The complex nature of translation, as a hermeneutic, culture-dependent gesture, has been extensively studied by Translation Studies scholars as well as by scholars in many adjacent fields of science and art.^[1] The reflection has embraced various turns as the recently ‘post-translation studies’ which is defined by Siri Nergaard and Stefano Arduini as ‘a sort of new era […], where translation is viewed as fundamentally transdisciplinary, mobile, and open-ended’ (2011, p. 8), and is further explored by Gentzler (2017).

Following the scientific and cultural developments, translation and post-translation studies seek constantly to explore more in depth all possible dimensions of translation’s dynamics, opening in new, under-represented in the past research areas, and work on an interdisciplinary basis with many other fields of study. This is the case of translations published in periodicals and anthologies, whose study is of interest for both Translation and Periodical Studies, which seem recently to burgeon (Van Remoortel et al., 2016). Our project aims to contribute to this debate with data from the Modern Greek field and this paper will tend to briefly describe it.

In Greece, noteworthy work has been done since the 1990s concerning periodicals but also anthologies. Apart from papers, monographs, and dissertations in the field, several research projects are helping us map the area and reveal its fruitfulness. From the many projects that focused on the Greek periodicals of the 19th and 20th centuries, we must mention the notable ones that have helped pave the way. One of the most significant projects was the Literary and Art Periodicals (1901–1940), under the scientific supervision of Professor Ch. L. Karaoglou from the Aristotle University of Thessaloniki. The project started in 1994, and its outcomes were published in four volumes, from 1996 until 2007, listing important publications and indexing their collaborators. Another important landmark capitalising on the trove held by the Greek libraries was the digital collections Kosmopolis and Pleias.^[2] The project was implemented by the Library and Information Centre of the University of Patras, under the scientific supervision of the Department of Philology and in collaboration with the holders of the primary sources, namely the Library of the Hellenic Parliament and the Hellenic Literary and Historical Archive (ΕΛΙΑ).^[3] This listing remains indicative and does not do justice to the numerous others developed by many Greek libraries and other institutions over the years.

Literary translations in the periodicals are a compelling area of study for various researchers who try to reach ‘deeper into an interdisciplinary field that considers how culture, translation, and media theory conceptualise textual, contextual, and paratextual dynamics, as well as how sociological approaches to translation and publishing history can help us understand the journal as a network of social practice’ (Fólica, Roig-Sanz, & Caristia, 2020, p. 2). Since it is widely accepted that translation has played a major role ‘in the crystallization of national cultures’ (Even-Zohar, 1990, p. 45) studying the Greek publishing history can help us better understand the country’s literary history too.

In this regard, monumental bibliography research, such as the electronic catalogue Greek Bibliography of the 19th Century,^[4] a database that relies heavily on work from various Greek scholars,^[5] and the two-volume bibliographic work by K. G. Kasinis, who has indexed the Greek translations of foreign literature published between 1801 and 1950, are a valuable tool for research. However, these works only cover one area, that of book publishing. The large number of literary works in periodicals is only partially indexed, and one must browse through and combine many different sources to obtain meaningful results. The only database that synthesises the research work in the field and is currently accessible is the Translations of Literature to Modern Greek: The Translation Production of the Eastern Mediterranean Hellenism During the Period 1880–1930.^[6] This database not only combines translations printed in books and periodicals alike but also harvests material exclusively from the publishing scene of Hellenism in Asia Minor, Cyprus, and Egypt.

The Idea Behind Μετάdata

The University of Patras has a profound interest in Greek periodicals. In addition to the digital collections that the Library and Information Centre administers, a large collection of periodicals exists in the physical format (partly due to donations). Moreover, the Department of Philology has incorporated periodical studies into its curricula and founded EATT,^[7] the Laboratory of Modern Greek Philology, which aims to use literary archives and the press for studying literature.^[8]

The need for a tool that combines all the scattered data about the standalone translations of literary texts (ie that were not published in a book) was evident. Μετάdata^[9] – a wordplay on the prefix μετά, after the Greek word for ‘translation’ (μετάφραση) – aspires to become the portal to all the translated literature published in Greek magazines, newspapers, and anthologies, from 1801 until 1974.^[10] The dates are not arbitrary: 1801 marks the beginning of a period during which Greek publishing progressively flourished and the Greek press began to emerge. The first Greek periodical, Hermes o Logios [Ερμής ο Λόγιος], appeared in 1811, followed by the first long-running Greek newspaper, Hellenikos Telegraphos [Ελληνικός Τηλέγραφος], in 1812, which lasted for 24 years. It is worth highlighting that both periodicals were published in Vienna, a major centre of the Greek diaspora prior to the establishment of the Greek state in 1830. Furthermore, the choice of the early 1800s ensures that the project aligns with the major bibliographic work of Philippos Iliou and the records compiled by K. G. Kasinis, which constitute fundamental documentation tools for the study of Greek book production from the 19th century onwards. The year 1974 is also a significant landmark in the history of modern Greece (the transition to democratic rule called Metapolitefsi [Μεταπολίτευση]), with implications for the country’s cultural formations.

As we were fully aware that this task was gigantic and almost unattainable, we had to set another limit for the time being. Given the personnel resources available at the moment, the decision not to study the primary sources was inevitable. Therefore, the indexing follows the work already done by Greek scholars and postgraduate students, which has already been reviewed and published. The general rule is to adhere to the original study, but we have made minor non-invasive corrections or additions after consulting with the primary sources for validation.

Structuring the Datasets

The indexing of periodicals can be quite problematic, especially of those published more than a century ago or of the popular ones with more opaque editorial practices. During our research on the popular magazine Bouketo, we were misled several times, particularly by popular weeklies and other Greek pulp magazines (Gkogkou, 2023, pp. 68–69). Translated material is even more problematic due to the distance between the source text and the published article, which may contain numerous iterations, most commonly through intermediary translations. Another puzzling case is the phenomenon of pseudotranslations, which refer to ‘texts which have been presented as translations with no corresponding source texts in other languages ever having existed – hence no factual ‘transfer operations’ and translation relationships’ (Toury, 1995, p. 40).^[11] Apart from questionable translations, there are many other challenges, from the inability to identify the author or the original work to the lack of sufficient information about the translating process and its agents. Nevertheless, since our goal is to record precisely and accurately, but not to study intensively, whatever is labelled as a ‘translation’ is included. These works include adaptations presented as such despite being controversial (Bassnett, 2011), mediated translations, self-translations, and pseudotranslations.

The Sources

In plain terms, our database is a synthesis of indexes. Without the volume of research conducted by Greek scholars in the field, our eagerness may have been challenged by the difficult task of collecting, classifying, listing and documenting, evaluating, verifying, and identifying a huge amount of data. We are fortunate that the interest shown by Greek academia has been immense, providing a sizable corpus of quality work. We found that the literature mapping of the existing indexes was not very difficult: the titular standardisation (usually the word ‘index’ is present, along with the periodical’s name and dates of circulation) facilitates retrieval through institutional repositories (many doctoral dissertations and master theses on extensive studies of specific periodicals include indexes) and library catalogues. Moreover, bibliographic endeavours (such as the Neohellenistis digital collection)^[12] and the inclusion of the study of periodicals in the University of Patras’s Department of Philology curriculum since 2008 proved useful.

By relying on already published and proofread work, we could not only eliminate the required time to find the translations by parsing the pages but also avoid possible errors, at least for works that incorporated years of workmanship and dove into the Greek press.^[13] Not everything was perfect, however; certain easily identified mistakes were silently corrected, and minor alterations (mainly, additions) were made based on each re-indexer’s discretion.

Defining the Necessary Fields

According to standard practice in periodical indexing, it is essential to include all the necessary information for the publication’s identification. Following indexing in chronological order by issue, it is common practice among Greek indexers to include the following fields when working on a single periodical:^[14]

• volume and/or issue number;
• date of issue;
• title(s) – subtitles or other paratextual elements such as column titles may be included;
• name of the author(s);
• short description of the subject or the translation or other relevant information (this is also an editing work);
• page number(s) – the page range for the article;
• genre; and
• information about instalments (indication about the starting and ending of a serial). The most common way to indicate this information is with the symbol → after the record for the first instalment and ← after the last.

Specifically for translations, the following fields are included:

• name of the translator(s);
• language or literary tradition; and
• title of original work (a task that, at times, requires in-depth research, especially for poetic texts or short stories).

Minimising Data to Meet Our Needs

By carefully examining the available descriptive data and considering the purpose of our work (a stepping stone for more thorough research), we concluded that it was not viable to keep all the fields. In a large-scale indexing project like Μετάdata, high specificity and extensiveness might be confusing and impractical. As long as we were generally referring to original and exhaustive works, we decided that it would be reasonable, if not convenient, to avoid certain fields. One of these fields was pages. Even in the cases where the index is not (yet) published, the rest of the information suffices for only browsing through a few pages. Another field we excluded was genre, an admittedly ‘problematic and unstable’ concept that ‘still awaits its Linnaeus’ (Duff, 1999, p. 1, 17). Genres change over time, and hybridity in literature published in the press can blur boundaries. The subjectivity on this matter can also be found among the indexers, who must adjust to their source material and the perceptions of certain eras. The third field was the paratext. We retained information relevant to the act of translation, except when referring to mediation, although indirect translation has a long tradition in the reception of foreign literature in Greece, with translations from French gradually increasing over the Italian in the 18th century, as Kasinis (2006) has shown.^[15] Undoubtedly, mediation in translation is a cultural phenomenon with a complex and intriguing character (Toury, 1995), and our decision was dictated merely by the incongruous information gathered and the inconsistency with which periodicals mention their source texts.

One other differentiation concerned the names: we dismissed the names of authors and translators as typed in the periodical, and we only used controlled vocabularies to ensure the consistency and uniformity of the database. For those unfamiliar with the peculiarity of the Greek alphabet, we must outline some points about depicting foreign names in Greek. Most of the time, foreign names are transliterated in the Greek alphabet by editors or translators. At times, transliteration reflects a decision based on pronunciation, for example, when the French poet Charles Baudelaire was transcribed as Σαρλ Μπωντλέρ; or a decision based on the spelling (Τσαρλς Μπωδελαίρ); or even a more radical choice to translate the given name as its equivalent Greek Κάρολος. Another common practice in the past was the Hellenisation of names, resulting in fairly odd correspondences, including Chateaubriand being translated as Σατωβριάνδος and Goethe as Γοήθιος. This information may have been interesting to historical orthography, but it would have also overloaded our data. Therefore, to have an accurate and unique record, we used authority records from either VIAF or BnF^[16] for the authors (and translators) that we could identify.

We also partly retained the indication about serials. The mark follows each title only once, for the first instalment.

Before continuing to present the database and its features, some basic information will give a general idea of the workflow. The database was implemented in Google Sheets (Figure 1), which is convenient for rapid indexing, allows collaboration and simultaneous work, and does not require significant database expertise. Each periodical was analysed in a different sheet according to a specific set of rules and practices. Whenever the indexing was completed, the data was pulled in with the rest of the information into a single consolidated sheet, which was then visualised through Looker Studio, a freely available visualisation tool offered by Google.

Data Visualisation

The richness of the dataset and the results of the copious indexing require visualisations to be appealing and comprehensive to users. To visualise our data, we needed an easy-to-use, feature-rich tool, given the lack of technical expertise within the team and constraints set by the funding. We decided to utilise Looker Studio, a tool that allows complex visualisation through a variety of interactive charts and tables and data cross-filtering. These visualisations were implemented intuitively, and the result was seamlessly integrated into the University of Patras Library and Information Centre’s existing website.

Tables

The central module of the visualisation dashboard consists of a table that contains all the paginated records, displaying the following three main fields (Figure 2): the title of the publication (as it was printed on the page), the name of the author, and the name of the translator (after the authority control, if possible). Apart from browsing through records, alphabetical sorting is also supported for each one of the fields, and automatic filtering is possible by clicking on one of the displayed records.

Supplementary indexes (Figure 3) allow users to efficiently browse and retrieve authors, translators, and indexed periodicals. In these indexes, as in the titles list, all the records associated with that value are displayed by clicking on a specific name or magazine.

Nevertheless, more fields have yet to be displayed in our dataset, and this analytic depiction of all descriptive elements has proven to be the biggest challenge for our visualisation. We have had to find ways to balance between the inconsistencies and irregularities of the publishing practice, the need to provide information as precisely and in as much detail as possible, and the requirement to adjust to the needs of researchers. This challenge has been especially apparent for the time series data, that is the information about the date of issue and its visualisation in the timeline chart (Figure 4). We have handled all these difficulties by adopting an external module, the templated record.

Templated Record

To display all the details of a selected record, we customised and configured a templated record module that would visualise the rest of the information present in the dataset. Thus, by clicking on one of the records in the titles index, all details are displayed and formatted via an HTML template, as shown below (Figure 5).

This module was essential for several reasons, one of which was dates and times. Visualisation demands a specific data format, but the nature of our material does not always allow for compatible data types. Publishing at irregular rates is common for magazines. However, the main problem arises when the cover date does not have a single numerical value but rather a span or even a phrase that has no numerical connotation whatsoever, such as ‘festive issue’, ‘extra’, or ‘summer issue’. Moreover, in Greek, the names of some months can appear with varying spellings, often due to changes in suffixes, though sometimes the differences are more substantial.^[17] These chronological ‘labels’ can carry important information about the nature of the periodicals and should be transcribed exactly as they appear. A representative example is the semi-monthly magazine Xekinima (Ξεκίνημα [Start]). Xekinima was a leftist publication that first appeared in 1944 in Thessaloniki, at a time when the country was still under German occupation. It was created by a group of students belonging to the prestigious University Cultural Club (EOP) supported by the United Panhellenic Organisation of Youth (EPON), the youth wing of the Greek National Liberation Front. The young fighters expressed their rebellion through the radical use of the Modern Greek language, as is clearly stated by Manolis Anagnostakis, one of the most eminent Greek post-war poets and main contributors to Xekinima (Varon, 1987). The spelling of ‘October’ on the cover as Οχτώβρης, instead of Οκτώβριος or Οκτώβρης, is indicative not only of their linguistic extremity but also of their support for communism.^[18]

Another aspect we considered was the linkage between the data in our database and the sources, either the indexing essay or the source periodical, as it was not feasible to give immediate access to full texts. In addition to the fact that many of the sources are not yet digitised or that it is not possible to provide direct access since the file consists of an entire issue or volume, allowing such access would demand considerable effort. Nevertheless, immediate access to the sources was deemed essential, partly because a few elements were not included and because cross-checking has always been a desideratum in research practice. Thus, whenever there is access to an electronic resource, the URL is embedded in the title of the periodical.

The templated record displays the data and field labels for three features, namely cover date, volume/issue, and translator. The original title follows the title as it appears on the page in a smaller typeface, and the translator’s pseudonym is distinguished using square brackets (see Figure 5).

Observations on How to Take Advantage of the Data

We need to provide specific numbers to obtain an idea of the size of our database and discuss some ways to make use of the data. As of July 2025, our spreadsheet has 22,000 records after indexing 146 prints (90 periodicals, 54 anthologies, and two supplements). This number is not insignificant considering that the database was launched in January 2024 and that all records have been added by hand (to ensure the quality of metadata) by a handful of volunteers. The amount of data constitutes only a small percentage of the translated literature published in Greek periodicals in the 19th and 20th centuries. Nevertheless, the diversity of the types of indexed periodicals and the inclusion of representatives of almost all periods^[19] allow some initial conclusions. On the other hand, their heterogeneity demands closer scrutiny when we seek answers to more complicated enquiries.

The identification of authors allows the application of geospatial information, which is captured on a map, offering a macroscopic picture of the translation dispersion and translation representation of countries. First, we must clarify that the geographic/national classification of a translation is based on an author’s nationality.^[20] There are cases wherein authors hold dual nationality, write in multiple languages, or have been affected by changes in nations and states throughout history, where the degree of complexity increases. At times, we have had to make several methodological compromises and risked inaccuracy since only one value (nationality/country affiliation) is permitted by the visualisation tool. By analysing the data at hand, we have reached some already-known conclusions: French literature occupies the biggest space in Greek periodicals, followed by literature in German, British, Russian, and Italian (Figure 6).

Several research projects dictate grouping translation into larger categories (Caristia, 2020, p. 183), but we have found that, in our case, even a minimal representation of a country’s literary tradition deserves to be highlighted. After the mid-20th century, and especially after the 1960s, the representation of works, mainly poetic, from South America (Colombia, Argentina, Chile, Peru, and Venezuela); Africa (including countries such as Madagascar, Senegal, and Algeria); or countries such as Haiti and Vietnam cannot stay unnoticed. The increased interest in the peripheral literatures can be interpreted through the lens of postcolonial theory (Bandia, 2016).

Another element where the results of the database can be combined to lead to fruitful conclusions is the translator’s presence. The translators’ choices as mediators are significant and are dictated by their ‘intellectual horizon’ (Espagne, 2016, p. 61). A macroscopic analysis can only focus on the fact that most of the translations are unassigned, a common practice for translations in the press justified by factors such as the type of magazine (less common in little magazines than in miscellanies), the genre (poetry translators tend to sign their creations whereas novels often do not come with a translator’s signature), the era, and the source text (especially when suspected plagiarism is involved).^[21] Nevertheless, a closer look and synthetic queries can serve dedicated research needs.

Conclusion

We hope this first brief and general presentation provides an idea of our database’s possibilities to a broader/international audience, despite the language barrier. Translations can enrich, empower, and even change the target language literary system. The collective data in Μετάdata may serve to determine whether, or at which point, Greek translations were a ‘channel through which fashionable repertoire [was] brought home’ or established a central position allowing agents ‘to violate the home conventions’ (Even-Zohar, 1990, p. 48, 50). In addition to this purpose, the data will allow researchers to look in reverse for the reception of foreign literature in Greece and estimate the dissemination of specific source literature beyond their borders and across time. The questions can be endless; we must keep feeding the need for carefully selected and organised data and, hopefully, map all Greek translations of literary texts exhaustively.

Acknowledgements

Many thanks are due to several people for the present paper. First, to Professors Stessi Athini and Titika Dimitroulia for their scientific support and overall supervision. We must also acknowledge the valuable contributions of Dr. Ioannis Tsakonas in the deployment of Μετάdata and its integration with the library portal and Dr. Kyriakos Stefanidis in the design and implementation of the database and visualisations. Finally, we would like to thank all the people working on data enrichment and, of course, the colleagues at ULiège Library for their endorsement.

References

Abstract

Translations scattered in periodicals and anthologies are valuable sources for understanding the shaping forces within the literary field of the target culture. However, while the translations attest to the dynamics of the source culture, their dispersion limits their validity. Μετάdata is a project that aspires to accumulate bibliographical data of all literary translations published in Greek periodicals and anthologies from 1801 to 1974 into one database. This article presents the steps we followed to implement the database and the various challenges we faced. The indexing relies mainly on secondary sources, and the fields used to describe the records have been customised to serve the database’s needs. We present the visualisations implemented to make the data appealing to users and offer some possibilities to maximise its use.

Keywords

Greek periodicals; Literary translations; Looker Studio; Data visualisation; Periodical indexes