← The findings  ·  Research tools  ·  johnholbein.org

Christ Mentions · replication

The data and the code

Everything behind the findings: the coded corpora, the coding rule, every derived table, and the scripts that turn one into the other. 85.1 MB in all. No account, no request, no embargo.

Start here. Download code.zip, put the scripts beside an output/ directory holding the corpora below, and run them in numeric order. 00_setup.R sets the seed; every figure and table regenerates from the corpora without touching the network. To rebuild from the original sources instead, run 01_download.R first — it fetches each text from its published home.

Run order

ScriptWhat it does
00_setup.RPaths, fetch helpers, text cleaning, the random seed (20260903).
01_download.R – 30_*.RDownload every source and build one row per verse.
10_christ_lexicon.RThe codebook as regex: 169 patterns in 38 categories, the false-positive list, the frame restrictions.
11_classify.RApplies the lexicon, the speech spans and the pronoun rules; emits the three nested indicators.
12_book_names.R, 13_canon_membership.RCanonical order, testament and section labels, and which communities hold each book.
40_*.R – 45_*.RSummary tables, the canon membership table, worked examples, the cumulative series.
69_analysis_common.RShared theme, palettes and statistics for everything below.
70_*.R – 77_*.RThe twenty-two analyses, families A to H, and the two hand audits.
80_*.py – 82_*.pyBuild the findings page, the shareable copies, and the deploy.
90_*.py – 97_*.pyThe walkthrough video, the link card, this release.

Files

Sizes are the download. Gzipped CSVs open directly in R (fread("file.csv.gz")), Python (pd.read_csv) and Stata 18+.

The passages

FileRowsSizeWhat it is
jesus_passages_all.csv.gz157,0498.6 MBEvery verse the coder marks as referring to Jesus Christ, across all 67 editions. One row per verse per edition.
jesus_passages_one_edition.csv16,2165.4 MBThe same passages restricted to one English edition per work -- the set the passage browser shows.
jesus_passages_browse.json16,2164.5 MBThe passage browser's payload, column-oriented JSON.

Reference tables

FileRowsSizeWhat it is
reference/00_MASTER_INDEX.csv678 KBEvery edition: books, chapters, verses, mentions, rate.
reference/00_LEXICON.csv1697 KBThe coding lexicon: 169 patterns in 38 categories.
reference/00_CANON_MEMBERSHIP.csv93875 KBAll 938 books mapped to the communities that hold them as scripture.
reference/00_EXAMPLES.csv426 KBOne named, titled and speaking example per work.
reference/00_VERSES_BY_BOOK.csv90182 KBBook-level verse and mention counts, one edition per work.
reference/00_VERSE_ACCOUNTING.csv675 KBHow the 1.22 million coded rows divide across works and editions.
reference/00_CUMULATIVE_SERIES.csv41,9952.5 MBPer-verse cumulative counts behind the animated sweep.
reference/00_TRADITION_SUMMARY.csv11347 BTotals by religious tradition.

Code

FileRowsSizeWhat it is
code.zip142 KBEvery script in the pipeline, in run order.

Coded corpora

FileRowsSizeWhat it is
corpora/01_LDS_standard_works.csv.gz2.3 MBCoded corpus, one row per verse, gzipped CSV.
corpora/03_Quran_quranacademy.csv.gz339 KBCoded corpus, one row per verse, gzipped CSV.
corpora/04_Quran_abdullahyusufal.csv.gz377 KBCoded corpus, one row per verse, gzipped CSV.
corpora/05_Quran_mohammedmarmadu.csv.gz335 KBCoded corpus, one row per verse, gzipped CSV.
corpora/06_Quran_mohammadhabibsh.csv.gz325 KBCoded corpus, one row per verse, gzipped CSV.
corpora/07_Quran_ajarberry.csv.gz309 KBCoded corpus, one row per verse, gzipped CSV.
corpora/08_Quran_muhammadtaqiudd.csv.gz421 KBCoded corpus, one row per verse, gzipped CSV.
corpora/09_Quran_mustafakhattaba.csv.gz349 KBCoded corpus, one row per verse, gzipped CSV.
corpora/10_Quran_abdelhaleem.csv.gz327 KBCoded corpus, one row per verse, gzipped CSV.
corpora/11_Quran_muhammadasad.csv.gz427 KBCoded corpus, one row per verse, gzipped CSV.
corpora/12_Quran_georgesale.csv.gz357 KBCoded corpus, one row per verse, gzipped CSV.
corpora/13_Quran_johnmedowsrodwe.csv.gz345 KBCoded corpus, one row per verse, gzipped CSV.
corpora/14_Hadith_bukhari.csv.gz1.3 MBCoded corpus, one row per verse, gzipped CSV.
corpora/15_Hadith_muslim.csv.gz911 KBCoded corpus, one row per verse, gzipped CSV.
corpora/16_Hadith_abudawud.csv.gz669 KBCoded corpus, one row per verse, gzipped CSV.
corpora/17_Hadith_tirmidhi.csv.gz485 KBCoded corpus, one row per verse, gzipped CSV.
corpora/18_Hadith_nasai.csv.gz542 KBCoded corpus, one row per verse, gzipped CSV.
corpora/19_Hadith_ibnmajah.csv.gz439 KBCoded corpus, one row per verse, gzipped CSV.
corpora/20_Hadith_malik.csv.gz320 KBCoded corpus, one row per verse, gzipped CSV.
corpora/21_Hadith_nawawi.csv.gz10 KBCoded corpus, one row per verse, gzipped CSV.
corpora/22_Hadith_qudsi.csv.gz11 KBCoded corpus, one row per verse, gzipped CSV.
corpora/23_Tanakh_EN.csv.gz1.2 MBCoded corpus, one row per verse, gzipped CSV.
corpora/24_Tanakh_HE.csv.gz1.6 MBCoded corpus, one row per verse, gzipped CSV.
corpora/25_Mishnah.csv.gz654 KBCoded corpus, one row per verse, gzipped CSV.
corpora/26_Bhagavad_Gita.csv.gz45 KBCoded corpus, one row per verse, gzipped CSV.
corpora/27_Guru_Granth_Sahib.csv.gz2.1 MBCoded corpus, one row per verse, gzipped CSV.
corpora/28_Dhammapada.csv.gz24 KBCoded corpus, one row per verse, gzipped CSV.
corpora/29_Tao_Te_Ching.csv.gz19 KBCoded corpus, one row per verse, gzipped CSV.
corpora/30_Analects.csv.gz60 KBCoded corpus, one row per verse, gzipped CSV.
corpora/31_1_Enoch.csv.gz66 KBCoded corpus, one row per verse, gzipped CSV.
corpora/32_Kojiki.csv.gz95 KBCoded corpus, one row per verse, gzipped CSV.
corpora/bible/47.4 MBThe Bible corpus as 36 editions, one gzipped CSV each.

Provenance and terms

Each text came from a public source, named in 01_download.R and in the edition column of every corpus file: scrollmapper's Bible database, Tanzil and the Quranic Arabic Corpus for the Qur'an, Sunnah.com for the hadith, Sefaria for the Tanakh and Mishnah, Project Gutenberg and the Internet Sacred Text Archive for the rest. Public-domain translations throughout, except the Mishnah (Joshua Kulp, CC-BY) and where an edition's own licence is recorded in the master index.

The coding, the scripts and the derived tables are yours to use; please cite the project and, where an edition carries its own licence, that edition too. Corrections are welcome and will be recorded rather than quietly patched — the findings page carries the seven false positives already fixed this way.

These are textual-reference rates. They are not evidence of belief, worship, or doctrinal centrality, and they should not be reported as such.