Content uploaded by Stephan Karl
Author content
All content in this area was uploaded by Stephan Karl on Dec 02, 2024
Content may be subject to copyright.
Gabrielle Kremer – Eduard Pollhammer – Julia Kopf – Franziska Beutler (Hrsg.)
ZEIT(EN) DES UMBRUCHS
Akten des 17. Internationalen Kolloquiums zum provinzialrömischen
Kunstschaffen
Wien – Carnuntum, 16.–21. Mai 2022
01-25_CRPA_n.indd 1 20.02.24 13:19
Veröffentlichungen aus den Landessammlungen Niederösterreich
Nr. 7
01-25_CRPA_n.indd 2 20.02.24 13:19
Österreichisches Archäologisches Institut
Sonderschriften Band 64
Gabrielle Kremer – eduard Pollhammer – Julia KoPf –
franzisKa beutler (hrsG.)
ZEIT(EN) DES UMBRUCHS
Akten des 17. Internationalen Kolloquiums
zum provinzialrömischen Kunstschaffen
Wien – Carnuntum, 16.–21. Mai 2022
01-25_CRPA_n.indd 3 20.02.24 13:19
Herausgeber
Österreichisches Archäologisches Institut
Reihenherausgabe: Sabine Ladstätter, Martin Steskal, Alice Waldner, Barbara Beck-Brandt
Dominikanerbastei 16
A-1010 Wien
<www.oeaw.ac.at/oeai/>
Das Österreichische Archäologische Institut ist eine Forschungseinrichtung der
Österreichischen Akademie der Wissenschaften
Veröffentlichungen aus den Landessammlungen
Niederösterreich, Nr. 7
Herausgegeben von Armin Laussegger Gedruckt mit Unterstützung des Landes Niederösterreich
Eigentümer & Verleger
Verlag Holzhausen GmbH
Traungasse 14–16
A-1030 Wien
<https://shop.verlagholzhausen.at/collections/archaeologia>
Lektorat und Redaktion: Barbara Beck-Brandt, Franziska Beutler, Julia Kopf, Gabrielle Kremer, Wien
Englisches Lektorat: Sarah Homan-Cormack, Wien
Satz und Layout: Andrea Sulzgruber, Wien
Alle Rechte vorbehalten
1. Auage 2024
Verlagsort: Wien – Printed in Austria
Die verwendete Papiersorte ist aus chlorfrei gebleichtem Zellstoff hergestellt, frei von säurebildenden Bestandteilen und alterungs-
beständig.
ISSN 1998-8931
ISBN 978-3-903207-86-8
Copyright © 2024
Verlag Holzhausen GmbH
Bibliograsche Information der Österreichischen Nationalbibliothek und der Deutschen Nationalbibliothek: Die ÖNB und die DNB
verzeichnen diese Publikation in den Nationalbibliograen; detaillierte bibliograsche Daten sind im Internet abrufbar. Für die Öster-
reichische Bibliothek: <https://onb.ac.at>, für die Deutsche Bibliothek: <http://dnb.dnb.de>.
Alle Rechte, insbesondere das Recht der Vervielfältigung und Verbreitung sowie der Übersetzung, sind dem Verlag vorbehalten. Kein
Teil des Werks darf in irgendeiner Form (durch Fotokopie, Mikrolm oder ein anderes Verfahren) ohne schriftliche Genehmigung des
Verlags reproduziert oder unter Verwendung elektronischer Systeme gespeichert, verarbeitet, vervielfältigt oder verbreitet werden.
01-25_CRPA_n.indd 4 20.02.24 13:19
5
INHALT
Vorwort .......................................................................................................... 9
Isabel Rodà de Llanza
Keynote lecture: Roman sculpture as a multidisciplinary research focus ...................... 11
Methoden der Forschung
Erich Draganits – Beatrix Moshammer – Gabrielle Kremer – Andreas Rohatsch –
Michael Doneus
Die Steinbruchlandschaft von Bad Deutsch-Altenburg (Ostösterreich). Eine Rohstoff -
quelle seit der Römerzeit ................................................................................... 31
Roland Dreesen – Eric Goemaere – Gabrielle Kremer
Provenance analysis of the natural stones in funerary monuments from the western
part of the civitas Treverorum ............................................................................. 43
Anique Hamelink – Nicolas Delferrière – Ursula Rothe
Polychromie auf Porträtgrabsteinen: clavi auf gallischer Kleidung? ............................ 51
Sophie Insulander
Prokonnesian marble in the architecture of imperial Ephesos. Attempting an
archaeological evaluation ................................................................................... 63
Stephan Karl – Paul Bayer – Kerstin Bauer
The Roman stone monuments of Seggau Castle revisited. On the potential of the
spatial recording and analysis of ancient stone monuments ....................................... 71
Veselka Katsarova – Vasiliki Anevlavi – Sabine Ladstätter – Walter Prochaska
Roman sculptures from Kasnakovo, Bulgaria. Archaeological and archaeometric
investigations .................................................................................................. 89
Gabrielle Kremer – Robert Linke – Georg Plattner – Eduard Pollhammer –
Marina Brzakovic – Robert Krickl – Nirvana Silnović
Colours revealed: First results on a polychrome Mithras relief from Carnuntum ........... 101
Sébastien Laratte – Véronique Brunet-Gaston – Christophe Gaston – Régis Bontrond –
Céline Schneider – Gilles Fronteau – Patrick Huard
L’Arc de Mars à Reims: modèle 3D et SIG ............................................................ 111
Alexandra S. Rodler-Rørbo – Barbara Tober
Colourful walls of Noricum. Mineral pigment characterization for provenance
evaluation ....................................................................................................... 123
01-25_CRPA_n.indd 5 20.02.24 13:19
6Inhalt
Zeiten des Umbruchs: Denkmäler der fortgeschrittenen Kaiserzeit und Spätantike
Cristina-Georgeta Alexandrescu
Late Roman funerary monuments from Scythia ...................................................... 133
Stefan Ardeleanu
Ritualized funerary mensae of the western late antique world. Typology and use
spectrum between micro-regional and global trends ................................................ 145
Luca Bianchi
Un’ultima testimonianza di arte romana di stato a Mursa ......................................... 163
Lorenzo Cigaina
Zirkus und spectacula im spätantiken Aquileia. Stadtrömische Einüsse und
provinziale Rücküsse am Befund einiger Steindenkmäler ....................................... 175
Montserrat Claveria
The sarcophagus of Covarrubias (Burgos, Spain). Images of eternity between
paganism and Christianity .................................................................................. 189
Maria-Pia Darblade-Audoin
Les hermès de Welschbillig. Style et technique au IVe siècle dans le Nord des Gaules .... 199
Nicolas Delferrière – Anne-Laure Edme
Le sarcophage romain de Mantoche conservé au musée Baron Martin de Gray
(Haute-Saône, France). Un exemple atypique en Gaule du Centre-Est ........................ 211
Anne-Laure Edme
»Time of change«: Le traitement des monuments païens à l’heure de l’essor du
christianisme ................................................................................................... 221
Nadežda Gavrilović Vitas
Late antique mythological statuary in the Roman Central Balkans. Its function and
meaning ......................................................................................................... 235
Stylianos E. Katakis
The twilight of the Asklepios cult in Epidauros. The evidence of the building activity,
inscriptions, and sculptures ................................................................................ 249
Panagiotis Konstantinidis
Religious syncretism in late Roman Achaea. Reconsidering the identity of
»Isthmia IS 445« .............................................................................................. 263
Aleksandra Nikoloska
Statuary collections from the late antique residences in Stobi .................................... 277
Neue Funde und Forschungen
Jeanine Abdul Massih – Frédéric Alpi – Zeina Fani Alpi
Cyrrhus, place militaire de l’armée romaine en Syrie du Nord. Indices archéologiques,
épigraphiques et iconographiques ........................................................................ 291
01-25_CRPA_n.indd 6 20.02.24 13:19
7
Inhalt
Lucia Carmen Ardeț – Adrian Ardeț
Statue group of Liber Pater accompanied by Pan and panther from Roman Dacia ......... 301
Georgia Aristodemou
A Polyphemus group in Crete? An old nd reconsidered .......................................... 309
Fabian Auer
Ad limbum diducti – Entrückungsdarstellungen in der Sepulkralkunst der Donau-
provinzen. Überlegungen zu Bildchiffren und deren Ursprung .................................. 323
Domagoj Bužanić
Examples of Roman ornamental waterspouts from Croatia ....................................... 335
Fulvia Ciliberto – Paola Ventura
Nuove sculture funerarie da Aquileia .................................................................... 343
Chloé Damay
Preliminary research on output from one or more limestone sculpture workshops
in Thugga (Tunisia) .......................................................................................... 355
Zdravko Dimitrov
New stone monuments from Colonia Ulpia Traiana Ratiaria .................................... 363
Michael Eisenberg – Arleta Kowalewska
The Flowers Mausoleum at Hippos of the Decapolis. A rst glance into one of
the nest Roman provincial architectural decorations in basalt .................................. 371
Carlos Fabião – Trinidad Nogales – Nova Barrero – Amílcar Guerra – Joaquim
Carvalho – José María Murciano – Rafael Sabio – Catarina Viegas – Soa Borges –
Ricardo Laria Machado – Daniel Moreno – João Aires – Sandro Barradas
Anteatro de Ammaia (Lusitania). Nuevo ejemplo de modelo provincial .................... 387
Sabrina Geiermann – Hannelore Rose
Das Römergrab Weiden. Aspekte seiner Präsentation vom 19. bis in das 21. Jahrhundert ... 405
Emmanouela Gounari
Roman portraits from Philippi ............................................................................ 415
Jochen Griesbach
»Über Geld spricht man nicht!«? Unterschiede in der Zurschaustellung von Reich-
tum und Status in römischen Grabdenkmälern Italiens und der Nordwestprovinzen ....... 427
Tibor Grüll ‒ Nándor Agócs ‒ János Jusztinger ‒ Ernő Szabó
The iconographic motif of book-scrolls on funerary reliefs in Noricum ....................... 445
Craig A. Harvey
A marble statue fragment of Victoria/Nike from Humayma, Jordan (Nabataean
Hawara, Roman Hauarra) ................................................................................. 457
Melissa Kays
Monuments of Aurelia Paulina and her portrayal of social change in Roman
Asia Minor ...................................................................................................... 469
01-25_CRPA_n.indd 7 20.02.24 13:19
8Inhalt
Ute Kelp – Anja Klöckner
Das Große Weinschiff aus Neumagen. Neue Überlegungen zu Rekonstruktion
und Bedeutung ................................................................................................. 483
Martin Kemkes
Neue Statuenfragmente aus dem Westkastell von Öhringen am Obergermanischen
Limes ............................................................................................................. 499
Pierre-Antoine Lamy – Christine Louvion, avec la collaboration de Marie-Laure
Florent-Michel et Charlie Mairel
Sous l’œil de Junon. Nouveau regard sur le programme décoratif du second forum
de Bagacum (Bavay, Nord) ................................................................................ 517
Katja Lembke
Stone monuments of Roman Egypt as monuments of state ....................................... 535
Ana Zora Maspoli – Örni Akeret – Cornelia Alder – Debora Brunner –
Sabine Deschler-Erb – Claudia Gerling – Natalie Schmocker – Ulrich Stockinger
Hic sitae sunt. Interdisziplinäre Auswertung der frühkaiserzeitlichen Gräber der
Maxsimila Cassia und Heuprosinis im Gräberfeld Brugg/Remigersteig in Vindonissa .... 547
Sorin Nemeti
The Danubian Riders. Art, myth and ritual of a regional cult ..................................... 559
Christine Ruppert – Gabrielle Kremer – Andrea Binsfeld
Grabbauten des 1. Jahrhunderts in der westlichen civitas Treverorum ......................... 571
Mirjana Sanader
Eine Skulptur des Apollo Kitharodos aus Dalmatien ................................................ 581
Alfred Schäfer
Zwei Gebälkblöcke mit römischem Opferzug ........................................................ 589
Astrid Schmölzer
Goddesses of Germania inferior. Investigations into the iconography of the
Rhineland Matronae .......................................................................................... 601
Kathrin Schuchter
Die Enthauptung Medusas auf norischen und pannonischen Grabreliefs. Überlegungen
zu Musterbüchern, Werkstätten und Bildschemata .................................................. 613
Nedjma Serradj-Remili – Leila Benchernine
Stèles inédites de la Numidie et de la Maurétanie Césarienne et nouvelle lecture .......... 621
Nirvana Silnović
A new lion statuette from the Mithraeum in Jajce .................................................... 637
Katarina Šmid
The curious bust, found in the third Poetovian Mithraeum in Poetovio, Pannonia
superior .......................................................................................................... 647
Jakob Unterhinninghofen
Grabaltäre mit Meerwesendekor aus dem Treverergebiet. Untersuchungen zu
Chronologie, Typologie und Ikonograe ............................................................... 659
01-25_CRPA_n.indd 8 20.02.24 13:19
71
st e P h a n Ka r l – Paul b a y e r – Ke r s t i n ba u e r
THE ROMAN STONE MONUMENTS OF SEGGAU CASTLE
REVISITED
ON THE POTENTIAL OF THE SPATIAL RECORDING AND ANALYSIS OF
ANCIENT STONE MONUMENTS
Abstract
In 2019–2021, the so-called Römersteinwand in Seggau Castle in Styria/Austria was renovated. It was built as a lapidar-
ium in 1831 from the spoliae derived from a late antique tower and contains more than 100 relief and inscription stones,
mostly made of local marble. The removal of the plaster covering the edges allowed an extensive examination of the
stones. In addition to the conventional documentation, the stones were also recorded in 3D using Structure-from-Mo-
tion in order to obtain a uniform and comprehensive database for stylistic, technical and quantitative evaluations. The
intensive study of the stones, furthermore, led to new discoveries beyond the information gained by removing the plas-
ter, including new adaptations of fragments, revised readings of inscriptions and also an architectural reconstruction
attempt. The quantitative analysis of the stone dimensions reveals a strong tendency to standardisation of formats based
on the pes Romanus in integers.
INTRODUCTION
The south-eastern part of the Roman province Noricum is characterised by a rich legacy of stone
monuments, mostly from the sepulchral sphere, for which primarily a local white marble from the
eastern Alps was preferred1, but also less precious stones like limestone were used. This produc-
tion started already in the late 1st century BC, ourished in the 2nd century AD, boomed during the
Severan era and declined from the mid 3rd century onwards2. This artistic craftsmanship mirrors,
after all, periods of economic prosperity of the Roman settlements in south-eastern Noricum. Re-
search on these Roman stone monuments has long been a specic focus within provincial Roman
archaeology; they were evaluated intensively in epigraphical, iconographical, typological and
chronological terms.
The analysis of Roman stone monuments, especially by visual comparison, is mainly based
on personal physical observations and descriptive or photographic documentation, such as those
published by the »Corpus Signorum Imperii Romani« (CSIR) or in other printed media. Within
publication activities, the image database of ancient stone monuments »Ubi Erat Lupa« (<http://
lupa.at>) is a milestone towards digital Cultural Heritage and applications of new digital technol-
ogies. The open accessible database, on the one hand, improves the quality of data for scientic
research by continuously expanding and bringing it up to date. On the other hand, it shares this
data for educational purposes and with the wider public. Additionally, the web-based database is
not restricted to ancient or modern borders.
Advances in digital technologies have led to an increasing creation of 3D cultural heritage
models, e.g. of excavated remains of buildings, but also of single nds in the course of archaeo-
logical eldwork. However, computational analysis of this 3D data of archaeological nds is still
1 Djurić 2019.
2 See e.g. Piccottini 1994, 5–6; Kremer 2001, 17–18; Hudeczek 2008, 7; Pochmarski 2016. For a chronological
distribution of stones from Styria see Marko 2021, 100 g. 3.
01-25_CRPA_n.indd 71 20.02.24 13:20
72 Stephan Karl – Paul Bayer – Kerstin Bauer
in its infancy. Nevertheless, there are sporadic and occasional approaches using methods of ge-
ometric analysis from computer science for tackling specic problems in archaeological research.
With respect to ancient stone monuments this wide eld in archaeology can be roughly divided
into four groups as follows (giving some selected references):
1. Architecture: Sapirstein and Psota 2018 used 3D models for the analysis and estimation
of the original dimension of the not completely preserved column drums of the Hera Temple at
Olympia. For this, a modied ICP algorithm, which calculates an ›inside‹ parameter for tting
fragments with partly lost original surfaces compared to an idealised model, was developed. A
penetration free alignment of matching blocks was also addressed by Thuswaldner et al. 2009 for
the digital anastylosis of the Octagon in Ephesos. Additionally, the authors integrated constraints
into the matching process by extracting features like straight edges (e.g. of the cornice prole) or
clipping/clamping holes.
2. Sculpture: 3D shape comparison is able to quantify similarity and to ensure traceability.
Such an approach was proposed by Zhang et al. 2013 for classical sculptures and Roman copies,
by Lu et al. 2013 for portrait sculptures of Augustus or by Langner 2021 for emperors of the
Julio-Claudian dynasty. Furthermore, Hölscher et al. 2020 investigated the head’s asymmetries
using geodesic distances and Thiessen polygons, exemplied on an archaic Greek marble head of
a bearded man, the so-called Sabouroff head.
3. Inscriptions/bas-reliefs: Shape perception enhancement is an important factor in the eld
of epigraphy. Sapirstein 2019 presented a hybrid approach by combining advantages of 2D and
3D analytical techniques. Transformed and projected in 2.5D raster, the segmentation algorithm
restores the original (uninscribed) plane of the inscription and classies incisions and breaks. Fi-
nally, a displacement map was generated and visualised by different modes. Elevation raster maps
generated from the 3D geometry was also proposed by Monna et al. 2018 for the documentation
of carved stones in bas-relief, in this case of the deer stones from Mongolia. These raster maps
were treated by common algorithms developed for geomorphological studies like sky-view factor
or positive openness3.
4. Tool marks: Studying working traces (tool marks) on stone monuments can be strongly im-
proved by using 3D data and non-photorealistic visualisation techniques. Already early on, Levoy
et al. 2000 used accessibility shading to visualise chisel marks on sculptures by Michelangelo. 3D
models and subsequent transverse and longitudinal sections were used by Trefný et al. 2022 for
the classication of different working traces regarding the sculptor’s tools and applied working
techniques.
This paper presents and discusses results of a systematic application of 3D technologies on a
large and diverse complex of Roman stone monuments, with the nd context Seggau.
THE FIND CONTEXT SEGGAU
The Roman stone monuments from Seggau Castle near Leibnitz in Styria/Austria represent one
of the most important contexts of this nd category in the north-western provinces. In terms of
marble objects, it includes more than two thirds of the known total of stone monuments from the
nearby municipium of Flavia Solva and its surrounding area4. The stones preserved in Seggau or
removed from there in modern times can mainly be assigned to sepulchral architecture; only a
small proportion represent honorary monuments and votive altars.
The reason for this unusually high concentration of stone monuments from the Roman period
at Seggau Castle is related to the fact that in Late Antiquity a mighty tower with a side length of
3 Kokalj – Hesse 2017, 22–24.
4 Flavia Solva and its surrounding area include the cadastral communities of Seggauberg, Leibnitz, Altenmarkt,
Wagna, Kaindorf an der Sulm, Leitring and Grottenhofen. Restricted to marble, 217 monuments are recorded in
the database »Ubi Erat Lupa«, of which 158 originate from Seggau (148 were found in the Old Tower).
01-25_CRPA_n.indd 72 20.02.24 13:20
73
The Roman stone monuments of Seggau Castle revisited
15 m and a wall thickness of 3.8 m was built on this hilltop. For its construction, the stone build-
ings and monuments along the necropolis roads and in the public spaces of Flavia Solva, which
were apparently abandoned and no longer maintained, were extensively demolished5. As is usual-
ly the case in quarries, the blocks were cut into suitable ashlars or other formats on the spot. Ex-
tensive demolition layers along the Flavia Solva cemetery road (Spitalsgelände and Altenmarkt
cemeteries), which have been archaeologically excavated and can be dated to Late Antiquity, such
as layers of marble chippings and mortar, rubble layers with cuttings and the use of cut-off frame
and cornice parts in wells testify to this large-scale stone recycling6.
This tower, mentioned in a document from 1219 as turris antiqua, already existed when the
archbishops of Salzburg built Leibnitz Castle close to the south of the turris in the 1130s, thereby
occupying a deeper position on this hill. With the foundation of the diocese of Seckau, its own
castle, Seggau Castle, arose around the Old Tower. Whether the tower ever served as a residential
tower for the prime bishops is not clear, but already about 120 years later it was described as ruin-
ous (turris desolata). However, the written sources report for the year 1341 that the turris antiqua
was raised and extended to a fortied watch-tower. Overall, however, it was not fully integrated
into the castle. In the course of the combination of the Archiepiscopal castle Leibnitz and the
Episcopal castle Seggau, realised after 1594, an attempt was made to demolish the tower, but this
was not successful due to its massiveness. It was eventually used as a bell tower and more or less
left to its fate. Ultimately, it had to be demolished in two campaigns in 1815/1816 and between
1826 and 1831 due to its acute danger of collapse7.
In the masonry of the tower, numerous Roman stone monuments with their reliefs and inscrip-
tions came to light again. A representative and considered selection of a total of 108 stones8, most-
ly marble, was visibly built in in 1831 at the ground oor level of a new corridor in front of the
mediaeval wing of the building originally adjacent to the tower. This museum-like arrangement
built partly on the site of the Old Tower is the present so-called Römersteinwand (g. 1).
Since the construction of this corridor, the built in stone monuments as well as the walls
have been subject to various changes, which have affected not only the appearance but also the
substance of the stones. In 2016, the Episcopal Estate Administration of Seggau Castle and the
Austrian Federal Monuments Ofce decided to carry out a complete restoration of the »Römer-
steinwand«, which, in addition to conservation measures on the stones, included a complete re-
newal of the plastering according to monument preservation guidelines in a close approximation
to the aesthetic and design concept of the rst plastering in 1831. After extensive preliminary
work, this complete restoration was nally carried out in 2019 and, after a break in 2020 due to
the COVID-19 pandemic, resumed in 20219. After the removal of the plaster, the appearance of
the »Römersteinwand« largely resembled its condition as a shell construction in 1831. This res-
toration project offered the rare opportunity to closely examine and document the wall surfaces
including the foundation zone as well as the individual stones in an unplastered state.
5 For the Old Tower see Karl – Wrolli 2011; Karl 2013.
6 Karl 2013, 281–286; Lamm 2016, 187–197. A publication about a late antique well within the cemetery »Spitals-
gelände« consisting of some hundreds of spoliae, excavated in 1981/1982 but published only in short notes (Fuchs
1983), is currently in process by the rst two authors. For the phenomenon of large-scale recycling see Kremer –
Kitz 2018, 362–365.
7 Karl 2013, 173. 176–177. A small remnant of this late antique tower, namely its eastern corner including the foun-
dation zone, has survived in the mediaeval masonry of the so-called Seckauer Haus in the Upper Castle, which
was formerly attached to it; see Karl – Wrolli 2011, 65–73. 82–85; specically on the building technique, see Karl
2013, 210–227.
8 For the stone monuments of Seggau Castle from the Roman period, see Hainzmann – Pochmarski 1994. For the
number, which has changed up to the present day, see Karl 2018, 14 tab. 1.
9 Bayer et al. 2020.
01-25_CRPA_n.indd 73 20.02.24 13:20
74 Stephan Karl – Paul Bayer – Kerstin Bauer
METHODS AND WORKFLOW
In both restoration campaigns, only a short period of a few weeks was available to document the
Roman stone monuments incorporated in the wall, both those normally visible and those that
were fully plastered, mostly undecorated ashlars. The aim was to document both the walls and the
individual stones as well as possible, both photographically and three-dimensionally.
First of all, the stones were carefully cleaned from recent plaster, wall paint and dirt. The
cleaning of the gaps between the stones was of particular importance in order to document and
measure the full dimensions of the otherwise concealed edges and to obtain as much information
as possible from the sides turned into the wall.
Photographic documentation
The rst step in the documentation was a conventional photographic acquisition of the stones:
frontal photos were taken of all of them, while only angled photos were possible for the details
that appeared in the spandrels of the wall. The pictures were taken in the dark with articial light-
ing from an angle in order to reveal as much detail as possible. The stones on the outer façade
were photographed in 2019 by Ortolf and Friederike Harl, the stones in the corridor in 2021 by the
authors of this article. The photos are available in the database »Ubi Erat Lupa«.
3D documentation and data processing
In order to capture as much information as possible, all stones were documented in 3D with Struc-
ture-from-Motion (SfM) using Agisoft Metashape10. A series of typically 30–100 photos from dif-
10 For taking the photos, a Nikon D7000 was used in 2019 and 2020 and a Nikon Z 5 in 2021.
1 The so-called Römersteinwand of the Seggau Castle after the restorations 2019–2021 (© K. Bauer [Archaeogon,
Bayer & Karl GesbR])
01-25_CRPA_n.indd 7401-25_CRPA_n.indd 74 04.03.24 15:2504.03.24 15:25
75
The Roman stone monuments of Seggau Castle revisited
ferent angles was taken from each stone for the 3D reconstruction, the models were georeferenced
and thereby also scaled with tachymetric measurements of targets glued onto the wall next to the
stones. Inside the corridor fewer targets were used and larger portions of the wall were document-
ed at once due to the lack of suitable space for the measuring instrument; however, this should not
affect the quality of the result. Finally, this work resulted in 3D models of 156 stones – counting
only the built in stones (per fragment) within the »Römersteinwand« – created from 9477 photos.
The processing of the created 3D models involved several steps (g. 2). Firstly, the georef-
erenced and therefore already scaled models were clipped to an adequate area showing also the
surrounding masonry. This data can be used for conservation purposes. Secondly, the models
were virtually cut out of the surrounding wall. At this stage, the les were cleaned (outliers, er-
roneous vertices etc.) and manually orientated in GigaMesh, which was also used to measure the
dimensions of the stones and to create textureless grey orthographic renderings. In some cases,
adjoining fragments were assembled in CloudCompare. Thirdly, the stones were reconstructed
based on the 3D model and observations made on site. The reconstruction primarily serves to deal
with the stone object in more detail; it is limited to the essentials and is exclusively predictive
based on existing information on the stone itself. For some, the measurements taken by Kaspar
Harb before the construction of the wall and corridor could be taken into account11. Finally, traces
of reuse were mapped on the stones, such as the lifting holes and reworkings from Late Antiquity
associated with the construction of the Old Tower, but there are also later traces, particularly from
the construction of the »Römersteinwand« in 1831.
In addition to the stones, the entire exterior wall, the corridor behind it and the adjacent stair-
case to the south were recorded by a terrestrial laser scanner in 201912. The laser scan served to
measure the »Römersteinwand« affected by the restoration together with the adjacent building
parts as a whole in a larger frame in order to obtain a 3D documentation for historical building
research as well as a fusion with the SfM models (g. 3). However, the architectural history and
the current presentation of the stone monuments will be published elsewhere13.
Comparing the two applied acquisition techniques (Laser and SfM) requires some discussion
of the accuracy and precision achieved. As is known, the marble translucency and its heteroge-
neous granular structure produce a signicant bias and increased noise in the measurements by
optical scanners, such as laser or, to a lesser extent, structured light scanners. The light penetrates
and encounters a number of crystalline domains, resulting in a wide scattering but also in a bias
in the depth measurement. This effect is called subsurface scattering and can only be avoided if
the surface is covered with a matt coating. The effect of error was quantied and evaluated on
Carrara marbles with a grain size below 1 mm14, but to the best of our knowledge never on coarse-
grained marbles from the eastern Alps. For these we have to assume a much stronger bias. SfM
is therefore the better choice for marbles, as here the geometric data is calculated from the pixels
of the photos taken. SfM has undergone a development in recent years that now makes this doc-
umentation method equal in accuracy and precision to laser scanning techniques. The resolution
of the SfM models from Seggau is below 0.5 mm for all stones.
Potential of 3D data and geometric analysis
The basic requirement for working with 3D data of stone monuments is high data quality: the
entire process chain must be coordinated with the desired result. Both the data acquisition (in the
11 For the edition of Kaspar Harb’s handwritten manuscript of 1837, see Karl – Wrolli 2011, 155–234.
12 We would like to thank Boris Stummer and Josef Schauer from the Department of Hydrology and Geoinformation
at the Ofce of the Provincial Government of Lower Austrian for the laser scan, and Eduard Pollhammer (Depart-
ment of Art and Culture, Archaeological Park Carnuntum) for the mediation.
13 The construction of the »Römersteinwand« and its changes till today are the focus of a separate paper by the rst
two authors for the »Fundberichte aus Österreich« (FÖ).
14 Godin et al. 2001; Garcia-Fernandez 2016.
01-25_CRPA_n.indd 75 20.02.24 13:20
76 Stephan Karl – Paul Bayer – Kerstin Bauer
2 Process chain of the 3D models (all ortho views): a) view of the immured stones, b) cropped, aligned and, if necessary, assembled stones, c) reconstruction based on the visible features,
d) analysis of reuse (especially in course of the construction of the tower in Late Antiquity); exemplied on lupa 1269 (© B. Bayer [Archaeogon, Bayer & Karl GesbR])
3 Orthoview of the exterior wall of the »Römersteinwand«; SfM models from the wall and the foundation zone, fused partly with Laserscan data from adjacent buildings (© S. Karl
[Archaeogon, Bayer & Karl GesbR])
01-25_CRPA_n.indd 76 20.02.24 13:20
77
The Roman stone monuments of Seggau Castle revisited
case of SfM camera, lighting, scaling, photographic knowledge) and the data processing (soft-
ware, parameters, data handling) have to function perfectly from a technical point of view in order
to achieve good results. A certain technical know-how is therefore essential. The impact of these
new technologies can mainly be grouped into following categories:
uniform rePresentation
»Classical archaeology is at its core a visual discipline«, Martin Langner once stated15. As a
scientic discipline, archaeology needs appropriate documentation for the comprehensibility of
the results. Compared to photographs, 3D models of stone monuments have the great advantage
that the stone surfaces can be represented uniformly without texture and with consistent virtual
lighting. The true-to-scale models can be oriented very precisely, and the renderings of the 3D
data are orthographic. These uniform representations are a great help in recognising similarities,
nding matching fragments, positioning of interrelated components (cf. gs. 4. 5. 7) or discussing
questions concerning the workshops16.
enhancement of visibility
Different ltering techniques make it possible to highlight certain properties of the surface and
thus increase the visibility of certain traits. Some of these methods have been borrowed from
geography, others have been developed specically for archaeological questions17. Especially for
the re-reading of poorly preserved inscriptions, such visualisations are very helpful, ideally using
several methods in parallel to extract as much information as possible from the stones (cf. g. 8).
An autopsy of the inscriptions remains indispensable to verify the results. Methods used on the
Seggau stones include curvature visualisation, ambient occlusion, height maps and distance cal-
culations to a smoothed surface (an idealised uninscribed surface).
intersections and reassembly
Besides the complex ltering procedures, section lines can be extracted from 3D models at ar-
bitrary positions. Applications are, for example, the systematic investigation of the shapes of
proled frames or other architectural details, but also – especially in the context of Seggau – the
lifting holes and other modications from the reuse of the stones. Another signicant advantage
of 3D models over photos is the possibility of virtually assembling matching fragments and vis-
ualising them without distortion (cf. gs. 2. 5)18. Possible matches can already be recognised from
the renderings and preliminarily arranging, while the nal visualisations and measurements are
created from the merged 3D models. Particularly in the case of immured stones or stones stored at
different places, the matching of fragments can be excellently checked and represented virtually.
RESULTS AND DISCUSSION
The proposed workow turned out to be very efcient, especially in the process toward more
thoughtful interpretative tasks based on the acquired 3D data. 156 stones – counting only the
stones (per fragment) built into the »Römersteinwand« – were recorded in the main phase, re-
sulting after the matching process in 148 highly accurate 3D models. Due to the fast acquisition
15 Langner 2021, 379: »Die Klassische Archäologie ist im Kern eine visuelle Disziplin.«
16 Cf. e.g. Pochmarski 2007, 93 gs. 1 and 2 with the renderings of 3D models concerning the similarity in regard
to lupa 1238 and lupa 1317.
17 For geometric analysis in the eld of Cultural Heritage see the survey by Pintus et al. 2016.
18 With the exception of lupa 1269, all new matchings were achieved by working with the renderings or 3D models.
01-25_CRPA_n.indd 77 20.02.24 13:20
78 Stephan Karl – Paul Bayer – Kerstin Bauer
4 Typological grouping of all 115 stones from the nd context Seggau immured or stored at Seggau and already
integrated into the database »Ubi Erat Lupa«, mostly visualised by their front view; marble in grey, limestone in
sepia (© S. Karl [Archaeogon, Bayer & Karl GesbR])
01-25_CRPA_n.indd 78 20.02.24 13:20
79
The Roman stone monuments of Seggau Castle revisited
method the prime work on the »Römersteinwand« was extended to all Roman stone monuments
built into or stored at Seggau Castle. The database of 3D models comprises a total of 183 objects,
whereby fragments belonging together are counted as one piece. By far the largest part of these
other stones are directly related to the demolition material of the Old Tower, but there are also oth-
er objects such as two loricae of local limestone which were incorporated in a wall on the ground
oor of the so-called Vizedomhaus dated to the 13th or 14th century.
Several new observations could be made regarding these stones that have been known and
intensively researched since 1831, partly due to the removal of the covering plaster, partly due to
the uniform and well-comparable renderings of the 3D models and, above all, due to close obser-
vation. In the course of a successful cooperation with Ortolf and Friederike Harl, all new results
including the newly measured dimensions of the stones as well as photos and renderings can be
found in the database »Ubi Erat Lupa«, improving and enlarging the set of stone monuments at
Seggau to 115 objects19. Within this section, only the new ndings on four selected monuments
will be shown in detail. Additionally, the following list briey summarises the most signicant
new observations:
Tab. 1 The most signicant new discoveries and observations made during the restoration campaigns 2019 and 2021
lupa number Short description
lupa 1268 Two crouching lions could be documented on the pediment.
lupa 1319 A plant candelabra is on the left side of the portrait niche and identies it as part of a grave
stele.
lupa 1320 A librarius on the right side of the portrait niche identies it as part of a grave stele, cf.
lupa 1220.
lupa 1321 The presence of an inscription on the front side turned into the wall conrms that the stone
with an already known servant relief is a grave altar: ---] / [---]ns / / - / - / - [---]r / [--- .
lupa 1325 A framed eld could be observed on the left side.
lupa 1260, 1329 The medallions are formed by two cornucopiae, cf. lupa 1204.
lupa 1330 The left side has a framed eld, the right side is at for connecting to an adjacent stone.
lupa 1331 The framed inscription eld could be documented on the right side of the visible relief.
lupa 1285 The left side also bears a relief.
lupa 4840, 5107, 5147 These honorary monuments have framed elds on the sides.
lupa 1259 (+ 5751) Matching fragments of a pediment with Ganymede sitting on a rock, offering a cup to an
eagle; behind the rock is a dog on a leash and behind the eagle a bow (cf. lupa 13331).
lupa 33202 One of three fragments of uted pilasters built into the walls ts with the already known
fragment S-31 (Karl – Wrolli 2011, 73 no. S 31).
DIMENSIONS AND STANDARDISATION
Epigraphic sources, but also depictions on stone monuments (e.g. lupa 14590) prove that the
Roman foot (pes Romanus, p.R.) was considered as a unit of measurement within the imperial
marble trade20. A major advantage of 3D models is that accurate values for height, width and
depth can be extracted using the bounding box function. By converting all completely preserved
or securely reconstructed dimensions within the set of Seggau stones to the Roman measurement
system and evaluating their discrete distribution according to intervals of 1 unica (= 1/12 p.R. =
2.47 cm) peaks at positions 29.6 ± 1.85 cm (~ 1 p.R.), 59.2 ± 1.85 cm (~ 2 p.R.), 88.8 ± 1.85 cm
(~ 3 p.R.) and 118.4 ± 1.85 cm (~ 4 p.R.) are clearly standing out (diagram 1). Stones that cannot
19 lupa 33200–33211 were newly added to the database. All new results from the total of 183 stones affected by this
BDA project from 2019–2021 will be part of a detailed and more comprehensive publication by the three authors,
intended for the »Fundberichte aus Österreich« (FÖ).
20 Toma 2020, 35–37.
01-25_CRPA_n.indd 79 20.02.24 13:20
80 Stephan Karl – Paul Bayer – Kerstin Bauer
be oriented are excluded from this conversion. Within the 183 stones only two stones are larger
than 177.6 cm (~ 6 p.R.): lupa 1322 with a width of 195 cm and lupa 1264 with a width of 207.9
cm. Also in the Roman marble quarries in the eastern Alps the measurement of 1 p.R. and its mul-
tiple can most often be deduced, both on the quarry faces and the raw products21.
The typological assignment of the stones could be strongly improved by the scaled models and
the visual enhancement based on the uniform and normalised renderings (g. 4). Such classi-
cation reveals certain formats and implies a certain kind of standardisation within the production
process. Ortolf Harl has already pointed out the urge to standardise the measurements of Roman
burial structures22. Standardisation is particularly evident in the grave altars, where groups of
different sizes can be seen. The most extensive group, the so-called grave altars of the standard
format, has heights between 114 and 121 cm, widths between 85 and 93 cm and depths between
57 and 62 cm which correspond approximately to 4, 3 and 2 p.R. Groups of different size formats
can also be seen in the portrait medallions.
Grave inscription of Marcus Annius Iunianus and his family (lupa 1269)
The grave inscription of Marcus Annius Iunianus and his family (cf. g. 3) has recently been
published23. In addition to the already known three tting fragments, a fourth one could be iden-
tied, which completes the reading. The reconstruction of the text allows a good estimation of
the original dimensions of the stone, 155 × 91 cm. Another stone from Seggau (lupa 1285) has
almost the same dimensions. These two blocks have a proled frame on three sides, while the
lower frame is simple square-shaped. With this special frame design, but also in the dimensions,
21 Karl – Bayer 2021, 70. 99–102.
22 Harl 1997, 189 n. 13.
23 Karl – Bayer 2020.
0
2
4
6
8
10
12
14
16
18
20
0,0
2,5
4,9
7,4
9,9
12,3
14,8
17,3
19,7
22,2
24,7
27,1
29,6
32,1
34,5
37,0
39,5
41,9
44,4
46,9
49,3
51,8
54,3
56,7
59,2
61,7
64,1
66,6
69,1
71,5
74,0
76,5
78,9
81,4
83,9
86,3
88,8
91,3
93,7
96,2
98,7
101,1
103,6
106,1
108,5
111,0
113,5
115,9
118,4
120,9
123,3
125,8
128,3
130,7
133,2
135,7
138,1
140,6
143,1
145,5
148,0
150,5
152,9
155,4
157,9
160,3
162,8
165,3
167,7
170,2
172,7
175,1
177,6
Amount (n = 189)
Dimensions (in cm)
height
width
depth
diameter
Diagram 1 Sequence of completely preserved or securely reconstructed dimensions according to height, width,
depth and diameter (in case of portrait medaillons); the representation is limited to 177.6 cm (~ 6 p.R.)
01-25_CRPA_n.indd 80 20.02.24 13:20
81
The Roman stone monuments of Seggau Castle revisited
5 Portrait niches lupa 1336 (left) and lupa 1337 with the joining satyr relief lupa 1281 (right). Although very similar
in style and dimensions, the portrait niches do not belong together. Hardly visible is the carved-off satyr relief next
to the woman (© S. Karl [Archaeogon, Bayer & Karl GesbR])
the two Seggau blocks correspond to the inscription block in the base level of the Priscianus tomb
in Šempeter v Savinjski dolini (lupa 11198), so that it is likely that they once belonged to such
grave monument types.
Portrait niches (lupa 1336. 1337+1281)
Since Erna Diez’ guide to the stone monuments of Seggau Castle, it has been assumed that two
fragments of portrait niches, one with a woman with a melon hairstyle (lupa 1336), the other with
a bearded man (lupa 1337), could belong together (g. 5)24. The break in the middle, however,
does not allow any clear conclusions about whether the two belong together, even without the
covering by the plaster. Previous supporting arguments include the very similar heights of the
stones and the similarities of the portraits and garments.
In the carved-off relief to the left of the woman, the visualisations of the 3D data helped to
identify a gure rushing to the right in a framed eld, holding an object in the lowered right hand
and with the left hand raised above the head. It is obviously a satyr in a very similar posture to
those in the centre elds of the two three-gure reliefs from Hartberg (lupa 6070) and Bad Wal-
tersdorf (lupa 6069) or like the one on a relief also walled in Seggau (lupa 1281). For the second
portrait niche lupa 1337, the adjoining side eld on the right could be identied in the relief stone
lupa 1281, representing a satyr in a similar dancing pose as the one next to the woman. A drawing
by Kaspar Harb from 182825, when this stone had not yet been walled in, shows the course of the
fracture on the left, which corresponds to the fracture on the portrait niche. The matching is also
conrmed by the same bedding surface’s plane which slightly slopes backwards. In contrast to
the side reliefs, both portrait niches are not framed and open at the top. This and the raw, partly
still bossed joint surface on the side of the broken-off satyr relief speak for a use in the masonry
structure of grave buildings.
Despite all the similarities, the fact that the woman is looking down to the right and the man is
looking forward speaks against the two portraits belonging together. Such a difference cannot be
observed in any of the other grave portraits in Solva and is also not to be expected, since the mar-
24 Diez 1959, 58 nos. 115. 116; Hainzmann – Pochmarski 1994, 254–257 nos. 95. 96; Pochmarski 2011, 67–68 nos.
34. 35 pl. 20, 2–3.
25 Karl – Wrolli 2011, 167 no. 18 pl. 48.
01-25_CRPA_n.indd 81 20.02.24 13:20
82 Stephan Karl – Paul Bayer – Kerstin Bauer
ried couples immortalised in stone either look
forward in parallel or are turned slightly to-
wards each other so that their gazes cross, pos-
sibly at the viewpoint of a visitor to the grave.
Also, the height of the satyr on lupa 1281 is
higher than on lupa 1336. In all probability, we
are dealing with two very similar portrait nich-
es with adjoining satyr reliefs, likely resulting
in the existence of two almost identical tomb
buildings.
Heracles relief blocks (lupa 1322. 5913.
5914)
Among the few mythological reliefs in Seg-
gau, three blocks with scenes of the labours
of Heracles are particularly noteworthy. Ga-
brielle Kremer and Erwin Pochmarski have
recently dealt with their undisputed afliation
as parts of the socle storey of a funerary ae-
dicula26. New insights have now been gained
through the removal of the plaster.
Contrary to previous assumptions, it turned
out that the protruding eld with the taming
of Diomedes’ horses (lupa 1322) is not the left
side eld of a three-part block, but the central
eld. To the left of this central eld is the rem-
nant of a eld receding as on the right, with
the remains of a Noric-Pannonian volute orna-
ment. The block with the cleaning of the Au-
gean stables (lupa 5914) could now be clearly
identied as a corner block. In addition, anoth-
er labour of Heracles was revealed on the side
in the right-hand gap (g. 6). A gure is depict-
ed in a standing position to the right, holding
his right arm strongly bent and raised above his chest. A quiver with arrows can be seen above the
right shoulder; to the left of the body, below the elbow, remains of the lion’s skin hanging down
from the back are recognisable. This is clearly a depiction of the killing of the Stymphalian birds.
In 2003, during the creation of a new entrance in the room behind the corridor of the »Römer-
steinwand«27, the block with the theft of the apples of the Hesperides (lupa 5913) was removed
from the wall, revealing that it must be the left rear corner block of the socle and conrming that
the reliefs covered only three sides of the socle.
With this new data a reconstruction is proposed for the »Heracles aedicula«. Assuming sym-
metry within the three sides of the socle storey, the corner elds on the left and right side must
have the same width (62 cm) as the one with the Hesperidean episode (Labour XII) on the left rear
corner. The eld with the cleaning of the stables of Augeas (VI), however, is narrower (58 cm),
which is why this most likely belongs to the front side and forms the right corner eld with the
26 Kremer 2001, 214–215 no. 136 g. 127; Kremer 2014, 74 g. 30–32; Pochmarski 2021, 38–41 no. 28–30 pl.
20–22.
27 Karl – Wrolli 2011, 64–65. 111 no. Harb-Nr. 33 pl. 25.
6 The corner block (lupa 5914) with the cleaning of
the stables of Augias (Labour VI) on the front side
and the newly discovered killing of the Stymphalian
birds on the right side (Labour V) (© P. Bayer [Ar-
chaeogon, Bayer & Karl GesbR])
01-25_CRPA_n.indd 82 20.02.24 13:20
83
The Roman stone monuments of Seggau Castle revisited
eld depicting the killing of the Stymphalian birds (V) on the right side. The clockwise sequence
of the preserved labours of Heracles (VIII and VII) on the triptych lupa 1322 and its wider elds
(each 65 cm) suggest that lupa 1322 forms the centre of the front. This results in a reconstructed
width of the socle storey of 3.11 m, while the depth cannot be reconstructed with certainty due to
the missing middle blocks of the secondary sides. The height of the blocks and the closed front
(without the otherwise usual grave inscription) indicate a two-tiered socle, with the Labours of
Heracles on the lower tier.
If the secondary sides of the socle also had ve elds as the quinquepartite front side, the total
number of elds would be 15, in which not only a complete dodekathlos – in clockwise read-
ing – but also non-canonical labours28 or other scenes may have been depicted. The order of the
preserved scenes corresponds to that given by Diodorus and the »Tabula Albana«29. The canonical
order of Diodorus recurs only rarely, for example on Roman Asiatic columnar sarcophagi from
the second half of the 2nd century AD30. However, a composition with 15 elds requires a socle on
a square ground, which is quite unusual in comparison to other preserved grave aediculae in the
south-eastern part of Noricum31. Taking into account the aspect ratio of the aedicula socle of the
Spectatii tomb in Šempeter v Savinjski dolini (lupa 13256), a depth of the socle storey of the Her-
acles aedicula with a front width of 3.11 m would be 2.45 m. A similar result can be reached by
considering the width of a single eld which corresponds to approximately 2 Roman feet (p.R.).
By inserting middle blocks on the left and right side with a width of two elds (ca. 118 cm; 4 p.R.)
the socle depth reaches 2.43 m (g. 7). On these middle blocks further Heraclean labours could be
displayed – divided by a frame or not –, but it remains speculative which labours, in which order
and if they are replaced or supplemented with other scenes. Only the rst labour, the slaying of
the Nemean lion, can be assumed as displayed on the right rear corner block.
Votive altar for Jupiter Optimus Maximus (lupa 5148)
Two votive altars are built into the »Römersteinwand«, one for Mars (lupa 5108), the other for
Jupiter (lupa 5148). Both show typical traces of their reuse in an ashlar masonry structure32, in
our case the Old Tower. For this purpose, the cornices above the shaft with the inscription and
any protruding proled bases were carved off. In relation to the scope of our paper the one for
Jupiter deserves more attention. The altar is complete in width and height, only parts of the top are
broken off. The shaft slightly tapers towards the top. On the top above the carved-off cornice the
remnant of the right lateral acroteria is preserved. A lifting hole for a forceps (the opposite hole
has to be assumed on the not visible right side of the altar) was deepened into the inscription eld,
damaging the inscription in the middle left area.
The altar dedicated to Jupiter Optimus Maximus was erected and consecrated by two cult com-
missioners as well as by the civitas (of the Solvensians [?])33. Below this, two persons are listed
in each of eleven lines. The twelfth and last line was lled with only one name. The listed donors,
all of them men, have both Latin and foreign names.
On the basis of the 3D visualisations (g. 8) in combination with close observation, it was now
possible to correct letters of the previous readings, especially in the marginal areas of the stone
as well as in more abraded places within the inscription eld, and to recognise new ligatures that
change the reading of some names. The different way in which the names were shortened in the
28 Cf. the reliefs with Heracles and Alcestis in Piber (lupa 1393) and Ptuj (lupa 4216). The relief from Piber has an
estimated width of ca. 107 cm.
29 Diod. 4, 8–27; IG XIV 1293. For Heracles Dodekathlos see RE Suppl. III (1918) 1020–1028 s. v. Herakles (Otto
Gruppe).
30 LIMC V (1990) 5–16; 11 no. 1730–1731 s. v. Herakles (John Boardman).
31 Cf. Kremer 2001, 377–378.
32 Cf. e.g. the stone monuments from the late antique bridge head fortication at Bölcske: Beszédes et al. 2003.
33 For the most recent work see Hainzmann 2016.
01-25_CRPA_n.indd 83 20.02.24 13:20
84 Stephan Karl – Paul Bayer – Kerstin Bauer
8 Collegium inscription of the altar lupa 5148 in different visualisations: a) photo, b) ambient occlusion, c) height
map, d) depth map to an idealised uninscribed plane (© P. Bayer [Archaeogon, Bayer & Karl GesbR])
7 Reconstruction of the ground plan of the socle storey of the »Heracles aedicula« (© S. Karl [Archaeogon, Bayer
& Karl GesbR])
01-25_CRPA_n.indd 84 20.02.24 13:20
85
The Roman stone monuments of Seggau Castle revisited
two columns is notable. While the left column is shortened by means of ligatures, in the right
column the nal letters of the names are not written out.
The newly discovered numbering at the beginning of each of the twelve name lines from unus
to duodecim (tab. 2) is far not documented by any other example on Noric collegium inscriptions.
These were previously overlooked or read as part of the names, yet could now be recognised with
the help of the renderings from the 3D models. This discovery can now also explain the shift
of the dividing ET column from the central axis to the right; both name columns have the same
width.
The meaning of the numbering remains unknown at this point. However, it could be an indica-
tion of a pairwise distribution of men to tasks within this group. The division into twelve sections
could refer on the one hand to the number of months, and on the other hand to the mythological
sphere, for example the number of Olympian gods or the Heraclean labours34.
Tab. 2 The new reading of the collegium inscription (lupa 5148)
[I(ovi)] O(ptimo) M(aximo)
Attius Vital(is)
e•t Senius Iuvenis
[c]ura(m) agent(es) civitasq(ue)
I Iul(---) Cogitatus e•tCunaito Titia[n](---)
[II ---]inius Ursinus e•t Polius Ianuriu(s)
[III ---]estus e•tSurillio Appia[n](---)
[IIII ---]tinus e•t Priscianus Sab[-](---)
[V ---] Vibianus e•tUrsinus Vitalis
[VI ---] Aqilinus e•t Ceser(---) Iulianu[s]
VI[I ---]n Sammio e•t Ermadio
VIII V•alen•tinus Ia•nu•a•ri•s e•t Iul(ius) Primitiv(---)
VIIII U•rsinus Secund(---) e•tTriclo Veran(---)
X Uran•i(---) Dom•estic[---] e•t Aur(---) Martius
XI Appul(---) Ursus e•t Aur(---) Secun•din[---]
XII Malius Decian•us e•t(vacat)
CONCLUSION
Finally, we would like to summarise the added value of 3D documentation of Roman stone mon-
uments. Especially in the case of a large-scale and unwieldy group of materials such as building
stones, which usually have to be visited and documented on site, the advantages of uniformly
displayable 3D data become apparent, without, however, denying the relevance of conventional
photographic documentation.
Many ndings, such as the newly discovered tting fragments in four cases, are the result of an
interaction between a detailed autopsy and the 3D model, which changes and expands the previ-
ous visual perception. The scaled and uniformly displayable 3D models increase the comprehen-
sibility of the results and allow further investigations and visualisations that are not feasible using
photos or the originals. The models offer a broad eld for the development of computer-aided
methods for the analysis of geometric features such as letter shapes, frame proling, garment
folds, anatomic details and the like.
34 A publication of the new ndings concerning the inscriptions of the »Römersteinwand« is in planning together
with Manfred Hainzmann.
01-25_CRPA_n.indd 85 20.02.24 13:20
86 Stephan Karl – Paul Bayer – Kerstin Bauer
The benet of these methods is comparable to the revolution in landscape archaeology through
LiDAR data and its analytical methods, which enable a completely new way of seeing and dis-
covering.
BIBLIOGRAPHY
Bayer et al. 2020 P. Bayer – J. Fürnholzer – B. Hebert – S. Karl – J. Nimmrichter – M. Yasar, Die
Römersteinwand von Schloss Seggau, BDA wiederhergestellt 56 (Vienna 2020)
Beszédes et al. 2003 J. Beszédes – Zs. Mráv – E. Tóth, Die Steindenkmäler von Bölcske – Inschriften und
Skulpturen – Katalog, in: Á. Szabó – E. Tóth (eds.), Bölcske. Römische Inschriften und
Funde, Libelli Archaeologici Ser. Nov. 2 (Budapest 2003) 103–218
Diez 1959 E. Diez, Flavia Solva. Die römischen Steindenkmäler auf Schloß Seggau bei Leibnitz
²(Vienna 1959)
Djurić 2019 B. Djurić, The Logistics behind Ancient Art. The Case of Noricum and Pannoniae.
Festvortrag, in: B. Porod – P. Scherrer (eds.), Der Stifter und sein Monument.
Gesellschaft – Ikonographie – Chronologie. Akten des 15. Internationalen Kolloquiums
zum Provinzialrömischen Kunstschaffen Graz 14.–20.06.2017 (Graz 2019) 8–30
Fuchs 1983 G. Fuchs, Ein spätantiker Brunnen aus Flavia Solva, PAR 33, 1983, 7–9
Garcia-Fernandez 2016 J. Garcia-Fernandez, An Assessment of Errors and Their Reduction in Terrestrial Laser
Scanner Measurements in Marmorean Surfaces, 3D Research 7, 2016, Article 77, DOI:
10.1007/s13319-015-0077-0 (29.03.2023)
Godin et al. 2001 G. Godin – M. Rioux – J.-A. Beraldin – M. Levoy – L. Cournoyer – F. Blais, An
assessment of laser range measurement on marble surfaces, 5th Conference on Optical
3D Measurement Techniques Vienna 01.–04.10.2001 (Vienna 2001) 49–56, <https://
graphics.stanford.edu/papers/marble-assessment/> (29.03.2023)
Hainzmann 2016 M. Hainzmann, CVRA(M) • AGENT(ES) • CIVITASQ(VE) adnotationes ad CIL III,
5319 (Noricum, Flavia Solva), Antichità Altoadriatiche 85, 2017, 347–360
Hainzmann – M. Hainzmann – E. Pochmarski, Die römerzeitlichen Inschriften und Reliefs von
Pochmarski 1994 Schloß Seggau bei Leibnitz, Die römerzeitlichen Steindenkmäler der Steiermark 1
(Graz 1994)
Harl 1997 O. Harl, Norikum oder Oberpannonien? Die Grabädikula eines Kaiserpriesters aus
Bad Waltersdorf/Steiermark, in: G. Erath – M. Lehner – G. Schwarz (eds.), Komos.
Festschrift Thuri Lorenz (Vienna 1997) 185–202
Hölscher et al. 2020 T. Hölscher – S. Krömker – H. Mara, Der Kopf Sabouroff in Berlin: Zwischen
archäologischer Beobachtung und geometrischer Vermessung, in: A. Delivorrias –
E. Vikela – A. Zarkadas – N. Kaltsas – I. Trianti (eds.), ΣΠΟΝΔΗ. Αφιέρωμα στη
μνήμη του Γιώργου Δεσπίνη (Athens 2020) 125–138
Hudeczek 2008 E. Hudeczek, Die Rundskulpturen des Stadtgebietes von Flavia Solva, CSIR Österreich
IV 1 (Vienna 2008)
Karl 2013 S. Karl, Turris antiqua in castro Leybentz. Zur frühesten Baugeschichte der
Burgenanlage Leibnitz/Seggau im Kontext der spätantiken Ostanke der Provinz
Noricum mediterraneum (Diss. Karl-Franzens-University Graz 2013)
Karl 2018 S. Karl, Visualisierung historischer Erscheinungsbilder der Römersteinwand im
Schloss Seggau, Sprechende Steine 32, 2018, 7–15
Karl – Bayer 2020 S. Karl – P. Bayer, Die Annii. Eine epigraphische Familienzusammenführung an der
Seggauer Römersteinwand, in: E. Steigberger (ed.), Von den Alpen bis ans Meer.
Festschrift Bernhard Hebert, Forschungen zur geschichtlichen Landeskunde der
Steiermark 86 (Vienna 2020) 101–104
Karl – Bayer 2021 S. Karl – P. Bayer, Schrämspuren, in: S. Karl, Das römerzeitlich genutzte
Marmorsteinbruchrevier Spitzelofen/Kärnten – Montanarchäologische Forschungen.
Mit Beiträgen von P. Bayer, M. Grabner, M. Hainzmann, R. Haubner, A. G. Heiss,
K. Layr, D. Modl, W. Prochaska, S. Strobl, B. Toškan, E. Wächter, M. Weißl und
S. Wiesinger, FÖ Beiheft 1 (Vienna 2021) 67–75
Karl – Wrolli 2011 S. Karl – G. Wrolli, Der Alte Turm im Schloss Seggau zu Leibnitz. Historische
Untersuchungen zum ältesten Bauteil der Burgenanlage Leibnitz in der Steiermark,
Forschungen zur geschichtlichen Landeskunde der Steiermark 55 (Graz 2011)
01-25_CRPA_n.indd 86 20.02.24 13:20
87
The Roman stone monuments of Seggau Castle revisited
Kokalj – Hesse 2017 Ž. Kokalj – R. Hesse, Airborne laser scanning raster data visualization. A Guide to
Good Practice (Ljubljana 2017), DOI: 10.3986/9789612549848 (05.05.2023)
Kremer 2001 G. Kremer, Antike Grabbauten in Noricum. Katalog und Auswertung von Werkstücken
als Beitrag zur Rekonstruktion und Typologie, SoSchrÖAI 26 (Vienna 2001), <https://
e-book.fwf.ac.at/view/o:183> (29.03.2023)
Kremer 2014 G. Kremer, Wiederverwendete Teile von Grabanlagen aus Carnuntum. Zu ausgewählten
Neufunden aus dem Bereich südlich der Zivilstadt, CarnuntumJb 2014, 67–78
Kremer – Kitz 2018 G. Kremer – I. Kitz, Reuse of Roman ornamental stones in the hinterland of Carnuntum,
in: C. Coquelet – G. Creemers – R. Dreesen – É. Goemaere (eds.), Roman ornamental
stones in north-western Europe: natural resources, manufacturing, supply, life & after-
life, Études et Documents Archéologie 38 (Namur 2018) 361–372
Langner 2021 M. Langner, Das Gesicht hinter dem Marmor. Computergestützte Maßvergleiche an
römischen Kaiserporträts, in: J. Lang – C. Marcks-Jacobs (eds.), Arbeit am Bildnis.
Porträts als Zugang zu antiken Gesellschaften. Festschrift Dietrich Boschung
(Regensburg 2021) 377–384
Levoy et al. 2000 M. Levoy – K. Pulli – B. Curless – S. Rusinkiewicz – D. Koller – L. Pereira – M. Ginzton –
S. Anderson – J. Davis – J. Ginsberg – J. Shade – D. Fulk, The Digital Michelangelo
Project: 3D Scanning of Large Statues, in: J. R. Brown – K. Akeley (eds.), SIGGRAPH‚00.
Proceedings of the 27th annual conference on Computer graphics and interactive
techniques (New York 2000) 131–144, DOI: 10.1145/344779.344849 (29.03.2023)
Lu et al. 2013 M. Lu – Y. Zhang – B. Zheng – T. Masuda – S. Ono – T. Oishi – K. Sengoku-Haga –
K. Ikeuchi, Portrait Sculptures of Augustus: Categorization via Local Shape Comparison,
in: Proceedings of the 2013 Digital Heritage International Congress (DigitalHeritage)
Marseille 28.10.–01.11.2013, Vol. 1 (IEEE Xplore) 661–664, <https://ieeexplore.ieee.
org/document/6743812> (29.03.2023)
Marko 2021 P. Marko, Good Times, Bad Times? An Overview of Findings on the 3rd Century in the
Territory of Flavia Solva/Wagna, in: M. Auer – C. Hinker (eds.), Roman Settlements
and the ›Crisis‹ of the 3rd Century AD, Ager Aguntinus 4 (Wiesbaden 2021) 93–105
Monna et al. 2018 F. Monna – Y. Esin – J. Magail – L. Granjon – N. Navarro – J. Wilczek – L. Saligny –
S. Couettee – A. Dumontet – C. Chateau, Documenting carved stones by 3D modelling –
Example of Mongolian deer stones, Journal of Cultural Heritage 34, 2018, 116–128,
DOI: 10.1016/j.culher.2018.04.021 (29.03.2023)
Piccottini 1994 G. Piccottini, Grabstelen, Reiter- und Soldatendarstellungen sowie dekorative Reliefs
des Stadtgebietes von Virunum und Nachträge zu CSIR Österreich II 1–4, CSIR
Österreich II 5 (Vienna 1994)
Pintus et al. 2016 R. Pintus – K. Pal – Y. Yang – T. Weyrich – E. Gobbetti – H. Rushmeier, A Survey of
Geometric Analysis in Cultural Heritage, Computer Graphics Forum 35, 2016, 4–31,
DOI: 10.1111/cgf.12668 (29.03.2023)
Pochmarski 2007 E. Pochmarski, Werkstätten von Porträtreliefs aus Flavia Solva, RÖ 30, 2007, 91–105
Pochmarski 2011 E. Pochmarski, Die Porträtmedaillons und Porträtnischen des Stadtgebietes von Flavia
Solva, CSIR Österreich IV 2 (Vienna 2011)
Pochmarski 2016 E. Pochmarski, Die typologischen, ikonographischen und chronologischen Probleme der
Grabstelen und Grabaltäre im Territorium von Flavia Solva, in: F. Humer – G. Kremer –
E. Pollhammer – A. Pülz (eds.), Akten der 3. Österreichischen Römersteintagung in
Carnuntum 02.−03.10.2014 (Vienna 2016) 143–161
Pochmarski 2021 E. Pochmarski, Die Grabbaureliefs (erster Teil) des Stadtgebietes von Flavia Solva,
CSIR Österreich IV 4 (Vienna 2021)
Sapirstein 2019 P. Sapirstein, Segmentation, Reconstruction, and Visualization of Ancient Inscriptions
in 2.5D, Journal on Computing and Cultural Heritage 12/2, 2019, Article 15, <https://
dl.acm.org/doi/10.1145/3286977> (29.03.2023)
Sapirstein – Psota 2018 P. Sapirstein – E. Psota, Pattern Matching and the Analysis of Damaged Ancient
Objects: The Case of the Column Drum, Journal on Computing and Cultural Heritage
9/3, 2018, Article 13, <https://dl.acm.org/doi/10.1145/2901297> (29.03.2023)
Thuswaldner et al. 2009 B. Thuswaldner – S. Flöry – R. Kalasek – M. Hofer, Q. Huang – H. Thür, Digital
Anastylosis of the Octagon in Ephesos, Journal on Computing and Cultural Heritage
2/1, 2009, Article 1, <https://dl.acm.org/doi/10.1145/1551676.1551677> (29.03.2023)
01-25_CRPA_n.indd 87 20.02.24 13:20
88 Stephan Karl – Paul Bayer – Kerstin Bauer
Toma 2020 N. Toma, Marmor – Maße – Monumente: Vorfertigung, Standardisierung und
Massenproduktion marmorner Bauteile in der römischen Kaiserzeit, Philippika 121
(Wiesbaden 2020)
Trefný et al. 2022 M. Trefný – D. Mischka – M. Cihla – A. G. Posluschny – F. R. Václavík – W. Ney –
C. Mischka, Sculpting the Glauberg ›prince‹. A traceological research of the Celtic
sculpture and related fragments from the Glauberg (Hesse, Germany), PLoS One 17,
2022, e0271353, DOI: 10.1371/journal.pone.0271353 (29.03.2023)
Stephan Karl, Dr.-Emperger-Weg 14, 8052 Graz, Austria.
[e] stephan.karl@chello.at
Paul Bayer, Rosenhaingasse 5, 8010 Graz, Austria.
[e] paulbayer@gmx.net
Kerstin Bauer, Rosenhaingasse 5, 8010 Graz, Austria.
[e] ke.bauer@edu.uni-graz.at
01-25_CRPA_n.indd 88 20.02.24 13:20
