Principle
Dil Dergisi attaches importance to the verifiability and reproducibility of linguistic claims. The journal supports the open sharing of research data in line with the FAIR principles and aims for data to be opened not only for verification but for reuse in new research. Data citation makes claims easier to verify and data easier to use in further work.
Coverage
Research data here includes corpora and corpus subsets; audio, video and eye tracking recordings; phonetic measurements; annotated and parsed data; grammaticality and acceptability judgement data; experimental stimuli, instructions and raw response logs; survey and interview transcripts; lexicon and gloss files; statistical scripts and analysis code.
Repositories
Data should be deposited in a trusted public repository that issues a persistent identifier. For linguistic data and statistical code the journal recommends TROLLing, the Tromsø Repository of Language and Linguistics. TROLLing is a FAIR aligned, fully open access repository in which every dataset carries searchable metadata identifying the researchers, the languages and linguistic phenomena involved, the statistical methods applied and the publications based on the data. It is recommended by leading journals in the field.
OSF and Zenodo may be used for general project based sharing, IRIS for second language acquisition instruments, TalkBank and CHILDES for acquisition and interaction data, ELAR, PARADISEC and AILLA for documentation and endangered language data, and CLARIN for European language resources.
Personal websites, institutional servers, cloud sharing links and article supplementary files are not substitutes for a persistent repository.
Data licensing and reuse
Reusability is the basis of the journal’s open science policy. Authors are encouraged to license datasets and code under CC BY 4.0 or CC0 1.0. The CC BY-NC-ND 4.0 licence applied to the article text does not extend to the dataset and a separate, more open licence may be applied to data. The no derivatives condition applies to the text alone and does not restrict the processing, combination or reanalysis of data.
Data citation
Datasets used or produced are cited in full in the reference list with a DOI and are referred to in the text as any other source. The journal follows the Austin Principles of Data Citation in Linguistics, which adapt the FORCE11 Joint Declaration of Data Citation Principles for linguistic scholarship, and the Tromsø recommendations for citation of research data in linguistics, which are based on those principles. Studies drawing on data produced by others are obliged to credit the data creators.
The reference format is as follows:
Surname, A., & Surname, B. (2026). Title of the dataset (Version 1.0) [Data set]. TROLLing. https://doi.org/10.18710/XXXXXX
Data Availability Statement
Every article must complete the Data Availability Statement section of the template, using one of the four forms below.
Where data are openly available:
The data and analysis scripts for this study are openly available in TROLLing under a CC BY 4.0 licence: https://doi.org/…
Where access is restricted:
The data are under restricted access for reasons of participant confidentiality. Metadata are publicly available at https://doi.org/… and access requests are assessed by the relevant committee.
Where an existing data source was used:
This study generated no new data. The corpus used is available at https://doi.org/…
Where data are shared on request:
The data are available from the corresponding author upon reasonable request.
The last option is accepted only where ethical or legal constraints apply and the reason must be stated.
Access during review
Authors are encouraged to provide reviewers with an anonymised, view only link to the data. OSF and Dataverse based repositories support this. The link must not contain author information that would compromise blind review.
Ethical limits
Openness does not override the rights of participants or communities. Data containing personal information, permitting re identification, or lacking the consent of the speaker community are not shared openly. In such cases a deposit form with public metadata and access controlled data should be used. For endangered language and fieldwork data, community consent and local access conditions must be observed.
Code and reproducibility
Where a study involves statistical analysis, scripts should be shared together with software and package version information. Files should be provided in open, durable formats such as UTF-8 plain text, CSV, TextGrid, ELAN eaf, XML or CoNLL-U.
Role in review
Data sharing is not a condition of submission. Reviewers are asked to assess whether the claims of the study can be verified against the data presented.