The Inscription between text and object: The deconstruction of ...

HAL Id halshs-02457550httpshalshsarchives-ouvertesfrhalshs-02457550

Submitted on 28 Jan 2020

HAL is a multi-disciplinary open accessarchive for the deposit and dissemination of sci-entific research documents whether they are pub-lished or not The documents may come fromteaching and research institutions in France orabroad or from public or private research centers

Lrsquoarchive ouverte pluridisciplinaire HAL estdestineacutee au deacutepocirct et agrave la diffusion de documentsscientifiques de niveau recherche publieacutes ou noneacutemanant des eacutetablissements drsquoenseignement et derecherche franccedilais ou eacutetrangers des laboratoirespublics ou priveacutes

The Inscription between text and object Thedeconstruction of a multifaceted notion with a view of a

flexible digital representationEmmanuelle Morlock Eleonora Santin

To cite this versionEmmanuelle Morlock Eleonora Santin The Inscription between text and object The deconstructionof a multifaceted notion with a view of a flexible digital representation First EAGLE InternationalConference on Information Technologies for Epigraphy and Cultural Heritage The Europeana networkof Ancient Greek and Latin Epigraphy Sep 2014 Paris France halshs-02457550


Submitted on 14 Apr 2015






22 The Inscription between text and objectThe deconstruction of a multifaceted notion with aview of a flexible digital representation

Emmanuelle Morlock Eleonora Santin

AbstractIn scholarly use the term lsquoinscriptionrsquo is not always unambiguous The sameconcept can designate either the signifiers on a support regardless of theirmeaning and textual function or can be used to distinguish different texts In adigital representation a distinct markup is utilised to encode the material andtextual dimensions In order to combine them in an adequate representationwe submit a definition of some epigraphic notions which supports the theo-retical model of an encoding schema compliant with the EpiDoc guidelinesdesigned as a part of the IGLouvre project

Keywords

Inscription (notion of) archaeological dimension of text-bearing objects epi-graphic edition text representation TEI EpiDoc digital edition

221 Introduction and purposesFor a long time epigraphic editions have approached inscriptionsmostlyas texts almost ignoring their physical nature For example refer-ence corpora like the Inscriptiones Graecae were not illustrated withphotographs This period is fortunately over although it left someconsequences in editorial practices

The 14th international symposium of Greek and Latin epigraphywhose main theme was Publicum Monumentum Textus has provedonce again that any modern survey must regard an inscription asexposed writing inseparable from its physical support (monumentobject vase mosaic) and its context whether certain or hypotheticalIncorporated into its support the inscription regains its primary valueas a semantic system to describe read and interpret by incorporating at

326 Information Technologies for Epigraphy and Cultural Heritage

least a threefold approach archaeological textual and historical1 Anedition which strives for completeness must take all these aspects intoaccount

The following questions lie at the core of this approach to epigraphicobjects and define a series of challenges in the editing of inscriptionsand their contexts How can earlier editorial practices be taken furtherin order to reduce the misinterpretations that arose in past and mightarise in the future from a fragmented presentation or a partial analysisof a text-bearing artifact How can epigraphic edition be properly re-structured in order to show a three-dimensional object which requiresa multidisciplinary investigation Can digital representation digitalencoding and digital edition help achieve such a difficult endeavor

In the last decade the digital edition of Greek and Latin inscriptionsmarked-up using the EpiDoc schema has gone through at least threeimportant changes and gave rise to three types of publications

1 Electronic republications enhanced and expanded versions ofprinted books with a new presentation improved particularlyfrom the point of view of data availability and data query quan-tity and quality of illustrations (eg Vindolanda Tablets on line2

[Terras 2006] and the addition Vindolanda tablets online 23 - Aphro-disias in Late Antiquity 20044 expanded version of the 1989 printedbook by Charlotte Rouecheacute)

2 New editions of corpora (eg Inscriptions of Aphrodisias 2007)5 thattook advantage of the digital environment but are still close tothe paper editions model [Bodard 2008] In these first essays theapparatus criticus and textual commentary have been reduced orsometimes omitted in order to mind the encoding aspects

3 Critical editions of new epigraphic corpora whose editors wereable to give a more extensive and accurate representation of the

1 For a similar approach see M Lameacute and P Kossmann From paper browser to digitaledition of inscriptions a new conceptual model for a global historical approachposter presented at the TEI Conference (Rome October 2014)httpeerhypothesesorgposters

2 httpvindolandacsadoxacuk3 httpvto2classicsoxacuk4 httpinsaphkclacukala20045 httpinsaphkclacukiaph2007

22 The Inscription between text and object 327

text its restitution and commentary taking advantage of previ-ous experiences (eg Monumenta Asiae Minoris Antiquae XI6)

Currently in enhancing the archeological dimension of the inscriptionsgreat results can be observed in some ongoing and quite advancedepigraphic projects in SanskritCam-language The Corpus of the Inscrip-tions of Campā) and in Celtic language Ogham in 3D7 The latter hasrevealed the great potential of the TEI-XML encoding associated withthe 3D scanning process [Devlin et al 2014a Devlin et al 2014b] Nowprogress remains to be made in order to create an encoding modelthat could combine the textual as well as the material dimension of anarcheological object bearing text and help us to determine

1 The arrangement of an inscription on the support

2 The textual cuts made by epigraphers on the base of differentcriteria

In this endeavor we have to bear in mind three basic values structuralearness flexibility and reversibility

The diplomatic transcription of an inscription is the result of an actof interpretation even if it is to some extent meant to be a neutral actAnd reading and recognizing different texts and subtexts is a fortiori aninterpretative process Their order and their presentation in a printedor digital edition is an editorial choice depending on the aim of thepaper as well as from the scholarly habits of its author Hence one canunderstand the importance of creating a model that provides a clearlsquomaprsquo of all the texts (coeval or not) readable on an object and the benefitof linking them to one or several high quality images At the sametime such amodel should be able to represent and graphically displaythe editorrsquos choices This would allow readers to follow the editorrsquosinterpretative path backwards and allow for the easy introduction ofmodifications if they want to reuse the file

The major challenge is finding an encoding structure that takesinto account not just one but several common epigraphic scenariosa composite text on a single support a simple or composite text on acomposite support and the rather common case of the supportrsquos re-use

6 httpmamacsadoxacukindexhtml7 httpoghamceltdiasiemenuphplang=en


The solution found within the project IGLouvre8 should be consideredas a suggestion and a starting point for a wider discussion

Starting from four possible configurations of interaction betweentext and object we first suggest a clarification of the notions embeddedin our model and then we propose a method to encode and thusbetter represent the main relationships between the inscription in itsmaterial dimension and the text

222 Interaction between text and object four possibleconfigurations

2221 One simple text written on a single object

Let us start from the one-to-one relationship the most linear and for-tunately the most common The prevalence of this configuration isperhaps the reason why some epigraphic projects do not need a wayto encode more complex configuration

A round funerary altar bearing epitaphs for three members of thesame family is a good example to start with for two kinds of reasonsthe arrangement of the writing on the round surface and the internalchronology of the inscription

22211 Epitaph of Damophon Epaphroditos and Theudoris

avoid indent appearence without adding a tableMonument description Funerary altar decorated with bucraniaPresent location Paris Louvre Museum (MA 2327)Original location KosLast recorded locations Athens then Toulon arsenalDate 2nd half of the second century BCBibliographyMonument Hamiaux et al 1998 205 n 221 Berges 1996 115-116 n26 (the text of the inscription is not the right one) pl 12 3Editions Dain 1933 17-18 n 9 (reviewed by L Robert Revue Archeacuteo-logique 2 1933 123 n 9)

8 French project lead by Michegravele Brunet Professor of Greek Epigraphy UniversityLumiegravere-Lyon 2 selected for funding by the ANR (French National Research Agencyref number ANR-12-BSH3-0012) It aims to publish a digital edition of the Louvrecollection of Greek Inscriptions


avoidΔαmicroοφῶντοςτοῦἘπαφροδίτουavoidἘπαφροδίτουτοῦἘπαφροδίτουπρεσβυτέρουΘευδωρίδος ltτᾶςgtτᾶς[Ἐπαφ]ροδίτου ζών[των]

l 10 Dain [Εὐmicro]όλπου ζῶν[τος]

Firstly in order to give a precise idea of the text layout and letteringa 3D image would be far more effective than a two-dimensional photo9

Secondly the decision to present the inscription as just one text withoutany further divisions or alternately as one text divided in three textualcomponents is a scientific statement coming from the assumption thatthe three names have either been carved at the same time or not ADain assumes that there are three inscriptions carved in three differentstages10 Revising the stone and the context of its fabrication the mod-ern editors will be able to confirm Dainrsquos opinion and in this case theymightwant to divide the text in three sections (for this configuration seethe next paragraph) Instead they might assume that the monumentwas commissioned by all the people mentioned on the stone duringtheir lifetime and that the inscriptions have been carved all at the sametime This last hypothesis seems to be supported by some epigraphic

9 See the photo in Hamiaux et al 1998 205 n 221 A program of 3D scanning andimaging of all these kind of monuments (altars with bucrania) is underway withinthe IGLouvre pro-ject

10 In the description of the stone Dain writes ldquoau-dessous de la guirlande reliantdeux tecirctes de beacuteliers premiegravere inscription de trois lignes au-dessous de la mecircmeguirlande deux autres inscriptionsrdquo In the critical notes he adds ldquoLrsquoinscription a eacuteteacutegraveacutee agrave trois reprises diffeacuterentesrdquo Maybe he hesitates between singular and pluralinscriptionsinscription because the singular represents the neutral point of view ofa contemporary reader (what we can see today avoiding any assumption) and theplural his interpretation


parallels found in the round altars from Kos11

Fig 221 Dain Inscriptions grecques du Museacutee du Louvre n 9

11 See Berges 1996 Katalog ns 1-111 (Rundaltaumlre aus Kos) in particular the monumentn 32 where the word ζωντων after two personal names at the genitive case is welllegible


2222 One structured text consisting of multiple textual com-ponents written on a single object or on one object-part

The textual components can be homogeneous with respect to their texttype or function (several epitaphs see 222) or heterogeneous (e g adedication and a signature an epitaph and a defixio a dedication anda decree) An ancient reader or a modern observer could see them asparts of a composite text

Within thematically classified epigraphic editions the observanceof strict classification rules leads editors to sometimes split into twodifferent entries what has been conceived and realised as a cohesiveensemble Such a practice could result in misinterpretations espe-cially when the necessary cross-references are omitted An editorialpresentation that compromises the overall view of an inscription evenin a thematic corpus is fortunately less and less common But oneof the most valuable advantages of a digital edition is the possibilityto markup different text forms (ie different taxonomies) withoutcompromising the overall view Giving that fact it would be betterto publish these composite texts as a whole while at the same timeshowing that they consist of heterogeneous components In that way itwould be possible to link every component to the previous epigraphiceditions in which it has been treated as an independent text includedinto different thematic groups (eg dedications vs decrees)

This case is exemplified by the editorial history of a marble slabfrom Delos (after 166 BC) bearing a dedication and a decree of thedionysiac artists honouring the aulos-player Craton son of Zotichosfrom Calcedonia now in the collection of the Louvre Museum12 Theinscription published by W Froehner [1865 n 67 Duumlrrbach 1921 n75] starting from the stonersquos autopsy was then edited as an unitarytext by all the principal editors except P Roussel13 who followingthematic criteria splits it into two different texts and puts them intodistinct sections of the IG volume (decreta collegiorum IG XI 4 1061 anddedicationes artificum dionysiacum IG XI 4 1136)

12 See the full bibliography in Le Guen 2001 231-239 n 45 and Aneziri 2003 D1013 Like G Daux reminds in his edition of 1935 ldquodans les IG la deacutedicace et le deacutecret

proprement dit sont placeacutes dans deux sections diffeacuterentes (nos 1136 et 1061) et queleurs lignes ont reccedilu une numeacuterotation indeacutependanterdquo


Fig 222 Louvre Museum (MA 841) - Dedication and decree in honor of Kraton son ofZotichos from Delos after 166 BC (IG XI 4 1061 + 1136) From a photograph providedby the Louvre Museum clicheacute Lebeacutee-Deacuteambrosis

2223 One structured text consisting of multiple textual com-ponents written on multiple objects that are themselvesparts of a composite object

Every object is a complete part or a broken part of a composite objectassembled or disassembled and scattered in different or in the samerepositories and archaeological sites This is the situation that epigra-phers have to describe every time they publish an inscription written as an example on different parts of a composite funerary monument(eg a sarcophagus) or on different blocks of a wall The textualcomponents can be homogeneous or heterogeneouswith respect to texttype or function


Once again the collection of the Louvre Museum provides us withan interesting example three funerary epigrams written on two slabsthat were parts of the same funerary monument perhaps a sarcopha-gus

22231 Funerary epigrams for Antiphon and Eurymenides sons ofSophocles

A Monument description A rectangular white marble slab cutagain in the modern age at the top there is a plate frame slightlyprominentDimensions H 55 x W 100 x D 11 cmText layout 8 lines one l per verse flush left second line indentedPresent location Paris Louvre Museum (MA 905-1)Findspot Thasos loc Μούργινα

B Monument description A rectangular white marble slab cutagain in the modern age largely damaged at the upper left corner andbroken into two parts stuck back togetherDimensions H 61 x W 94 x D 8 cmText layout 18 lines one l per verse Two lines groups the firstconsisting of 8 lines (flush left second line indented) and the secondconsisting of 10 lines (flush left no indentation)Present location Paris Louvre Museum (MA 905-2)Findspot Thasos loc ΜούργιναOriginal location ThasosDate about 100 BCBibliographyEditions Conze 1860 pp 18-21 [textual order a c b] (Kaibel EpigrGr 208 add p 519 Demitsas n 1161-1162) IG XII 8 441 [textualorder a c b] from a squeeze Peek 1955 GV 2038 [textual order a c b](Peek Griechische Grabgedichte 1960 n 47) Dunant et al 1958 160pl 40Studies Mendel 1900 p 281 Lane 1988


A Marble slab (MA 905-1)a Epigram for Antiphon

Meter elegiac coupletNarrative form 1st person

avoid indentἄρτι microε νυmicroφιδίων ἀπὸ δύσmicroορον ἅρπασε παστῶνspδαίmicroων ἐς τριτάταν νισόmicroενον δεκάδαἄρτι βίου περόωντα κατrsquo εὐκλέα θέσmicroια δόξαςspστυγνὸς ἄπαιδα δόmicroοις ἀmicroφεκάλυψrsquo ἈίδαςἈντιφόωντα γοναῖσι Σοφοκλέος ὃν τέκε microάτηρspἩρώ τᾶι λιπόmicroαν οὐ τέκος ἀλλὰ τάφοναἰαῖ τίπτε Τύχα microε τὸν εὐκλέα πατρίδι κόσmicroονspτλάmicroονα δυσπενθής ὠρφάνισας βιότου

B Marble slab (MA 905-2)b Epigram for Eurymenides


avoid indentοὐ γάmicroον οὐχ ὑmicroέναιον ἐmicroοὶ [c 6 - 7]spἩρώ ἀποφθίmicroενον δrsquo ἐστενάχησε γό[οις]εἰκοστὸν τανύσανθrsquo ἐτέων δρόmicroον ἄ micromicroε δrsquo ὁmicro[αίmicroους]spτλάmicroονας ἐν δισσοῖς microησὶν ὅδrsquo ἔσχε τάφοςπατρὸς δrsquo εὐόλβοιο Σοφοκλέος ἄρσενα γέν[ν]ανspὠκύmicroορον φθιmicroέναν ἐστενάχησε Θάσοςmicroάτηρ δrsquo ἁ microεγάλαυχος ἐφrsquo υἱάσιν ἁ πάρ[ο]ς εὔπαιςspοὐχὶ τέκη κω[φ]ọ[ὺς δrsquo]ἀντὶ δέδορκε τάφους(vac 2 lines)

c Epigram for EurymenidesMeter iambic trimeterNarrative form 3rd person

avoid indentὁ τύmicroβος ἐσθλὸν υἷα τὸν ΣοφοκλέοςΕὐρυmicroενίδην κέκευθεν ὧι βίου microόνα


ἐτῶν διεξάmicroειπτο διπλόα δεκάςκατεῖδε δrsquo οὔτι νυmicroφικῶν ἐφίmicroερονπαστὸν γάmicroων πάρεδρον ἀλλrsquo ἀπrsquo ὀλβίωνσφαλεὶς microελάθρων στυγνὸν ἦλθrsquo ὑπrsquo Ἀίδανδυσπενθὲς Ἡροῖ microατρὶ καὶ συναίmicroοσιλιπὼν φίλαισιν ἄλγος ἁ δrsquo ἁλιστεφὴςδόξας ἕκατι τῶιδε πατρία Θάσοςτὰ σεmicroνὰ τιmicroᾶς δῶρrsquo ἔνειmicroεν εὐκλέος

How many text structures how many sequences are acceptable fora composite text like this As many as the perspectives which an editormight hold as possible and worthy of notice

bull The chronological sequence of recorded events (is the text chrono-logically structured)

bull The poetrsquos perspective (is the text based on a poetic project and aconsequent poetic arrangement)

bull The lsquoengraving perspectiversquo (what was the order of engraving Isthere a connection between this order and the inner chronology)

If the display context and the mutual position of the marble slabs canbe reconstructed one might also add the ancient readerrsquos point of view

2224 Multiple distinct texts consisting of one or several tex-tual components written on a single object (no linkswith one another apart from the support)

It is the case of the supportrsquos re-use In order to show various scholarlyapproaches in publishing this particular occurrence we will comparetwo editions in which editors decided to present the inscription fromtwo different perspectives In IG IX 2 1040 a-d (Fig 223) O Kernhad an object-perspective since he published under the same text-entryall that is readable on the stone and performed text divisions both inthe diplomatic and in the critical transcription by means of a sequenceof lower case letters (elsewhere in the same volume he used romannumbers) In the inscriptions of Gonnoi (Gonnoi nos 114 115 122 123127 19814 see Fig 224 225 and 226) B Helly adopted thematic and

14 In the epigraphic archive of HiSoMA at Lyon the number of the object is GHW 4348see photos the inventory number in Larissa Museum is 318


IG IX 2 n Gonnoi n Type of text Date Position1040 a 198 votive

dedication2nd c BC front face

1040b ll 1-5 114 manumission end 1st c BC front face1040b ll6-15 115 manumission about 25 BC front face

1040b ll16-19 122 manumission 1st half of the

first c ADfront face

1040c 123 manumission Tiberiusreign

left side

1040d 127 manumission 45-46 AD right side

Tab 221 Bibliographical concordance

chronologic collecting criteria and so decided to split the lsquoinscriptionrsquointo six different text-entries

On the one hand it is reasonable to separate texts that have norelation with one another on the other hand it would be important toshow the history of the different uses and reuses of an object andmakereaders able to verify the fact that there are really no links betweenthe texts apart from the fact that they are on the same support all thematerial aspects of the writing changes of hands andwriting style textlayout etc

223 Defining concepts key entities for the material andtextual dimensions

Our first attempts to represent these configurations involving a one-to-many textobject relationship by means of an EpiDoc markup stum-bled upon the ambiguity of the notion of lsquoinscriptionrsquo If the notionmainly describes a lsquotextrsquo is it correct practice to use the EpiDoc lsquotextpartrsquosubdivision of the lsquotextrsquo element to encode material parts of an objectSince the term is often used as a substitute for a unique lsquoobjectrsquo orlsquodocumentrsquo15 bearing a unique text what should we do with texts that

15 See Cayless et al 2009 The authors restate the historical and theoretical backgroundof the creation of EpiDoc The dual use of the term lsquoinscriptionrsquo throughout the articleto designate the source alternatively as an object and as a text must be related to thefact that ldquothe collaborators were seeking a digital encoding method that preservedthe time-tested combination of flexibility and rigor in editorial expression to whichclassical epigraphers were accustomed in print while bringing to both the creatorand the reader of epigraphic editions the power and reusability of XMLrdquo


Fig 223 Kern IG IX 2 1040 a-d


Fig 224 Larissa Museum inv n 318 front face - IG IX 2 1040 a-b


Fig 225 Larissa Museum inv n 318 - IGIX 2 1040 c

Fig 226 Larissa Museum inv n 318 - IGIX 2 1040 d

run across several objects or fragments The recommended practicetaught in the EpiDoc training sessions16 is very flexible permitting theuse of the textpart subdivision both for purely textual units or text areas

16 See Bodardrsquos slides Structure of the Epigraphic Text from the Digital Classicist wikipage httpwikidigitalclassicistorgEpiDoc_Summer_School


visible on specific parts of the object17 But as D Buzzetti demonstratesit the process of text encoding in a scholarly context is at the same timethe building of a representation and of the representation of a representation[Buzzetti 2002] It requires the clarification of the underlying textmodel necessarily used (knowingly or not)

In order to properly represent these configurations we tried tomodelthe distinctions we needed to clarify the relationships between theabstract and material dimensions leaving aside for the moment theambiguous notions18

These distinctions help us clarify certain structural issues that ap-peared in our first attempts to provide an EpiDoc transcription for theseconfigurations presented in section 222 The way an entity can beidentified described and represented by means of markup is never adirect consequence of its intrinsic nature but depends on the perspec-tive adopted For example if a standing statue is entirely preserved ina museum in one piece its base would be described as what we call lsquoatypological object partrsquo but not as a lsquophysical object partrsquo In contrastif a similar statue is broken into two different parts (eg one being thebase and the other the body) and is then located in different museumsboth the base and body parts will then be described as lsquophysical objectpartsrsquo according to this typology From the textual perspective parallelexamples can be explored A composite text consisting of heteroge-neous components [2222] may not be considered as an abstract textualunit fitting into existing literary genres As an existing unit of the sourceyet implicit its identification is subject to interpretation Its inclusionin the representation as a logical textual unit depends on the decisionof the editor If it is represented it must then be seen as an editorialunitwhich materializes an entity that is implicitly present in the sourceThe nature of this editorial decision is structural As it operates at thehighest level of the hierarchy (the text that encompasses the others) itimpacts the way the entities are defined All of these key entities havefound a corresponding element in the EpiDoc schema Does this allowus to build a coherent encoding strategy

17 No more than the TEI EpiDoc is meant to be a prescriptive standard with respect tothe use of the elements

18 For a comprehensive exploration of the definition of what an inscription is from anontological perspective see Panciera 2012


Entity name Definition

Text-bearing object

A material object (artifact) that bears one orseveral inscribed texts The material objectcan consist of one single piece or severaldistinct physical elements

Physical object part

A detachable physical part of a materialobject that can be physically isolated such asa slab a bloc or a fragment Several objectsparts originating from the same object(whether single or composite) may be kept indifferent institutions

Typological objectpart (or rsquophysicalfeaturersquo)

A non detachable part of an object identifiedwith reference to a given epigraphic orarcheological typology (eg base front-faceside etc)

Inscribed entity The set of marks that were inscribed on amaterial support

Abstract text

An abstract entity corresponding to thersquoobject of thoughtrsquo that is the denotata of theinscribed entity or its intellectual content Itcan be classified into a textual genre such asa decree a dedication a manumission etc Itmay be structured as a unified or compositetext

Textual componentof a composite text

A distinct text that pertains to a definedgenre and that structurally functions as acomponent of an overall composite text

Edited text

A representation of the inscribed textintended for publication As the result of ascholarly process involving interpretationand editorial choices it is supposed torespect some shared standards orconventions for both the structure and thedistinctions represented

Tab 222 Key entities and their definitions


224 The encoding strategy of the IGLouvre projectThis encoding strategy is being defined within the framework of

the IGLouvre project Our first objective is to offer a system compliantwith the EpiDoc schema which should be able to coherently representwith markup all the configurations we have identified in our corpus Asecond and derived objective is to enable the highest possible flexibilityin the exploitation and representation of these relationships in the webinterface The contours of the final web application that will give aninterface to to the digital publication are not specified yet But sincethe aim is to exploit thoroughly the material and textual dimensions ofthe various items present in the Louvre collection we need to be ableto define a precise connection between these entities As is highlightedin table 223 the mapping between the EpiDoc schema and the entitiesof our model has been established rather easily However we need tosay that the decision to use the lsquomsPartrsquo element to represent the entitylsquophysical object partrsquo is currently under discussion19 within the EpiDocand TEI communities

19 In her feature request ticket (httpsourceforgenetpteifeature-requests505)posted on 2014 April 29th C Schroeder asks for a re-definition of the element in theguidelines for exactly the same kind of use for the element


Entityname

Functionofthe

elem

ent

Path(position

intree)

Possibleelem

ents

Atext-bearin

gobject

Identifi

cation

TEIteiHeadersourceDesc

msD

esc

ltmsIdentifiergt

ltaltIdentifiergt

Description

TEIteiHeadersourceDescmsDescphysDesc

objectDesc

supp

ortD

esc

ltsupportgt

Aphysicalobjectpart

Identifi

cation

TEIteiHeadersourceDescmsDesc

msP

art

ltmsIdentifiergt

ltaltIdentifiergt

Description(object)

TEIteiHeadersourceDescmsDescmsPart

physDescobjectDesc

supp

ortD

esc

ltsupportgt

Description(history)


msPart

hist

ory

ltorig

ingt

ltprovenancegt

Atypologicalobject

part

Identifi

cation

TEItextbodydiv[type=rsquoeditionrsquo]

div[

ty

pe=rsquo

text

part

rsquo][xm

lid]

Categorization


div[

ty

pe=rsquo

text

part

rsquo][

subt

ype]

Controlled

vocabulary

Description(layout)


physDescobjectDesc

layo

utD

esc

ltlayoutgt

Content

(transcrip

tion)


div[type=rsquotextpartrsquo]

abmixed

content

Theabstracttext

Identifi

cation


msC

onte

nts

ltmsItem[xm

lid]gt


Categorization


msC

onte

nts

ltmsItem[class]gt

Nam

eTE

IteiHeadersourceDescmsDescmsContents

msI

tem

lttitlegt

Atextualcom

ponent

ofacompositetext

Identifi

cation


div[

ty

pe=rsquo

text

part

rsquo][xm

lid]

Categorization


div[

ty

pe=rsquo

text

part

rsquo][

subt

ype]

Controlled

vocabulary

Theeditedtext

Acollectionofweb

pagesd

esigned

from

the

transformationof

theEp

iDoc

XML

representation

Tab

223

Amapping

betweenourtypologyandthekeyentitiesw

ithintheEpiDoc

schema


In order to achieve our second aim we intend to use the linkingmechanisms provided by the TEI framework Once the entities areidentified with an xmlid attribute the markup can establish withprecision the relations between one another using attributes like tar-get or corresp To give a detailed example the case illustrated inparagraph 2223 is developed in table 224 the encoding distinguishestwo msParts elements in the teiHeader and four textual units one foreach epigram and one for the group they constitute20

Identified elements can be pointed to using the xmlid attributeMore precise linking between the abstract texts listed in the msContentelement (teiHeader) can be provided using as many ltlocusgt elementsas needed with a target attribute The same pattern may also beused to record distinct stonecutters (in handNote elements) or differentdates (in origDate elements) and associate themwith the relevant partsin the transcribed text21 For cases such as those we treated in theexamples commented in 2221 2222 and 2224 where the inscriptionis carved on a single object we decided to use a lsquodefault msPartrsquo to drawa symmetry with the case were several msParts are used

In cases where the textual structure overlaps the physical agency ofinscribed texts areas the use of an empty element milestone assortedwith the relevant unit attribute (eg lsquosectionrsquo)22 resolves the problemcaused by the need to represent two overlapping structures in a singleXML tree It should however not be denied that this approach impactsthe workload of the task of encoding But in our point of view it provesto be worthwhile as soon as you consider the range of possibilitiesoffered in the digital web interface In some cases like the re-use ofthe same support for the engraving of successive texts this strategy isalso entirely necessary in order to link them to the same object

20 The ODD file which formalises the schema and its documentation is supposed toinclude the typology used for the subtype attribute The EpiDoc documentationstates that subtype is not constrained but common values might include ldquofrag-mentrdquo ldquocolumnrdquo ldquosectionrdquo etc We consider that any categorisation can be used

21 It is also possible to record data related to illustrations of the objects or inscribedportions of the objects (eg drawings photographs etc) in a facsimile element toprovide links to an image or a region of an image via the facs attribute but it is aquite awidespread practice which doesnrsquot require special comments for our purpose

22 The term lsquosectionrsquo denotes the abstract nature of the entity considered It can beopposed to another kind of milestone unit like lsquoblockrsquo which can be used when aphysical structure overlaps a textual one


Entities Type Encoding

Slabs physicalobject part

lt- - in the teiHeader - -gt ltmsPartxmlid=rdquomspat01rdquo n=rdquoArdquocorresp=rdquomilst01rdquogt ltmsPartxmlid=rdquomspat02rdquo n=rdquoBrdquocorresp=rdquomilst02rdquogt

Epigrams Abstracttexts

lt- - in the teiHeader - -gt ltmsItemxmlid=rdquomsi01rdquogt lttitlegtEpigram forAntiphonlttitlegt ltlocustarget=rdquolg01rdquogt ltmsItemgtltmsItem xmlid=rdquomsi02rdquogtlttitlegtEpigram forEurymenideslttitlegt ltlocustarget=rdquolg02rdquogt ltmsItemxmlid=rdquomsi03rdquogt lttitlegtEpigram forEurymenideslttitlegt ltlocustarget=rdquolg03rdquogt ltmsItemgt

A group ofepigrams forthe sons ofSophocles

The editedoverall text

lt - - in the TEItext element- -gt ltdivtype =rdquoeditionrdquogt ltdiv type=rdquotextpartrdquosubtype=rdquogroup-of-epigrams25rdquogt ()ltdivgt ltdivgt

Eachepigram

Textualcomponentsof acompositetext

lt - - in theTEItextdivtype=rsquoeditionrsquo element--gt ltmilestone xmlid=rdquomilst01rdquounit=rdquoblockrdquo corresp=rdquomspart01rdquogt() ltmilestone xmlid=rdquomilst02rdquounit=rdquoblockrdquo corresp=rdquomspart02rdquogt()

The versesof eachepigram

Textinscribed ineach textualcomponent

ltlg xmlid=rdquolg01rdquo gt ltlgtltlbgtοὐγάmicroον ()ltlgt ltlgtltlbgt()ltlgtltlggt

Tab 224 EpiDoc markup of the example presented in 2223


225 Conclusions and perspectivesThis encoding strategy permits us to meet the following requirements

bull the material and abstract dimensions of the items in the Louvrecollection are taken into account in a compliant EpiDoc markupexploiting its capacity to provide fine grained identifiers and link-ingmechanisms that are required to build on an interface showinginscriptions not just as decontextualized texts

bull the scientific editors keep full control on the editorial choices theymade beyond the structure of the printed or digital publication

bull the deconstruction of the notion of lsquoinscriptionrsquo will also providehelp for designing and implementing several extractions and dataexports that will have to be developed in the near future to ensurethe interop-erability of the digital collection and its re-use forother projects

Further work needs to be done to make explicit this encoding strategyin the form of an ODD schema and documentation file One of theimportant next steps of the IGLouvre project will be the specification ofthe web interface of the digital edition But before this further stage itwould be interesting to reformulate our model of what an inscription isusing the CIDOC-CRM metamodel23 This work may provide critiqueand opportunity for enhancements It also may help see to what extentour work can be useful for other projects In conclusion even thoughthe material and the textual dimensions cannot be separated in theeditorial representation they need to be precisely distinguished in theabstract model of the source that must be clarified before structuringthis representation Finally is the ambiguity of the notion of lsquoinscrip-tionrsquo a hurdle impossible to avoid What is an inscription It is aninscribed text an inscribed object in a given state of preservation oran edited text We think that in order to escape ambiguity we haveonly two ways stepping back to the ancient meaning of the Greekepigramma (ἐπίγραmicromicroα) and state that an inscription is nothing else butletters on a support or accept that in the epigraphic field an inscription

23 Other authors have already explored this perspective [Ore et al 2009] In the lastmeeting of the TEI consortium the same authors suggested the introduction of newelements for the entities physicalObject and conceptualObject httpwwwtei-corgSIGOntologiesmeetingsm20131003html


is above all an editorial unit which results from individual scientificchoices and disciplinary criteria The need for a clear understanding ofthis underlying model may be considered as one of the most fruitfulcontributions of the digital edition

Bibliography

Aneziri S (2003) ldquoDie Vereine der dionysischen Techniten im Kontextder hellenistischen Gesellschaft Untersuchungen zur GeschichteOrganisation und Wirkung der hellenistischen TechnitenvereinerdquoIn Historia Einzelschriften 163 pp 3ndash542 url httpcatinistfraModele=afficheNampcpsidt=15281450 (visited on 07292014) (citon p 331)

Berges D (1996) Rundaltaumlre aus Kos und Rhodos GebruumlderMann Verlag(cit on pp 328 330)

Bodard G (2008) ldquoThe Inscriptions of Aphrodisias as Electronic Pub-lication a userrsquos perspective and a proposed paradigmrdquo In DigitalMedievalist 4 url httpwwwdigitalmedievalistorgjournal4bodard (cit on p 326)

Buzzetti D (2002) ldquoDigital representation and the textmodelrdquo InNewLiterary History 331 pp 61ndash88 url httpmusejhuedujournalsnlhsummaryv033331buzzettihtml (visited on 07292014) (citon p 340)

Cayless H C Rouecheacute T Elliott and G Bodard (2009) ldquoEpigraphyin 2017rdquo In Digital Humanities Quarterly 31 url http www digitalhumanitiesorgdhqvol31000030000030html (visitedon 02212013) (cit on p 336)

Conze A (1860) Reise auf den Inseln des thrakischen Meeres Hannover(cit on p 333)

Dain A (1933) Inscriptions grecques du Museacutee du Louvre les textes ineacuteditsSocieacuteteacute drsquoeacuteditionrdquo Les Belles Lettresrdquo (cit on p 328)

Devlin G and R Shaw (2014a) 2013 UPDATE ON THE 3D TECHNOL-OGY SIDE OF OGHAM IN 3D url http oghamcelt dias ieresourcesdocs2013-11_ScanningProcess_updatepdf (visited on07292014) (cit on p 327)

mdash (2014b) Scanning processing and modeling of Ogham Stones url httpoghamceltdiasieresourcesdocs2013-01_ScanningProcess_


TheDiscoveryProgramme pdf (visited on 07292014) (cit onp 327)

Dunant C and J Pouilloux (1958) ldquoRecherches sur lrsquohistoire et lescultes de Thasos II De 196 avant Jrdquo In Eacutetudes thasiennes pp 170ndash172 (cit on p 333)

Duumlrrbach F (1921) Choix drsquoinscriptions de Deacutelos Vol 1 1-2 Eacuteditions ELeroux (cit on p 331)

FroehnerW (1865) Les Inscriptions grecques Deacutepartement des antiqueset de la sculpture moderne Paris Mourgues (cit on p 331)

Hamiaux M and A Pasquier (1998) Les sculptures grecques Vol 2Reacuteunion des museacutees nationaux (cit on pp 328 329)

Lane E N (1988) ldquoΠΑΣΤΟΣrdquo inGlotta 66 pp 100ndash123 (cit on p 333)Le Guen B (2001) Les Associations de Technites dionysiaques agrave lrsquoeacutepoque

helleacutenistique Vol 1 Corpus documentaire Vol 2 Synthegravese EacutetudesdrsquoArcheacuteologie Classique XI-XII Nancy (cit on p 331)

Mendel G (1900) ldquoInscription de Thasosrdquo InBulletin de correspondancehelleacutenique 241 pp 263ndash284 url http www persee fr webrevueshomeprescriptarticlebch_0007-4217_1900_num_24_1_3412 (visited on 07292014) (cit on p 333)

Ore C-E and Oslash Eide (2009) ldquoTEI and cultural heritage ontologiesExchange of informationrdquo In Literary and Linguistic Computing 242pp 161ndash172 url httpllcoxfordjournalsorgcontent242161short (visited on 07292014) (cit on p 347)

Panciera S (2012) ldquoWhat Is an Inscription Problems of Definition andIdentity of an Historical Sourcerdquo In Zeitschrift fuumlr Papyrologie undEpigraphik 183 pp 1ndash10 url httpwwwdigitalmeetsculturenetwp-contentuploads201310Panciera-Inscription-ZPE-2012pdf(visited on 07292014) (cit on p 340)

Peek W (1955) Griechische Vers-Inschriften - Grab-Epigramme Vol 1Berlin Akademie-verlag (cit on p 333)

Terras M M (2006) Image to interpretation an intelligent system to aidhistorians in reading the Vindolanda texts OxfordUniversity Press urlhttpdiscoveryuclacuk12320 (visited on 07292014) (cit onp 326)


Submitted on 14 Apr 2015









Keywords






















































































left side

























Text-bearing object







Abstract text




Edited text








Entityname

Functionofthe

elem

ent

Path(position

intree)

Possibleelem

ents

Atext-bearin

gobject

Identifi

cation


msD

esc

ltmsIdentifiergt

ltaltIdentifiergt

Description


objectDesc

supp

ortD

esc

ltsupportgt

Aphysicalobjectpart

Identifi

cation


msP

art

ltmsIdentifiergt

ltaltIdentifiergt

Description(object)


physDescobjectDesc

supp

ortD

esc

ltsupportgt



msPart

hist

ory

ltorig

ingt

ltprovenancegt

Atypologicalobject

part

Identifi

cation


div[

ty

pe=rsquo

text

part

rsquo][xm

lid]

Categorization


div[

ty

pe=rsquo

text

part

rsquo][

subt

ype]

Controlled

vocabulary

Description(layout)


physDescobjectDesc

layo

utD

esc

ltlayoutgt

Content

(transcrip

tion)



abmixed

content

Theabstracttext

Identifi

cation


msC

onte

nts

ltmsItem[xm

lid]gt


Categorization


msC

onte

nts

ltmsItem[class]gt

Nam

eTE


msI

tem

lttitlegt

Atextualcom

ponent

ofacompositetext

Identifi

cation


div[

ty

pe=rsquo

text

part

rsquo][xm

lid]

Categorization


div[

ty

pe=rsquo

text

part

rsquo][

subt

ype]

Controlled

vocabulary

Theeditedtext

Acollectionofweb

pagesd

esigned

from

the

transformationof

theEp

iDoc

XML

representation

Tab

223

Amapping


ithintheEpiDoc

schema

















Eachepigram
















Bibliography


























Keywords






















































































left side

























Text-bearing object







Abstract text




Edited text








Entityname

Functionofthe

elem

ent

Path(position

intree)

Possibleelem

ents

Atext-bearin

gobject

Identifi

cation


msD

esc

ltmsIdentifiergt

ltaltIdentifiergt

Description


objectDesc

supp

ortD

esc

ltsupportgt

Aphysicalobjectpart

Identifi

cation


msP

art

ltmsIdentifiergt

ltaltIdentifiergt

Description(object)


physDescobjectDesc

supp

ortD

esc

ltsupportgt



msPart

hist

ory

ltorig

ingt

ltprovenancegt

Atypologicalobject

part

Identifi

cation


div[

ty

pe=rsquo

text

part

rsquo][xm

lid]

Categorization


div[

ty

pe=rsquo

text

part

rsquo][

subt

ype]

Controlled

vocabulary

Description(layout)


physDescobjectDesc

layo

utD

esc

ltlayoutgt

Content

(transcrip

tion)



abmixed

content

Theabstracttext

Identifi

cation


msC

onte

nts

ltmsItem[xm

lid]gt


Categorization


msC

onte

nts

ltmsItem[class]gt

Nam

eTE


msI

tem

lttitlegt

Atextualcom

ponent

ofacompositetext

Identifi

cation


div[

ty

pe=rsquo

text

part

rsquo][xm

lid]

Categorization


div[

ty

pe=rsquo

text

part

rsquo][

subt

ype]

Controlled

vocabulary

Theeditedtext

Acollectionofweb

pagesd

esigned

from

the

transformationof

theEp

iDoc

XML

representation

Tab

223

Amapping


ithintheEpiDoc

schema

















Eachepigram
















Bibliography









































































































left side

























Text-bearing object







Abstract text




Edited text








Entityname

Functionofthe

elem

ent

Path(position

intree)

Possibleelem

ents

Atext-bearin

gobject

Identifi

cation


msD

esc

ltmsIdentifiergt

ltaltIdentifiergt

Description


objectDesc

supp

ortD

esc

ltsupportgt

Aphysicalobjectpart

Identifi

cation


msP

art

ltmsIdentifiergt

ltaltIdentifiergt

Description(object)


physDescobjectDesc

supp

ortD

esc

ltsupportgt



msPart

hist

ory

ltorig

ingt

ltprovenancegt

Atypologicalobject

part

Identifi

cation


div[

ty

pe=rsquo

text

part

rsquo][xm

lid]

Categorization


div[

ty

pe=rsquo

text

part

rsquo][

subt

ype]

Controlled

vocabulary

Description(layout)


physDescobjectDesc

layo

utD

esc

ltlayoutgt

Content

(transcrip

tion)



abmixed

content

Theabstracttext

Identifi

cation


msC

onte

nts

ltmsItem[xm

lid]gt


Categorization


msC

onte

nts

ltmsItem[class]gt

Nam

eTE


msI

tem

lttitlegt

Atextualcom

ponent

ofacompositetext

Identifi

cation


div[

ty

pe=rsquo

text

part

rsquo][xm

lid]

Categorization


div[

ty

pe=rsquo

text

part

rsquo][

subt

ype]

Controlled

vocabulary

Theeditedtext

Acollectionofweb

pagesd

esigned

from

the

transformationof

theEp

iDoc

XML

representation

Tab

223

Amapping


ithintheEpiDoc

schema

















Eachepigram
















Bibliography








































Text-bearing object







Abstract text




Edited text








Entityname

Functionofthe

elem

ent

Path(position

intree)

Possibleelem

ents

Atext-bearin

gobject

Identifi

cation


msD

esc

ltmsIdentifiergt

ltaltIdentifiergt

Description


objectDesc

supp

ortD

esc

ltsupportgt

Aphysicalobjectpart

Identifi

cation


msP

art

ltmsIdentifiergt

ltaltIdentifiergt

Description(object)


physDescobjectDesc

supp

ortD

esc

ltsupportgt



msPart

hist

ory

ltorig

ingt

ltprovenancegt

Atypologicalobject

part

Identifi

cation


div[

ty

pe=rsquo

text

part

rsquo][xm

lid]

Categorization


div[

ty

pe=rsquo

text

part

rsquo][

subt

ype]

Controlled

vocabulary

Description(layout)


physDescobjectDesc

layo

utD

esc

ltlayoutgt

Content

(transcrip

tion)



abmixed

content

Theabstracttext

Identifi

cation


msC

onte

nts

ltmsItem[xm

lid]gt


Categorization


msC

onte

nts

ltmsItem[class]gt

Nam

eTE


msI

tem

lttitlegt

Atextualcom

ponent

ofacompositetext

Identifi

cation


div[

ty

pe=rsquo

text

part

rsquo][xm

lid]

Categorization


div[

ty

pe=rsquo

text

part

rsquo][

subt

ype]

Controlled

vocabulary

Theeditedtext

Acollectionofweb

pagesd

esigned

from

the

transformationof

theEp

iDoc

XML

representation

Tab

223

Amapping


ithintheEpiDoc

schema

















Eachepigram
















Bibliography




























Entityname

Functionofthe

elem

ent

Path(position

intree)

Possibleelem

ents

Atext-bearin

gobject

Identifi

cation


msD

esc

ltmsIdentifiergt

ltaltIdentifiergt

Description


objectDesc

supp

ortD

esc

ltsupportgt

Aphysicalobjectpart

Identifi

cation


msP

art

ltmsIdentifiergt

ltaltIdentifiergt

Description(object)


physDescobjectDesc

supp

ortD

esc

ltsupportgt



msPart

hist

ory

ltorig

ingt

ltprovenancegt

Atypologicalobject

part

Identifi

cation


div[

ty

pe=rsquo

text

part

rsquo][xm

lid]

Categorization


div[

ty

pe=rsquo

text

part

rsquo][

subt

ype]

Controlled

vocabulary

Description(layout)


physDescobjectDesc

layo

utD

esc

ltlayoutgt

Content

(transcrip

tion)



abmixed

content

Theabstracttext

Identifi

cation


msC

onte

nts

ltmsItem[xm

lid]gt


Categorization


msC

onte

nts

ltmsItem[class]gt

Nam

eTE


msI

tem

lttitlegt

Atextualcom

ponent

ofacompositetext

Identifi

cation


div[

ty

pe=rsquo

text

part

rsquo][xm

lid]

Categorization


div[

ty

pe=rsquo

text

part

rsquo][

subt

ype]

Controlled

vocabulary

Theeditedtext

Acollectionofweb

pagesd

esigned

from

the

transformationof

theEp

iDoc

XML

representation

Tab

223

Amapping


ithintheEpiDoc

schema

















Eachepigram
















Bibliography
























Categorization


msC

onte

nts

ltmsItem[class]gt

Nam

eTE


msI

tem

lttitlegt

Atextualcom

ponent

ofacompositetext

Identifi

cation


div[

ty

pe=rsquo

text

part

rsquo][xm

lid]

Categorization


div[

ty

pe=rsquo

text

part

rsquo][

subt

ype]

Controlled

vocabulary

Theeditedtext

Acollectionofweb

pagesd

esigned

from

the

transformationof

theEp

iDoc

XML

representation

Tab

223

Amapping


ithintheEpiDoc

schema

















Eachepigram
















Bibliography







































Eachepigram
















Bibliography
































Bibliography























The Inscription between text and object: The deconstruction of ...

Documents

Transcript of The Inscription between text and object: The deconstruction of ...