All counts machine-reproducible — the walls on which the book stands
7 min read
In-depth:
This book has repeatedly distinguished between walls (what is countable, well-documented) and pillars (what must be weighed). This appendix lays the walls bare: the complete, machine-generated counts of the Day-field in the Hebrew Old Testament and the Greek New Testament, with the scripts and sources alongside, so that anyone can redo them. Not a single number in this book asks for trust; each asks for recounting. That is the promise of the honest scale (chapter 4), here redeemed.
The original text is twofold and has remained the same throughout. For the Old Testament: the Westminster Leningrad Codex (OSHB/morphhb, CC-BY 4.0), counted on a lemma basis for yom (H3117) and YHWH (H3068) with the script `scripts/tel_hebreeuws.py`; the raw rows stand in `data/master_hebreeuws.csv`. For the New Testament: the Scrivener 1894 Textus Receptus (beta-code, public domain, held locally in `data/scrivener_tr/`), counted on the word-stem hēmer- with hand-checking (the genitive requirement for the modifier) via `scripts/tel_grieks.py`; raw rows in `data/master_grieks.csv`. Where the machine caught too much, it was excepted by hand — and transparently; those false hits are marked separately, not omitted.
The Hebrew picture first. The word yom occurs in the Old Testament 2303 times; after hand-checking, 379 hits of these fall in the Day-field proper (yom with a modifier), with 13 false hits set apart. Those 379 distribute across subfields, and precisely that distribution is the first proof that the Day is not a loose theme but a spectrum — from the neutral "that day" to the charged core.
Subfield
Torah
Early Proph.
Isaiah
Jeremiah
Ezekiel
Lam.
The Twelve
Writings
TOTAL
That day
31
86
46
21
15
0
50
27
276
Calamity
1
1
0
5
0
0
4
11
22
Core (yom YHWH)
0
0
3
0
3
0
13
0
19
Distress
1
1
1
1
0
0
5
6
15
Wrath
0
0
1
0
1
4
4
4
14
Battle/slaughter
1
1
1
1
0
0
3
3
10
Darkness
1
0
0
0
2
0
4
3
10
Vengeance
0
0
3
1
0
0
0
1
5
Devastation
0
0
1
0
1
0
1
0
3
Visitation
0
0
1
0
0
0
1
0
2
Trumpet
1
0
0
0
0
0
1
0
2
Great day
1
0
0
0
0
0
0
0
1
The core of the core are the eighteen classic yom YHWH hits — the exact construction "day of the LORD" (chapter 2). The counter machine-tallies nineteen, but the nineteenth, Ezekiel 48:35, is on hand-checking the city-name YHWH-shammah ("the LORD is there"), not a day-phrase, and is therefore set apart. The eighteen, in canonical order: Isa 2:12; 13:6; 13:9; Ezek 13:5; 30:3; Joel 1:15; 2:1; 2:11; 2:31 (Heb. 3:4); 3:14 (Heb. 4:14); Amos 5:18 (2×); 5:20; Obad 15; Zeph 1:7; 1:14 (2×); Mal 4:5 (Heb. 3:23). The most striking pattern springs at once from the table: thirteen of the nineteen core hits stand in the Twelve — the Minor Prophets are the heart of the Day, not the Major. That is a wall, not an opinion.
The Greek picture mirrors the Hebrew in a surprising way. The word hēmera occurs in the New Testament (TR) 389 times, but the greater part is non-technical: Luke and Acts alone account for 179 occurrences (46%), almost always the narrative "day(s)." The technical Day-phrases must therefore, just as in the Old Testament, be brought out by hand-checking. Do that, and there remain fourteen genuine core phrases, and their distribution is the hinge of Part III.
Three walls rise from this. First: Paul is the New Testament center of the technical Day — of the fourteen genuine core phrases, nine stand with him, exactly as the Twelve formed the center of gravity in the Old Testament; the Day shifts from the prophets to the apostle. Second: the comforting "day of Christ" is uniquely Pauline — all four of the hēmera Christou texts are his, over against the terror of the hēmera kyriou (the heart of crux 2). Third: 2 Thessalonians 2:2 reads in the Textus Receptus "day of Christ" (where the critical texts read "kyriou") — a text-critical fact that sharpens the distinction between the two days and on which both crux 2 and crux 3 turn (chapters 28, 29). That this is a textual variant and no certainty honestly bears the hallmark contested.
That is the nature of this appendix: it gives not the interpretation but the data. The interpretation — what the counts mean — has been weighed throughout the book, with hallmarks and counter-positions. But the counts themselves stand here bare, with source and script, so that no one need take them on authority. Whoever runs the scripts on the same sources gets the same numbers. That is the ground on which everything rests: not "mathematical proof" — that word is and remains forbidden (chapter 4) — but machine-recountable. The walls are real; on those walls the pillars were permitted to be set.
Expert noteThe hand-check: why the machine does not have the last word
A pure count is not the same as a naive count, and it is honest to show where the human eye had to correct. The machine counts forms; it does not know whether a genitive in the window is really the modifier of yom/hēmera. Three kinds of false hit are therefore excepted by hand — and transparently. (1) City-names and proper names: Ezekiel 48:35 (YHWH-shammah, "the LORD is there") machine-counts as yom YHWH but is the name of the city. (2) Dative subjects: in the Greek the counter caught, among others, 2 Timothy 1:18 (where kyrios is the subject, not the modifier), Romans 14:6 (kyriō = "to the Lord"), and Revelation 1:10 (kyriakē hēmera, "the Lord's day" with an adjective instead of a genitive, so not the yom YHWH form; whether that day is Sunday, the Sabbath or the future day of YHWH, the text does not decide — open). (3) Narrative days: "some days" (Acts 10:48), "the days of Noah" (1 Pet 3:20). All of these stand outside the core count, but they are not concealed — they are marked separately in the concordance, so that the reader can check the boundary for himself. That is the difference between a wall and a mirage: a wall lets itself be viewed from both sides. The 18 (Hebrew) and the 6 + 4 + 2 + 2 (Greek) are what remains after that check — and precisely for that reason reliable. (The method of counting-plus-hand-checking: well-documented and reproducible.)
This book has repeatedly distinguished between walls (what is countable, well-documented) and pillars (what must be weighed). This appendix lays the walls bare: the complete, machine-generated counts of the Day-field in the Hebrew Old Testament and the Greek New Testament, with the scripts and sources alongside, so that anyone can redo them. Not a single number in this book asks for trust; each asks for recounting. That is the promise of the honest scale (chapter 4), here redeemed.
The original text is twofold and has remained the same throughout. For the Old Testament: the Westminster Leningrad Codex (OSHB/morphhb, CC-BY 4.0), counted on a lemma basis for yom (H3117) and YHWH (H3068) with the script `scripts/tel_hebreeuws.py`; the raw rows stand in `data/master_hebreeuws.csv`. For the New Testament: the Scrivener 1894 Textus Receptus (beta-code, public domain, held locally in `data/scrivener_tr/`), counted on the word-stem hēmer- with hand-checking (the genitive requirement for the modifier) via `scripts/tel_grieks.py`; raw rows in `data/master_grieks.csv`. Where the machine caught too much, it was excepted by hand — and transparently; those false hits are marked separately, not omitted.
The Hebrew picture first. The word yom occurs in the Old Testament 2303 times; after hand-checking, 379 hits of these fall in the Day-field proper (yom with a modifier), with 13 false hits set apart. Those 379 distribute across subfields, and precisely that distribution is the first proof that the Day is not a loose theme but a spectrum — from the neutral "that day" to the charged core.
The core of the core are the eighteen classic yom YHWH hits — the exact construction "day of the LORD" (chapter 2). The counter machine-tallies nineteen, but the nineteenth, Ezekiel 48:35, is on hand-checking the city-name YHWH-shammah ("the LORD is there"), not a day-phrase, and is therefore set apart. The eighteen, in canonical order: Isa 2:12; 13:6; 13:9; Ezek 13:5; 30:3; Joel 1:15; 2:1; 2:11; 2:31 (Heb. 3:4); 3:14 (Heb. 4:14); Amos 5:18 (2×); 5:20; Obad 15; Zeph 1:7; 1:14 (2×); Mal 4:5 (Heb. 3:23). The most striking pattern springs at once from the table: thirteen of the nineteen core hits stand in the Twelve — the Minor Prophets are the heart of the Day, not the Major. That is a wall, not an opinion.
The Greek picture mirrors the Hebrew in a surprising way. The word hēmera occurs in the New Testament (TR) 389 times, but the greater part is non-technical: Luke and Acts alone account for 179 occurrences (46%), almost always the narrative "day(s)." The technical Day-phrases must therefore, just as in the Old Testament, be brought out by hand-checking. Do that, and there remain fourteen genuine core phrases, and their distribution is the hinge of Part III.
Three walls rise from this. First: Paul is the New Testament center of the technical Day — of the fourteen genuine core phrases, nine stand with him, exactly as the Twelve formed the center of gravity in the Old Testament; the Day shifts from the prophets to the apostle. Second: the comforting "day of Christ" is uniquely Pauline — all four of the hēmera Christou texts are his, over against the terror of the hēmera kyriou (the heart of crux 2). Third: 2 Thessalonians 2:2 reads in the Textus Receptus "day of Christ" (where the critical texts read "kyriou") — a text-critical fact that sharpens the distinction between the two days and on which both crux 2 and crux 3 turn (chapters 28, 29). That this is a textual variant and no certainty honestly bears the hallmark contested.
That is the nature of this appendix: it gives not the interpretation but the data. The interpretation — what the counts mean — has been weighed throughout the book, with hallmarks and counter-positions. But the counts themselves stand here bare, with source and script, so that no one need take them on authority. Whoever runs the scripts on the same sources gets the same numbers. That is the ground on which everything rests: not "mathematical proof" — that word is and remains forbidden (chapter 4) — but machine-recountable. The walls are real; on those walls the pillars were permitted to be set.
Expert noteThe hand-check: why the machine does not have the last word
A pure count is not the same as a naive count, and it is honest to show where the human eye had to correct. The machine counts forms; it does not know whether a genitive in the window is really the modifier of yom/hēmera. Three kinds of false hit are therefore excepted by hand — and transparently. (1) City-names and proper names: Ezekiel 48:35 (YHWH-shammah, "the LORD is there") machine-counts as yom YHWH but is the name of the city. (2) Dative subjects: in the Greek the counter caught, among others, 2 Timothy 1:18 (where kyrios is the subject, not the modifier), Romans 14:6 (kyriō = "to the Lord"), and Revelation 1:10 (kyriakē hēmera, "the Lord's day" with an adjective instead of a genitive, so not the yom YHWH form; whether that day is Sunday, the Sabbath or the future day of YHWH, the text does not decide — open). (3) Narrative days: "some days" (Acts 10:48), "the days of Noah" (1 Pet 3:20). All of these stand outside the core count, but they are not concealed — they are marked separately in the concordance, so that the reader can check the boundary for himself. That is the difference between a wall and a mirage: a wall lets itself be viewed from both sides. The 18 (Hebrew) and the 6 + 4 + 2 + 2 (Greek) are what remains after that check — and precisely for that reason reliable. (The method of counting-plus-hand-checking: well-documented and reproducible.)