Introduction

This repository seeks to document proper names in various Bible translations aligned by verse reference.

What this data is

ref	text	macula_lemma	macula_eng	hcb	cu2010s	cu2010t	rso94	sruv06	tbi	lsg10	rv09	tpb08	bbc
LUK 1:3!11	Θεόφιλε	Θεόφιλος	Theophilus	Tiyofilus	提阿非罗	提阿非罗	Феофил	Theofilo	Teofilus	Théophile	Teófilo
LUK 1:5!5	Ἡρῴδου	Ἡρῴδης	Herod	Hiridus	希律	希律	Ирода	herode	herodes	Hérode	Herodes	herot	Herodes
LUK 1:5!8	Ἰουδαίας	Ἰουδαία	Judea	Yahudiya	犹太	猶太	Иудейского	Yudea	yudea	Judée	Judea	judia	Judea
LUK 1:5!12	Ζαχαρίας	Ζαχαρίας	Zechariah	Zakariya	撒迦利亚	撒迦利亞	Захария	Zakaria	zakharia	Zacharie	Zacarías	sekaraia	Sakarias
LUK 1:5!15	Ἀβιά	Ἀβιά	Abijah	Abiya	亚比雅	亞比雅	Авиевой	Abiya	abia	Abia	Abías	abiya	Abia

Languages currently included:

• ARBNAV: Arabic (NAV)
• BBC: Toba
• BSB: English
• CU2010S: Chinese Simplified
• CU2010T: Chinese Traditional
• ESVUK16: English
• ENGULB: English
• HCB: Hausa
• LSG10: French
• RSO94: Russian
• RV09: Spanish
• SRUV06: Swahili
• TBI: Indonesian
• TGLULB: Tagalog
• TPB08: Tok Pisin

Why this is useful

...

Data Sources

This data uses Macula Greek and Hebrew combined with alignments provided by UBS and generated using eflomal. Note that stopwords are used to filter out poor alignments. The names.tsv file pulls together these sources to produce a list of names in the Bible to support Bible translation work.

Usage

Extraction Helper

The scripts/extract_name_pairs.py script exports columns of name pairs, dropping any rows with empty values.

python scripts/extract_name_pairs.py [-h] [-l] [column_one] [column_two]

Extract name pairs from a TSV file

positional arguments:
  column_one  Name of the first column to extract
  column_two  Name of the second column to extract

options:
  -h, --help  show this help message and exit
  -l, --list  List available columns

Note: You are expected to run python from the project root so that the relative path to ./names.tsv is valid.

Sqlite

Using sqlite, you can create a list of mappings from one language to another for a given langauge pair.

For example, if you wanted the TBI-BBC mappings for Luke, open sqlite3:

.mode tabs
.import names.tsv names

This will load the tsv file into a names table. To get a table like the one above in order to explore the data, you could try something like:

SELECT * FROM names WHERE ref LIKE "LUK 1:%";

A more interesting/useful query would be one that groups equivalents across a language pair. For example, if we wanted to map TBI proper nouns to BBC translations and count the references, we could do something like this (I've included a commented column that will list every verse in which the name appears):

SELECT
    tbi,
    GROUP_CONCAT(DISTINCT bbc) FILTER (WHERE bbc <> "") AS bbc,
    GROUP_CONCAT(DISTINCT macula_eng) FILTER (WHERE macula_eng <> "") AS macula_eng,
    COUNT(ref) AS refs
    -- GROUP_CONCAT(DISTINCT ref) AS refs
FROM
    names
WHERE
    ref LIKE "LUK%" AND
    tbi <> ""
GROUP BY
    tbi
ORDER BY
    tbi;

Note: If you use a filter like tbi <> "", it may be wise to also check what names map to an empty slot for tbi. The reason the filter is there is that the GROUP_CONCAT function can produce a long line which makes the text much less readable on the command line.