Corpus of Portuguese Historical and Metalinguistic Grammars (1536–1864)

Description This dataset presents a Corpus of Portuguese Historical Metalinguistic Grammars (1536–1864), a machine-readable collection of six foundational normative works published between the 16th and 19th centuries. The corpus serves as a primary source for research in Diachronic Linguistics, specifically regarding the "Competition of Grammars" and the evolution of normative syntax in European and Brazilian Portuguese. Files Included in this Dataset The dataset is distributed as a ZIP file containing the following components: portuguese_grammars.db: The core relational database containing the full text and metadata. /facsimiles: A folder containing 6 PDF files with embedded ATR text layers. /raw_txt: A folder containing raw plain text exports for non-database usage. These are digital surrogates of the original prints, named by [BirthYear]_[AuthorName].pdf. Database Structure (Variables) The following variables correspond to the columns in the main grammars table of the SQLite database (portuguese_grammars.db): id: Unique identification of each record, assigned chronologically based on the author's birth year. author_name: Name of the grammarian responsible for the work. author_birth_year: Year of birth of the author, used for generational sorting and diachronic analysis. title: Full title of the grammatical work as cataloged. publication_year: Year in which the specific edition used in this corpus was published. text_content: Full, machine-readable text of the grammar. txt_filename: Filename of the corresponding raw text export. pdf_filename: Filename of the digital surrogate (facsimile) associated with the record. Methodology & Transcription Criteria Transcription & ATR: Text recognition was performed on the Transkribus platform using the "Early Portuguese Printing" (EPP) model (released Oct 2024). Curation: The text underwent manual curation to verify layout segmentation and content integrity. Despite these efforts, the corpus may contain residual ATR errors. Users are encouraged to report errors to the corresponding author at: [e-mail]. Editorial Principles: For a detailed description of the paleographic transcription rules and editorial principles applied to this corpus, please refer to the article: PACHECO ROCHA, S. R. A Corpus of Portuguese Historical and Metalinguistic Grammars (1536–1864). Journal of Open Humanities Data, v. 12, n. 65, p. 1-8, 2026. DOI: https://doi.org/10.5334/johd.542. Authors and Works Included João de Barros (1496–1570): Grammatica da Lingua Portuguesa (1540) Fernão de Oliveira (1507–1581): Grammatica da Lingoagem Portuguesa (1536) Duarte Nunes de Leão (1530–1608): Origem da Lingoa Portuguesa (1606) Luís Caetano de Lima (1671–1757): Orthographia da Lingua Portugueza (1736) Bernardo de Lima e Melo Bacellar (1736–?): Grammatica Philosophica (1783) Francisco Júlio Caldas Aulete (1823–1878): Grammatica Nacional (1864) Source of Facsimiles The digital surrogates (PDFs) used in this corpus were obtained from the digital collection of the National Library of Portugal (Biblioteca Nacional de Portugal - BNP). We gratefully acknowledge their work in digitizing and making these public domain materials accessible. License This dataset is dedicated to the public domain under the Creative Commons Zero (CC0) license. Citation If you use this corpus, please cite the associated Data Paper/Dissertation: Pacheco Rocha, Saulo Rogério. 2025. O Português das “Obras Proueitosas” e dos “Ilustres & Muito Reuerendos Senhores”: uma análise da mudança nos padrões de colocação pronominal clítica em gramáticas portuguesas do século 15 ao 19 e seus autores. Florianópolis: Dissertação de Mestrado da Universidade Federal de Santa Catarina. Available at: https://repositorio.ufsc.br/handle/123456789/265186