Kagawa Entrance Japanese common mistakes
Kagawa Entrance Japanese · Understand the traps and how to avoid them.
When asked 「傍線部の理由として最も適当なものを選べ」, the candidate picks the option that repeats words from the passage near the line rather than the option that states the causal relation the passage actually develops, so a paraphrased correct option loses to a verbatim-sounding distractor.
Under time pressure the eye matches surface vocabulary instead of reconstructing the logic. Expository passages in this exam typically place the reason across a paragraph boundary (a claim followed by しかし/つまり/したがって), and the distractor is built from the same nouns as the line, so pattern-matching feels safer than reasoning.
What to do instead
Before looking at the choices, write a one-sentence answer in your own words in the margin using the connectives around the line (原因→結果, 主張→根拠). Then eliminate any option that contains a word from the passage but not the relation you wrote. Practise on 200–300 character expository paragraphs, timing 90 seconds per item.
The candidate treats 具体例 and 一般論 as the author's main claim, choosing a choice that summarises an anecdote or a quoted opposing view instead of the 筆者の主張.
Japanese expository writing often opens with an example or a counter-view and delays the thesis until after しかし/むしろ/大切なのは. Readers who read linearly and stop at the first strong statement anchor on the wrong sentence.
What to do instead
After the first read, mark with brackets where the 意見 (opinion) starts and where 例示 (for example, たとえば, ～という例がある) starts. Confirm the final paragraph's 結論 and check that your chosen option is compatible with it. Drill by summarising each paragraph in seven characters or fewer and labelling it 主張/例/対比.
In 文学的文章 the candidate answers 心情 questions with the emotion named in the text (うれしい, さびしい) and ignores the gap between what the character says and what the narration shows, missing irony or suppressed feeling.
Literary items in this exam frequently hinge on 心情の変化 and on 言動の裏側. Candidates read dialogue literally and do not connect it to the preceding scene description or to the 比喩 that carries the real feeling.
What to do instead
For every 心情 question, locate three things: the triggering event, the character's 言動 (speech or action), and the narrative description (情景・比喩). The answer must be supported by the third. Write the change as 「A→B」 before choosing, and reject any option that only restates the dialogue.
The candidate translates 古文 word by word into modern Japanese and then answers from that translation, producing answers that ignore 係り結び, 敬語の方向, and 已然形の働き.
Classical grammar is taught as lists, so candidates recognise forms in isolation but do not use them as reading signals. A ぞ/なむ/や/か sentence changes its meaning at the 連体形/已然形 ending, and honorifics tell you who is doing what to whom - both invisible in a literal gloss.
What to do instead
Train a fixed routine: circle 係助詞 and check the ending; underline 敬語 and draw an arrow from giver to receiver (尊敬=subject acts, 謙譲=speaker acts toward a superior, 丁寧=addressee). Read the 主語補助 line before answering. Do ten short 古文 passages applying only this routine before attempting full questions.
In 漢文 the candidate reads 返り点 in the wrong order - especially 一二点 versus 上下点 and レ点 combined with 一 - and then interprets the sentence with the wrong subject or object.
The reordering rules are mechanical and easy to memorise passively but slow to apply live. Candidates also skip 置き字 and mis-handle 再読文字 (未・将・当・応・須・猶・盍), which changes tense and modality.
What to do instead
Physically number the reading order above each character before translating (1, 2, 3…), then read the numbers aloud. Keep a one-page card of 再読文字 with their two readings and meanings, and apply it to every sentence containing one. Practise five-character and eight-character lines until numbering takes under 20 seconds.
On 漢字の読み書き the candidate writes a homophone with the right sound but wrong meaning (e.g. 対象/対照, 関心/感心, 適応/適応 vs 適用) or mis-strokes a 音読み/訓読み distinction in a compound.
Preparation focuses on rare kanji while the exam mostly tests common kanji in context. Candidates memorise readings without the collocation, so the 送り仮名 and the neighbouring kanji do not constrain the choice.
What to do instead
Study kanji in two-character compounds with a short phrase showing meaning, and always write the whole compound, never the single character. Keep an error log of only the pairs you personally confused, and re-test them after 3 days and after 10 days. Check 送り仮名 last, since it is the most common silent loss.
On 語句・文法 items the candidate judges correctness by how natural a sentence sounds rather than by the grammatical category being tested, so ら抜き言葉, れ足す言葉, and 誤用 of 敬語 pass unnoticed.
Spoken Japanese tolerates forms the written standard rejects. Candidates also confuse 品詞 classification (形容動詞 vs 名詞＋だ, 連体詞 vs 形容詞) because they were taught labels without tests for them.
What to do instead
For every item, name the category being tested before answering (活用形, 品詞, 敬語の種類, 慣用句). Apply a two-question test: can it take だ? can it take い? For 敬語, ask who is honoured. Build a personal list of the ten 誤用 you actually miss and review it weekly.
In the 作文 section the candidate answers the prompt with a general essay about the topic, without taking a clear position, without a supporting example from personal experience, and without the required 字数 or 段落 structure.
Under exam conditions candidates write what is easy rather than what the 課題 asks, and they run out of time for the final check. The result loses marks on 課題対応 and 構成 even when the Japanese is correct.
What to do instead
Spend 3 minutes on a 構成メモ: 意見 (one sentence) → 理由 → 具体例 → まとめ, with a target 字数 per section. Write the conclusion sentence early so the essay cannot drift. Reserve 3 minutes to count 字数 and check です・ます consistency and 主語述語 correspondence.
On 総合読解・情報の扱い items with multiple materials (a graph, a table, a notice, a short article) the candidate reads the prose first and treats the data as decoration, so the answer contradicts a number in the figure or ignores a condition in the notes.
The prose feels like the real text and the figure feels like an illustration. But these items are designed so that the correct option requires combining a stated condition with a value or date in the material.
What to do instead
Read the question first, then the figure's title, axis labels, units, and 注記 before the prose. Write the decisive number or condition beside each option as you eliminate it. If an option cannot be checked against the material, it is a distractor.
The candidate allocates time evenly across sections and arrives at 古文・漢文 and 作文 with minutes left, so the last items are answered by guessing and the essay is unfinished.
The reading sections expand to fill whatever time they are given, and candidates do not know their own per-item pace. Classical and composition items are disproportionately costly to rush.
What to do instead
Set fixed checkpoints in a practice run (e.g. 現代文 within 45% of total time, 古文漢文 by 70%, composition start by 75%) and record actual split times after each mock. Practise skipping a stuck item within 60 seconds and returning, so no section is entered late.