Vers une analyse automatique du discours en histoire de l’architecture
Bibliographic record
Abstract
This contribution examines the methodological potential of automated discourse analysis in architectural history. Language, a fundamental component of architectural practice, can now be approached through digital methods derived from corpus linguistics and natural language processing. Starting with a historiographical overview of the relationship between architecture and language since the 1960s, the paper shows how figures such as Summerson, Zevi, and Jencks conceptualized architecture as a linguistic system. Yet, despite these early insights, few studies have applied computational methods to the analysis of architectural texts. The pioneering work of Alexander Tzonis in the 1970s on “conceptual systems” in French architectural texts laid the groundwork for the computerized study of architectural discourse. Today, the widespread digitization of sources and major technological advances make it possible to assemble large textual corpora suitable for automated discourse analysis. Approaches developed in the digital humanities – such as textometry, topic modeling, argument analysis, and the use of large language models – have proven particularly fruitful. They have opened up a promising interdisciplinary field, offering new ways to trace the evolution of deontic discourse and theoretical conceptions of architecture, as well as the formulation of aesthetic judgments and the documentation of the reception of architectural works.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.006 | 0.025 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.009 | 0.006 |
| Science and technology studies | 0.002 | 0.003 |
| Scholarly communication | 0.013 | 0.007 |
| Open science | 0.002 | 0.002 |
| Research integrity | 0.002 | 0.005 |
| Insufficient payload (model declined to judge) | 0.019 | 0.010 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".