(1) Overview
Stone carved inscriptions are a commonly surviving group of archaeological evidence. This paper introduces the public inscriptions from Palmyra, a corpus which consists of 1134 systematically recorded public inscriptions dating between the 1st century BC and the 4th century CE. These public inscriptions are just one part of the written record that survived from ancient Palmyra; however, it is these inscriptions that hold the most information on the socio-economic development of the city.
The ancient city of Palmyra, located in the Syrian desert, approximately 150 km east of the modern city of Homs (ancient Emesa), has been the focus of study by archaeologists, historians, and epigraphers since the 18th century. Large monumental structures, tombs, and funerary portraits have shaped modern insights into the ancient city and are the foundation of numerous studies of Palmyra’s inhabitants and the activities of daily life [1, 2].
Inscriptions were a prominent part of Palmyra’s visual culture, in the public, religious, and funerary spheres. They have been found across the city, inscribed and displayed on different objects, such as funerary reliefs, altars, consoles (Figure 1), column drums, door lintels, and architraves. Today many of them have lost their original contexts but can be assigned to a specific building or building type based on the medium and content of the inscription.
Figure 1
Column console discovered in the sanctuary of Allat. Fourteen lines of Greek are inscribed on the front face, followed by five lines on the left side, and then ten lines of Aramaic continuing after the Greek text (IGLS 127; Yon 2012).
The dataset presented in this paper compiles only the inscriptions that can be assigned to the public and religious sphere. These would have been visible to anyone in the city, and most frequently name the benefactor(s) of the monuments, honour citizens, or address religious, military or trade topics. Funerary inscriptions can be divided into two categories: those directly associated with a portrait inside the tomb, and founder and cession inscriptions, which name the founder of the tomb, their families, or indicate which part of the tomb had been sold or assigned for use by other families or extended kin. The former category of inscriptions on individual portraits primarily served to commemorate the deceased and provide information only about the paternal family line. As these inscriptions and portraits were made for the eyes of the immediate family in private familial tombs, they have been excluded from this study. Conversely, the latter category of inscriptions was included in the dataset of public inscriptions, as these were intended not only for the immediate family but also for passersby, and they provide insight into the socio-economic status of the families.
Not all inscriptions bear a date or can be dated with security. However, it was possible to assign dates to 557 inscriptions within a range of 25 years, using a combination of their context (e.g., association with a well-dated building) or content (e.g., explicitly mentioned date in c. 500 inscriptions). The earliest year mentioned in the inscriptions is 44 BC and the latest is 328 CE, with the majority of the inscriptions stemming from the time of Palmyra’s prosperity from the mid 1st to the end of the 3rd century CE.
While some of the recorded inscriptions are in situ [3] or stored on site [4], most can be found in museums world-wide. Especially moveable objects such as altars are part of museum collections in Europe [5] or America [6].
Spatial coverage
Description: Syria, Homs, Palmyra
Latitude: 34 33 12 N degrees minutes/34.5530 decimal degrees
Longitude: 038 16 05 E degrees minutes/38.2680 decimal degrees
Temporal coverage
App. 1st century BC – 4th century AD
(2) Methods
Data collection
Within the framework of the project “Circular economy and urban sustainability in Antiquity” [7], 1134 public inscriptions were compiled in an excel spreadsheet and saved as a CSV file. This dataset was compiled from the corpora of Aramaic inscriptions, Greek and Latin inscriptions and bilingual inscriptions [8, 9, 10]. In addition, the recently published inscriptions from the Temple of Allat were included [11]. When information was contradictory in the literature, the newest reading and dating were used, based on the most recent literature and corrections made by the authors of this article.
Corrections to the dating of the inscriptions have been undertaken based on an annotated index of dated Palmyrene Aramaic texts by Taylor [12]. Jean-Baptiste Yon reviewed the collected information (dates, readings, etc.) and, where necessary, corrected these. In cases of wide date spans, Yon narrowed them down, based on the content and context of the inscription.
The data collection included all inscriptions from the public sphere, meaning the public and religious buildings in the city, inscribed into the architectural elements (e.g. column drums) as well as objects (e.g altars). The founder and cession inscriptions from the necropoleis surrounding the city were also collected in the dataset. These inscriptions were prominently placed over the entrance of the tombs making them public inscription (for more explanation, see overview).
The aim of the data collection was to gain a complete overview of the public inscriptions in all three (Latin, Greek and Palmyrene Aramaic) languages, allowing for the study of the Palmyrene socio-cultural development through the lens of its inscriptions. This compilation of all data available from the entire set of known public inscriptions is the first attempt at creating a comprehensive file that allows for statistical analyses, bringing together centuries of research in one place and allowing others to use the dataset for further research.
The CSV file includes: a full overview of the dating of the inscriptions, the language, the medium, object type it was written on, the name of the commissioner(s) or commissioning group and their professions, the topic(s) of the inscription, as well as information on find context, original context, and secondary context (see Table 1).
Table 1
Short description of the data collected.
| VARIABLE | DESCRIPTION |
|---|---|
| ID | Individual unique ID assigned by the project |
| PAT_number | Unique ID assigned in Hillers and Cussini in 1996 |
| IGLS_number | Unique ID assigned in Yon in 2012 |
| CIS | Unique ID assigned in Corpus Inscriptionum Semiticarum |
| RES | Unique ID assigned in Répertoire d’épigraphie sémitique |
| Year_min | Earliest possible dating of the inscription; terminus post quem |
| Year_max | Latest possible dating of the inscription; terminus ante quem |
| Language | Language of the inscription (Palmyrene Aramaic, Greek, Latin); in case of multiple languages used within one inscription a compound assignation was used, e.g., GrPal (Greek-Palmyrene) |
| Medium | Stone type (e.g. limestone) |
| Made by | Inscription funded by a private individual/private group or by a public body (e.g. the council; tribe) |
| Indiv? | Inscription funded by an individual or a group |
| Founder | Name of the founder, group of founders, or public body |
| Topic_primary | Main topic of the inscription (e.g. honorific) |
| Topic_secondary | Secondary topic of the inscription, if applicable (e.g. military) |
| Building_type | Type of building or object inscription relates to (e.g. religious_building; religious_column) |
| Object_type | Type of object the inscription is placed on (e.g. statue console) |
| Original_context | Context the inscription securely relates to, either due to content of inscription or found insitu (e.g. Agora) |
| Latitude | Latitude of building the inscription relates to |
| Longitude | Longitude of building the inscription relates to |
| Insitu? | Information whether the inscription was found insitu or not |
| Insitu_context | Insitu context of the inscription |
| Secondary_context | Context the inscription was found or used in which is not the original placement |
| IndA – IndQ | Name of individual(s) named in the inscription |
| IndA_function – IndQ_function | Information on the individual(s) gained from the inscription (e.g. honoree; son; centurio) |
| Names | All names mentioned in the inscription in a list |
| Places | Places (e.g. cities) mentioned in the inscription |
| Gods | Deities mentioned in the inscription |
The data
The data mainly represents inscriptions from religious contexts (see also constraints discussed below), numbering 234 inscriptions. The three best represented religious contexts are the Temple of Baalshamin (86 inscriptions), the Temple of Bel (78 inscriptions), and the Temple of Allat (32 inscriptions). The second largest context are the founder and cession inscriptions, of which 129 have been recorded. From most tombs only one inscription has been preserved, with a few exceptions (e.g. Tomb of Malkû, 11 inscriptions; Tomb of Julius Aurelius Malê, 8 inscriptions). The third largest context is the agora with 123 inscriptions.
Palmyrene Aramaic? was the predominant language with a total of 677 inscriptions, while Greek was used in 223 inscriptions. Bilingual inscriptions in both Palmyrene and Greek were documented in 199 cases. Only 26 Latin inscriptions were documented and eight cases present trilingual inscriptions. Bilingual inscriptions in Greek/Latin and Latin/Palmyrene are only represented once each in the dataset. An analysis of the chronological distribution shows the prominence of each language at certain points in time (Figure 2).
Figure 2
Top: This plot displays 573 inscriptions (single dots) and their density over time. The date is based on each individual mean date. All inscriptions can be dated to within 25 years period (573 inscriptions, as the four Palmyrene/Greek/Latin inscriptions are not included). Bottom: Chronological distribution of languages dated to within 25 years period (573 inscriptions, as the four Palmyrene/Greek/Latin inscriptions are not included).
While a detailed analysis of the data lies beyond the scope of this data paper, the authors have explored this in a forthcoming article in the Journal of Urban Archaeology, which will be published later this year. [13].
Constraints
The first constraint in the dataset is the lack of information about original context, affecting 541 inscriptions, 47.7% of the entire dataset. In these cases, it is not possible to claim anything about the original intention or use of the inscription.
The second constraint is the limitation of dating. Only just over half of the inscriptions (577/50.9%) can be dated to within 25 years; the remaining inscriptions (557/49.1%) do not have characteristics to narrow their date to this degree of accuracy.
Third, as with most datasets pertaining to ancient cities, the original excavation was often not systematic, meaning data comes from a few dozen well-researched structures of Palmyra. The dataset is therefore biased through the targeted excavation and documentation of only a few contexts in the city. This bias is toward large representative public buildings in the centre of the city and away from domestic architecture.
(3) Dataset description
Object name
Public inscriptions from Palmyra
Data type
CSV file
Format names and versions
CSV files: Palmyra_Public_Inscriptions.xlsx + Palmyra_Public_Inscriptions_Names.xlsx
Creation dates
The inscriptions were published in corpora and monographs between 1889 and 2017. The compilation of the data from the literature was undertaken in various steps between 2020–2021. The analysis of the data has been undertaken in various steps between 2022–2023.
License
CC-BY 4.0
Repository location
Zenodo; DOI: 10.5281/zenodo.18934260
Publication date
10/03/2026
(4) Reuse Potential
This dataset makes Palmyrene epigraphic data available for the first time in a digitized form and can thus be utilized for further studies of the inscription practices in the public sphere.
Now available online, researchers can access and use the data for their research on Palmyra. The format of the data allows users to import data to programs such as Jupyter Notebook for further analysis. The format of the data allows for easy addition to, and correction of the corpus, in case of future publications of so far unpublished inscriptions or reinterpretations of the published inscriptions.
Acknowledgements
The authors thank everyone who was involved in the data collection within the Palmyra Portrait Project and Circular Economy and Urban Sustainability in Antiquity project.
Competing Interests
The authors have no competing interests to declare.