Doing a PhD in a Language You Can Barely Order Lunch In

I grew up speaking Arabic and French. I do my research in English. I live in a city where daily life happens in Mandarin. On a good day this feels like a superpower. On a bad day I stand in front of a vending machine losing an argument with it.

Research English is its own dialect

Nobody warns you that academic English is not really English. It is a specialised dialect with strange customs, where "we note that" means "pay attention", "somewhat limited" means "bad", and "future work will address this" means "we ran out of time".

Learning to read it took months. Learning to write it took considerably longer, and I am still working on it. The trick that helped most was collecting sentence patterns rather than words. How do people phrase a limitation? How do they introduce a contribution without sounding like they are bragging? I keep a file of these. It is not glamorous and it works.

The lab runs on a third language

Papers and code are in English. Lunch, logistics, and the important gossip about which printer is lying to you are all in Mandarin. My Mandarin arrived slowly and in a strange order: I could discuss signal processing before I could confidently buy fruit.

What genuinely helped was giving up on being correct. I spent my first months silent, constructing grammatically perfect sentences in my head that I never said out loud. The moment I started cheerfully producing broken sentences at people, I improved. Being understood beats being correct. Correct comes later, and it comes faster if you were talking the whole time.

What multilingualism actually gives you

The practical benefit is not the languages. It is the tolerance for not understanding.

When you spend years being the person in the room who caught roughly 70% of what was said, you develop a specific skill: staying calm, holding the gaps open, and reconstructing the meaning from context instead of panicking. That is very close to what research feels like. You read a paper you only partly understand. You inherit code with no comments. You stare at a result that makes no sense yet.

Being comfortable in that fog is a real advantage, and I did not get it from a course. I got it from four years of nodding politely and figuring it out afterwards.

And the small things

I still count in French. I still do arithmetic in French, actually, which is mildly inconvenient in a lab meeting. I dream in whichever language I spent the most time in that week. And I have made peace with the fact that there are jokes I will never fully land in any of my four languages.