Skip to navigation – Site map

HomeIssuesIssue 14What is Text, Really? TEI and Bey...

Index terms

Top of page

Full text

1The 2019 annual TEI conference in Graz was titled “What is text, really? TEI and beyond.” The development of the TEI and the discussion around it has shown how manifold text(s) can be, and how far-reaching the TEI approach is. Thus, the theme of the conference, “What is text, really?” posed a fundamental question, which goes beyond the pure reference to the seminal paper by DeRose, Durand, Mylonas, and Renear (1990). In 1990 they answered the question somewhat pragmatically by proposing a model for text as an ordered hierarchy of content objects which can easily be formalised with digital technologies, but, as they said later on: text can be much more than that. Encoding can make various aspects of texts explicit, enabling scholars to examine their nature and their relationship with other objects. In this context, the power of TEI rests in its technological interactions, through supporting software of all kinds operating upon texts, from visualisation to annotation tools, digital publishing systems, or statistical analysis. The TEI framework is a way of modelling knowledge and engaging in a dialogue with ontologies, conceptual models, and recent approaches such as text as graph. For text-centric disciplines the TEI offers a range of solutions that address core research needs. However, text and its encoding is only a small part of the data modelling ecosystem and digital humanities practices. XML/TEI competes with other formats, models, and approaches to processing. This issue tries to give some answers as to how TEI relates to the surrounding ecosystem.

2This volume makes available a selection of the papers that were presented at the 2019 conference. The authors took their time to incorporate inspiration from the conference itself and various comments of the peer reviewers. One could arrange the contributions to this issue into three groups, each one reflecting on a different aspect of “What is text, really?”:

  1. Which consequences result from the decision of the TEI to model texts in the XML ecosystem and how do TEI files relate to graph-based technologies?

  2. Where are the limits of the Guidelines in covering specific textual phenomena?

  3. How should the digital text, which the reader receives only via algorithmic mediation, be performatively conceived?

1. What are the consequences of the decision of the TEI to model texts in the XML ecosystem, and how do TEI files relate to graph-based technologies?

3Several papers discuss this topic. Most prominently, Elli Bleeker, Ronald Haentjens Dekker and Bram Buitendijk (https://​journals.​openedition.​org/​jtei/​3919) dive into non-XML based text encoding. They describe the details of the TAGML mark-up language which supports an understanding of text that includes layers, non-linearity, discontinuity and overlapping structures. Several papers consider the relationship to graph-based data from the Linked Open Data and Semantic Web community. The two papers dealing with parliamentary corpora consider the relationship between XML/TEI and other annotation and data description standards: Tomaž Erjavec and Andrej Pančur (https://​journals.​openedition.​org/​jtei/​4133) describe the application profile proposed by the CLARIN consortium. While this seems to be standard use case for a TEI customisation, they provided pipelines to formats for computational linguistic processing and have argued for ways to align their work with standards like Akoma Ntoso/Legal Document XML and RDF. Tanja Wissik (https://​journals.​openedition.​org/​jtei/​4214) touches the border of texts encoded in TEI with non-verbal or at least non standardised participation in parliamentary debates—interruptions and interjections—and proposes to leave the pure TEI ground for a SKOS taxonomy of these communicative acts. Parliamentary records are also subject of Naomi Truan’s and Laurent Romary’s contribution (https://​journals.​openedition.​org/​jtei/​4164), who derive interpretations of the TEI from an interpretation of the linguistic setting, making clear how the utterances in this context are different for playwright’s texts. Their small scale project resulted in a generalised proposal for encoding corpora of parliamentary debate. Martin de la Iglesia, Franziska Diehr, Uwe Sikora, Sven Gronemeyer, Maximilian Behnert-Brodhun, Christian Prager, and Nikolai Grube (https://​journals.​openedition.​org/​jtei/​3336) study the encoding of texts in a partially deciphered writing system. They have to deal with unknown graphematics and thus heavily use glyph representation and region description in zones. However, much of the knowledge about the texts and the script is managed in non-TEI technologies like RDF and CIDOC-CRM. Finally, we have to acknowledge that the afterlife of texts depends on their preservation – and the research data community has built its own standards. Andrew Creamer, Gaia Lembi, Elli Mylonas, and Michael Satlow (https://​journals.​openedition.​org/​jtei/​4324) take up the challenge of how to make preservation a central part of TEI projects. Their paper discusses how the EpiDoc-encoded data of the Inscriptions of Israel/Palestine project inserts itself in the FAIR ecosystem and they evaluate their solution against FAIR metrics.

2. Where are the limits of the Guidelines in covering specific textual phenomena?

4TEI is, for instance, a well-established standard for linguistic corpora. However, linguistics requires rule sets that are much stricter than the generous latitude given in the TEI Guidelines (Burnard 2019). Consider the case for processing of etymological information in dictionaries: Jack Bowers, Axel Herold, Toma Tasovac, and Laurent Romary (2022) describe an encoding profile serving this purpose. Stylometry and standard distant reading conceptualises texts as a bag of words. The linguistic community has regarded detailed textual annotation to be superfluous, so it has not considered TEI based encoding at large. Nevertheless, the ELTEC consortium has decided to build their corpus with TEI. Lou Burnard, Christof Schöch, and Carolin Odebrecht (https://​journals.​openedition.​org/​jtei/​3500) report on their attempts to build a bridge to the stylometry community by creating restrictive customizations implementing a “No Surprises” principle. The contribution by Kazuhiro Okada, Satoru Nakamura, and Kiyonori Nagasaki (https://​journals.​openedition.​org/​jtei/​4403) demonstrates how non-European textual traditions include textual features worth encoding with the TEI. It will certainly trigger a discussion, how to use the ruby encodings described in this contribution for European glossing traditions like ancient scholiae (Mastronarde 2020, Dué et al. 2020, Callimachus 2016) or medieval commentaries on religious and legal texts (Monella 2019, Rehbein 2014, Aris and Wiener 2012, Witt 2018). Finally, text and its interpretation are subject to various forms of uncertainty. Michał Kozak, Alejandro Rodríguez, Alejandro Benito-Santos, Roberto Therón, Michelle Doran, Amelie Dorn, Jennifer Edmond, Cezary Mazurek, and Eveline Wandl-Vogt (https://​journals.​openedition.​org/​jtei/​4239) describe the PROVIDEDH Open Science annotation platform that stores these forms in TEI.

3. How should the digital text, which the reader receives only via algorithmic mediation, be performatively conceived?

5Text is more than a static set of data. The reception of a digital text involves algorithms, which are discussed in a final group of three papers: Punctuation seems to be a minor detail of text, but Elisa Cugliana and Gioele Barabucci (https://​journals.​openedition.​org/​jtei/​3715) can describe the computational, philological, and epistemological implications of their encoding. They propose detailed encoding combined with a rule-based transformation stack to convert the initial encoding. The combination of text as data and processing instructions records the full knowledge of the editors. Another well-established technology stack beyond TEI is relational database management. Such a database is combined with XML/TEI encoded texts in the Digital Victorian Periodical Poetry project. Martin Holmes (https://​journals.​openedition.​org/​jtei/​3874) describes the architecture that handles this setup, where the metadata is maintained by a RDBMS and the canonical source of textual data are the TEI files. The automatic conversion methods the project deploys demonstrate that peaceful coexistence of both stacks is realistic. Text is not a given object; text is produced and can be constantly under revision. When Anne Baillot and Julie Giovacchini (https://​journals.​openedition.​org/​jtei/​3419) discuss the use of the TEI for the management of a scholarly journal, it becomes visible how TEI reflects practices in text production and what is kept outside the text encoding.

6This issue is a collaborative effort—by the participants in the conference, who discussed the papers presented; the authors, who directed the free-flowing discussions into a stable scholarly text; the reviewers, who checked to see if the arguments made by the authors meet scholarly standards; and the editorial team of the jTEI, who converted the manifold formats and styles into jTEI conformant XML. I want to thank all of them, and name some in particular: Walter Scholger, the heart of the conference; and Joel Kalvesmaki, Tanja Wissik, Pietro Luzzi, Tiago Sousa Garcia, and Ron Van den Branden, who as the jTEI editors did the hard work of the review and publishing process, managed by Joel. Being named as the editor of this issue, I feel like a tiny drop in the stream of work, engagement, and efforts by all of you. Thanks!

SVN keywords: $Id: jtei-vogeler-introduction-source.xml 1182 2023-03-01 15:11:53Z ron $

Top of page


Aris, Marc-Aeilko, and Claudia Wiener, project leaders. 2012. “Editing Glosses.” https://​www.​mueze.​uni-muenchen.​de/​editing_glosses/​index.​html.

Bowers, Jack, Axel Herold, Toma Tasovac, and Laurent Romary. 2022. “TEI Lex-0 Etym: Toward Terse Recommendations for the Encoding of Etymological Information.” Journal of the Text Encoding Initiative, Rolling Issue. doi:10.4000/jtei.4300.

Burnard, Lou. 2019. “What Is TEI Conformance, and Why Should You Care?” Journal of the Text Encoding Initiative 12. doi:10.4000/jtei.1777.

Callimachus. 2016. Aetia. Edited by Susan Stephens, Chris Francese, Meagan Ayer. Carlisle (PA): Dickinson College Commentaries. http://​dcc.​dickinson.​edu/​callimachus-aetia/​preface.

DeRose, Steven J., Renear, Allen H., David G. Durand, and Elli Mylonas. 1990. “What Is Text, Really?” Journal of Computing in Higher Education 1 (2): 3–26. doi:10.1007/BF02941632.

Dué, Casey, Mary Ebbott, Neel Smith, and Christopher Blackwell, eds. 2019-2020. Homer Multitext. Houston: University of Houston. Release 2020h. urn:cite2:hmt:publications.cex.2020h:al, https://​www.​homermultitext.​org/​hmt-digital/​index.​html.

Mastronarde, Donald J. 2020. “Euripides Scholia.” https://​euripidesscholia.​org/​index.​html.

Monella, Paolo. 2019. “A Digital Critical Edition Model for Priscian: Glosses, Graeca, Quotations.” Analecta Romana Instituti Danici 44: 135–149.

Rehbein, Malte. 2014. “From the Scholarly Edition to Visualization: Re-using Encoded Data for Historical Research.” International Journal of Humanities and Arts Computing 8 (1): 81–105. doi:10.3366/ijhac.2014.0121.

Witt, Jeffrey. 2018. “Digital Scholarly Editions and API Consuming Applications.” In: Digital Scholarly Editions as Interfaces, edited by Roman Bleier, Martina Bürgermeister, Helmut W. Klug, Frederike Neuber, and Gerlinde Schneider, 219–47. Schriften des Instituts für Dokumentologie und Editorik. Norderstedt: BoD.

Top of page


Top of page


Electronic reference

Georg Vogeler, “What is Text, Really? TEI and Beyond – An Introduction”Journal of the Text Encoding Initiative [Online], Issue 14 | April 2021- March 2023, Online since 27 February 2023, connection on 26 February 2024. URL:; DOI:

Top of page

About the author

Georg Vogeler

Universität Graz, Zentrum für Informationsmodellierung

Top of page


The text only may be used under licence For this publication a Creative Commons Attribution 4.0 International license has been granted by the author(s) who retain full copyright. . All other elements (illustrations, imported files) are “All rights reserved”, unless otherwise stated.

Top of page
Search OpenEdition Search

You will be redirected to OpenEdition Search