Downloads
The Accented Scrivener's 1894 Textus Receptus
In 2009 there was no accented copy of Scrivener's 1894 Textus Receptus in the public domain, so one was made here; from 2011 it was offered on this site and copied widely online. In 2026 every word of it was checked against the Cambridge printing itself. The Downloads page shows what changed and why the smallest mark matters: every Greek letter is a number, so every verse has a weight — and James 1:6 shows what one movable letter and two iota subscripts do to it.
Where to find it
- Strong's Dictionary
- Webster's Search
- Downloads
Sidebar → Library → Downloads
Downloads · Scrivener's 1894 Textus Receptus
The TR-1894,
read from the printed page.
A new accented Greek New Testament, checked word for word against the Cambridge printing of 1894, with every word tagged. It replaces the text I made available in 2011.
No account yet? Sign up free
- 2011
- οὐρανῳ
- 1894
- οὐρανῷ
- 2011
- τὴς γὴς
- 1894
- τῆς γῆς
Small marks, but they are the text.
2,979
words corrected
in 2,183 verses of the 2011 text
140,768
printed words tagged
each with Strong's number, morphology and sense
684
printed pages read
in two scanned copies of the 1894 printing
Where it came from
2009
In 2009 I could not find an accented copy of Scrivener's 1894 text in the public domain, so I made one. Its letters were checked against Alan Bunning's collation of the electronic Scrivener texts, and against the printed page where they differed. Its accents were filled in mostly by comparison with the accented Stephanus 1550, and where the 1894 differs from the 1550, those accents could not be checked.
That text, offered here from 2011, has been copied widely online. In 2026 I set out to check every word of it against the printing itself. This is the result.
Why it matters
Every letter is a number.
Each Greek letter has a numerical value, and so each verse has a weight: the sum of its letters. For the structure behind the text, that weight is like a checksum. You may not know when it is wrong, but you can know when it is right, because the right weight carries a coherent thought.
That is why the text has to be exact, down to the smallest mark. An accurate text brings an accuracy to the structural design behind it, a design no man could have known.
Take James 1:6. Two small things decide its weight: the iota subscript written under two of its words (ῳ, worth 10 each), and the movable ν that most texts add to ἔοικε (worth 50). Counted four ways, the verse gives four weights.
James 1:6
But let him ask in faith, nothing wavering. For he that wavereth is like a wave of the sea driven with the wind and tossed.
αἰτείτω δὲ ἐν πίστει, μηδὲν διακρινόμενος· ὁ γὰρ διακρινόμενος ἔοικε(ν) κλύδωνι θαλάσσης ἀνεμιζομένῳ καὶ ῥιπιζομένῳ.
- 8,040 with the ν, without the subscripts H8040 the left hand
- 8,060 with the ν and the subscripts H8060 Shammai
- 8,010 without the ν, with the subscripts H8010 Solomon, “peaceful”As printed
- 7,990 without the ν or the subscripts H7990 ruler; lawful (Aramaic)
Which weight carries the thought? Read the verse before it:
James 1:5
If any of you lack wisdom, let him ask of God, that giveth to all men liberally, and upbraideth not; and it shall be given him.
Solomon asked God for wisdom (2 Chr 1:10), and it was granted him (2 Chr 1:12).
This connection was found years ago by structural reasoning, and the printed text confirmed 8,010 mark for mark: ἔοικε without the ν, and both subscripts. It is a plain case for why an accented text has to be exact.
Count on from Solomon
- 19,068 verses From Solomon's request for wisdom, 2 Chr 1:10, to James 1:6.
- 19,068 Webster's In the public-domain Word Index of Webster's 1828, Volume 2 word #19,068 is SUN.
- 19,068 verses After 1 Kings 16:34 stands 1 Cor 1:22: “For the Jews require a sign, and the Greeks seek after wisdom.”
- 1,887 verses From 1 Cor 1:22 to James 1:6.
What was corrected
2,979 words now follow the printing.
Measured against the 2011 text, in 2,183 verses. Most are the marks that were weakest in 2011: the accents, and the printing's own punctuation.
By kind
- Accents and breathings 1,531
- Punctuation 1,274
Including 299 stops missing at verse ends.
- Parentheses the printing sets 81
- Spelling 58
- A word and its punctuation together 22
- Word division 9
- Movable ν 4
Before and after
Matt 8:22
εῖπεν → εἶπεν
The circumflex, as printed.
Mark 12:5
τοὓς → τοὺς
A slip in the 2011 text.
2 Tim 2:19
Χριστού → Χριστοῦ
The circumflex, as printed.
Rev 2:21
εκ’ → ἐκ
ἐκ is never elided.
Rom 14:8
ἀποθνῄσκωμεν → ἀποθνήσκωμεν
The 2011 text spelt the word two ways in one verse.
83 places
Ἰερουσαλήμ → Ἱερουσαλήμ
The rough breathing everywhere, as printed; ten had the smooth.
The verse weights
What the corrections changed in the numbers
Most of the corrections are accents and punctuation, which carry no value. But an iota subscript is worth 10, and a letter is a letter, so I owe a plain account of what changed. Of the 7,957 verses of the New Testament, the weight of 65 changed; the other 7,892 are exactly as they were. (A colophon is weighed with its epistle's last verse.)
- 62 An iota subscript. In 24 verses the printing sets a subscript my text lacked, mostly datives such as ἡμέρᾳ and τῷ. In 38 it leaves out one my text had, most of them whole families the 1894 printing sets plain: ἀποθνήσκω, μιμνήσκω, σώζω, εἰκῆ; and three colophons, where the 1887 printing has δευτέρα.
- 2 A letter. Eph 1:7 τὸ → τὸν (+50), the only letter in the verses where my text and the 1894 printing differ; and the colophon of 1 Timothy, ἐστίν → ἐστὶ (−50), as the 1887 printing has it.
- 1 My own arithmetic. Rev 21:17 (−290): a merged cell in my workbook added the wrong pair of words. The text was right; the sum was not. In every other verse my weights matched my text.
How it was read
The printing is the standard.
The Cambridge University Press printing of 1894, in two scanned copies: Google Books (University of Michigan) and archive.org (UCLA, scanned at 500 ppi). Same book. Two resources. Two different scans.
- 1
Two independent witnesses
Compared word by word with a modern typeset edition and with Maurice Robinson's unaccented Scrivener. Every difference was read on the printed page.
- 2
A reading of every page
Agreement was not taken as proof. All 684 pages were read by OCR independant of AI which would have seen my text, and every word where it differed was verified on the page image.
- 3
A second look
Small marks were re-read enlarged, hard words on the full printed line, parentheses in a pass of their own, and the remaining questions that software analysis couldn't resolve I decided myself.
As printed, flaws included
Two missing tittles.
The text follows the printing exactly, its own accent conventions included, such as an acute on a final syllable before a comma (πολλοί,). Where the type itself failed, the failure is kept: twice the relative pronoun is printed bare, without breathing or accent, in both copies.
“One jot or one tittle shall in no wise pass” (Matt 5:18). Five other places where the printing is irregular are kept too, each with a note in the tagged file.
Matt 20:23
τὸ βάπτισμα ο ἐγὼ βαπτίζομαι
Two lines above, in verse 22, the same word is printed ὃ.
Matt 21:44
ἐφ’ ον δ’ ἂν πέσῃ
In verse 42 the same word is printed ὃν.
What you get
One text, four ways to use it.
All with the KJV verse division and the 14 colophons at the end of their books. The 2011 files, including the Excel file, are retired.
- XML TR-1894 OSIS, tagged Every printed word with its number in the verse, Strong's number, dictionary form, morphology, transliteration and sense. It replaces the 2011 Excel file.
- KEY Morphology key What every morphology code means, as a page to read or print and as a CSV of all 979 codes.
- ZIP SWORD module For Bible software that reads SWORD modules. The same module drives the site's Verse Analysis interlinear.
- XML Plain OSIS, accented and unaccented The Greek alone, without word tags.
The files and their formats are described on the TR-1894 downloads page.
How to get a copy
Free, by request.
- 1
Sign in
Or create a free account.
- 2
Complete your profile
Your name, address and phone number.
- 3
Send a request
A few words on how you found this text and how you will use it.
- 4
Download
I read every request myself. You will get an email when I answer, and the button takes you to the files.
No account yet? Sign up free
Other resources will be offered here as they are ready. — Wade Balzer
What you can do here
2,979 words corrected
In 2,183 verses of the 2011 text — most of them accents and punctuation.
Every word tagged
140,768 printed words, each with its Strong's number, morphology and sense.
Read from the page
All 684 printed pages, in two scanned copies of the 1894 printing.
Free, by request
Sign in, complete your profile, and send a request. Every one is read.
Like what you see? Create a free account and every tool is yours.