At a glanceKey from adjacent plaintext · partial (88% of the cipher)
- The drafts
- Five drafts of telegrams from the Army General Staff in Tokyo (Maj.-Gen. Tamura) to Major-General Fukushima in Peking, February to May 1901.
- Where
- National Institute for Defense Studies, file 参謀本部 雑, 通報綴 (January–June 1901), via JACAR: C09122808300, C09122808400, C09122809400, C09122810100, C09122812300. Catalogue entry 715.
- The cipher
- The telegrams are plain romanised Japanese; only the sensitive words are in digits, two per kana, on a grid where the tens digit gives the vowel and the units digit the consonant. Five values lie outside the grid.
- How it was solved
- No contemporary translation was found. Three words guessed from the romaji right after the digits (ノゾミヲ before mitasigatasi, オオシマノ, リコウシヨウ) fixed 13 values, and the grid rule gave the rest.
- Credit
- Harumi-Matsuba, with Claude, Codex and Sonnet (issue #371, issue #374, kagigo1901-public, November 2026).
- What it says
- Moves of officers (Kamio, Ōshima, Kusunose, Shiba); Li Hongzhang, Zhang Zhidong and a secret agreement; a visit to Japan by “the Field Marshal” from Kiaochow Bay; relief troops.
- Still open
- Five off-grid values (ten occurrences), a handful of doubtful digits and kana, and the rest of the file. The digits rest on AI transcriptions; no person has compared them with the images, and the images could not be reached from here.
01 The five drafts
In the General Staff’s file of communications for January–June 1901 are five drafts to Fukushima written in romanised Japanese, with runs of digits in groups of five where the names and other sensitive words stand. The October 2026 sweep of JACAR catalogued them as entry 715 and took the Latin strings for enciphered romaji. Harumi-Matsuba read them from the images as plain romaji (JACAR’s record text garbles the handwriting) and noticed that all 23 digit runs have an even number of digits: two digits per kana, cut into fives.
| JACAR | Frames | Date | Pairs | In the table | Counted as read |
|---|---|---|---|---|---|
| C09122808300 | 0306–0307 | February 1901 | 11 | 11 | 10 |
| C09122808400 | 0308–0309 | 14 February 1901 | 15 | 15 | 15 |
| C09122809400 | 0331–0332 | 2 March 1901 | 59 | 52 | 48 |
| C09122810100 | 0348–0350 | 8 March 1901 | 42 | 40 | 37 |
| C09122812300 | 0412–0414 | 1 May 1901 | 21 | 20 | 20 |
| Total | 148 | 138 | 130 | ||
The images could not be fetched from this session, so no crop is shown.
02 The table
A grid. In the lower half (01–50) the tens digit is the vowel row (0 a, 1 i, 2 u, 3 e, 4 o) and the units digit the consonant column: 1 the bare vowel, 2 k, 3 g, 4 s, 5 z, 6 t, 7 d, 8 n, 9 h, and 0 b, which takes the slots 10, 20, 30, 40, 50. In the upper half (51–95) the tens digit runs 5–9 for a–o and the units are 1 p, 2 m, 3 y, 4 r; 55 is ワ and 65 is ン. So ア 01, カ 02, シ 14, オ 41, ノ 48, マ 52, ミ 62, リ 64, ヨ 93. ヲ is sent as オ. Sino-Japanese words keep the kana spelling of the time (条約 ゼウヤク, 張 テウ).
This is not the table of the incoming telegrams of the same months that Harumi-Matsuba reconstructed first (issue #371: a serial count from ア 00 to ン 67, rebuilt from telegrams whose translations were filed with them). Tried on these drafts, it puts 21 of the 148 pairs out of range and gives no words. Nor is it the 1902–04 table of the Army’s numeric kana cipher, a different cipher with its own page.
03 How the table was found
No contemporary translation of these drafts was found. The way in was the clear romaji right after each run. In C09122812300, 48456 241 mitasigatasi invites ノゾミヲ (望ミヲ充タシ難シ, “the wish is hard to meet”); its spacings, ノ to ゾ 3 and ゾ to オ 4, match the o-row of the incoming table. Then 41411 45248 kotowa fits オオシマノ (“the matter of Ōshima”), and 64422 11493, whose third pair repeats as the sixth, fits リコウシヨウ, Li Hongzhang. Those three words fix 13 values; the rule of the grid fills the rest, and everything else came out of it: ヒメジ, クスノセ, シバ, テウシドウ (Zhang Zhidong), セイアン (Xi’an), ゲンスイハコウシウワンヨリニホンニ (“the Field Marshal … from Kiaochow Bay to Japan”).
Controls by the contributor: counting the kana covered by Japanese dictionary words (JMdict, three kana or more), the real table covers 58, against at most 37 for 1,000 grids with permuted digits and at most 36 for 1,000 random reassignments. A blind second transcription by OpenAI Codex from the images agrees on 285 of 296 digits.
04 The reading
Digits in code type, the kana from the table, and the clear romaji in italics. [nn] is a value outside the table, 〔 〕 a doubtful kana. Translations are Harumi-Matsuba’s.
C09122808300, February 1901. 02624 14832 4221 カミオノ〔ケ〕コウ yurusazu 1982 1518 ヒメジニ yarutumori. “Kamio’s 〔…〕 is not permitted; we intend to send him to Himeji.”
C09122808400, 14 February 1901, No. 21. オオシマノ kotowa ima soodantiu, mosi ikeneba クスノセハ ikaga シバハトニ kaku kaesu. Tamura. “The matter of Ōshima is under discussion now. If he cannot go, how about Kusunose? Shiba will be sent back in any case.”
C09122809400, 2 March 1901. The longest: Li Hongzhang’s answer to Japanese advice that a secret agreement was disadvantageous (“the advice is hard to accept”), Zhang Zhidong’s request “that this treaty should not come into being”, and “a telegram of the same meaning has come from Xi’an”. Four of the five off-grid values stand here.
C09122810100, 8 March 1901. ゲンスイハコウシウワンヨリニホンニ kitaru nozomi a… “The Field Marshal wishes to come to Japan from Kiaochow Bay …. We are sounding out how 〔…〕[89][80] would be placed to receive him. If it looks as though we shall receive him suitably, recommend the 〔pleasure〕 trip to him.” The contributor takes the Field Marshal to be Count Waldersee.
C09122812300, 1 May 1901. キテウスル yooni … nareri; mata コウタイヘイノスウハ[80]ガ kimarazareba ノゾミオ mitasigatasi. Tamura. “… is to return home. Also, the number of relief troops: unless [80] is settled, the wish is hard to meet.”
Two checks against content, from the contributor: Fukushima’s own telegram of 22 April 1901 (C09122801000, read with the incoming table) says the Field Marshal has received the Emperor’s permission to go to Japan, and his telegram of 25 February (C09122805900) says Shiba is to leave at the beginning of the next month. Not re-checked here.
05 What was checked here
A decoder was written here from the stated rule alone, without the contributor’s code. Over the first transcription (made before the table was known) it finds 23 runs, none of odd length, 148 pairs, 136 of them in the table; The contributor’s four re-read digits bring it to 138: two that the blind second reader also has (42291→42211, 94427→94421, which change kana but not the count) and two it does not (43962→43162, 16598→16593, the doubtful 〔チ〕 and 〔ヨ〕), which bring the out-of-table pairs 96 and 98 into the table. Every kana string agrees with the contributor’s. The incoming table, decoded beside it, gives nothing. The JACAR images could not be reached from this session, so no digit was checked against them, and the dictionary control was not re-run (there is no Japanese model in the project’s language library).
06 Still open
- Values 59, 60, 80, 89, 99, ten occurrences, outside the grid. They look like word codes (the contributor suggests countries or offices: Russia, China, Germany, a minister), but nothing confirms them.
- C09122809400: the first group (99506, or 77306 for the second reader); 22652, which gives マ where ヤ would make ミツヤク, 密約; two digits re-read with the table in hand that the blind reader did not confirm (〔チ〕, 〔ヨ〕).
- C09122810100: 09626 gives ミ where マ would make マンユウ, 漫遊; 〔モチ〕. C09122808300: 〔ケ〕コウ, word not identified.
- The rest of the file after C09122810000 has not been viewed; which officers Kamio, Ōshima, Kusunose and Shiba are has not been checked.
- The transcriptions are AI work (Claude first, Codex blind second); no person has compared them with the images.
Full texts
The transcription and the decryption of every document on this page, so that the reading can be used and checked. Short files are shown here; all are in the repository.
February 1901 (undated)
Tamura to Fukushima, Feb 1901: Kamio's (unidentified kekō) is not permitted, and the General Staff means to send him to Himeji.
Transcription
# C09122808300 (undated, Feb 1901) - digit runs from ct_outgoing1901.txt (first transcription), one run per line 02624148324221 19821518
Decryption
# C09122808300 (February 1901 (undated), frames 0306-0307): reading Digits as in ct_outgoing1901.txt (re-read groups as noted), kana from decode.py (grid table), romaji clear text after each run, and the reading in kanji-kana. [nn] = value not in the table; 〔 〕 = doubtful kana. Reading and kanji forms after Harumi-Matsuba, kagigo1901-public docs/READING1901B.en.md (commit 9d53bd6); kana checked here with decode.py. - `02624 14832 4221` カミオノ〔ケ〕コウ | yurusazu | 神尾ノ〔ケコウ〕許サズ - `1982 1518` ヒメジニ | yarutumori | 姫路ニ遣ル積リ Japanese: 神尾ノ〔ケコウ〕許サズ姫路ニ遣ル積リ Holes: 〔ケ〕 = 32, clear on the image per the contributor; the word ケコウ is not identified.
Reading
# C09122808300 (February 1901 (undated), frames 0306-0307): reading Digits as in ct_outgoing1901.txt (re-read groups as noted), kana from decode.py (grid table), romaji clear text after each run, and the reading in kanji-kana. [nn] = value not in the table; 〔 〕 = doubtful kana. Reading and kanji forms after Harumi-Matsuba, kagigo1901-public docs/READING1901B.en.md (commit 9d53bd6); kana checked here with decode.py. - `02624 14832 4221` カミオノ〔ケ〕コウ | yurusazu | 神尾ノ〔ケコウ〕許サズ - `1982 1518` ヒメジニ | yarutumori | 姫路ニ遣ル積リ Japanese: 神尾ノ〔ケコウ〕許サズ姫路ニ遣ル積リ Holes: 〔ケ〕 = 32, clear on the image per the contributor; the word ケコウ is not identified.
Translation
# C09122808300: English Kamio's 〔kekō〕 is not permitted; we intend to send him to Himeji. (Translation after Harumi-Matsuba, READING1901B.en.md; [nn] = value not in the table.)
14 February 1901
Tamura to Fukushima, 14 Feb 1901: the matter of Ōshima is under discussion; if he cannot go, how about Kusunose; Shiba will be sent back in any case.
Transcription
# C09122808400 (14 Feb 1901, No. 21) - digit runs from ct_outgoing1901.txt (first transcription), one run per line 4141145248 2224483409 1410094618
Decryption
# C09122808400 (14 February 1901, No. 21, frames 0308-0309): reading Digits as in ct_outgoing1901.txt (re-read groups as noted), kana from decode.py (grid table), romaji clear text after each run, and the reading in kanji-kana. [nn] = value not in the table; 〔 〕 = doubtful kana. Reading and kanji forms after Harumi-Matsuba, kagigo1901-public docs/READING1901B.en.md (commit 9d53bd6); kana checked here with decode.py. - `41411 45248` オオシマノ | kotowa ima soodantiu, mosi ikeneba | 大島ノ事ハ今相談中、若シ行ケネバ - `22244 83409` クスノセハ | ikaga | 楠瀬ハ如何 - `14100 94618` シバハトニ | kaku kaesu. Tamura | 柴ハ兎ニ角返ス。田村 Japanese: 大島ノ事ハ今相談中、若シ行ケネバ楠瀬ハ如何柴ハ兎ニ角返ス。田村 Holes: No holes.
Reading
# C09122808400 (14 February 1901, No. 21, frames 0308-0309): reading Digits as in ct_outgoing1901.txt (re-read groups as noted), kana from decode.py (grid table), romaji clear text after each run, and the reading in kanji-kana. [nn] = value not in the table; 〔 〕 = doubtful kana. Reading and kanji forms after Harumi-Matsuba, kagigo1901-public docs/READING1901B.en.md (commit 9d53bd6); kana checked here with decode.py. - `41411 45248` オオシマノ | kotowa ima soodantiu, mosi ikeneba | 大島ノ事ハ今相談中、若シ行ケネバ - `22244 83409` クスノセハ | ikaga | 楠瀬ハ如何 - `14100 94618` シバハトニ | kaku kaesu. Tamura | 柴ハ兎ニ角返ス。田村 Japanese: 大島ノ事ハ今相談中、若シ行ケネバ楠瀬ハ如何柴ハ兎ニ角返ス。田村 Holes: No holes.
Translation
# C09122808400: English The matter of Ōshima is under discussion now. If he cannot go, how about Kusunose? Shiba will be sent back in any case. Tamura. (Translation after Harumi-Matsuba, READING1901B.en.md; [nn] = value not in the table.)
2 March 1901
Tamura to Fukushima, 2 Mar 1901: Li Hongzhang's answer to Japanese advice against a secret agreement, Zhang Zhidong's request that the treaty should not come into being, and a telegram of the same sense from Xi'an; four word codes unread.
Transcription
# C09122809400 (2 Mar 1901) - digit runs from ct_outgoing1901.txt (first transcription), one run per line 99506226522248 5503 644221149321096080 5918 9960352109 43962142220911840306 59183621144721 3521532248 34110165986492 3765912112
Decryption
# C09122809400 (2 March 1901, frames 0331-0332): reading Digits as in ct_outgoing1901.txt (re-read groups as noted), kana from decode.py (grid table), romaji clear text after each run, and the reading in kanji-kana. [nn] = value not in the table; 〔 〕 = doubtful kana. Reading and kanji forms after Harumi-Matsuba, kagigo1901-public docs/READING1901B.en.md (commit 9d53bd6); kana checked here with decode.py. - `99506 22652 2248` [99]〔ボ〕ミツ〔マ〕クノ | furiekinaru | [99]〔ボ〕ミツ〔マ〕クノ不利益ナル - `5503` ワガ | tiukokuni taisi | 我ガ忠告ニ対シ - `64422 11493 2109 6080` リコウシヨウハ[60][80] | osite | 李鴻章ハ[60][80]ヲシテ - `5918` [59]ニ | tugesasuruni | [59]ニ告ゲサスルニ - `99603 52109` [99][60]ゼウハ | motukano ... riekito kangoureba | [99][60]ゼウハ目下ノ…利益ト考フレバ - `43962 14222 09118 40306` ゴ〔チ〕ウコクハイレガタ | ki i o motte ...; mata waga | ゴ〔チ〕ウコクハイレガタキ意ヲ以テ…、又我ガ - `5918 36211 44721` [59]ニテウシドウ | yori kono | [59]ニ張之洞ヨリ此ノ - `35215 32248` ゼウヤクノ | naritatanu yoo nito tanomi kitareri; mata | 条約ノ成リ立タヌ様ニト頼ミ来レリ。又 - `34110 16598 6492` セイアン〔ヨ〕リモ | onadi imino | 西安〔ヨ〕リモ同ジ意味ノ - `37659 12112` デンポウキ | tari. Tamura | 電報来タリ。田村 Japanese: [99]〔ボ〕ミツ〔マ〕クノ不利益ナル我ガ忠告ニ対シ李鴻章ハ[60][80]ヲシテ[59]ニ告ゲサスルニ[99][60]ゼウハ目下ノ…利益ト考フレバゴ〔チ〕ウコクハイレガタキ意ヲ以テ…、又我ガ[59]ニ張之洞ヨリ此ノ条約ノ成リ立タヌ様ニト頼ミ来レリ。又西安〔ヨ〕リモ同ジ意味ノ電報来タリ。田村 Holes: First group 99506 (second reader 77306) not settled; 22652 gives マ where ヤ (53) would give 密約; 〔チ〕 and 〔ヨ〕 rest on digits re-read with the table in hand (43962->43162, 16598->16593), not confirmed by the blind second reader. [59] [60] [80] [99] are values outside the table. The romaji of the first transcription is partly unread (motukano..., ki i o motte...).
Reading
# C09122809400 (2 March 1901, frames 0331-0332): reading Digits as in ct_outgoing1901.txt (re-read groups as noted), kana from decode.py (grid table), romaji clear text after each run, and the reading in kanji-kana. [nn] = value not in the table; 〔 〕 = doubtful kana. Reading and kanji forms after Harumi-Matsuba, kagigo1901-public docs/READING1901B.en.md (commit 9d53bd6); kana checked here with decode.py. - `99506 22652 2248` [99]〔ボ〕ミツ〔マ〕クノ | furiekinaru | [99]〔ボ〕ミツ〔マ〕クノ不利益ナル - `5503` ワガ | tiukokuni taisi | 我ガ忠告ニ対シ - `64422 11493 2109 6080` リコウシヨウハ[60][80] | osite | 李鴻章ハ[60][80]ヲシテ - `5918` [59]ニ | tugesasuruni | [59]ニ告ゲサスルニ - `99603 52109` [99][60]ゼウハ | motukano ... riekito kangoureba | [99][60]ゼウハ目下ノ…利益ト考フレバ - `43962 14222 09118 40306` ゴ〔チ〕ウコクハイレガタ | ki i o motte ...; mata waga | ゴ〔チ〕ウコクハイレガタキ意ヲ以テ…、又我ガ - `5918 36211 44721` [59]ニテウシドウ | yori kono | [59]ニ張之洞ヨリ此ノ - `35215 32248` ゼウヤクノ | naritatanu yoo nito tanomi kitareri; mata | 条約ノ成リ立タヌ様ニト頼ミ来レリ。又 - `34110 16598 6492` セイアン〔ヨ〕リモ | onadi imino | 西安〔ヨ〕リモ同ジ意味ノ - `37659 12112` デンポウキ | tari. Tamura | 電報来タリ。田村 Japanese: [99]〔ボ〕ミツ〔マ〕クノ不利益ナル我ガ忠告ニ対シ李鴻章ハ[60][80]ヲシテ[59]ニ告ゲサスルニ[99][60]ゼウハ目下ノ…利益ト考フレバゴ〔チ〕ウコクハイレガタキ意ヲ以テ…、又我ガ[59]ニ張之洞ヨリ此ノ条約ノ成リ立タヌ様ニト頼ミ来レリ。又西安〔ヨ〕リモ同ジ意味ノ電報来タリ。田村 Holes: First group 99506 (second reader 77306) not settled; 22652 gives マ where ヤ (53) would give 密約; 〔チ〕 and 〔ヨ〕 rest on digits re-read with the table in hand (43962->43162, 16598->16593), not confirmed by the blind second reader. [59] [60] [80] [99] are values outside the table. The romaji of the first transcription is partly unread (motukano..., ki i o motte...).
Translation
# C09122809400: English ... to our advice that the ... secret agreement is disadvantageous, Li Hongzhang had [60][80] tell [59] that, [99][60] ... considering it advantageous at present ..., the advice is hard to accept .... Also, Zhang Zhidong has asked our [59] that this treaty should not come into being. Also a telegram of the same meaning has come from Xi'an. Tamura. (Translation after Harumi-Matsuba, READING1901B.en.md; [nn] = value not in the table.)
8 March 1901
Tamura to Fukushima, 8 Mar 1901: the Field Marshal (Waldersee, per the contributor) wishes to come to Japan from Kiaochow Bay; Tokyo is sounding out how he would be received and asks Fukushima to recommend the trip if it looks favourable.
Transcription
# C09122810100 (8 Mar 1901) - digit runs from ct_outgoing1901.txt (first transcription), one run per line 3365241109422914215565936418496518 9216898048 0284411912 5584094427 93212408840962657321
Decryption
# C09122810100 (8 March 1901, frames 0348-0350): reading Digits as in ct_outgoing1901.txt (re-read groups as noted), kana from decode.py (grid table), romaji clear text after each run, and the reading in kanji-kana. [nn] = value not in the table; 〔 〕 = doubtful kana. Reading and kanji forms after Harumi-Matsuba, kagigo1901-public docs/READING1901B.en.md (commit 9d53bd6); kana checked here with decode.py. - `33652 41109 42211 42155 65936 41849 6518` ゲンスイハコウシウワンヨリニホンニ | kitaru nozomi a... | 元帥ハ膠州湾ヨリ日本ニ来タル望ミア… - `92168 98048` 〔モチ〕[89][80]ノ | tugoo ikani | 〔モチ〕[89][80]ノ都合如何ニ - `02844 11912` カレオヒキ | ukurukao saguritiunari | 彼ヲ引キ受クルカヲ探リ中ナリ - `55840 94421` ワレハソウ | tooni hikiukuru | 我ハ相当ニ引キ受クル - `93212 40884 09626 57321` ヨウスナレハ〔ミ〕ンユウ | wo susumerareyo. Tamura | ヨウスナレバ〔ミ〕ンユウヲ勧メラレヨ。田村 Japanese: 元帥ハ膠州湾ヨリ日本ニ来タル望ミア…〔モチ〕[89][80]ノ都合如何ニ彼ヲ引キ受クルカヲ探リ中ナリ我ハ相当ニ引キ受クルヨウスナレバ〔ミ〕ンユウヲ勧メラレヨ。田村 Holes: 42291 and 94427 of the first transcription read 42211 and 94421 (re-read with the table, and so read by the blind second reader). 09626 gives ミ where マ (52) would give 漫遊. [89] [80] outside the table; 〔モチ〕 doubtful.
Reading
# C09122810100 (8 March 1901, frames 0348-0350): reading Digits as in ct_outgoing1901.txt (re-read groups as noted), kana from decode.py (grid table), romaji clear text after each run, and the reading in kanji-kana. [nn] = value not in the table; 〔 〕 = doubtful kana. Reading and kanji forms after Harumi-Matsuba, kagigo1901-public docs/READING1901B.en.md (commit 9d53bd6); kana checked here with decode.py. - `33652 41109 42211 42155 65936 41849 6518` ゲンスイハコウシウワンヨリニホンニ | kitaru nozomi a... | 元帥ハ膠州湾ヨリ日本ニ来タル望ミア… - `92168 98048` 〔モチ〕[89][80]ノ | tugoo ikani | 〔モチ〕[89][80]ノ都合如何ニ - `02844 11912` カレオヒキ | ukurukao saguritiunari | 彼ヲ引キ受クルカヲ探リ中ナリ - `55840 94421` ワレハソウ | tooni hikiukuru | 我ハ相当ニ引キ受クル - `93212 40884 09626 57321` ヨウスナレハ〔ミ〕ンユウ | wo susumerareyo. Tamura | ヨウスナレバ〔ミ〕ンユウヲ勧メラレヨ。田村 Japanese: 元帥ハ膠州湾ヨリ日本ニ来タル望ミア…〔モチ〕[89][80]ノ都合如何ニ彼ヲ引キ受クルカヲ探リ中ナリ我ハ相当ニ引キ受クルヨウスナレバ〔ミ〕ンユウヲ勧メラレヨ。田村 Holes: 42291 and 94427 of the first transcription read 42211 and 94421 (re-read with the table, and so read by the blind second reader). 09626 gives ミ where マ (52) would give 漫遊. [89] [80] outside the table; 〔モチ〕 doubtful.
Translation
# C09122810100: English The Field Marshal wishes to come to Japan from Kiaochow Bay .... We are sounding out how 〔...〕[89][80] would be placed to receive him. If it looks as though we shall receive him suitably, recommend the 〔pleasure〕 trip to him. Tamura. (Translation after Harumi-Matsuba, READING1901B.en.md; [nn] = value not in the table.)
1 May 1901
Tamura to Fukushima, 1 May 1901: someone is to return home; the number of relief troops cannot be met until a matter (word code 80) is settled.
Transcription
# C09122812300 (1 May 1901) - digit runs from ct_outgoing1901.txt (first transcription), one run per line 1236212474 422106113911482421098003 48456241
Decryption
# C09122812300 (1 May 1901, frames 0412-0414): reading Digits as in ct_outgoing1901.txt (re-read groups as noted), kana from decode.py (grid table), romaji clear text after each run, and the reading in kanji-kana. [nn] = value not in the table; 〔 〕 = doubtful kana. Reading and kanji forms after Harumi-Matsuba, kagigo1901-public docs/READING1901B.en.md (commit 9d53bd6); kana checked here with decode.py. - `12362 12474` キテウスル | yooni ... nareri; mata | 帰朝スル様ニ…ナレリ。又 - `42210 61139 11482 42109 8003` コウタイヘイノスウハ[80]ガ | kimarazareba | 交代兵ノ数ハ[80]ガ決マラザレバ - `48456 241` ノゾミオ | mitasigatasi. Tamura | 望ミヲ充タシ難シ。田村 Japanese: 帰朝スル様ニ…ナレリ。又交代兵ノ数ハ[80]ガ決マラザレバ望ミヲ充タシ難シ。田村 Holes: [80] outside the table; a few romaji letters of the first line unread (yooni ... nareri).
Reading
# C09122812300 (1 May 1901, frames 0412-0414): reading Digits as in ct_outgoing1901.txt (re-read groups as noted), kana from decode.py (grid table), romaji clear text after each run, and the reading in kanji-kana. [nn] = value not in the table; 〔 〕 = doubtful kana. Reading and kanji forms after Harumi-Matsuba, kagigo1901-public docs/READING1901B.en.md (commit 9d53bd6); kana checked here with decode.py. - `12362 12474` キテウスル | yooni ... nareri; mata | 帰朝スル様ニ…ナレリ。又 - `42210 61139 11482 42109 8003` コウタイヘイノスウハ[80]ガ | kimarazareba | 交代兵ノ数ハ[80]ガ決マラザレバ - `48456 241` ノゾミオ | mitasigatasi. Tamura | 望ミヲ充タシ難シ。田村 Japanese: 帰朝スル様ニ…ナレリ。又交代兵ノ数ハ[80]ガ決マラザレバ望ミヲ充タシ難シ。田村 Holes: [80] outside the table; a few romaji letters of the first line unread (yooni ... nareri).
Translation
# C09122812300: English ... is to return home. Also, the number of relief troops: unless [80] is settled, the wish is hard to meet. Tamura. (Translation after Harumi-Matsuba, READING1901B.en.md; [nn] = value not in the table.)
Key: key.md
07 Sources
- National Institute for Defense Studies, 参謀本部 雑, 自明治34年1月5日至同年6月29日 通報綴, via JACAR: C09122808300, C09122808400, C09122809400, C09122810100, C09122812300.
- Harumi-Matsuba, kagigo1901-public (commit 9d53bd6):
docs/READING1901B.en.md,docs/OPEN.en.md,data/outgoing1901.txt,src/table1901b.py; code MIT, documents and data CC BY 4.0. Issues #371 and #374.
Files: targets/army1901/ (transcription, decoder, readings).
Cite this page
Harumi-Matsuba and Daniel Bourdeau, “General Staff drafts to Fukushima, 1901”, Cyphersolver, posted 10 November 2026, https://cyphersolver.com/army1901.html.