How to Reference Sources in Non-Latin Scripts on Wikipedia

Wikipedia is built on sources. But what happens when the best source for a fact is written in Arabic, Japanese, or Russian? Many editors assume that if a source isn’t in English, it can’t be used. That’s wrong. Wikipedia’s policy explicitly allows sources in any language, as long as they’re reliable and relevant. The real challenge isn’t whether you can use them-it’s how to cite them correctly so other editors can verify them.

Why Non-Latin Scripts Matter

Over 60% of the world’s population uses writing systems that aren’t based on the Latin alphabet. That includes Chinese, Arabic, Hindi, Cyrillic, Thai, and dozens more. If Wikipedia only accepted sources in English or other Latin-script languages, it would ignore the vast majority of human knowledge. Think about it: a fact about the history of the Qing Dynasty might be best documented in a Chinese academic journal. A study on Islamic law might be published in Arabic. A report on Russian military mobilization might come from a regional newspaper in Moscow. Excluding these sources isn’t neutrality-it’s bias.

Wikipedia’s guideline on reliable sources doesn’t require English. It requires verifiability. That means any reader, no matter their language, should be able to check your claim. If you cite a Japanese source, you’re not just helping Japanese speakers-you’re making the article more trustworthy for everyone.

How to Cite a Non-Latin Source Correctly

There are three basic rules for citing sources in non-Latin scripts:

1. Keep the original title in its native script
2. Provide a transliteration in Latin letters
3. Include an English translation of the title

For example, if you’re citing a Russian article titled «Роль космических технологий в развитии сельских регионов», you’d write it like this:

Роль космических технологий в развитии сельских регионов (Role kosmicheskikh tekhnologiy v razvitii selskikh regionov) - “The Role of Space Technologies in the Development of Rural Regions”

This format lets anyone-whether they read Cyrillic or not-find the source. A Russian speaker can copy-paste the original title into a search engine. An English speaker can search the translated title. And someone who knows Russian but not English can use the transliteration to look it up.

Transliteration Systems You Must Know

Not all transliterations are equal. Using the wrong system can make a source impossible to find. Wikipedia follows standardized systems for each script:

• Chinese (Mandarin): Use Hanyu Pinyin. Don’t use Wade-Giles or older systems.
• Arabic: Use the ALA-LC Romanization system.
• Russian, Ukrainian, Bulgarian (Cyrillic): Use the BGN/PCGN system.
• Japanese: Use Hepburn romanization for names and titles.
• Korean: Use Revised Romanization of Korean.
• Greek: Use the Greek Transliteration System (ISO 843).

These systems aren’t optional. If you write “Moskva” instead of “Moskva” for Russian, you’re using an outdated form. If you write “Beijing” instead of “Peking” for Chinese, you’re using the correct modern form. Getting this right matters. Editors will flag incorrect transliterations as unreliable.

What to Do When You Can’t Find an English Translation

Sometimes, a source doesn’t have an official English translation. That’s okay. You still cite it-but you translate the title yourself.

Don’t guess. Use tools like Google Translate or DeepL to get a rough translation, then refine it. For example, if you find a Thai article titled “ผลกระทบของการเปลี่ยนแปลงสภาพภูมิอากาศต่อการเกษตรในภาคเหนือ,” a machine translation might say “The impact of climate change on agriculture in the north.” That’s fine. But if you know the region is “Northern Thailand,” update it to: “The impact of climate change on agriculture in Northern Thailand.”

Always note that the translation is your own. Add “(translated by editor)” after the English title. This tells others you didn’t just copy a machine translation without checking.

How to Handle Non-Latin Script URLs

Many websites use non-Latin characters in their URLs. A news site in Arabic might have a link like https://example.com/أخبار/المناخ. That’s fine. Wikipedia allows internationalized domain names (IDNs). But here’s the catch: you must also provide a Punycode version.

Punycode turns Unicode URLs into ASCII. For the example above, the Punycode version would be https://example.com/xn--4gq1a3d5a/. Why? Because some Wikipedia templates and bots can’t handle Unicode URLs properly. Always include both:

https://example.com/أخبار/المناخ (Punycode: https://example.com/xn--4gq1a3d5a/)

This ensures the link works across all systems, even if someone’s browser or device doesn’t support non-Latin URLs.

Using Non-Latin Sources in Article Text

You can quote directly from non-Latin sources-but only if you’re fluent. If you’re not, don’t translate it yourself unless you’re confident. Instead, use a secondary source: cite a published English-language book or journal that references the original.

For example, if you’re writing about a 1987 Korean protest and the best source is a Korean memoir, but you don’t read Korean, find a scholarly article in English that analyzes that memoir. Cite the English article instead. This is called “secondary sourcing,” and it’s perfectly acceptable.

But if you do quote directly, always provide the original text in its native script, followed by your translation in square brackets:

“우리는 자유를 위해 싸웠다” [We fought for freedom].

This keeps the source authentic and lets others verify the translation.

Common Mistakes to Avoid

Even experienced editors mess this up. Here are the top three errors:

1. Using only the English translation - If you cite “The History of China” without giving the original Chinese title, you risk citing the wrong book. There are dozens of books with similar English titles.
2. Using inconsistent transliteration - Mixing “Beijing” and “Peking” in the same article confuses readers and breaks consistency.
3. Ignoring script order - Always put the original script first, then transliteration, then translation. Reversing this order makes the citation look sloppy.

Also, don’t assume a source is unreliable just because it’s in a non-Latin script. A peer-reviewed journal in Japanese is just as credible as one in English. The language doesn’t determine reliability-peer review, publisher reputation, and author credentials do.

Tools That Help

Several tools make citing non-Latin sources easier:

• Wikidata - Many non-Latin sources have entries in Wikidata with standardized transliterations and translations. Link to them.
• Google Translate - Use it for quick translations, but always double-check with native speakers or academic sources.
• Transliteration converters - Sites like Transliteration.org or ALALC Converter auto-convert text into the correct format.
• Wikipedia’s citation templates - Use {{cite web}}, {{cite journal}}, etc. They support non-Latin scripts and auto-format transliterations.

Don’t rely on automated tools alone. Always review the output. A machine might transliterate “Татьяна” as “Tatiana” when the correct form is “Tatyana.” Small details matter.

Why This Matters for Global Knowledge

Wikipedia isn’t just an English website. It’s a global project. In 2025, over 300 language editions of Wikipedia exist. The Chinese edition has more articles than the English one. The Arabic edition is growing faster than any other. If we only use Latin-script sources, we’re not just excluding non-English speakers-we’re excluding entire cultures from being represented accurately.

When you cite a source in Hindi, Arabic, or Korean, you’re not just following a style guide. You’re helping to correct centuries of Western-centric knowledge gaps. You’re saying: this fact matters, no matter what alphabet it’s written in.

And if you’re an editor who’s nervous about citing non-Latin sources? Start small. Pick one article. Find one source in another script. Cite it right. Then do it again. The more you do it, the easier it becomes. And the more accurate Wikipedia becomes for everyone.