When I was preparing my proposal, I searched with Chinese keywords, and everything that came up was Chinese review articles, with the same few references over and over.
My advisor said, “In your field, there is ten times more English literature than Chinese literature.”
I switched to English keywords, and the very first article was two years newer than the Chinese review.
The Problem Isn’t the Database—It’s the Keywords
In Chinese databases, a concept may have three or four translations; in English, the same concept often has only one mainstream term.
Searching in Chinese means looking only within “the portion of the literature that has been translated into Chinese.”
The key to cross-language searching is to translate Chinese concepts into the “accepted terms” in English, not to translate them literally.
Three Steps to Translate Keywords
My approach:
1. Pick out the core concepts from Chinese review articles (three to five)
2. Use paper translation to translate them into English, and look at English synonyms at the same time
3. Take the English terms back to the database and search, checking the number of hits and title relevance
The second step is the most error-prone: the academic community may not use the literally translated term.
For example, the literal translation of “interfacial bonding strength” is interface bonding strength, but this field more commonly uses interfacial adhesion. After translating it, you also need to see which term is used in the titles in the search results.
MuchSci’s paper translation gives you a translation, but you have to judge the synonyms yourself—I usually throw the translated term back into the search box and see which one returns more relevant hits.
Do It in Reverse, Too
After finding a batch of English literature, I do something in reverse: I write down the words that repeatedly appear in English titles.
These words are “the accepted terms in this field,” more accurate than any dictionary.
Translating forward once gets you through the door; only by noting a batch of terms in reverse do you truly get inside.
A Specific Example
I wanted to find literature on “material embrittlement in low-temperature environments.”
· Literal translation: low temperature embrittlement
· After searching, I found that this field more often uses cryogenic embrittlement and ductile-to-brittle transition
I “spotted” the latter two terms in the titles of the search results; they weren’t translated.
Combining Search Terms
Single-word searches are too broad, so I combine three blocks:
| Block | Content | Function |
| Object | Material/structure name | Limit scope |
| Condition | Temperature, load, environment | Limit scenario |
| Phenomenon | Failure mode | Limit the problem |
Take two or three synonyms for each block, connect them with OR, and connect the blocks with AND.
With a search string like this, the number of hits usually drops from several thousand to a few dozen.
It’s Not That Chinese Literature Is Useless
The value of Chinese review articles is that they “help you find terms”: they translate English literature into Chinese, so you can see the common translations of a set of concepts.
My current order is: first read two Chinese review articles to find terms, then search with English terms, and finally go back to the original English articles to verify.
A Few Takeaways
Cross-language searching is essentially “translating twice”: once translating concepts into English, and once translating English article titles back into Chinese to confirm whether they are what I need.
If the first translation is wrong, the search goes off track; if the second translation is wrong, you may read for a long time before realizing it’s not the same thing.
I run both translations through MuchSci’s paper translation, especially titles—**prepositions and modifying relationships in titles are the easiest to mistranslate.** MuchSci supports submitting multiple documents at once; I usually run a dozen or so titles together, which is faster than checking them one by one.