alex67mtv/CatechismOfTheCatholicChurch

Dataset Card for Catechism of the Catholic Church (CCC)

Table of Contents

• —Dataset Description
• —Dataset Summary
• —Supported Tasks
• —Languages
• —Dataset Structure
• —Data Instances
• —Data Fields
• —Data Splits
• —Dataset Creation
• —Source Data
• —Limitations
• —Additional Information
• —Licensing Information
• —Citation Information
• —Contributions

Dataset Description

Source material: Catechism of the Catholic Church - Official Vatican Website

Dataset Summary

This dataset contains the full text of the Catechism of the Catholic Church (CCC) structured as a collection of articles. Each row corresponds to a single numbered article of the Catechism.

The Catechism of the Catholic Church, promulgated in 1992 under Pope John Paul II, is an official summary of the doctrines and teachings of the Catholic Church. It is organized into numbered paragraphs (articles) that systematically present Catholic beliefs, moral teachings, sacramental theology, and prayer.

In this dataset, each paragraph is represented as a separate entry, allowing for easy use in natural language processing tasks such as:

• —information retrieval
• —semantic search
• —question answering
• —religious studies research

The dataset preserves the article numbering used in the official Catechism.

Supported Tasks

The dataset can support several NLP tasks, including:

• —`text-retrieval` / `semantic-search` Models can retrieve relevant doctrinal passages in response to theological questions.

• —`question-answering` The articles may be used as a knowledge source for systems answering questions about Catholic doctrine.

• —`text-embedding` / `representation-learning` Each article can serve as a unit for embedding or semantic similarity tasks.

• —`topic-classification` Articles may be classified by doctrinal theme (e.g., sacraments, moral theology, prayer).

Languages

The dataset text is in English.

The original Catechism is available in several languages, but this dataset currently includes the English version.

Dataset Structure

Data Instances

Each instance corresponds to one numbered article of the Catechism.

Example:

{
  "text": "27 The desire for God is written in the human heart, because man is created by God and for God..."
}

Data Fields

• —`text`: string The official paragraph number in the Catechism of the Catholic Church followed the full text of the corresponding Catechism article. The paragraph number can easily be parsed by splitting the field on the first space character.

Data Splits

The dataset currently contains a single split:

The Catechism contains 2 865 numbered paragraphs.

No validation or test splits are provided since the dataset is primarily intended as a reference corpus rather than a supervised learning dataset.

Dataset Creation

This dataset was created to provide a structured, machine-readable version of the Catechism of the Catholic Church suitable for natural language processing and AI research.

The Catechism is widely used in theology, religious education, and academic research. Structuring it as a dataset allows developers and researchers to:

• —build theological search systems
• —experiment with retrieval-augmented generation (RAG)
• —perform linguistic or doctrinal analysis
• —create educational AI tools related to Catholic teaching

Source Data

Initial Data Collection and Normalization

The text originates from the official English edition of the Catechism of the Catholic Church, available through Vatican publications and archives.

During dataset creation:

• —the Catechism text was extracted from the official vatican website
• —paragraph numbering was preserved
• —footnotes were not included, but their reference weren't remove in the paragraphs
• —each numbered article was stored as a separate dataset row

Who are the source language producers?

The text was written by theologians and scholars commissioned by the Catholic Church under the authority of the Vatican during the drafting of the Catechism.

Limitations

• —The dataset does not include footnotes or references, which are abundant in the original text.
• —Some parsing artifacts may remain due to HTML extraction.
• —Paragraph-level segmentation may not always align perfectly with semantic boundaries for NLP tasks.
• —The dataset does not include labels for supervised learning tasks.
• —The dataset does not include the original sections.

Additional Information

Licensing Information

The text in this dataset is derived from the Catechism of the Catholic Church, originally published by Libreria Editrice Vaticana.

The dataset curator does not own the copyright to this material. The text is redistributed for research and educational purposes only. This dataset was created independently and is not affiliated with, endorsed by, or approved by the Vatican or Libreria Editrice Vaticana.

No permission or formal acknowledgment was obtained from the original publisher when compiling this dataset. Users of the dataset are responsible for ensuring that their use complies with applicable copyright laws and the licensing terms of the original publication.

Citation Information

If you use this dataset in research, please cite the original source:

Catechism of the Catholic Church. Libreria Editrice Vaticana, 1992.

Example BibTeX:

@book{catechism1992,
  title={Catechism of the Catholic Church},
  author={{Catholic Church}},
  year={1992},
  publisher={Libreria Editrice Vaticana}
}

Contributions

Contributions, corrections, and improvements are welcome via pull requests on the dataset repository.