How to Copy Mixed English and Japanese Text Without Corrupting Characters

Summary

• Mixed English/Japanese text gets corrupted mainly due to the wrong character encoding, font substitution, or a “plain text” paste that strips Unicode formatting.
• First confirm the text is actually Unicode (UTF-8/UTF-16) end-to-end: source app, clipboard, destination app, and file encoding if you are saving.
• Use the right paste mode for the destination: “Keep Source Formatting” (rich text) vs “Keep Text Only” (plain text) can change what survives.
• Browser and desktop apps can behave differently; test in a neutral target (Notepad/TextEdit) to isolate whether the problem is the source or destination.
• A quick result check is to paste into a Unicode-safe editor and verify Japanese characters (kana/kanji) remain intact and searchable.

When you copy text that mixes English and Japanese (for example: “Meeting at 3pm - 会議は15時です”), the characters can turn into question marks (????), boxes (□), or garbled symbols. That corruption almost always comes from one of four places: the source app copied in a non-Unicode encoding, the destination app expects a different encoding, the paste mode strips or transforms characters, or the file you save uses the wrong encoding.

This guide walks you through a reproducible test, then gives application-specific steps for Windows, macOS, and common apps (Word, Google Docs, Excel/Sheets, email, code editors). It ends with a simple “result check” so you can confirm the fix.

Reproducible example (use this to test your setup)

Copy this exact text (it includes ASCII, kana, kanji, punctuation, and full-width characters):

What “good” looks like: After pasting, you should still see 次の会議, 資料, 了解, and the half-width katakana ｱｲｳ without turning into ??? or boxes.

Step 1: Isolate where corruption happens (source, clipboard, or destination)

A. Paste into a neutral target first

This tells you whether the text is already broken before it reaches your destination app.

• Windows: Paste into Notepad (plain text). If Japanese is corrupted here, the issue is likely the source app or the copy operation.
• macOS: Paste into TextEdit set to plain text mode (Format > Make Plain Text). If it breaks here, suspect the source app or copy operation.

B. Compare “normal paste” vs “paste as plain text”

Many apps offer multiple paste modes. If one mode corrupts characters and another doesn’t, you’ve found the lever to fix it.

• Normal paste may bring rich text (RTF/HTML) and preserve typography.
• Paste as plain text strips formatting; it should still preserve Japanese characters if the destination supports Unicode, but it can expose encoding limitations in older apps.

C. If you are saving to a file, test the file encoding

If copy/paste looks fine in-app but becomes corrupted after saving and reopening, the file encoding is the culprit (for example, saving as ANSI/“Western” instead of UTF-8).

Why mixed English/Japanese text gets corrupted

• Wrong encoding on save/export: A file saved as “ANSI”, “Windows-1252”, or another legacy encoding cannot represent Japanese characters, so they become ? or boxes.
• Destination app is not Unicode-safe in that field: Some older tools or specific input fields only accept limited character sets.
• Font substitution: The characters are present but displayed as tofu (□) because the chosen font lacks glyphs for Japanese. This is display corruption, not data corruption.
• Paste mode transforms content: Pasting “unformatted text” can remove markup that the destination relied on, or the destination may sanitize characters.
• Browser vs desktop differences: Web apps may sanitize pasted HTML; desktop apps may prefer RTF; the same clipboard content can be interpreted differently.

Fixes by scenario (copy/paste vs saving)

Scenario 1: Japanese turns into “????” or random symbols

This is usually an encoding mismatch (not just a font issue).

• If you are pasting into an older app: Try pasting into a modern Unicode editor first (Notepad on Windows, TextEdit on macOS in plain text), then copy again from there into the destination. This can force a clean Unicode copy.
• If you are saving to a file: Save as UTF-8 (or UTF-8 with BOM if a specific Windows tool requires it). Avoid “ANSI” for mixed Japanese text.
• If you are copying from a PDF: Some PDFs copy text with odd encodings. Try copying from the original source (Word/Docs) if available, or use a different PDF viewer and compare results.

Scenario 2: Japanese shows as boxes (□) but copying elsewhere works

This is often a font/glyph problem.

• Change the destination font to one that includes Japanese glyphs (many system fonts do; the exact name varies by OS).
• If the destination app allows per-field fonts (forms, design tools), set that field’s font explicitly.
• Confirm by copying the “boxed” text and pasting into a Unicode editor. If it appears correctly there, the underlying characters are intact.

Scenario 3: It looks fine after paste, but breaks after saving/reopening

This points to file encoding on save/export.

• In code editors, choose “Save with encoding” and select UTF-8.
• In CSV exports, prefer UTF-8 and re-import using UTF-8 in the target program.
• If a tool offers multiple CSV encodings, test by reopening the saved file in a Unicode-capable editor and verifying the Japanese characters are still correct.

Application-specific steps (desktop vs browser)

Google Docs (browser)

• Copy source: Copy normally from the source app.
• Paste into Docs: Use normal paste first. If formatting causes issues, try “paste without formatting” (the shortcut varies by OS and browser).
• Result check: Use Docs search (Ctrl/Cmd+F) for a Japanese substring like “会議” to confirm the characters are real text (not images) and weren’t replaced.

Microsoft Word (desktop)

• Try different paste options: After pasting, Word may show a paste options control. Test “Keep Source Formatting” vs “Keep Text Only”.
• If boxes appear: Select the pasted text and change the font to one that supports Japanese glyphs.
• Result check: Copy the pasted Japanese text from Word and paste into Notepad/TextEdit plain text. If it remains correct there, the data is fine.

Excel / Google Sheets

• Direct paste: Paste into a cell and confirm Japanese characters render correctly.
• CSV warning: If you export to CSV and reopen in another program, corruption can happen on import. When importing, choose UTF-8 if prompted.
• Result check: Copy the cell content and paste into a Unicode editor; verify characters remain intact.

Email clients (webmail and desktop)

• Webmail compose windows: Try “paste without formatting” if the email editor mangles punctuation or spacing.
• Desktop email: If Japanese becomes boxes, change the compose font or switch the message format (HTML vs plain text) and retest.
• Result check: Send a test email to yourself and verify the received message displays Japanese correctly on both desktop and mobile.

VS Code / code editors

• Paste is usually safe: If characters corrupt after saving, check the file encoding setting and resave as UTF-8.
• Result check: Close and reopen the file, then search for a Japanese substring to confirm it persisted.

Browser vs desktop: what to do when they behave differently

If copying from a browser into a desktop app corrupts characters (or vice versa), do a three-way test:

• Copy from the source.
• Paste into a neutral Unicode editor (Notepad/TextEdit plain text).
• Copy again from the neutral editor into the destination.

If the neutral editor path works, the issue is likely how the source or destination interprets rich clipboard formats (HTML/RTF). Using the neutral editor forces a simpler plain-text Unicode transfer.

Result check: confirm you did not silently lose characters

• Visual check: Look for kanji/kana turning into ?, □, or missing characters.
• Search check: Search for a Japanese substring (e.g., “資料”). If search can find it, it is real text and likely intact.
• Round-trip check: Copy the pasted text again and paste into a second app (Notepad/TextEdit). If it survives the round trip, corruption is less likely.
• Normalization check (optional): If you rely on exact matching, compare half-width/full-width forms (e.g., ｱ vs ア, １ vs 1). Some transformations are not “corruption” but can affect matching.

Troubleshooting quick map (symptom → likely cause → fix)

Frequently Asked Questions

Back to FAQ Table of Contents

Back to FAQ Table of Contents

Back to FAQ Table of Contents

Back to FAQ Table of Contents

Back to FAQ Table of Contents

Back to FAQ Table of Contents

Back to FAQ Table of Contents

Back to FAQ Table of Contents

Related Guides

• What Is an AI Context Pack?
• How to Prepare Better Context for ChatGPT
• How to Organize Copied Text for AI Work
• Source-Labeled Context for AI: Why It Matters
• ChatGPT Context Pack for Consultants