Gunma Entrance Japanese common mistakes
Gunma Entrance Japanese · Understand the traps and how to avoid them.
When asked 傍線部「〜」とあるが、どういうことか, the candidate copies a phrase that appears near the line verbatim, or picks the choice that shares the most words with the passage, instead of restating the line's content in their own words.
The line usually contains a concrete image or example (e.g. 比喩や具体例), while the answer choices paraphrase it abstractly. Word-matching feels safe and is fast, so under time pressure the candidate anchors on surface overlap and never checks whether the chosen option actually explains the mechanism the line illustrates.
What to do instead
Before looking at the choices, cover them and write a 15–25 character paraphrase of the line in the margin, including the subject and the predicate (「〜が〜だということ」). Then eliminate any option that only repeats vocabulary from the passage without adding the 'what does this mean' content. Train by re-labelling 10 past 傍線部 questions: mark which option is a 言い換え and which are 本文の言葉のコピー.
For 理由説明 questions the candidate selects a true statement about the topic that is not the reason the author gives, or picks a cause-and-result pair with the direction reversed (taking the effect as the cause).
Expository passages contain many correct background facts, and the candidate treats 'this is true and related' as 'this is the reason'. Reversal happens because Japanese reason sentences put the conclusion first and the reason after (「〜のは…からだ」), so a reader tracking only content, not the 因果の向き, flips the arrow.
What to do instead
For every 理由 question, locate the explicit marker (から・ので・ため・なぜなら・というのは…だ) and draw an arrow from reason to conclusion in the margin. Reject options that are factually correct but not connected by a marker to the target claim. Practise with 5 passages, listing each 因果 pair as 'AだからB' and checking the option says A→B, not B→A.
In 小説 the candidate answers 心情 questions by choosing the option that names the emotion most strongly (悲しみ・怒り・不安), ignoring the physical descriptions, dialogue, and silences the author actually uses to convey feeling.
Emotion labels are easy to recognise and feel like the 'point' of the question, but literary texts rarely state feelings directly; they show them through 動作・表情・会話・情景. The candidate reads for plot and then guesses a mood instead of reading the descriptions as evidence.
What to do instead
Underline every 動作・表情・台詞・情景描写 in the relevant scene and write beside each what it implies (e.g. 「茶碗を置く音がした」→ 動揺・気まずさ). Choose the option that is supported by at least two of these textual signals, and reject options that no description backs. Do this on 3 short stories, then compare with the answer key to see which signals you missed.
The candidate ignores 場面の転換 and attributes a line of dialogue or thought to the wrong speaker, or reads a later scene as if the relationship were still the one from the opening.
Japanese fiction uses sparse dialogue tags and relies on 敬語, 呼称, and paragraph breaks (改行・一字下げ) to signal who speaks and when time or place shifts. A reader skimming for content misses these structural cues and carries an outdated emotional state into a new scene.
What to do instead
On first read, bracket each scene at every 場面転換 (time words like 翌朝・その夜, place changes, blank-line breaks) and note the speaker of each quoted line in the margin using 敬語 and 呼称 as clues. Answer 心情 questions only with evidence from within the same bracketed scene.
In 古文 the candidate translates word-by-word into modern Japanese and answers from that rough gloss, ignoring 敬語の種類 (尊敬・謙譲・丁寧) and who-does-what-to-whom, so subject and object are swapped.
Classical Japanese omits subjects and marks respect through verbs; without identifying 敬語, the candidate cannot tell who is honoured and therefore who is the actor. School grammar feels like memorisation, so it is skipped in favour of 'getting the gist', which produces confidently wrong answers.
What to do instead
For every 古文 question, first circle all 敬語 verbs and label them 尊敬 (主語を高める) or 謙譲 (目的語を高める), then assign 主語・目的語 explicitly in the margin before translating. Memorise the high-frequency 敬語 verbs (給ふ・侍り・参る・申す・仰す・御覧ず) with their type. Practise on 3 short passages, writing 'AがBに〜する' for each sentence.
In 漢文 the candidate reads 返り点 and 送り仮名 only partially, reading the kanji in printed order or misapplying レ点, so the sentence is parsed with the wrong word order and the 置き字 and 否定・反語 are misread.
返り点 require reconstructing a reading order that is not the visual order; under time pressure the eye follows the printed sequence. 不・無・非・未 and 反語 (豈…乎、安…哉) change meaning drastically and are easy to skim past as single characters.
What to do instead
Rehearse the fixed rules until automatic: レ点 returns one character, 一二点 jump back over a span, 上下点 handle nested spans; read bottom-up within each marked span. Before answering, write the 訓読文 in full with 送り仮名, then circle every 否定語 and 反語・疑問の終助詞 and state in modern Japanese whether the sentence affirms or denies. Do 5 sentences daily, checking 訓読 against the model.
In 作文 the candidate writes a general essay about the topic (e.g. about the importance of effort) without answering the specific 課題, and the 段落 are not organised around the required 意見・理由・体験, so the piece loses points on 構成 and 課題適合.
Under 時間制限 the candidate starts writing immediately from a familiar theme, and pre-learned phrases fill space. The prompt's conditions (字数, 段落数, 体験を入れる, 意見を述べる) are read once and not checked during writing, so the structure drifts.
What to do instead
Spend the first 3–4 minutes on a 構成メモ: state the 課題 in one line, then a 3-part plan (主張→理由・体験→まとめ) with a keyword for each 段落. Keep the memo visible and check after each paragraph that it answers the 課題. After writing, verify 字数 within ±10% and that 段落の数 and required elements are present before polishing 表記.
The candidate confuses 音読み and 訓読み in 漢字の読み questions, or writes a 同音異義語 with the wrong kanji (e.g. 干渉/感称, 対象/対称, 関心/感心) because they choose by sound alone.
Many exam kanji have both readings and appear in 熟語 whose reading is fixed by convention; the candidate applies a familiar reading from a different word. For 書き取り, homophones share pronunciation, so relying on sound rather than meaning and 送り仮名 produces plausible but wrong characters.
What to do instead
Build a 同音異義語 list organised by reading (かんしん・たいしょう・こうしょう etc.) with a short meaning phrase for each character, and review it weekly. For 書き取り, write the whole 文 and check the 送り仮名 and 意味 fit before finalising; for 読み, mark whether the word is 音読み or 訓読み and confirm against a 熟語 list rather than a single-character guess.
For 語句・文法 questions the candidate judges correctness by how natural a sentence 'sounds' rather than by the grammatical rule, so they accept ら抜き言葉, misused 敬語 (ご覧になられる, お伺いいたします used wrongly), and incorrect 接続 (「〜に関して」vs「〜に対して」) and fail 文法的に正しいものを選べ items.
Everyday speech contains forms the exam marks as incorrect, and fluency in casual Japanese makes these feel right. Grammatical terminology (活用形, 助詞の働き, 敬語の種類) is often only half-remembered, so the candidate substitutes intuition for rule-checking.
What to do instead
For each option, identify the specific rule being tested and name it: 敬語の種類 (尊敬・謙譲・丁寧), 助詞の用法, 活用形の接続, 慣用句の形. Reject an option only when you can state the rule it violates. Keep a one-page rule sheet (ら抜き, 二重敬語, 誤用の慣用句) and drill 20 items per session, writing the reason next to each answer.
In 漢字・語句の知識 items the candidate guesses at 四字熟語, 慣用句, and ことわざ by overall impression, mixing up pairs with similar meanings (「一朝一夕」と「一日千秋」, 「情けは人のためならず」の意味) or choosing a 類義語 that does not fit the sentence's register.
These are memorised units, and partial memory produces a feeling of recognition without exact meaning. Similar-looking four-character compounds and proverbs whose literal reading differs from their actual meaning are especially prone to confident errors.
What to do instead
Maintain a card list of 四字熟語・慣用句・ことわざ with the exact meaning in one short phrase and one example sentence, grouped by confusable pairs (e.g. 似た形, 似た意味). Review by self-testing meaning→expression as well as expression→meaning, and for 語句の使い方 items substitute the phrase into the sentence and check the resulting meaning matches the card.