Pilot Testing the Aesthetic-to-Risk Framework of Nihilistic Violence Aesthetics (ART-NVA) on Publicly Coded Recruitment Communication Targeting or Grooming Youth
Year: 2026
Type: dataset
Abstract: Background: The Aesthetic-to-Risk Framework of Nihilistic Violence Aesthetics (ART-NVA) offers an interpretive vocabulary for public online violent communication. Its revised formulation specifies three inclusion criteria that jointly define construct membership, five conceptual dimensions coded as interpretive lenses, a community, platform, and genre scope condition that governs whether the dimensions may be applied at all, an eight-register formal layer, and three Weberian ideal types. None of this has been tested empirically. Objective: This pilot evaluates whether the construct gate, the five dimensions, the baseline-referenced coding rule, and the ordered type logic can be applied transparently and independently by two analysts to public coded recruitment or grooming communication associated with nihilistic violent-extremist networks, while testing retrieval, safety, burden, de-identification, and governance procedures. Methods: The target main corpus is 24 posts, eight each from Telegram, Reddit, and X, with a registered platform fallback of six posts per platform if all retrieval pathways are exhausted and only nine or ten eligible posts are available. Twelve main posts, four per platform, are coded independently by both authors; remaining main posts are divided equally. Six initial calibration posts are double-coded and excluded from analysis, with three preselected balanced reserves activated together only on a registered trigger. A frozen, countersigned Source Baseline Record is required for each source before any dimension may be coded Present. Criteria are coded Yes, No, or Unclear; dimensions are coded Present, Absent, or Unclear; every Yes and every Present requires a de-identified evidence anchor. Units receive T1, T2, T3, U-TYPE, or OUT-OF-SCOPE through a fixed ordered rule that the construct gate governs. Analyses are descriptive and cover gate behavior, codability, co-occurrence, register distribution, typology behavior, retrieval yield, burden, and preliminary agreement on the 12 overlap posts across nine variables. Conclusions: The pilot is designed to refine instruments and procedures and to decide whether a larger study is warranted. It will not estimate prevalence, classify persons or groups, predict violence, or claim validation from a small pilot sample.
Author: Landress, H.N.
Language: English
Subfield: Sociology and Political Science
Field: Social Sciences
Domain: Social Sciences
Open Access status: green
OpenAlex ID: W7203949907