The AUGMANITAI compendium is situated within a constructivist-descriptive paradigm that foregrounds the contingency of terminological stabilization in emergent knowledge domains. Following the foundational insight that concept systems are never merely discovered but always co-constituted through the epistemic practices that render them legible (Wüster, 1979; Cabré, 1999; Temmerman, 2000), the compendium adopts an orientation in which terminological adequacy is understood as a dynamic property rather than a static correspondence. The ontological commitment remains deliberately minimal: the compendium describes phenomenal surfaces as they present themselves within the relevant discourse horizon, without presupposing reductive explanatory models.
This positioning entails that the compendium does not advance causal claims about the phenomena it names. Each term captures a describable regularity at the level of observable interaction, acknowledging that the same regularity may admit multiple competing theoretical framings. The epistemic modesty of this approach is not a limitation but a design decision rooted in the recognition that premature theoretical closure in a rapidly evolving field risks producing artifacts that constrain rather than enable subsequent inquiry.
The identification and individuation of terminological units follows from the sustained engagement with the domain under investigation. Candidate concepts are recognized through a process of progressive conceptual differentiation: an initially undifferentiated discursive field is iteratively resolved into distinct, nameable units by attending to the points at which semantic boundaries become salient against the background of accumulated domain expertise. This process is fundamentally hermeneutic in character — it presupposes, and simultaneously refines, a pre-understanding of the conceptual landscape that cannot be reduced to a finite set of procedural steps.
The author's approach is informed by what may be characterized as sustained immersive domain engagement: an extended, high-intensity interaction with the phenomena under investigation that permits the recognition of recurring patterns, anomalous configurations, and conceptual lacunae that would remain invisible to more superficial or intermittent modes of inquiry. The depth of this engagement constitutes the primary epistemic resource of the project and is, by its nature, not fully formalizable.
Terminological individuation in emergent domains resists algorithmic reduction. The capacity to perceive a conceptual boundary where none has previously been drawn is an exercise of trained judgment operating at the intersection of domain knowledge, linguistic sensitivity, and pattern recognition cultivated through sustained practice.
Each individuated term is assigned to one of 43 thematic domains reflecting the internal structure of the conceptual field as it has emerged through the research process. Domain boundaries are understood as heuristic rather than ontological — they organize the terminological space for navigational purposes without claiming that the underlying phenomena respect these divisions. Cross-domain membership is acknowledged through the compendium's relational infrastructure (cf. Section 6).
A secondary classification axis assigns each term a confidence level reflecting the epistemic status of the described phenomenon: Documented (D) for phenomena attested in existing literature or established practice, Inferred (I) for phenomena derivable from documented observations through principled reasoning, and Predicted (P) for phenomena extrapolated from observable trajectories. This tripartite scheme adapts established epistemological distinctions (cf. Popper, 1963; Lakatos, 1978) for a domain in which many phenomena precede their formal description.
Each term is formalized bilingually (English/German) following principles inspired by ISO 704, ISO 1087, and ISO 30042. The bilingual architecture addresses the observation that certain phenomena in the domain exhibit language-dependent conceptual valence: the German formulation may foreground aspects that the English rendering backgrounds, and vice versa. This is not merely a translational exercise but a substantive epistemological commitment to the view that conceptual content is partially constituted by the linguistic resources through which it is articulated (cf. Sapir, 1929; Whorf, 1956; Lucy, 1992). Where no established German equivalent exists, the English term is retained with a descriptive German gloss that preserves the relevant semantic dimensions.
In accordance with the compendium's transparency provisions (§12), the research process incorporates contemporary computational methods as epistemic instruments. These instruments serve an augmentative function: they extend the scope of inquiry beyond what would be achievable through unaided human cognition alone, without displacing the human judgment that governs all classificatory and definitional decisions. The precise configuration, sequencing, and parametric calibration of these augmentative processes constitute proprietary methodological knowledge developed over the course of the research program and are not disclosed in this document.
The author retains full editorial authority over term selection, definition formulation, classification assignment, and network topology. No terminological unit enters the published compendium without having passed through the author's evaluative framework. The augmentative components accelerate certain phases of the workflow but do not determine its outputs.
The compendium maintains a relational network in which terminological units are connected through typed associations reflecting their conceptual proximity, complementarity, or opposition. The topology of this network is an emergent property of the terminological individuation process itself: as each new term is situated within the existing conceptual space, its relational profile becomes apparent through the same hermeneutic engagement that governs its individuation. The resulting graph structure serves both navigational and analytical functions, revealing clustering patterns and conceptual neighborhoods that may not be apparent from the examination of individual terms in isolation.
Published terms pass through a multi-gate quality assurance pipeline that evaluates connotative safety, audience appropriateness, privacy compliance, medical and legal claim avoidance, and regulatory alignment. Terms that fail any gate are quarantined for review or permanently excluded. The pipeline operates with a zero-tolerance policy for content that could be misinterpreted outside its academic context. Gate specifications are maintained internally and updated with each compendium version.
The compendium is DOI-registered through Zenodo, ensuring persistent identification and long-term citability. Each major version is archived independently. Machine-readable exports are provided in multiple formats (JSON-LD/Schema.org, SKOS/RDF, Turtle, CSV, BibTeX, RIS, JSONL) to facilitate integration with existing knowledge infrastructure, bibliographic systems, and computational research workflows. The multi-format export strategy reflects a commitment to interoperability across heterogeneous scholarly and technical ecosystems.
This methodology reflects a single-author research program operating within the conventions of applied terminology science. Terms are not subject to external peer review in the traditional sense, and the compendium does not claim consensus status. The confidence classification system relies on the author's assessment of the available evidence base. Users are encouraged to evaluate each term independently against their own domain expertise and to consult the cited literature where applicable. The compendium is explicitly descriptive and makes no prescriptive, therapeutic, diagnostic, or regulatory claims (cf. §1, §3, §4, §5, §8, §9, §10).
Terminological suggestions, corrections, and domain expertise contributions are welcome. Contact: augmanitai [at] proton [dot] me. All contributions are reviewed within the established evaluative framework and credited in the changelog if incorporated.
Andreas Ehstand · Nepomukweg 7, 82319 Starnberg, Deutschland
ORCID: 0009-0006-3773-7796 · DOI: 10.5281/zenodo.14888381
Datenschutzerklärung · Term Index
© 2026 Andreas Ehstand. CC BY-NC-ND 4.0 International.