N1 2013-12 — audit report

70 questions (問題1–13, the non-listening half). Generated from the OCR and correction audit trails.

OCR confidence

Source: retypeset-digital — quality 4/5: Cleanly retyped Word-document conversion: crisp digital text, generous spacing, no scan noise or watermarks; includes a bonus full listening transcript (听力台词原文) and a complete-but-for-one-cell answer key. Docked a point for retyping defects: question 52 is misprinted as a duplicate '51.' on p13 (numbering confirmed via the answer key), duplicated 注1 labels on p14, the listening 問題2 7番 answer cell is blank, question numbers are plain text rather than boxed (losing the original layout cues), the 問題7 cloze page (p7) and a few reading pages are set in very small dense type, and the 問題6 example/answer-sheet illustration is dropped. Chinese title header on every page; otherwise content matches the official N1 structure.

Every section was transcribed twice independently (a full-page pass and a double-resolution half-page pass) and reconciled. Of 13 sections, 0 agreed on all content between the two passes (reconciled automatically, differing only in transcriber notes) and 13 had at least one content difference resolved by a third adjudication pass that re-read the page images. 38 leaf-level differences were examined in total.

Residual OCR risks (21):

• 問題1: transcribed from text witnesses, not the booklet image; see the section-specific audit notes
• 問題2: transcribed from text witnesses, not the booklet image; see the section-specific audit notes
• 問題3: transcribed from text witnesses, not the booklet image; see the section-specific audit notes
• 問題4: transcribed from text witnesses, not the booklet image; see the section-specific audit notes
• 問題5: transcribed from text witnesses, not the booklet image; see the section-specific audit notes
• 問題6: transcribed from text witnesses, not the booklet image; see the section-specific audit notes
• 問題7: transcribed from text witnesses, not the booklet image; see the section-specific audit notes
• 問題8: Passage (4) final clause 日記のとっては is grammatically anomalous (expected 日記にとっては) but is what the page prints at full resolution; kept verbatim as a likely source retype typo.
• 問題8: Passage (4) 配慮を動かす and the published original's 働かす differ; the rendered PDF unambiguously shows 動 so it is kept as printed, but this is a source-level retype defect.
• 問題8: Furigana over 盗み読ま could in principle exist at a resolution higher than both available renders, but neither the full-page nor the double-resolution half shows any legible kana there, so none was added.
• 問題8: Email (passage 2) paragraph segmentation relies on interpreting blank-line gaps as paragraph boundaries and intra-block line breaks as <br>; an alternate convention reading could merge or split differently, but the gap measurements strongly support the chosen structure.
• 問題9: transcribed from text witnesses; see the section-specific audit notes
• 問題10: 宙 ruby reads ちゅう and 浮世 ruby reads うきよ from the small inline rt glyphs; confident but rt are tiny.
• 問題10: Source contains several genuine retype typos (何度が, ためんに, とられている, 最もし) preserved as printed per convention; not corrected.
• 問題11: Passage B contains several apparent retype typos (説得; しばし; missing particle before 可能; 安全を安全に for 安心) that are transcribed as printed per convention; they are intentional fidelity to the flawed source, not OCR errors.
• 問題12: The book title is canonically 『新しい当たり前――暮らし…』 with a 2-em dash, but this retypeset clearly prints a single ― (one character cell) — kept as printed per convention; a downstream consumer expecting the canonical title should be aware.
• 問題12: 下を向いていればいるほど was reconstructed from the image after both transcribers erred; the いれば↔いいる region is low-contrast but the zoomed crop is unambiguous.
• 問題12: The ① underline target mismatch (passage ご馳走様 vs stem ごちそうさま) is a genuine source inconsistency, preserved verbatim; not an OCR error.
• 問題13: The ② 西川 fee (5,000円) is shown for a row whose body no longer mentions a plot size; this matches the print but is internally odd for the source document (a retypeset artifact, not an OCR uncertainty).
• 問題13: 貸し出を in the ④ row is missing the trailing し of 貸し出し; transcribed as printed but could be a retype slip rather than an intended form.
• 問題13: Table top-left cell transcribed as empty <th></th>; if downstream tooling requires a placeholder it may need a full-width space, but the image shows no glyph there.

Source-text quirks the OCR transcribed verbatim (printed typos/recall artifacts; the ones judged to be retype errors were corrected in the layer below) — 27 noted at OCR time:

• 問題1: q5: stem reconstructed as 主張を貫いた (txt garbled 出張を員いた); answers confirm 貫いた→つらぬいた.
• 問題2: q8 answer 4: とりわけ (JSON garbled どりわけ).
• 問題3: q15 head word: 雑踏 (JSON garbled 秦踏).
• 問題5: q26: stem reconstructed as 時期（　）時期だけに (txt garbled first word to 時間; JSON and natural grammar 「時期が時期だけに」 support 時期).
• 問題5: q29: answers were missing in the JSON witness; recovered from the txt witness (思う/思っている/思われる/思わせる).
• 問題5: q32 answer 3: 待たされるかと思いきや (txt garbled 思いきゃ).
• 問題6: Exact position of ★ among the 4 blanks could not be confidently recovered from the OCR/retype witnesses (both heavily garbled the blank run). Blanks are rendered as four slots with ★ placed at the 3rd slot per the JSON witness layout; the printed position should be re-verified against the image if available.
• 問題6: q37 answer 4: チームだろうと (JSON garbled チームだろうとど).
• 問題7: Both witnesses garbled 吠える throughout as 咲える/惨える/咆える and 犬 as 大; normalized to 吠える / 犬.
• 問題8: Passage (3): the source prints 'スレ違い' with katakana スレ (likely a retype typo for すれ違い); transcribed as printed. Verified against p11-t.
• 問題8: Passage (4): the inline reference mark is printed as （注） (no number) while the gloss is labeled （注1）; this inconsistent numbering is transcribed exactly as printed.
• 問題8: Passage (4): the source prints 様々な配慮を動かす (kanji 動), confirmed against p12-t; transcribed as printed even though the published original uses 働かす.
• 問題8: Passage (4): the printed clause is 「これは限りなく私的な記録であるはずの日記のとっては矛盾である」; 日記のとっては is an apparent retype typo for 日記にとっては but is transcribed exactly as printed (verified against full-resolution p12.png).
• 問題8: Q47 answer 2 prints '確認てほしい' (likely retype typo for 確認してほしい); transcribed as printed.
• 問題9: 問題9 has three passages: (1) q50-52, (2) q53-55, (3) q56-58. Reconstructed from two OCR witnesses: input/N1_2013-12/5Mcv4He-jlpt-freq-test-and-answers.txt (cleaner text layer, used as base) and input/N1_2013-12/5Mcv4He-jlpt-freq-exam-and-answers.txt field exam_sections.文法 (heavily garbled), cross-checked against the Chinese kaisetsu quotes. The exam is image-based; no furigana were printed over the 問題9 passages in either witness.
• 問題9: Passage (2) Q53 ささいな思いつき: the txt witness reads 難細な, freq-repo unreadable; reconstructed as ささいな (the kaisetsu paraphrases it as 小小想法). FLAG: moderate uncertainty on this single word; both witnesses are garbled here.
• 問題10: Source retype typos transcribed as printed: passage 第3段落 これまで何度が出会ってきた (が, expected も); われわれの発信が不足しているためんに (ためんに, expected ために); Q60 stem どのようにとられているか (expected とらえている); Q60 option 1 最もし必要とされる (stray し).
• 問題11: Passage B prints 「安全を安全に繋げる」 (安全, a likely retype typo for 安心); both transcribers had silently corrected it to 安心 — restored to the printed 安全 per the transcribe-as-printed rule.
• 問題11: Passage B citation prints a stray closing 」 before 』: 『…フードシステム入門」』 — transcribed as printed; likely a retype typo.
• 問題11: Passage B citation title dash is a single 2-em dash glyph (―, U+2015), not doubled ――; transcribed as printed.
• 問題11: Passage B 注 gloss prints 「（注）フードシステムしばし：…」; 「しばし」 appears to be a retype typo (expected something like 「とは」); transcribed as printed.
• 問題11: Passage B sentence 「コストさえかければ技術的に安全度を上げること可能である」 is missing a particle (が) before 可能; likely a retype typo, transcribed as printed.
• 問題12: Underline-reference mismatch (transcribed each as printed): the passage prints the ① underlined span as 「ご馳走様」 (kanji 馳走+様), but question 65 prints the reference as ①「ごちそうさま」 (all hiragana).
• 問題12: 「ごちそう様を言いなさい」 in the dialogue is printed with kanji 様; transcribed as printed (likely a source retype slip for ごちそうさま).
• 問題13: Source typos retained as printed: 住復はがき (intended 往復はがき) and 市役所市民課市民農園可係 (the 可 is spurious; the 問い合わせ先 line correctly reads 市民農園係).
• 問題13: 貸し出を (in the ④ row) is printed without the trailing し of 貸し出し; transcribed as printed.
• 問題13: 問題13 contains only questions 69–70 as expected; plain (unboxed) numbers, consistent with this retypeset source. The 聴解 section (問題1…) follows on p24 but belongs to a different section and is ignored.

Corrections applied (18)

Evidence-based reconstructions of the original exam text. This section is the canonical correction log for the sitting.

• 問8 q47 answers[1]: 「確認てほしい」 → 「確認してほしい」
  • 確認てほしい is ungrammatical (verb 確認する requires して). Retype artifact dropping し.
  • evidence: Both text witnesses (5Mcv4He line 254 and jlpt-freq line 254) show 「確認してほしい」. [correct-retypes workflow (propose + adversarial verify), 2026-06-17]
• 問8 q49 passage: 「秘密な」 → 「秘密の」
  • Source uses 秘密の表現; 秘密な is a retype artifact.
  • evidence: Google Books (黒井千次『図書』2007年2月号): 「他人には読まれたくない秘密の表現であるともいえる」; text witness also shows particle の (「秘容の表現」). [correct-retypes workflow (propose + adversarial verify), 2026-06-17]
• 問8 q49 passage: 「配慮を動かす」 → 「配慮を働かす」
  • Original published source prints 働かす; 動かす is a substitution error (the downstream text witness shares the same error, but Google Books reflects the authoritative original).
  • evidence: Google Books (黒井千次『図書』2007年2月号): 「様々な配慮を働かす。万が一日記が盗み読まれたり、死後に他人の目に曝される…」 [correct-retypes workflow (propose + adversarial verify), 2026-06-17]
• 問8 q49 passage: 「日記のとっては」 → 「日記にとっては」
  • の is ungrammatical here; the idiom is 〜にとっては. Retype artifact.
  • evidence: Google Books (図書): 「日記にとっては矛盾である」; text witness line 270 also shows 「日記にとっては」. [correct-retypes workflow (propose + adversarial verify), 2026-06-17]
• 問10 q59 passage: 「何度が出会って」 → 「何度か出会って」
  • が is a substitution for か; 何度か (several times) is required.
  • evidence: jlpt-freq witness line 421: 「これまで、何度か出会ってきた」. [correct-retypes workflow (propose + adversarial verify), 2026-06-17]
• 問10 q59 passage: 「ためんに」 → 「ために」
  • ためんに is a non-word; correct form is ために. Stray ん is a retype artifact.
  • evidence: jlpt-freq witness line 424: 「われわれの発信が不足しているために」. [correct-retypes workflow (propose + adversarial verify), 2026-06-17]
• 問10 q60 question: 「どのようにとられているか」 → 「どのようにとらえているか」
  • とられている is nonsensical; intended verb is とらえる (to perceive/regard).
  • evidence: jlpt-freq witness line 439: 「近年の音楽研究者は、音楽をどのようにとらえているか」. [correct-retypes workflow (propose + adversarial verify), 2026-06-17]
• 問10 q60 answers[0]: 「最もし必要」 → 「最も必要」
  • Stray し; correct phrase is 最も必要とされる.
  • evidence: jlpt-freq witness line 440: 「社会の状況が悪化したときに最も必要とされる」. [correct-retypes workflow (propose + adversarial verify), 2026-06-17]
• 問11 q63 passageB: 「安全を安全に」 → 「安全を安心に」
  • Passage contrasts 安全 (objective) with 安心 (subjective); 安全を安心に繋げる is intended.
  • evidence: jlpt-freq witness line 470: 「安全を安心に繋げる難しさ」. [correct-retypes workflow (propose + adversarial verify), 2026-06-17]
• 問11 q63 passageB: 「上げること可能」 → 「上げることが可能」
  • Missing subject particle が before 可能; retype artifact.
  • evidence: jlpt-freq witness line 468: 「コストさえ掛ければ技術的に安全度を上げることが…」. [correct-retypes workflow (propose + adversarial verify), 2026-06-17]
• 問11 q63 passageB: 「入門」』」 → 「入門』」
  • Stray duplicate closing bracket 」 before 』 in the citation; retype artifact.
  • evidence: jlpt-freq witness line 474 shows a single closer; book title 『…フードシステム入門』. [correct-retypes workflow (propose + adversarial verify), 2026-06-17]
• 問11 q63 passageB: 「フードシステムしばし：」 → 「フードシステム：」
  • The 注 glosses the term フードシステム; しばし is a spurious artifact.
  • evidence: jlpt-freq witness line 475: 「(注) フードシステム : 食品が生産者から消費者に届くまでの流れ」. [correct-retypes workflow (propose + adversarial verify), 2026-06-17]
• 問12 q65 passage: 「ごちそう様を」 → 「ごちそうさまを」
  • Hybrid ごちそう様 (hiragana stem + kanji 様) is a retype slip; original prints all-hiragana ごちそうさま.
  • evidence: Google Books (松浦弥太郎『あたらしいあたりまえ。』および『今日もていねいに。』): 「ごちそうさまを言いなさい」. [correct-retypes workflow (propose + adversarial verify), 2026-06-17]
• 問13 q69 passage: 「住復はがき」 → 「往復はがき」
  • 住復 is a non-word; standard term is 往復はがき (return postcard).
  • evidence: jlpt-freq witness line 614: 「往復はがきに…」; standard term 往復はがき. [correct-retypes workflow (propose + adversarial verify), 2026-06-17]
• 問13 q69 passage: 「市民農園可係」 → 「市民農園係」
  • Spurious 可; the office is 市民農園係, as printed in this document's own 問い合わせ先 line.
  • evidence: Internal consistency: same passage's 問い合わせ先 line reads 市民農園係; jlpt-freq witness line 617 also shows 市民農園係. [correct-retypes workflow (propose + adversarial verify), 2026-06-17]
• 問12 q68 passage: 「何度も試した僕の保障つきです」 → 「何度も試した僕の保証つきです」
  • 文脈は「私が請け合う・保証する」の意味で、guaranteeを表す語は『保証』。『保障』は権利・安全などを守る意味（社会保障）で、ここでは不適。同音異義の取り違えで、リタイプ系統で共有された誤りと判断。
  • evidence: Category homophone, confidence medium. [retype-corrections-pass2 (propose + adversarial verify), 2026-06-20]
• 問12 q65 passage: 「大事な話しをするのはやめましょう」 → 「大事な話をするのはやめましょう」
  • ここでの『話』は名詞（するの目的語）なので標準の送り仮名は『話』。同じ問12本文中で『相手の目を見て話をすると』『な話をして』のように名詞は一貫して『話』と表記されており、『話し』は内部不整合。リタイプの送り仮名付与ミス。
  • evidence: Category orthography, confidence medium. [retype-corrections-pass2 (propose + adversarial verify), 2026-06-20]
• 問13 q69 passage: 「農業用大型機械の貸し出を行っています」 → 「農業用大型機械の貸し出しを行っています」
  • 名詞『貸し出し』の末尾『し』が欠落。『貸し出を行う』は非標準で文法的に不自然。行政の募集案内文として標準形は『貸し出し』。脱字によるリタイプ誤り（quirk 42該当）。
  • evidence: Category dropped-char, confidence medium. [retype-corrections-pass2 (propose + adversarial verify), 2026-06-20]