijo in Ilirikom. Slovenski prostor in njegovo sosedstvo v pozni antiki (Between Italy and Illyricum. Slovene territory and its neighbourhood in late Antiquity). – Zbirka Zgodovinskega časopisa 46. BRONK RAMSEY, C. 2001, Development of the Radiocar- bon Program OxCal. – Radiocarbon 43/2A, 355–363. BRONK RAMSEY, C. 2009, Bayesian analysis of radiocarbon dates. – Radiocarbon 51/1, 337–360. BUKO, A. 1990, Ceramika wczesnopolska. Wprowadzenie do badań (The Early Medieval Polish Pottery. Introduc- tion into Research). – Wrocław. CIGLENEČKI, S. 2000a, Archaeological Investigations of the Decline of antiquity in Slovenia (Arheološke raziskave zatona antike v Sloveniji). – V / In: R. Bratož (ur. / ed.), Slovenija in sosednje dežele med antiko in karolinško dobo. Začetki slovenske etnogeneze / Slowenien und die Nachbarländer zwischen Antike und karolingischer Epoche. Anfänge der slowenischen Ethnogenese 1, Situla 39 (Razprave 1. razreda SAZU 18), 119–134. CIGLENEČKI, S. 2000b, Tinje nad Loko pri Žusmu. Pozno- antična in zgodnjesrednjeveška naselbina / Tinje oberhalb von Loka pri Žusmu. Spätantike und frühmittelalterliche Siedlung. – Opera Instituti Archaeologici Sloveniae 4. 369Začetki zgodnjeslovanske poselitve Prekmurja CIGLENEČKI, S. 2001, Romani e Longobardi in Slove- nia nel VI secolo. – V / In: Paolo Diacono e il Friuli altomedievale (secc. VI - X). Atti del XIV Congresso internazionale di studi sull'Alto Medioevo, Cividale 1999, 179–209, Spoleto. CIGLENEČKI, S. 2005, Langobardische Präsenz im Sü- dostalpenraum im Lichte neuer Forschungen. – V / In: W. Pohl, P. Erhart (ur. / eds.), Die Langobarden. Herrschaft und Identität, Österreichische Akademie der Wissenschaften, Philosophisch-Historische Klasse, Denkschriften 329 (Forschungen zur Geschichte des Mittelalters 9), 265–280, Wien. CIGLENEČKI, S. 2017, Versuch einer Interpretation des Siedlungswechsels im südlichsten Teil von Noricum Mediterraneum. – V / In: I. Dörfler, P. Gleirscher, S. Ladstätter, I. Pucker (ur. / eds.), Ad amussim. Festschrift zum 65. Geburtstag von Franz Glaser, Kärntner Muse- umsschriften 82 (v tisku / in print). CROKE, B. 2001, Count Marcellinus and his Chronicle. – Oxford. CURTA, F. 2001a, The Making of the Slavs. History and Archaeology of the Lower Danube region c. 500–700. – Cambridge. CURTA, F. 2001b, The “Prague type”: a critical approach to pottery classification. – Archaeologia Bulgarica 5, 73–106. CURTA, F. 2010, Etnicitet u ranosrednjovjekovnoj arheo- logiji. Primjer ranoslavenskih nalaza u jadranskoj regiji. – Starohrvatska prosvjeta, III. ser., 37, 17–48. EICHERT, S. 2010, Die frühmittelalterlichen Grabfunde Kärntens. Die materielle Kultur Karantaniens anhand der Grabfunde vom Ende der Spätantike bis ins 11. Jahrhundert. – Aus Forschung und Kunst 37. EICHERT, S. 2013a, Archäologische und historische Evi- denzen für soziale Strukturen im frühmittelalterlichen Ostalpenraum. – V / In: Soziale Gruppen und Gesell- schaftsstrukturen im westslawischen Raum, Beiträge zur Ur- und Frühgeschichte Mitteleuropas 70, 385–404. EICHERT, S. 2013b, Zur Absolutchronologie des Ostal- penraums im Frühmittelalter unter besonderer Berüc- ksichtigung 14C-datierter Grabinventare. – Berichte der Bayerischen Bodendenkmalpflege 54, 419–428. FUSEK, G. 1994, Slovensko vo včasnoslovanskom období. – Archaeologica Slovaca Monographiae. Studia 3. FUSEK, G. 2008, Frühe Slawen im Mitteldonaugebiet. – V / In: Bemmann, Schmauder (ur. / eds.) 2008, 645–656. FUSEK, G. 2015, Drevnee slavjanskoe naselenie na territorii Slovakii (Earliest Slavic Population on the Territory of Slovakia). – V / In: Rabinovich, Gavritukhin (ur. / eds.) 2015, 151–162. FUSEK, G., J. ZÁBOJNÍK 2003, Príspevok do diskusie o počiatkoch slovanského osídlenia Slovenska. – Slovenská archeológia 51, 319–340. FUSEK, G., J. ZÁBOJNÍK 2010, Frühslawisches Siedlungs- objekt aus Suchohrad. Zur Problematik der langobar- disch-slawischen Beziehungen. – V / In: J. Beljak, G. Březinová, V. Varsik (ur. / eds.), Archeológia barbarov 2009. Hospodárstvo Germánov / Archaeology of the Bar- barians 2009. Economy of the Germans, Archaeologica Slovaca Monographiae, Communicationes 10, 155–180. FUSEK, G., L. OLEXA, J. ZÁBOJNÍK 2010, k nové sídliskové nálezy z Nižnej Myšle. K problematike včasnoslovanského osídlenia horného Potisia. – Slovenská archeológia 58/2, 337–363. GAVRITUHIN, I. O. 1997, Chronologija pražskoj kultury. – V / In: Trudy VI Meždunarodnogo Kongressa slavjanskoj archeologii 3, 39–52, Moskva. GAVRITUHIN, I. O. 1998, Gorizont pražskoj kul'tury poselenija Zimno 4 (Horizont der Prager Kultur in der Siedlung von Zimme, Fst. 4). – Archaeoslavica 3, 171–207. GAVRITUHIN, I. O. 2005, Kompleksy Pražskoj kul’tury s datirujuščimi veščami (Die Fundkomplexe der Prager Kultur mit sicher datierenden Fundstücken). – V / In: P. Kaczanowski, M. Parczewski (ur. / eds.), Archeologia o początkach Słowian. Materiały z konferencji, 403–461, Kraków. GAVRITUHIN, I. O. 2009, Ponjatie Pražskoj Kul'tury. – V / In: Složenie Russkoj Gosudarstvennosti v kontekste rannesrednevekovoj istorii starogo sveta. Trudy gosu- darstvennogo jermitaža 49, 7–25, Sankt Petersburg. GAVRITUKHIN I. O. 2015, Dunajskij “strat” ranneslav- janskoj arheologii. Ot redaktorov jetogo nomera (The Danube “Stratum” of the Early Slavic Archaeology. A Word from the Editors). – V / In: Rabinovich, Gavri- tukhin (ur. / eds.) 2015, 17–26. GRAČANIN, H. 2008, Slaveni u ranosrednjovjekovnoj južnoj Panoniji. – Scrinia Slavonica 8, 13–54. GRAFENAUER, B. 1971–1971, Naselitev Slovanov v Vzhodnih Alpah in vprašanje kontinuitete. – Arheološki vestnik 21–22, 17–32. GUŠTIN, M. (ur. / ed.) 2002, Zgodnji Slovani. Zgodnje- srednjeveška lončenina na obrobju vzhodnih Alp / Die frühen Slawen. Frümittelalterliche Keramik am Rand der Ostalpen. – Ljubljana. GUŠTIN, M. 2003, Nova Tabla pri Murski Soboti. – V: B. Djurić et al., Zemlja pod vašimi nogami. Arheologija na avtocestah Slovenije. Vodnik po najdiščih, 199–201, Ljubljana. GUŠTIN, M. 2004a, Nova Tabla near Murska Sobota. – In: B. Djurić et al., The Earth Beneath Your Feet. Archa- eology on the Motorways in Slovenia. Guide to Sites, 205–207, Ljubljana. GUŠTIN, M. 2004b, Steklene posode z Nove Table pri Murski Soboti (Glass vessels from Nova Tabla near Murska Sobota). – V / In: I. Lazar (ur. / ed.), Drobci antičnega stekla / Fragments of ancient Glass, Annales Mediterranea, 71–79, Koper. GUŠTIN, M. (ur. / ed.) 2008, Srednji vek. Arheološke raz- iskave med Jadranskim morjem in Panonsko nižino  / Mittelalter. Archäologische Forschungen zwischen der Adria und der Pannonischen Tiefebene. – Annales Me- diterrranea, Ljubljana. GUŠTIN, M., D. PAVLOVIČ 2009, Karakteristike najstarije slavenske keramike. Primjer lonca iz objekta SO 11 s nalazišta Murska Sobota – Nova tabla (Characteristics of the Oldest Slavic Pottery. The Example of a Pot from Structure SO11 from theMurska  Sobota-Nova Tabla Site). – Archaeologia Adriatica 3, 2009 (2011), 293–303. GUŠTIN, M., D. PAVLOVIČ 2013, Die slawische Einwan- derung ins Prekmurje- Mura Gebiet (Pannonisches südostrand) auf Grund der 14C Daten. – V / In: M. Dulinicz, S. Moździoch (ur. / eds.), The Early Slavic 370 Daša PAVLOVIČ Settlement in Central Europe in the light of new dating evidence, Interdisciplinary medieval studies 3, 217–221. GUŠTIN, M., G. TIEFENGRABER 2002, Oblike in kro- nologija zgodnjesrednjeveške lončenine na Novi tabli pri Murski Soboti / Formen und Chronologie frühmit- telalterlicher Keramik in Nova tabla bei Murska Sobota. – V / In: Guštin (ur. / ed.) 2002, 46–62. GUTJAHR, C. 2002, Pregled raziskav zgodnjesrednjeveške keramike na Štajerskem / Ein Überblick zur frühmittel- alterlichen Keramikforschung in der Steiermark. – V / In: Guštin (ur. / ed.) 2002, 146–160. JELÍNKOVÁ, D. 1990, K chronologii sídlištníh nálezů s keramikou pražského typu na Moravé. – V / In: Pravéke a slovanské osídlení Moravy, 249–281, Brno. JELÍNKOVÁ, D. 2015, K izučeniju kul'tury s keramikoj prazhskogo tipa na territorii Moravii: formirovanie i osobennosti (On the problem of Prague-Type Pottery Culture in Moravia: the origins and specifics). – V / In: Rabinovich, Gavritukhin (ur. / eds.) 2015, 117–150. KARO, Š. 2012, Zgodnjesrednjeveško obdobje. – V / In: I. Šavel, Š. Karo, Popava pri Lipovcih 1, Zbirka arheologija na avtocestah Slovenije 30, 53–67. [http://www.zvkds.si/ sites/www.zvkds.si/files/uploads/files/publication/30_po- pava_pri_lipovcih_1.pdf]. KERMAN, B. 2011, Tü mo. Slovanska poselitev Prekmurja / Slavic settlement of Prekmurje. – Murska Sobota. KNIFIC, T. 1994, Vranje near Sevnica: A Late Roman Settlement in the Light of Certain Pottery Finds. – Arheološki vestnik 45, 211–237. KOBYLIŃSKI, Z. 2005, The Slavs. – V / In: P. Fouracre (ur. / ed.), The New Cambridge Medieval History, Vol. 1: c.500–c.700, 524–544, Cambridge. KUNA, M., N. PROFANTOVÁ et al. 2005, Počátky raného středověku v Čechách / The Onset of the Early Middle Ages in Bohemia. – Praha. KUNA, M., N. PROFANTOVÁ 2011, Prague-type cul- ture houses. Aspects of form, function and meaning (Nastambe praške kulture – Aspekti izgleda, svrhe i značenja). – Histria Antiqua 20, 415–429. LADSTÄTTER, S. 2000, Die materielle Kultur der Spätantike in den Ostalpen. Eine Fallstudie am Beispiel der westlichen Doppelkirchenanlage auf dem Hemmaberg. – Mitteilungen der Prähistorischen Kommission 35. MICHCZYŃSKI, A. 2007, Is it possible to find a good point estimate of a calibrated radiocarbon date? – Ra- diocarbon 49/2, 393–401. MILAVEC, T. 2007, Prispevek h kronologiji S-fibul v Slo- veniji / A contribution to the chronology of S-fibulae in Slovenia. – Arheološki vestnik 58, 333–355. MODRIJAN, Z., M. NOVŠAK 2015, Lonec z glajenim okrasom iz rimskodobnega vodnjaka v Dobovi / A pot with burnished decoration from a Roman period well near Dobova. – Arheološki vestnik 66, 23–45. OBLOMSKIJ, A. M., O. V. PETRAUSKAS , R. V. TERPI- LOVSKIJ 1999, Environmental reasons of migrations of the south-eastern Europe population in the 1st–5th centuries AD. – Archaeologia Polona 37, 71–86. PARCZEWSKI, M. 1991, Origins of Early Slav Culture in Poland. – Antiquity 65/248, 676–681. PARCZEWSKI, M. 1993, Die Anfänge der frühslawischen Kultur in Polen. – Wien. PAVLOVIČ, D. 2008, Novi izsledki arheoloških terenskih raziskav na Novi tabli pri Murski Soboti (Neue Ergeb- nisse der archäologischen Feldforschungen in Nova tabla bei Murska Sobota). – V / In: M. Guštin (ur. / ed.), Srednji vek. Arheološke raziskave med Jadranskim morjem in Panonsko nižino / Mittelalter. Archäologische Forschungen zwischen der Adria und der Pannonischen Tiefebene, 49–52, Ljubljana. PAVLOVIČ, D. 2012, The settlement and organization of early Slavic hamlets in Prekmurje, Slovenia (Naselja- vanje i organizacija naselja ranih Slavena u Prekmurju, Slovenija). – Dani Stjepana Gunjače 2, 317–332, Split. PAVLOVIČ, D. 2013, Zgodnjesrednjeveška poselitev se- verovzhodne Slovenije: primer najdišča Nova tabla pri Murski Soboti. – Neobjavljena disertacija / Unpublished PhD dissertation, Univerza na Primorskem, Koper. PAVLOVIČ, D. 2015, The beginning of Slavic settlement in north-eastern Slovenia and the relation between “Slavic” and “Lombard” settlement based on new interpretations of the archaeological material and radiocarbon dating. – Fundberichte aus Österreich, Tagungsbände 1, 59–79. PAVLOVIČ, D. 2017, Peeking into early medieval kitchens. Functionality and appearance of the sunken features from the site Nova tabla by Murska Sobota (Slovenia) (Kukati v zgodnjesrednjeveške kuhinje ali poskus dolo- čanja namembnosti objektov na najdišči Nova tabla pri Murski Soboti (Slovenija). – V / In: T. Sekelj Ivančan, T. Tkalčec (ur. / ed.) Srednjevjekovna naselja u svjetlu arheoloških izvora / Mediaeval Settlements in the Light of Archaeological Sources. Zbornik drugog međunaro- dnog znanstvenog skupa srednjovjekovne arheologije, Zbornik Instituta za arheologiju 6, 27–48, Zagreb. – v tisku / in print. PLEINEROVÁ, I. 2000, Die altslawischen Dörfer von Březno bei Louny. – Praga, Louny. PLETERSKI, A. 1990, Etnogeneza Slovanov. – Ljubljana. PLETERSKI, A. 2008b, Zgodnjesrednjeveška naselbina na Blejski Pristavi. Najdbe / Frühmittelalterliche Siedlung Pristava in Bled. Funde. – Opera Instituti Archaeologici Sloveniae 14. PLETERSKI, A. 2009, The inventing of Slavs or inventive Slavs? O ideovém světě a způsobu bydlení starých Slo- vanů. – Archeologické rozhledy 61, 331–336. PLETERSKI, A. 2010a, Zgodnjesrednjeveška naselbina na blejski Pristavi. Tafonomija, predmeti in čas / Frühmit- telalterliche Siedlung Pristava in Bled. Taphonomie, Fundgegenstände und zeitliche Einordnung. – Opera Instituti Archaeologici Sloveniae 19. PLETERSKI, A. 2010b, Datiranje z metodo radioaktivnega ogljika 14C in zgodnji srednji vek na ozemlju Slovenije / 14C-dating and the Early Middle Ages in Slovenia. – Argo 53/1, 45–49. PLETERSKI, A. 2010c, Datiranje zgodnjesrednjeveške naselbine Lehen pri Mitterkirchnu v zgornji Avstriji kot kontrola nove datacijske metode s pomočjo refe- renčne tabele in korelacijske formule ustij loncev (Die Datierung der frühmittelalterlichen Siedlung Lehen bei Mitterkirchen in Oberösterreich als Kontrolle einer neuen Datierungsmethode mit Hilfe einer Referenztabelle und einer Korrelationsformel für Topfränder). – Vjesnik Arheološkog muzeja u Zagrebu, 3. ser., 43, 309–324. 371Začetki zgodnjeslovanske poselitve Prekmurja PLETERSKI, A. 2013a, Frühe slawische Besiedlung Sloweni- ens im Lichte der 14C-Datierung. – V / In: M. Dulinicz, S. Moździoch (ur. / eds.), The Early Slavic settlement of Central Europe in the light of new dating evidence, Interdisciplinary Medieval Studies 3, 207–216. PLETERSKI, A. 2013b, Etnogeneza Slavena – metode i proces (The ethnogenesis of the Slavs, the methods and the process). – Starohrvatska prosvjeta, 3. ser., 40, 7–32. PLETERSKI, A. 2015, Rannie slavjane v Vostočnyh Al'pah i na sosednih zemljah (The Early Slavs in the Eastern Alps and Their Periphery). – V / In: Rabinovich, Ga- vritukhin (ur. / eds.) 2015, 227–248. POHL, W. 2008a, Migration und Ethnogenesen der Lango- barden aus Sicht der Schriftquellen. – V / In: Bemmann, Schmauder (ur. / eds.) 2008, 1–12. POHL, W. 2008b, Die Langobarden – zwischen der Elbe und Italien. – V / In: Die Langobarden. Das Ende der Völkerwanderung. Katalog zur Ausstellung im Rhei- nischen LandesMuseum Bonn 22.8.2008–11.1.2009, 23–33, Darmstat. PROFANTOVÁ, N. 2008, Die frühsslawische Besidlung Böhmens und archäologische Spuren der Kontakte zum früh- und mittelawarischen sowie merowingischen Kulturkreis. – V / In: Bemmann, Schmauder (ur. / eds.) 2008, 619–644. PROFANTOVÁ, N. 2009, Kultura s keramikou pražského typu a problém šíření slavinity do středni Evropy. K článku Florina Curty. – Archeologické rozhledy 61, 303–330. PROFANTOVÁ, N. 2012, Cultural discontinuity and the migration hypothesis. The 6th-century Slavic migration in the light of new archaeological finds from Bohemia. – V / In: R. Annaert, T. Jacobs, I. In 't Ven, S. Coppens (ur. / eds.), The very beginning of Europe? Cultural and Social Dimensions of Early-Medieval Migration and Colo- nisation (5th-8th century). Archaeology in Contemporary Europe Conference Brussels - May 17-19 2011, Relicta Monografieën 7, 255–264. PROFANTOVÁ, N. 2015, Slavjane na territorii Čehii i ih kontakty v VI—VII vv. (Slavs on the Territory of Czechia and their Contacts in 6th–7th Centuries). – V / In: Rabinovich, Gavritukhin (ur. / eds.) 2015, 97–116. PROFANTOVÁ, N., M. BUREŠ 2013, Bedeutende frühe slawische Siedlungen in Böhmen (6.–8. Jh.) und Probleme der Absoluten Datenerfassung. – V / In: M. Dulinicz, S. Moździoch (ur. / eds.), The Early Slavic Settlement in Central Europe in the light of new dating evidence, 169–194, Wroclaw. RABINOVICH, R. A., I. O. GAVRITUKHIN 2015, Slavjane na Dunae. Obretenie Rodiny (The Slavs on the Danu- be. Homeland Found). – Stratum plus 5, Archaeology and Cultural Anthropology, Saint Petersburg, Kishinev, Odessa, Bucharest. REIMER et al. 2009 = (P. J. Reimer, M. G. L. Baillie, E. Bard, A. Bayliss, J. W. Beck, P. G. Blackwell, C. Bronk Ramsey, C. E. Buck, G. S. Burr, R. L. Edwards, M. Friedrich, P. M. Grootes, T. P. Guilderson, I. Hajdas, T. J. Heaton, A. G. Hogg, K. A. Hughen, K. F. Kaiser, B. Kromer, F. G. Mccormac, S. W. Manning, R. W. Reimer, D. A. Richards, J. R. Southon, S. Talamo, C. S. M. Turney, J. Van Der Plicht, C. E. Weyhenmeyer) 2009, INTCAL09 and MARINE09 radiocarbon age calibration curves 0-50.000 years CAL BP. – Radiocarbon 51, 1111–1150. SARANTIS, A. 2009, War and Diplomacy in Pannonia and the Northwest Balkans during the Reign of Justinian. The Gepid Threat and Imperial Respons. – Dumbarton Oaks Paper 63, 15–40. SARANTIS, A. 2010, The Justinianic Herules: From Allied Barbarians to Roman Provincials. – V / In: F. Curta (ur. / ed.), Neglected Barbarians, Studies in the Early Middle Ages 32, 361–402. SCHALLERT, J., M. L. GREENBERG 2007, The Prehistory and Areal Distribution of Slavic *gъlčěti ‘Speak’. – Slo- venski jezik / Slovene Linguistic Studies 6, 9–76. SEKEJ IVANČAN, T. 2010, Podravina u ranom sredjem vijeku. Rezultati arheoloških istraživanja ranosrednjov- jekovnih nalazišta u Torčecu. – Monografije Instituta za arheologiju / Monographiae Instituti archaeologici 2. SNOJ, M., M. L. GREENBERG 2012, O jeziku slovanskih prebivalcev med Donavo in Jadranom v srednjem veku (pogled jezikoslovcev) (On the Language of the Medieval Slavic Population in the Area between the Danube and the Adriatic (from a Linguistic Perspective)). – Zgodo- vinski časopis 66 (146), 276–305. STADLER, P., H. FREISINGER, W. KUTSCHERA, A. PRILLER, P. STEIER, E. M. WILD 2003, Ein Beitrag zur Absolutchronologie der Langobarden Aufgrund von 14C-Datierungen und ein Versuch zur Datirung der Beraubung langobardischer Gräber. − Archaeologia Austriaca 87, 265–278. STEINACHER, R. 2010, The Herules: Fragments of a Hi- story. – V / In: F. Curta (ur. / ed.), Neglected Barbarians, Studies in the Early Middle Ages 32, 320–360. STUIVER, M., P. J. REIMER 1993, Extended 14C data base and revised CALIB 3.0 14C age calibration program. – Radiocarbon 35, 215–230. ŠALKOVSKÝ, P. 2001, Häuser in der frühmittelalterlichen slawischen Welt. – Archaeologica Slovaca monographiae. Studia Instituti Archaeologici Nitriensis Academiae Scientiarum Slovacae 6. ŠAŠEL, J. 1976, Omemba Slovanov v pesmi Martina iz Brage na Portugalskem. – Kronika 24, 151–158. ŠAŠEL KOS, M. 1994, Romulovo poslanstvo pri Atilu (Ena zadnjih omemb Petovione v antični literaturi) (The Embassy of Romulus to Atila (One of the last citations of Poetovio in classical literature)). – Zgodovinski časopis 48/3, 285–295. ŠAVEL, I. 2002, Zgodnjesrednjeveški objekt z najdišča Pod Kotom – jug pri Krogu / Ein frühmittelalterliches Objekt vom Fundort Pod Kotom – jug bei Krog. – V / In: Guštin (ur. / ed.) 2002, 11–16. ŠAVEL, I. 2008b, Staroslovansko grobišče Popava II pri Lipovcih (Das altslawische Gräberfeld Popava II bei Lipovci). – V / In: M. Guštin (ur. / ed.), Srednji vek. Arheološke raziskave med Jadranskim morjem in Pa- nonsko nižino / Mittelalter. Archäologische Forschungen zwischen der Adria und der Pannonischen Tiefebene, Annales Mediterrranea, 65–70, Ljubljana. ŠAVEL, I. 2009, Pod Kotom - jug pri Krogu. – Zbirka Ar- heologija na avtocestah Slovenije 7. [http://www.zvkds.si/sites/www.zvkds.si/files/uploads/ files/publication/007_pod_kotom-jug_pri_krogu.pdf]. 372 Daša PAVLOVIČ ŠTIH, P. 2008, Slovenska zgodovina. Družba – politika – kultura. – E – Razpoznavanja / E - Recognitiones 1, Lju- bljana. [http://www.dlib.si/?URN=URN:NBN:SI:DOC- -INOR6GFF]. TEJRAL, J. 2008, Zur Frage langobardischer Funde nördlich der mittleren Donau. – V / In: Die Langobarden. Das Ende der Völkerwanderung, Katalog zur Ausstellung im Rheinischen LandesMuseum Bonn 22.8.2008–11.1.2009, 53–71, Darmstat. TERPILOVSKIJ, R. V. 2005, Nasledie kievskoj kul’tury v V– VI vv. Das Erbe der Kiever Kultur im 5.–6. Jh. – V / In: P. Kaczanowski, M. Parczewski (ur. / eds.), Archeologia o początkach Słowian. Materiały z konferencji, Kraków, 387– 402, Kraków. VIDA, T. 2008, Die Langobarden in Pannonien. – V / In: Die Langobarden. Das Ende der Völkerwanderung, Ka- talog zur Ausstellung im Rheinischen LandesMuseum Bonn 22.8.2008–11.1.2009, 73–89, Darmstat. VOJAKOVIĆ, P. et al. 2016 = P. Vojakovič, T. Verbič, I. Bekljanov Zidanšek, M. Novšak, P. Leben-Seljak, T. Leskovar, B. Toškan, A. Šemrov, M. Culiberg 2016, Končno poročilo o arheoloških raziskavah na lokaciji OŠ Cerklje ob Krki, parc. št. 3124/1, 3124/2, 3124/11, 3126/1, 3127/11 k.o. Cerklje ob Krki, Ljubljana 2016. – Neobjavljeno poročilo / Unpublished Report; Ministrstvo za kulturo RS). ŽUŽEK, A. 2007, Naselitev Slovanov v vzhodnoalpski pro- stor (Die Landnahme der Slawen im Ostalpenraum). – Zgodovinski časopis 61/3–4 (136), 261–287. INTRODUCTION The modern archaeological excavations that took place between 1998 and 2008 in advance of the motorway construction between Maribor and Lendava, in north-eastern Slovenia, revealed a number of previously unknown early medieval habitation sites dating between the 6th and the 9th century. As the state provided comprehensive organised funding for such investigations, it was possible to take a high number of charcoal samples from the fills of the unearthed sunken features and subject them to analysis measuring the decay of the radioactive isotope 14C (hence the terms 14C or radiocarbon dating), which ranks among the most widely used methods of dating archaeologi- cal contexts. The greatest number of 14C dates from charcoal samples was taken at the Nova tabla site near Murska Sobota.1 The obtained 2σ calibrated age 1 The site at Nova tabla near Murska Sobota was excavated in three campaigns (1999–2001, 2002–2003, 2007–2008) that recorded 193 sunken features and 13 graves ascribed to the early medieval settlement located south and east of the artificial lake of Kamešnica near Murska Sobota (Guštin, Tiefengraber 2002; Guštin 2003; Pavlovič 2008; 2013; 2015). Beginnings of the Early Slavic settlement in the Prekmurje region, Slovenia Translation ranges fall between the years 414 and 941. The dates form two distinct groups: the early dates span the period from 414 to the 670s, the later ones from the 670s to 941. This chronological framework corresponds with the historical and archaeologi- cal interpretative model used to explain how the material culture groups, believed to represent the Early Slavs, arrived in the area of the present-day Slovenia, most likely in the 570s or 580s, after the Langobards had left for northern Italy in 568 as recorded in literary sources. In the introduction, we should say a few words concerning the (numerous) problems facing any researcher of the early medieval settlement, particu- larly that associated with the Slavs. This contribution contains a multitude of terms, formulations and topics that may be viewed as problematic by scholars with differing views, who may even dismiss the argumentation employed here altogether. Leaving aside the questions of defining the beginning of the Early Middle Ages, there are two basic sets of questions viewed from two differing standpoints: one is based on the writings of Florin Curta, the other on the modified views of Lubor Niederle. These views are present in the work of most of the prominent researchers of the early medieval 373Beginnings of the Early Slavic settlement in the Prekmurje region, Slovenia period in Slovenia, the Czech Republic, Slovakia, Poland, Ukraine, Russia and Belarus.2 The first set of questions focuses on defining the Early Slavs and where they came from. Curta believes that the name Slavs does not denote an ethnically homogeneous community migrating from the areas of the present-day Ukraine and southern Belarus (Curta 2001a, 337), but a community that formed through a process of social differentiation within a group of people. This process of emerging elites, coupled with a political and military mobilisation, is seen as a response to specific historical circumstances created by the fortification of the border along the Danube (ib., 343–344). Sclavene, as the name of the community, is believed to be a Byzantine construct created so as to facilitate understand- ing of the complex situation involving different ethnic groups living beyond the northern border; in short, Curta believes that the Slavic ethnicity is a Byzantine invention (ib., 118–119). The other, more widespread view is that the Slavs originate from a relatively small area in the upper reaches of the River Dnepr, that their communication was based on a common language and that they came to inhabit vast areas stretch- ing from Hamburg in present-day Germany to the Peloponnese in Greece by the end of the 8th century (Pleterski 2013b with references). The second set of questions deals with the pos- sible material traces of the Slavs, whether these have been archaeologically identified and whether it is at all possible to tie a specific material cul- ture with the Early Slavs (also on the subject see Pleterski 2013a, 207). Apart from the topical, but general theoretical questions such as associating archaeological cultures with individual peoples and understanding a material culture as an expression of identity, which we will not tackle here, this set of questions mainly focuses on whether the so-called Prague-type pottery culture (or Prague culture), that Ivan Borkovský formulated in 1940, does in fact represent the archaeological remains of the Early Slavs and what are the elements that characterise this culture. 2 Detailed new studies on the origin and settlement of the Early Slavs from Russia to the Mediterranean, also based on archaeological and other scientific analyses and dedicated to Lubor Niederle, have been collectively presented in a special volume of the Stratum Plus journal (Rabinovich, Gavritukhin 2015). As the name reveals, its most clearly identifiable element is pottery. It mainly consists of hand- built and vase- or barrel-shaped jars without a pronounced shoulder and with a short vertical rim. Other common elements are the location of dwellings near rivers, unfortified settlements with sunken-featured buildings arranged in a semicircle and furnished with stone ovens, cemeteries with small numbers of cremation burials in urns or in simple pits. These elements, however, only rarely appear together in a single site; usually only two or three have been identified together (Profantová 2012, 255–256). At Nova table, for example, archae- ologists unearthed Prague-type pottery, within an unfortified settlement located in the lowland and in proximity to several waterways, with sunken- featured buildings presumably arranged in rows, some also in semicircular groups around the central yard. The site did not reveal the otherwise typical square sunken-featured buildings with vaulted stone ovens (see Pavlovič 2013, 135–139). The Prague culture is believed to have formed in the 5th century or after the Hun incursion into Europe in 375. By pushing against the Germanic peoples living along the coasts of the Black Sea, the Huns created the conditions that led to the formation of three Early Slavic cultures, which included the Prague culture, in the area of the earlier Kiev and Chernyakhov cultures (Pleterski 2013b, 21–22). The sites that exhibit the above-enumerated characteristics have been documented in Ukraine, Belarus, Moldova, Romania, Poland, Slovakia, the Czech Republic, Slovenia, Croatia and Germany; given that those in Ukraine and Belarus are earliest in date (late 4th century), it is presumed that groups bearing this material culture migrated westwards and southwards in the late 5th and the 6th century (e.g. Profantová 2012, 256; Gavrituhin 2009, 11). In the 6th and 7th centuries, there is habitation discontinuity and a complex transformation observable in much of central Europe and in the Pannonian Plain as well. The areas of this discontinuity and the areas where literary sources first mention the Slavs correspond with the distribution area of the Prague culture (Profantová 2012, 255). The simplicity of this culture can hardly be explained as technological regression on the part of the surviving Late Antique communities; it is much more likely that it reflects the arrival of new groups of people with a lower level of technological development (Pleterski 2013a, 207). The analyses of the literary sources (Fusek 2008; Bratož 2014, 481–503), the comparability of small finds from Ukraine and central Europe, 374 Daša PAVLOVIČ the continuity between Prague-type pottery and the pottery from the 8th- and 9th-century hillforts in the present-day Czech Republic, that can be more reliably connected with historically attested Slavic populations, all support the hypothesis that the Prague culture represents the archaeological record of the Early Slavs (Profantová 2012, 260). In the last two decades, a completely different view on both the existence of the Prague culture and its association with the Early Slavs has emerged. Many objections have been voiced against previous interpretations, ranging from methodologically inappropriate classification of the Prague pottery, intuitively determined typology and selective choice of areas of research to misinterpretation and uncritical reading of the ancient texts. The conclusions of Curta that pertain to this discus- sion include the observation that the (relevant) archaeological finds from sites in Slovenia and Croatia only date to the second half of the 7th century and that there is no clear chronological difference between the Prague-type pottery and the later pottery decorated with wavy lines (Curta 2010, 34–35). Detailed and well-founded critiques of the hypotheses put forward by Florin Curta have been published by other authors as well, from the ranks of archaeologists, linguists and historians (e.g. Biermann 2009; Profantová 2009; Pleterski 2009; id. 2013b, 22–23; Snoj, Greenberg 2012, 283; Bratož 2014, 484–487). The discussion below is based on two 14C dates from the Nova tabla site with the 2σ ranges end- ing in 543 and 547, respectively. I will present and evaluate the archaeological finds unearthed together with the two dated samples, which will be compared with the finds recovered at early medieval sites from a wider area, the publications of which have been available to me. Interpretation follows in the last part, where I argue that archaeological comparisons, distribution of small finds, historical circumstances, numerous 14C dates, literary sources and the results of linguistic analyses all suggest that groups historically and linguistically associated with the Early Slavs, who migrated from the area of Ukraine to the fringes of the Pannonian Plain, were already present in the Prekmurje3 region in the first half of the 6th century. 3 For an overview of the archaeological investigations of the early medieval sites in the Prekmurje region in Slovenia see: Guštin (ed.) 2002 and id. (ed.) 2008; Kerman 2011. THE SITE OF NOVA TABLA NEAR MURSKA SOBOTA Until recently, very little was known regarding the material culture of the first newcomers associated with the Slavs on the territory of the present-day Slovenia. This changed at the end of the 20th and in the first few years of the 21st century, when large-scale excavations took place in north-eastern Slovenia in advance of the construction of the motorway net- work. A series of sites was excavated south of the city of Murska Sobota, which revealed a hitherto poorly known pattern of early medieval settle- ment in the lowland. The largest investigated area, covering as much as 40 ha, was at Nova tabla near Murska Sobota (Guštin (ed.) 2008 with references; Pavlovič 2008; 2012; 2013). The site yielded remains ranging from the Eneolithic to the modern period, with those from the Early Middle Ages comprising 193 sunken features forming a settlement (Fig. 3), as well as 13 inhumation burials. The sunken features were arranged in three large clusters. The largest was located in the centre of the excavated area, the two smaller ones to the west of it. The sunken features in the smaller clusters were located along a palaeochannel of the Dobel stream. The burials were found along one of the Roman-period burial plots surrounded by a ditch and possibly once covered over by an earthen mound. As for the sunken features, it is very dif- ficult to interpret them as either the remains of sunken-featured buildings, outbuildings or refuse pits because of their uncharacteristic shapes and the small number of hearths and ovens that rep- resent the most reliable indicators of the function (for more detail on the subject see Pavlovič 2017). The recovered pottery is distinguished by a characteristic porous surface or incised decoration, comprising groups of wavy and straight lines, and offers vital evidence for dating the remains from the last third of the 6th to the 9th century. The chronological framework is supported by a great number of 14C dates (Guštin, Tiefengraber 2002; Guštin, Pavlovič 2013, 151; Pavlovič 2015, 61). The pottery from the early medieval sunken features at Nova tabla represents the bulk of the small finds and the only remains that allow for a cultural and chronological attribution, albeit a broad one. The site revealed 6038 sherds (95 kg) of early medieval pottery. All belong to hand-built kitchen and storage vessels, predominantly jars; baking dishes are less common. The pottery is comparable with the so-called Prague-type pottery 375Beginnings of the Early Slavic settlement in the Prekmurje region, Slovenia associated with the arrival of the first Slavs in the central Danube Basin (Fusek 2008, 645; Fusek, Zábojník 2003; Kuna, Profantová 2011, 415). The settlement at Nova tabla and most other settlements connected with the Early Slavs in the western Pannonian Plain and other parts of central and eastern Europe are located in the lowland and have specific sunken features. The unearthed features or buildings were separate from one another and only rarely overlapped. The sites of Early Slavic set- tlements do not exhibit complex stratigraphy4 that would allow us to establish relative chronological relationships between the features and consequently the small finds. The Nova tabla site only clearly re- vealed a pattern of so-called horizontal stratigraphy. The features with earlier pottery were located in the central part of the excavated area, those with later pottery in the western part; we may infer from this that habitation shifted westwards (Guštin, Tiefen- graber 2002, 62; Pavlovič 2012, 322–327). The pottery also suggests two settlement phases (early and late) at the site, possibly with a short interruption between them. Chronological dif- ferences have thus been observed in the pottery (Pavlovič 2015, 61, Figs. 2–4) and in the location of the dwellings (Pavlovič 2012, 325). Attributable to the early phase are mainly the buildings of the largest cluster of sunken features, with individual satellite structures located at some distance. The late phase mainly comprises the buildings of the two, roughly equal-sized, western clusters (Pavlovič 2015, Fig. 5). This two-phase division is supported by two distinct ranges of 14C dates (Pavlovič 2015, Fig. 6). Nova tabla: 14C dates At the Nova tabla site, 36 charcoal samples were taken from 34 (19%) buildings (Fig. 1) and 4 The problem of the absence of overlap and hence the lack of stratigraphy in the early medieval lowland settlements has already been raised by the investigators of the large settlement at Roztokách near Prague. They have recorded instances of two or even three sunken-featured buildings overlapping, more exactly eight such instances, which is a very low number given the extent of the settlement (Kuna, Profantová 2005, 208). The overlapping buildings yielded very few small finds and the differences between the clusters of buildings are very small. A single building of a late date yielded more numerous finds, but these all belong to one decorated jar and one ‘classic’ Prague-type jar (ib., 347), which do not enable a chronological division. subjected to a 14C analysis.5 All 36 dates pertain to the samples of charcoal taken from the fills of the sunken features attributed to the Early Middle Age during the excavation because the associated pottery was deemed comparable with that from the Czech Republic, Slovakia and Poland. The sampled charcoal probably represents the remains of the buildings’ structural wood or of firewood; in most cases it is not possible to de- termine how it transformed to charcoal.6 The quality of a 14C date depends on the ‘degree of confidence in the archaeological context from which the sample is recovered, the demonstrated purity of the material to be analysed and the known accuracy and precision of the analytical method’ (Boaretto 2009, 275). Reliable archaeological con- texts are, for example, mortar surfaces, layers of plaster or daub, layers of phytoliths and organic residues inside complete vessels or bones. Given the general practice and knowledge of the excavators investigating at Nova tabla between 1999 and 2008, and given the specific nature of the site without surviving ground surfaces with floors or layers of phytoliths, only the charcoal in the fills of the sunken features could be sampled. Samples were taken where archaeologists found a sufficient amount of charcoal. Although charcoal is less reliable because it is usually scattered across a layer with many opportunities for intrusion, it can provide a relatively reliable sample providing that the micro and macro contexts are secure (Boaretto 2009, 278). Macro context is the layer in which charcoal was found. Understanding the creation of a layer also reveals the quality and reliability of a sample. Micro context is the local environment of the sample within the layer. The samples that might be contaminated by bioturbation, most of all those in the vicinity of rodent disturbances, or by human intervention should be excluded (ib., 278). In this sense, the Nova tabla site comes with a certain number of problems. The main one is its exposure to intense land cultivation that dam- aged the upper parts of all sunken features. The only uncontaminated parts were thus the parts of the fills at the very bottom. The second problem is that large constructions that would provide a 5 Samples were analysed at the Leibniz Labor für Altersbestimmung und Isotopenforschung, Christian-Albrechts- Universität Kiel. 6 The literature dealing with 14C dates (e.g. Bayliss 2009, 129; Pleterski 2010a, 86–87) places increasing emphasis on the taphonomy of the sampled context. 376 Daša PAVLOVIČ chronological framework for the creation of the charcoal, such as roof constructions, ovens or fireplaces with charred firewood remains, layers of a burnt-down house, clusters of charred cereals in a silo and so forth, have only rarely survived. The third problem is that the samples come without the data on the exact location within a fill. As the charcoal samples from Nova tabla do not originate from a compact layer or a large construction, it is not at all possible to dismiss the presence of old wood. Having said that, knowing the excavators and their level of knowledge on a personal basis, but also based on my own experience as deputy head of excavations in the 2007/2008 campaign and as an active participant in sampling the sunken feature of SO 149A, I believe that charcoal samples were taken where its quantity was highest and never in the upper parts of the fills susceptible to bioturba- tion and damage from land cultivation. Despite the caveats noted above, I consider the number7 and, as I will demonstrate below, the homogeneity of the obtained dates as indicative of appropriate sampling that yielded reliable results.8 New inves- tigations and possible other analyses may either confirm or refute this assumption. I calibrated the 14C dates from Nova tabla us- ing a uniform calibration curve.9 In their analysis, only the 2σ range, at 95% confidence level, was considered; the use of narrower ranges has been proven to be unreliable and possibly leading to great errors and erroneous conclusions (Michczyński 2007, 401). One sample was revealed to have belonged to a building from the Roman period.10 Other 14C 7 Pleterski writes that the whole procedure is based on a great amount of data and that the reliability of dating increases with the number of samples (Pleterski 2010a, 86, chapter on dating groups of vessel rims using the 14C method (Datiranje skupin ustij z metodo radioaktivnega ogljika 14C). 8 I recognise that charcoal obtained from taphonomically questionable contexts represents a poor basis for building systematic chronologies (cf. e.g. Boaretto 2009, 276). 9 Calib Rev7.1.0, RADIOCARBON CALIBRATION PROGRAM*, Copyright 1986-2016 M Stuiver and PJ Reimer. *To be used in conjunction with: Stuiver, M., and Reimer, P.J., 1993, Radiocarbon, 35, 215–230 and P.J. Reimer et al. 2009, Radiocarbon 51, 1111–1150. The last calibration performed on 4 February 2017. 10 The sample from Building SO 8, 14C 2σ from the mid-2nd to the end of the 4th century. During excavation, the building was attributed to the Early Middle Ages on the basis of the porous surface on a base fragment of a ceramic dates are shown on Fig. 1. The date ranges span the period from the early 5th to the early 10th century. The dates fall into two groups (Fig. 1): twenty dates show earlier ranges, fifteen show later ranges. There is only slight overlap between the two groups, in the 670s. The first group, of earlier ranges, can be divided into two subgroups. The first comprises eight dates with ranges beginning in the first half of the 5th century (SO 149A, SZ 6, SO 75, SO 161, SO 105, SO 149, SZ 10 and SO 11). The second subgroup consists of twelve dates of shorter ranges from the second half of the 6th century to the 670s (SO 169, SZ 2, SZ 11, SO 73, SZ 1, SZ 9, SO 47, SZ 3, SZ 4, SO 60 × 2 and SO 58). For most dates, the ranges of the two subgroups partially overlap, suggesting a possible contemporaneity of the sampled build- ing remains. The ranges of the fifteen dates of the second group (SO 164, SO 34, SO 44, SO 39, SO 156, SO 144, SO 33, SO 43, SO 155, SO 141, SO 166, SO 168, SO 40, SE 189 and SO 46) extend from the 670s to the early 10th century. The 14C dates indicate at least two chronologi- cal horizons within the early medieval settlement. The locations of individual charcoal samples were plotted onto a map of the site (Pavlovič 2013, 199, Fig. 79), revealing that the buildings with earlier ranges concentrated in the central part and the buildings with later ranges mainly concentrated in the western part of the site. This corresponds with the results of the pot- tery analysis: the sunken features that contained plain hand-built pottery and those that yielded the predominantly decorated pottery finished on a slow wheel are spatially mutually exclusive.11 The buildings that revealed hand-built pottery mainly formed part of the large, central cluster, while vessel. Re-examination of the small finds has revealed that this base fragment shows traces of being thrown on the fast wheel, which is indicative of the Roman period. Moreover, the building yielded a fragment clearly belonging to a Roman-period vessel. Building SO 8 is thus not early medieval, but Roman in date. Roman habitation at Nova tabla is corroborated by small finds from other buildings, but also by the well-preserved goods from a contemporary cemetery (Guštin 2003; id. 2004; Pavlovič 2013, 58–62). 11 Mitja Guštin and Georg Tiefengraber also observed two chronological horizons and, after the first excavation campaign, even noted that the settlement most probably 'spread' westwards (Guštin, Tiefengraber 2002, 62). For greater detail and a slightly different view see Pavlovič (2012, 322–327; 2013, 203–228; 2015; 2017). 377Beginnings of the Early Slavic settlement in the Prekmurje region, Slovenia those with wheel-finished and decorated pottery appeared almost exclusively in the western part of the site (Pavlovič 2013, 186, Fig. 74). The body of evidence, provided by pottery (in the 2σ ranges of 14C dates) and the distribution of buildings, allows us to conclude with a fair amount of confidence that there were at least two distinct chronological phases of the early medi- eval site at Nova tabla, an early and a late phase. Given the current state of research, it is difficult to say whether there was continuity between the two phases. We are, however, more inclined to suppose that the changes observable in the small finds are not indicative of continuous development from the early to the late phase, but rather reflect a hiatus between the two phases, with the changes in the second habitation phase being ascribable to external factors. The earliest early medieval 14C dates from Nova tabla The two earliest early medieval 14C dates do not postdate the mid-6th century (Figs. 1; 2). It is prob- able that the fills of the sunken features of SO 149A and SZ 6 (Fig. 3), from which the charcoal samples were taken, were created in the same period. SO 149A with its small finds and the 14C date has al- ready been published (Pavlovič 2008), while SZ 6 is comprehensively presented here for the first time.12 The two dates lead us to rethink the time when the first Early Slavic groups arrived to the region of Prekmurje and to re-examine the existing hy- potheses that are based on literary sources. The two sunken features were similar in shape and size (Figs. 4; 5). They were sub-oval in plan with relatively gently sloping walls and a flat bot- tom. SO 149A measured 2.4 m in length, 2 m in width and 0.25 in depth, while SZ 6 measured 2.25 m in length, 2.2 m in width and 0.45 m in depth. A smaller pit, SO 14913 was dug into SO 149A (Fig. 1) (Pavlovič 2008, 50–51). This small sunken feature contained much more charcoal and pottery (55 sherds weighing 928 g) than SO 149A 12 The pottery fragment from the sunken feature SZ 6: Pavlovič 2015, Fig. 3: 12. 14C date: Guštin, Pavlovič 2013, Fig. 3. 13 The charcoal from the sunken feature SO 149 has also been radiocarbon dated, to between AD 533 and 635 at 91.6% credibility (KIA 35476). (three sherds weighing 134 g). SZ 6 revealed 167 pottery sherds (2206 g) and a piece of slag. Comparing the small finds from SO 149A and SZ 6 shows clear common characteristics. The pottery from both sunken features is hand-built, has a pitted and porous surface as a consequence of burnt-out organic temper or leached limestone sand, and is plain. The reconstructed jars are barrel- shaped with maximum diameter in the upper third. The shoulder is unpronounced. SO 149A revealed one rim fragment of a jar, SZ 6 revealed five such fragments (Fig. 6). They belong to the group of short unpronounced rims that are often vertical, rarely slightly everted with a rounded lip. Rim height is chronologically diagnostic: short rims are early, longer and more everted ones are later (e.g. Kuna, Profantová 2005, 340 with references). Such jars are ascribed to the Prague-type pot- tery present across wide areas from Russia to Slovenia and interpreted as the earliest pottery of the Early Slavs.14 The archaeological contexts that only revealed this pottery with the so-called archaic rims, such as those on the pottery from SO 149A and SZ 6, are seen as the earliest phase of Slavic settlement (Parczewski 1993, 56–58; Fusek 1994, 101; Kuna, Profantová 2005, 212, 213, Fig. 84; Pleinerova 2000, 147–149). At Nova tabla, the only sherd that stands apart in this sense is one found in SZ 6 (Fig. 6: 11) which differs from others in its incised decoration that is unique at the site. Description of the sunken features and catalogue of small finds SO 149A (Fig. 4): Clearly discernible dark grey patch of irregular plan. The sunken feature had two parts: one is the earlier building of SO 149A, measuring 2.4 × 2 × 0.25 m,15 filled with brown-grey sandy loam con- taining charcoal and several pottery sherds. This 14 Numerous authors have studied the Early Slavic pottery (see e.g. Gavrituhin 1997; Terpilovskij 2005; Buko 1990), mainly those from Ukraine, Poland, the Czech Republic and Slovakia. Several authors have also published the history of research, the pottery classification methods (e.g. Fusek 1994, 18–34 with references; Kuna, Profantová 2005, 154–159) and the genesis of Prague-type pottery (e.g. Parczewski 1993, 62–65). For a critical view of determining Prague-type pottery see Curta (2001b). 15 The measurements in the text are taken from the record sheets. The drawing of the pit is based on the digital 378 Daša PAVLOVIČ feature was cut by the smaller SO 149, measuring 1.4 × 1 × 0.25 m, with a fill of dark grey sandy loam containing large quantities of charcoal and pottery sherds. Both features had gently sloping walls and a flat bottom. Small finds: – 3 pottery sherds (134 g), – 2 sherds of Roman pottery (5 g),16 – half a bead of bluish translucent glass. Description of select small finds (Fig. 6): 1. Fragment of a baking dish. Fabric: fine-grained. Surface: both surfaces porous. Colour: light brown to orange-brown. Size: h. 2.2 cm; w. 7.2 cm; th. up to 1.3 cm; reconstr. diam. 23.6 cm. – Temporary Inv. No. SO 149A/2. 2. Rim and upper body fragment of a jar. Fabric: fine-grained. Surface: both surfaces porous and smooth. Colour: both surfaces mottled, from light brown to brown. Production manner: clay coils and not wheel-finished. Size: h. 8 cm; w. 9.6 cm; th. 0.9 cm; reconstr. diam. 15.2 cm. – Temporary Inv. No. SO 149A/1. 3. Part of a glass bead. Colour: blue, translucent. Size: h. 0.4 cm; th. 0.16 cm; reconstr. diam. 0.38 cm. – Temporary Inv. No. SO 149A/3. SZ 6 (Fig. 5): Sub-round feature measuring 2.25 × 2.20 m in plan and dug 0.46 m into the sterile soil. Filled with yellow-brown sandy earth containing charcoal, cobbles, pottery sherds and plant seeds. Small finds: – 167 pottery sherds (2206 g), – 6 sherds of prehistoric pottery (103 g), – 8 cobble fragments (232 g), – 1 piece of slag (70 g). recording of it using ACad software. For reasons unknown, the two sets of measurements differ slightly. 16 Most of the early medieval sunken features revealed artefacts attributable to either the Roman period or to prehistory, ranging from complete or broken Roman bricks, glass fragments, pottery, a prehistoric stone axe and so forth. It has been established that some of these items were reused in the Early Middle Ages, for a purpose other than originally intended, I believe that the items did not arrive into the pits accidentally, but were picked up by the inhabitants in the vicinity and brought to their dwellings. The clearest example of such reuse is the complete Roman bricks or their fragments being used as bedding of hearths. The recovered glass sherds, e.g. hollow rims of beakers or urns, may have been used as necklace pendants. Objects such as the prehistoric stone axe may also have had an apotropaic significance. Description of select small finds (Fig. 6): 4. Rim and upper body fragment of a jar. Fabric: medium-grained; individual coarse inclusions; po- rous. Surface: smooth. Colour: mottled grey-brown and light brown, the latter at places with a reddish hue, on the exterior; brown-red and dark red on the interior. Production manner: hand-built. Size: rim diam. 13.4 cm; h. 7.6 cm. – Temporary Inv. No. SZ 6/3x. 5. Rim and upper body fragment of a small jar. Fabric: coarse-grained; porous. Surface: smooth. Colour: brown on the exterior; dark brown-grey on the interior. Production manner: hand-built. Size: rim diam. 7.6 cm; h. 6.5 cm. – Temporary Inv. No. SZ 6/10. 6. Rim and upper body fragments of a jar. Fabric: medium-grained; individual coarse inclusions; po- rous. Surface: coarse, original surface not preserved. Colour: mottled orange-brown, dark brown and dark brown-grey on the exterior; brown on the interior. Production manner: hand-built. Size: rim diam. 15.1 cm; h. 10.9 cm. – Temporary Inv. No. SZ 6/6x. 7. Rim and upper body fragment of a jar. Fabric: coarse-grained; porous. Surface: coarse. Colour: orange-brown. Production manner: hand-built. Size: rim diam. 14.6 cm; h. 16.9 cm. – Temporary Inv. No. SZ 6/7x. 8. Rim and upper body fragments of a jar. Fabric: medium-grained; individual coarse inclu- sions; porous. Surface: coarse. Colour: dark brown to dark brown-grey on the exterior; dark grey on the interior. Production manner: hand-built. Size: max. diam. 16.2 cm; base diam. 9.4 cm; h. 18.3 cm. – Temporary Inv. No. SZ 6/8x. 9. Rim and upper body fragment of a jar. Fabric: coarse-grained; slightly porous. Surface: coarse, original surface not preserved. Colour: light brown with the odd light grey-brown patch on the exte- rior; dark brown-grey and brown on the interior. Production manner: hand-built. Size: 4.5 × 4.4 cm. – Temporary Inv. No. SZ 6/3. 10. Rim and upper body fragment of a jar. Fabric: fine-grained; individual small inclusions. Surface: smooth. Colour: mottled brown-orange, orange, brown-red, ochre and dark brown on the exterior; brown-red and dark brown-red on the interior. Production manner: hand-built. Size: base diam. 5.0 cm; h. 4.3 cm. – Temporary Inv. No. SZ 6/1x. 11. Fragment of a jar. Fabric: coarse-grained; porous. Surface: smooth, original surface mainly not preserved. Colour: light brown. Production manner: hand-built. Size: 7.0 × 8.2 cm. Decoration: incisions. – Temporary Inv. No. SZ 6/2x. 379Beginnings of the Early Slavic settlement in the Prekmurje region, Slovenia THE EARLIEST 14C DATES FROM OTHER EARLY MEDIEVAL SITES IN PREKMURJE Pod Kotom – jug near Krog The investigated features at this site included a sunken feature (SE [= SU] 095/096) that provided a 14C date comparable to the earliest two from Nova tabla (Šavel 2009, 157–159; Šavel 2002, 11–16).17 The date was first published in 2010 (Pleterski 2010a, 88; Pleterski 2010b, 47, Tab. 1). At that time, the end date of the range, AD 535, seemed ‘much too early’ and was not considered in further analyses (Pleterski 2010a, 127). The publication states that the sunken feature was filled with dark grey loamy earth containing stone cobbles, charcoal, house daub, bone remains, pottery sherds and an iron object (Šavel 2009, 157). The charcoal was sam- pled, but the exact location where the sample was taken is not given in any of the publications. The pottery is of two kinds: a – hand-built and plain; b – finished on the slow wheel and decorated with incisions. The hand-built pottery includes sherds of short vertical or slightly everted rims such as are characteristic of early Prague-type pottery (e.g. Šavel 2002, Pl. 13: 3,6). Popava near Lipovci 1 At this site, the charcoal from Building 17 (SE 390) was sampled and the 14C analysis revealed a 2σ range between 420 and 540. The 14C date was deemed not to correspond with the small finds in the building (Karo 2012, 53). The pottery from this stratigraphic unit is not hand-built and cannot be attributed to the Prague-type; it is later, finished on the slow wheel and decorated with incisions or wavy lines. The published data do not offer an explanation as to the presumed discrepancy between the 14C date and the estimated age of the small finds in the fill. It is not easy to understand how the sunken features and their fills were created at the early medieval sites in Prekmurje, primarily the length of time it took for the features to become filled, making it all the more difficult to attempt to explain the discrepancies in dating. It is easy to dismiss the samples on the grounds of them being old wood that found its way into the fill through unknown processes or later land cultivation, or being the core of an old tree trunk much earlier than the time of the wood actually being used. 17 For the date see: Pleterski 2013a, 211, 212, Fig. 1; id. 2010b, 47, Tab. 1; id. 2010a, 88. Having said that, it is also possible to see them as an indication of the site being inhabited in the actual time interval proposed by the 14C date. DISCUSSION OF THE DATING OF THE POTTERY The closest comparisons for the Early Slavic Prague-type pottery with ‘archaic rims’ from Prek- murje come from Slovakia. Gabriel Fusek (1994) collected, classified and determined the pottery from the Slovak sites attributable to the Early Slavic settlement.18 He determined three chronological phases, I, II and III. He divided Phase I to Ia and Ib (ib., 65–76). Phase Ia is determined with contexts that yielded exclusively hand-built plain pottery with ‘archaic rims’ (Fig. 7), such as the pottery from SO 149A and SZ 6 at Nova tabla (Fig. 6). Phase Ia lasted from the transition from the 5th to the 6th century to the beginning of the last third of the 6th century, i.e. to the departure of the Langobards from Pannonia to Italy. From the archaeological standpoint, he based the early settle- ment of the Slavs in western Slovakia on the absence of Germanic sites and on the comparability of the small finds with those in the original settlement area of the Slavs, i.e. east of the Carpathian flanks (Fusek 1994, 93, 101). At the same time, from the historical point of view, he re-evaluated the relevant literary sources considered as supportive of the early Slavic settlement (Fusek 2008, 645–656). Recent excavations at Suhograd in westernmost Slovakia unearthed a building with an oven (Building 1) and sherds of hand-built Prague-type pottery with archaic features. The pieces of charcoal from the fill of the sunken feature were sampled and the two 14C dates obtained placed the building in the first half of the 6th century and certainly prior to 550 or 570, respectively (Fusek, Zábojník 2010, 155–180).19 Both the pottery (ib., Figs. 4; 5) and the ranges of the 14C dates (ib., Figs. 7; 8) are thus comparable with those of SO 149A and SZ 6 from Nova tabla. Parallels for the early Prague-type pottery are also known from the sites at Nižná Myšľa-Alamenev and Ždaňa (eastern Slovakia), both ascribed to the group of Early Slavic sites in the upper Tisa 18 Most of the small finds recovered up to that point. 19 Fusek also deals with the issue of the earliest Slavic settlement in the area of the present-day Slovakia in his 2015 contribution. 380 Daša PAVLOVIČ valley (junction of the modern states of Slovakia, Hungary, Romania and Ukraine), where the first phase of Slavic settlement dates after 470 and before the arrival of the Avars in 567/568 (Fusek, Olexa, Zábojník 2010, 354, Figs. 7, 16; Béreš 2013, 31, Pls. I; II). Like Fusek for Slovakia, Darinka Jelínková established a relative chronology for Bohemia (Jelínková 1990, 273, Fig. 20) and Moravia (ib., Fig. 19). She observes that the pottery of the first phase in Moravia closely corresponds with the earliest pottery from Ukrainian sites, but also that the differences in the historic and cultural characteristics prevent us from directly transfer- ring the Ukrainian absolute dates to Moravia that would place the beginning of the Slavic settlement in the late 5th or the first half of the 6th century. She dates the Slavic settlement of Moravia to the second half of the 6th century (ib., 277). This dat- ing is believed to be corroborated by the newly obtained 14C dates from Pavlov – Gorní Pole (Building 953) spanning the period from 551 to 648 (Jelínková 2015, 144). The more widespread presence of Prague culture in southern Moravia only in the second half of the 6th century is con- nected with the fact that, prior to that period, the area was settled by people with Merovingian cultural affinities, namely the historically attested Langobards before their move to Italy (Jelínková 2015, 135–143; Fusek 1994, 93; Fusek, 2008, 648; Fusek, Zábojník 2010, 170, Fig. 10). The latest relative chronology for Bohemia was published by Nad’a Profantová (Kuna, Profantová 2005, 211–219, Figs. 83, 90). The closest parallels for the pottery from the sunken features SO 149A and SZ 6 at Nova tabla come from her earliest Phase RI (Fig. 8). Profantová believes the comparisons of the pottery from Phase RI with the pottery from other regions, the occurrence of Germanic finds in the contexts of Phases I and II at the site at Roztoky and the historical interpretations indicate Slavic settlement in the area of the Czech Republic dur- ing the first half of the 6th century (ib., 219–225), though the 22 14C dates mainly from the Roztoky, Liboc and Běchovice sites in the vicinity of Prague do not explicitly confirm that (Profantová 2015, Pls. 1, 101–102; Profantová, Bureš 2013). Geographically more distant – but no less com- parable – are the earliest Slavic ceramic assemblages from Poland dated to the second half of the 5th and first half of the 6th century (Parczewsky 1993, 56, 58, Fig. 13). The earliest is the group of sites in southern Poland, where the sunken features are sub-oval in plan20 (ib., 101–102, Fig. 26), like those in Prekmurje (Pavlovič 2017). We should also note the similarity of the Prague- type pottery from Prekmurje with that from the sites in the present-day Ukraine. It was Igor O. Gavrituhin21 (1997) who established a typology of the Prague-type pottery and the chronology of the Prague culture in Ukraine. He divided the culture into three phases (I, II, III) and a pre-phase (0). The pre-phase lasted to the transition to the 5th century, Phase I is dated from the mid-5th to the first half of the 6th century, Phase II from the mid- 6th to the early 7th century and Phase III from 650 to the end of the 7th century. Based on Gavrituhin’s typochronology, the pottery from SZ 6 and SO 149A at Nova tabla is most similar to the jars of rather simple profiles and with either vertical or slightly everted rims (Type II, Subtypes Б and Д with rims of forms б, в and г; Gavrituhin 1997, Fig. 1). Such forms are typical of his Phase I, i.e. the time from the mid-5th to the early 6th century (ib., 43). The researchers of the Prague culture in Ukraine and Belarus distinguish between regional variants as the consequence of the differences in the cultural background and the integration of different neighbouring cultures (Baran 1981, 86), and within them local traits of the Prague-type pot- tery; analysis has shown, however, that the pottery from different regions exhibits a similar sequence of changes. They see the development and variety within the Prague culture as a chronologically and spatially dynamic system (Gavrituhin 1998, 200). In spite of that, it is possible to compare the pot- tery from Nova tabla with the Prague-type pottery from Ukranian sites such as Zymne (Buildings 19a, 21, 23 – Gavrituhin 1998, Pls. 1; 2; 3), Zelenyj Gaj (Building 4 – Baran 1987, Fig. 2), Raškov III (Buildings 12 and 38 – Baran 1988, Pls. 31; 39), Kodyn I (Buildings 17, 21 and 26 – Baran 1987, Fig. 5), Kodyn II (Buildings 12, 21 and 30 – Baran 1987, Fig. 6) and others. The importance of these comparisons lies in the well-founded dating of the latter to the second half of the 5th and the early 6th century, which is based on the metal artefacts of the Byzantine-Germanic cultures, mainly brooches and bracelets, as well as on the finds of Byzantine coins unearthed in the sunken-featured buildings 20 Similarly also the early sites in eastern Slovakia, i.e. the upper Tisa valley (Fusek, Olexa, Zábojník 2010; Béreš 2013). 21 There are several different transcriptions of his name: Gavrituhin, but also Gavrituchin and Gavritukhin. 381Beginnings of the Early Slavic settlement in the Prekmurje region, Slovenia of the Prague culture (Baran 1981, 67 and Fig. 2; Baran 1987, Fig. 13; Gavrituhin 1997, 43, Fig. 3: 44–46; Gavrituhin 2005). The first typological determinations of the jars from Prekmurje relied, in a large measure, on the typologies established for the finds from Slovakia and Poland; what differed was the chronological aspect. Mitja Guštin and Georg Tiefengraber proposed a pottery typology for Nova tabla, as well as a chronology supported by 14C dates, taking the finds excavated in the 1999–2001 campaign as basis. They distinguished between two horizons: Murska Sobota 1 and Murska Sobota 2 (MS 1 and MS 2). The early horizon (MS 1), which includes the sherds of Prague-type pottery, was attributed to the second half of the 6th century or late 6th and the 7th century on the basis of the 14C dates (Guštin, Tiefengraber 2002, 58–60). In 2010, Andrej Pleterski established a chronol- ogy of all early medieval finds from the eastern Alpine area (Pleterski 2010a, 85–160). He based his chronology on pottery, classified according to rim groups (S1-S7, V1-V2) as first determined for the pottery from the cemetery at Pristava in Bled (Pleterski 2010a, 63–84), and on 14C dates (ib., Fig. 4.63) from relevant, published sites in Slovenia, Croatia, Hungary, Italy, Austria and Germany (ib., Fig. 4.2). He created a reference table (ib., Fig. 4.95) that presents the periods when individual group of rims were in use (ib., 158–160). The exact dates that begin and end these periods were gained from the ranges of calibrated 14C dates us- ing a special procedure (ib., 126). As the earliest, Pleterski identified the jar rims of group S1. The common criterion for this group is that that the rims were not finished on the wheel. According to his chronology, such hand-built rims occur between 584 and 674, though an earlier beginning is not excluded (ib., 127–129). The most recent researcher to attempt a typo- logical, cultural and chronological assessment of the pottery from Prekmurje as part of a wider area covering Croatia, Austria and Hungary is Luka Bekić (2011, 37; id. 2016, 85). He includes the sites from Prekmurje in the group called Drava-Mura- Sava seen as representing the traces of the earliest Slavic settlement in the western Pannonian Plain in the late 6th and first half of the 7th century. His chronology is based on the typology of jar rims and the 14C dates provided by charcoal samples taken at the sites of Stara ves, Pod lipom, Blizna, Brezje, Šarnjak and Brekinjova kosa from northern and north-western Croatia (Bekić 2016, 85–99, Figs. 51 and 52).22 Stefan Eichert proposed a similar development of Early Slavic pottery based on the finds from Austrian Carinthia (Kärnten) and a wider eastern Alpine area. He believes that the area of Carinthia, previously using the local, Late Antique pottery, witnessed an introduction of new inhabitants and Prague-type pottery in the transition from the 6th to the 7th century. He presumes that two pottery traditions joined together at this time to produce a transitional ware,23 which differs from Prague- type pottery in the fabric, but is similar to it in the hand-built production manner and absence of decoration (Eichert 2010, 133, Fig. 33). Eichert’s view of the development of pottery in Carinthia is all the more interesting when it is compared to his chronology (Eichert 2010, 155, Fig. 43). Based on literary sources and a historical analy- sis, but most of all on a detailed analysis of the archaeological finds, Eichert sees four phases with occasional slight overlap (Eichert 2013b, Fig. 6). He suggests that the transitional phase began immediately after the arrival of the Slavs to the eastern Alps, which occurred around 590, and ended around 660 with the onset of the Group A graves. This group includes no pottery, because it is defined by the burials of an elite group (graves of the Grabalja vas type). The men of this group were buried with Avaro-Byzantine belts, some with Frankish weapons, while only a handful of women were buried with pieces of jewellery from the Mediterranean. Cemeteries often include a single burial of this type, surrounded by burials without grave goods. Such a disposition is believed to indicate a population of a specific social and economic stratification (Eichert 2013a). Cemeter- ies where the general population was buried with goods only appeared around 740 (beginning of the Group B graves). The pottery in these graves was decorated with wavy and horizontal lines, and finished on the wheel. The beginning and end of individual phases in absolute terms are mainly based on the analysis and comparison of 22 On Prague-type pottery in Croatia see: Bekić 2012, 21–35. 23 The hypothesis of contact between the indigenous population and the Prague culture group is founded, rather unconvincingly, on the archaeological finds from Hemmaberg near Globasnitz, where hand-built jars, presumably of the Prague-type pottery, were found together with jars of a Late Roman tradition in the sacristy area of the fourth church, above a layer of burnt remains (Ladstätter 2000, 159–160). 382 Daša PAVLOVIČ the grave groups verified by the 14C dates from the graves with representative goods (Eichert 2013b). Because of a lack of goods that would date between the late 6th and the last third of the 7th century, the beginning of the transitional phase at around 590 is defined with the last recorded Late Antique goods (ib., 419), while the end is defined with the appearance of the Group A graves (ib., 421). DISCUSSION OF THE ORIGINS OF THE EARLY MEDIEVAL SETTLEMENT Bogo Grafenauer (1970–1971, 20) believed that the first Slavs came to the Pannonian Plain from the north and stopped at the Danube in the first half of the 6th century. Based on the literary sources mentioning the Slavs, the widespread opinion was that the Slavic settlement wave turned southwards around 550 and swept across north- eastern Slovenia prior to 577, the year when the bishop of Poetovio (present-day Ptuj) is no longer mentioned in literary sources. The second great Slavic wave into Slovenian territory presumably occurred during the Avar push from the south and east following the departure of the Langob- ards to Italy in 568, most probably in the last two decades of the 6th century (Grafenauer 1970–1971, 23; Štih 2008, 20; for a slightly different opinion see Žužek 2007, 266–268, 275–278). Grafenauer based his research into the Slavic settlement of Slovenian territory on the surviving synodical records of the Patriarchate of Aquileia, which, in his opinion, showed a gradual decline of dioceses. The absence of bishops’ signatures was taken as an indication of the downfall of their dioceses, which Grafenauer attributes to the advancing Slavs (Grafenauer 1970–1971, 24–25). Rajko Bratož (2014, 505–547)24 also studied the downfall of the dioceses in more detail. He, however, believes that the bishops from the endangered dioceses failed to attend the synods for reasons unknown and only mentions the threat of a Slavic invasion as a note in brackets (ib., 535).25 As for the issue of the 24 Here he also gives a history of research into the settlement of the Slavs in the eastern Alpine areas, which is a historical view that also takes into account the archaeological findings. 25 Hrvoje Gračanin offers a slightly different view of the reasons behind the downfall of the dioceses. He believes that the absence of the name of a bishop may be related to the formal abolition of the respective civitas due to a non-functioning city centre and an insufficient number of inhabitants in the diocese (Gračanin 2008, 25). Slavic advancement towards the eastern Alps, he states that the results of the analysis of the liter- ary sources largely correspond with the findings of the archaeological investigations of the early Slavic migratory settlements in the Prekmurje and Podravje regions, agreeing with the proposed dat- ing of the late 6th and early 7th century (ib., 493). There is, however, a hint of circular reasoning in this, as the first chronologies of the archaeologi- cal finds, in spite of the available 14C dates, were adapted to suit the historical interpretations of the literary evidence, while Bratož corroborates his analysis of the literary sources with the very same archaeological interpretations that are based on the historical ones. I have shown above that the earliest 14C dates and the comparable pottery from the sites in the Czech Republic, Slovakia, Poland and Ukraine indicate that the earliest Prague culture groups may have settled in Prekmurje region as early as the first half of the 6th century. Such dating corresponds neither with the hith- erto accepted dating of the early medieval sites in Prekmurje nor the chronologies for the neighbour- ing areas of Austrian Carinthia and Croatia, nor with the widely accepted historical views regard- ing Early Slav settlement of the western fringes of the Pannonian Plain. I have presented a different view in my doctoral dissertation (Pavlovič 2013, 203–210), while others have already discussed the implications of such a view (Pleterski 2015, 241). The absence of Late Antique finds from the lowlands of north-eastern Slovenia suggests that there must have been very few inhabitants living here in the 5th and partly also the 6th century (Ciglenečki 2000, Fig. 2). The political, economic and possibly also climatic instabilities (Oblomskij, Petrauskas, Terpilovskij 1999, 84), but mainly the incessant incursions and migrations of peoples resulted in some – even vast – areas between the Carpathian Basin and central Europe becoming depopulated. Literary sources report of such desolate areas. Procopius mentions that the Herules, who lived on the north bank of the Danube before their defeat at the hands of the Langobards in 509, ‘traversed all the nations of the Sclaveni (Sclavenoi) one after the other, and after next crossing a large tract of barren country, they came to the Varni’ living in the north of Germany. The report by Gregory of Tours pertains to a slightly later time, writing that the Avars marching against the Franks suffered from a lack of food supplies, which implied that 383Beginnings of the Early Slavic settlement in the Prekmurje region, Slovenia they crossed deserted lands (Kobyliński 2005, 531; Steinacher 2010, 353, Fn. 132). Marcellinus Comes, a chronicler of the Eastern Roman Empire up to 534, wrote that they settled the barbarians in different areas including the deserted areas and cities of the Romans. He illustrated this with the example of the Herules, part of which were set- tled, following their defeat by the Langobards in 509, in the area of the Gepids, and then, in 512, within the Empire (Croke 2001, 74, 131),26 most probably the deserted area in the vicinity and north of Singidunum (present-day Belgrade) (Steinacher 2010, 350, 351). Literary sources thus indicate the existence of vast uninhabited areas in the Pannonian Plain north and south of the Danube (Fig. 9). Comparable with the vacant land around Singidunum, the absence of archaeological finds suggests there were deserted areas also in the southern and western fringes of the Pannonian Plain, for example in the regions of Podravina in Croatia (Sekelj Ivančan 2010, 165) and Prekmurje in Slovenia (Ciglenečki 2000, Fig. 2). These empty areas enabled groups of Early Slavs to easily migrate from beyond the Carpathians to the Pannonian Plain, across the Danube and all the way to Prekmurje, without having to engage in battles and thereby also avoiding mention in Byzantine written records. Supporting this hypothesis, regarding the set- tlement of Prague culture groups associated with the Early Slavs in the area of present-day Prek- murje and northern Croatia in the first half of the 6th century, is the first piece of epigraphic evidence mentioning Slavs (Sclavus in singular) in Pannonia. In 558, Martin of Braga wrote an encomium to Martin of Tours in an epigraphic dedication of the cathedral, finished in that year, in the monastery of Dumio near Braga (Bracara Augusta) in Portugal. The dedication lists the peoples, which included the Slavs, who lived in Pannonia and came to know Christ through the efforts of Martin of Tours (Šašel 1976, 151, 152). Martin of Tours was born in Savaria (present-day Szombathely, Hungary), while Martin of Braga most likely originated from the area where the Rivers Drava and the Sava flew into the Danube (Bassianensis Pannonia). Jaroslav Šašel showed that Martin of Braga did not list the peoples in an orderly manner, but rather spontaneously, from the memories of his youth, most probably from 26 Alexander Sarantis (2010) discussed this period in the history of the Herules in detail. the 520s. Šašel justifiably asked the question of which Slavs Martin of Braga had in mind. Given the available knowledge, Šašel presumed the Slavs living in the area of the Gepids, in the present-day Romania, or north of the Danube, in the present- day Slovakia (Šašel 1976, 156, also 155 on the map). The new discoveries discussed above suggest that Martin of Braga may have referred to the Slavs living in the vicinity of the place of birth of Martin of Tours, i.e. in Prekmurje and northern Croatia.27 An important consideration that supports this hypothesis is provided by the study of the Lan- gobard habitation traces in the Pannonian Plain and the eastern Alps.28 Archaeological sources suggest that the Lan- gobards inhabited the areas between the Szom- bathely–Keszthely–Pécz line and the knee of the Danube, and also lived in hilltop settlements west of the Ptuj–Sisak line. The area between the two lines revealed no archaeological traces of either Langobard or Late Antique habitations. I argue below that the area between Lake Balaton and Ptuj was settled by the Slavs already in the first half of the 6th century. Towards the end of the 5th century, the Langob- ards moved from the area along the lower reaches of the River Elbe, in the present-day Germany, to the areas of Lower Austria and Moravia (Pohl 2008b, 26).29 Based on the recovered pottery re- mains (Fusek 1994, 102), the arrival of the Early Slavs from Carpathian Ruthenia in the area of Slovakia has been dated to the late 5th and the early 6th century. Procopius’ account of Ildiges, a Langobard pretender to the throne who also sought refuge with the Slavs (Fusek 2008, 646) suggests that the Slavs inhabited the area south of the Carpathian Mountains and in the vicinity of the Danube by the first half of the 6th century at the latest. The communities of the Langobards and the Slavs thus lived side by side, bordering each other, in the Pannonian Plain north of the Danube in the first half of the 6th century. Evidence 27 On Martin of Braga and later authors mentioning the Slavs also see Gračanin (2008, 23–24) with references (ib., Fn. 74). 28 This is not a completely new hypothesis. The earlier arrival of individual Slavic groups to the fringes of the Pannonian state of the Langobards has already been mentioned by other researchers (Pleterski 1990, 51; id. 2015, 241; Gavritukhin 2015, 20–21). 29 For an overview of the literary and archaeological evidence on the Langobard migrations from the River Elbe to Italy see e.g. W. Pohl (2008a, 1–12; id. 2008b, 23–33). 384 Daša PAVLOVIČ of this is the artefacts attributed to the Meroving- ian culture complex, i.e. the Langobards, which were found in sunken features together with the Prague-type pottery (Profantová 2008, 623–631; Fusek, Zábojník 2010, 172). Advancing Prague culture groups may have turned southwards after contact with the Lango- bards, continuing all the way to the present-day Prekmurje. Speaking in favour of such a hypothesis is the Prague-type pottery from Nova tabla that shows similarities with the earliest Slavic pottery from Ukraine from the late 5th or early 6th century (see above). Langobard migrations are well known from literary sources (see e.g. Pohl 2008a), but the key evidence for understanding the settlement of the Langobards in Pannonia is provided by archaeo- logical sources and the 14C dates of osteological finds. The settlement was gradual and began in the early 6th century. Cemeteries reveal that ini- tially, during the first two thirds of the 6th century, they inhabited the northern part of Pannonia, a belt south of the Danube. Here, the longer lasting Szentendre type cemeteries, with a greater number of burials, have been documented over a time span extending from 510 up to the Langobard depar- ture for Italy in 568. Dated between 535 and 568 are smaller Vörs-Kajdacs type cemeteries with a shorter duration, documented between (south) eastern Pannonia and the western part of the Lake Balaton area (Vida 2008, 76, Fig. 1; Stadler et al. 2003, 265–269, Pl. 2).30 The final movement of the Langobards in their Pannonian phase, recorded in literary sources, came soon after 546 (548 at the latest), when Byzantine Emperor Justinian, waging war with the Goths and in fear of the advancing Franks, made an official gift of pólis Norikón, of Pannonian forts and numerous other places together with a large sum of money (Bratož 2014, 454–458). Regardless 30 The attempt at absolutely dating the Langobard settlement north of the Danube and in the Pannonian Plain is based on the results of the 14C analyses of 63 mostly collagen samples of human bone remains from sites in Moravia, Lower Austria and western Hungary. The obtained 14C dates have been statistically analysed using the OxCal software taking into account the historically established settlement phases of the Langobards in the Pannonian Plain. The authors of the project conclude that the absolute time frame of the Langobard settlement of this area based on 14C dates is most likely comparable with the time frame as revealed by the archaeological investigations of the above-mentioned cemeteries (Stadler et al. 2003). of whether we agree with the interpretation that the name of pólis Norikón marked a city with its administrative territory, most probably of former Poetovio or Celeia (Šašel Kos 1994, 294), or the area between them, while the mention of Panno- nian forts referred to the hillforts between Sisak (Croatia) and Ajdna (NW Slovenia) (Ciglenečki 2005, 272, 274–275, Fig. 5; Ciglenečki 2001, 187, 188, Fig. 5), it is certain that the Byzantines gave the Langobards the area west of the Ptuj–Sisak line. The archaeological finds indicate small garrisons of Langobards with their families living in hillforts together with the Roman population west of the Ptuj–Sisak line and guarding the passages into Italy (Ciglenečki 2000a, 122–123; id. 2005, 271, 272). The earliest pottery with stamped and burnished decoration, associated with the Langobards and recovered at hilltop settlements (Ajdovski gradec above Vranje, Rifnik, Tinje above Loka pri Žusmu, Puštal above Trnje and Kranj; – Knific 1994, 222, Fig. 15; Modrijan, Novšak 2015, 30), dates to the first third of the 6th century (Knific 1994, 219). Similar dating has been obtained from a typologi- cal analysis of the S-brooches, characteristic of the Langobards, from Slovenian sites that also reveal connections with the Langobard pre-Pannonian phase, i.e. to a time prior to 510 (Milavec 2007, 333). Surprisingly, the Langobards settled a large part of Pannonia, even extending west of Lake Balaton to central and western Slovenia, but avoided Prek- murje and its vicinity. Langobard finds are also virtually absent in Austrian Styria (Steiermark), where a single Langobard brooch has been recorded and very little stamped pottery, all without clear contexts (Gutjahr 2002, 147). Tivadar Vida believes that the reasons for the absence of Langobard habitation traces west of the Szombathely–Keszthely–Pécz line should be sought in the provincial Roman population that presumably inhabited this area and deterred a westward expansion of the Langobards (Vida 2008, 76). This, however, seems hardly likely. The geographically closest and archaeologically proven remains of the Roman population from the first half of the 6th century come from the hills of the Styria region (Štajerska) in Slovenia (Ciglenečki 2000a, Fig. 2). Extensive archaeological investi- gations taking place over the last two to three decades in the lowland areas of Dravsko polje, Prekmurje and western Hungary have failed to reveal traces of Late Antique habitation, which is, to the contrary, well-documented on the hilltop settlements to the west of this area. Moreover, the 385Beginnings of the Early Slavic settlement in the Prekmurje region, Slovenia small finds attributed to the Merovingian cultural circle and identified with the historically attested Langobards, which have been recovered from the hilltop settlements in Slovenia, point to a cohabita- tion of the Langobards and the Roman population (Ciglenečki 2001; 2005). It is my opinion that the Langobard expansion in the said area was rather hindered by the pres- ence of Prague culture groups, who were already living in Prekmurje and northern Croatia in the first half of the 6th century. I believe that the Langobards respected the boundaries of the Slavic settlement in Prekmurje, as they did north of the Danube. Focusing on the area of the River Morava, Gabriel Fusek and Jozef Zábojník analysed the site at Suchohrad and mapped all the known contemporary Langobard sites, on the one side, and those Prague culture groups (Phase Ia, first half of the 6th century), on the other. This showed that the two communities must have respected each other’s territories (Fusek, Zábojník 2010, 172, Figs. 10; 11). A cohabitation of the bearers of the culture of the Prague-type pottery, i.e. the Slavs, and the Langobards on the territory of the present-day Slovenia may be indicated by a pottery sherd that shows a production manner characteristic of the Prague-type pottery, but bears stamped decoration (Fig. 10), typical of the Langobard pottery. It was unearthed at Cerklje ob Krki,31 in south-eastern Slovenia, in the sunken feature of SU 1224/1225 with three recorded fills. All three yielded sherds of hand-built pottery with short and either vertical or slightly everted rims typologically attributable to the Prague-type pottery.32 The stamps on the said sherd are simple circles that do not appear to form a pattern. Confirming the archaic nature of the Prague-type pottery and the chronological possibility of cohabitation is the 14C date obtained from a charcoal sample from the fill of SU 1225. The 2σ range ends with the year 560 (report by Vojaković et al. 2016, 120). Stamped decoration on Prague-type pottery has relatively rarely been recorded on sites in Bohemia and Moravia, it is slightly more frequent in western Moravia. It has 31 Unpublished report: Vojaković et al. 2016. Investigations conducted by: Arhej, d. o. o., head of excavation Matjaž Novšak, deputy Petra Vojaković. I sincerely thank the team of Arhej for kindly informing me of their discoveries. 32 This is a preliminary attribution and we hope that a detailed analysis will follow shortly. I would like to thank Petra Vojakovič for sharing the information with me. been ascribed to the influence of Germanic peo- ples, primarily the Langobards (Fusek 1994, 62; Kuna, Profantová 2005, 341, Fig. 69: A). Profan- tová believes that the greater occurrence of such decoration in western Slovakia is connected with a longer cohabitation of the Early Slavs with the Langobards and the Gepids. She attributes the pottery with such decoration from Roztoky in the Czech Republic to the earliest phase of that site (Kuna, Profantová 2005, 174). The Langobard pottery does not often bear such stamped circles. Some sherds of this kind have been documented in Moravia, more precisely on the sherds of hand-built pottery from sites such as Šaratice (Tejral 2008, Fig. 7: 5), which came to light in graves ascribed to the Langobards with a distinct Polabian Germanic tradition, dating to the second half of the 5th century (ib., 64–66). I presume that the sherd with stamped decoration from Cerklje ob Krki reflects the contacts between Germanic peoples, presumably the Langobards, and Prague culture groups, very probably the Early Slavs, that took place in the area of south-eastern Slovenia. Such an interpretation suggests a date prior to the Langobard move to Italy in 568. Similar contact between Prague culture groups and those of the Merovingian cultural circle south of the Danube has been observed in the An den Klostergründen cemetery in Großprüfening, Regensburg, which revealed cremation burials in urns identified as Prague-type jars (Losert 2011, 482–483, Fig. 7). The metal goods in the graves point to a date of the graves to the second half of the 6th and early 7th century. The possibility of a cohabitation of two communities already in the first half of the 6th century is indicated by two 14C dates, for Graves 11 and 20 with the 2σ ranges ending in 546 and 569, respectively (ib., 489; Pleterski 2014, 23–25). In connection with this newly proposed hypoth- esis, it is also interesting to consider analyses of the languages of the Slavic inhabitants living between the Danube and the Adriatic in the Middle Ages. There are two main views on the spread of the Slavic language to this area: a migration and a diffusion view (Snoj, Greenberg 2012, 282). Marko Snoj and Marc L. Greenberg firmly believe that the migration model of the linguistic development of the Slavic language on the territory of the present-day Slovenia explains the traces of differences from an earlier period, which is the time before 500. As one of the instructive examples, the authors cite the clas- 386 Daša PAVLOVIČ sification analysis of the lexeme *gъlčěti, meaning ‘to speak’, which is present in the Prekmurje and neighbouring Pannonian dialects, but not in the rest of the Slovenian language, and also appears in some of the dialects in Bulgaria. The semantic development of the verb *gъlčěti from meaning ‘to speak’ to meaning ‘to roar, buzz, rumble’ was an innovation in semantic-formal agreement with the antonym of *mъlčěti meaning ‘to be silent’. The same new meaning can be also found in central Russia, which points to a connection between the eastern Slavic areas and the population that migrated to the areas of the later Prekmurje and Bulgarian dialects (Snoj, Greenberg 2012, 283).33 From the linguistic point of view, it is thus pos- sible to see the arrival of Slavic speaking groups to Prekmurje prior to or around 500. CONCLUSION The last three decades have greatly advanced our knowledge of the Early Slavs on the territory of the present-day Slovenia. The Prague-type pottery with close parallels among the earliest pottery of this type from sites across Ukraine, Poland, the Czech Republic and Slovakia, coupled with the 14C dates of the char- coal samples from SO 149A and SZ 6 at Nova tabla and from the sunken feature of SE 095/096 at Pod Kotom – jug all point to a date prior to the mid-6th century. What such a date suggests is that the first Prague culture groups settled the fringes of the Pannonian Plain already in the first half of the 6th century. The area may have been uninhabited prior to their arrival, as deduced from the absence of Late Antique sites in Prekmurje and the lowlands of northern Croatia. This hypothesis concerning the Slavic settlement of Prekmurje in the first half of the 6th century is supported by the results of linguistic analyses, but also by the fact that the Langobards avoided the area between Lake Balaton and the plain of Dravsko polje in their set- tlement of the Pannonian Plain, implying that they respected the boundaries of a community, i.e. the Slavs, already living there. In light of a presumed Slavic settlement of the Prekmurje area in the first half of the 6th century, the first epigraphic record of the Slavs in Pannonia, provided by Martin of 33 A detailed analysis of the verb *gъlčěti can be found in the article by Joseph Schallert and Marc L. Greenberg (2007, 9–76). Braga, may actually relate the real demographic situation of Pannonia in that period, rather than merely reflecting influences from Constantinople, and should not be discounted as empty rhetoric or exaggeration intended to emphasise the rapid spread of Christianity.34 In short, mounting evidence, from a range of disciplines, strongly supports the conclusion that small-scale Prague culture groups, corresponding to the Early Slavs, were already beginning to settle the western fringes of the Pannonian Plain during the first half of the 6th century. Acknowledgement Special thanks go to Alan Rushworth for proofreading the English text. The contribution presents the results of the research (co) financed from the national budget pursuant to the contract between the Slovenian Research Agency and the National Museum of Slovenia (research programme P6-0283). Translation: Andreja Maver Daša Pavlovič Narodni muzej Slovenije Prešernova 20 SI-1000 Ljubljana Dasa.Pavlovic@nms.si 34 This is an opinion voiced by Florin Curta, who sees the writings of Martin of Braga as irrelevant for arguing such an early presence of the Slavs in Pannonia (Curta 2001a, 46). For a differing opinion on the relevance of the said writings see Bratož (2014, 285–286).