Translation Isn't Cheating. Bad Translation Is

At some point in your English journey, probably early, somebody handed you a commandment: stop translating in your head. Think in English! Real learners think in English! Translation is a crutch, a bad habit, training wheels you should be ashamed of.

And so, on top of all the normal difficulties of learning a language, you acquired a bonus difficulty: guilt. Every time you noticed your own language helping you (which, let's be honest, is constantly, because it's the language your entire mind is built in), you felt like you were doing English wrong.

If this is you, welcome, and let me offer some immediate relief: the commandment is half nonsense. I've been teaching English for over five years, and I've watched the "never translate" rule produce mostly anxiety and secret translators. Because here's the truth nobody puts on the poster: your first language is the single biggest asset you bring to English, AND it's responsible for your most stubborn errors. Both things. At the same time. Translation isn't a sin or a virtue. It's a tool, and like most tools, everything depends on which end you grab. Today: when your language is helping, when it's sabotaging, and how to tell the difference in the moment.

01 · The Big Idea

The Big Idea

Here's the whole distinction:

Unpack that slowly, because the entire post lives in it.

You already own a complete map of reality in your first language. You know what regret is, what a deadline feels like, the difference between borrowing and stealing. Twenty or thirty years of concepts, fully installed. When you use your language to reach a meaning fast (glancing at a translation to confirm that reliable means what you suspected), you're using that map, and it's magnificently efficient. An adult who refuses this shortcut and insists on learning every word from context alone, like a baby, is declining their biggest advantage out of principle. Babies take years. You don't have to.

But your language also contains a complete set of structures: word orders, verb patterns, which preposition goes where, which words count as one word. And these, exported wholesale into English, are the factory that produces your most persistent mistakes. "Can you borrow me your pen." "He said me that." "I have seen him yesterday." "An advice." Every one of those is a perfectly good structure from some other language, wearing English words. If you've read this blog for a while, you've met this villain in nearly every post; today it finally gets named properly. It's not translation. It's structural transfer: building English sentences on your language's blueprints, live, word by word.

So the commandment was aimed at something real. It just hit the wrong target. The problem was never that your language is present in your head. The problem is using it as a sentence factory instead of a meaning dictionary.

02 · Where Translation Genuinely Helps (Use Without Guilt)

Where Translation Genuinely Helps (Use Without Guilt)

As a comprehension shortcut. New word, quick glance at the translation, meaning confirmed, back to reading. Five seconds. The context-only alternative costs minutes and often lands on the wrong guess anyway. For abstract words especially (grief, policy, awkward), a translation delivers in one second what fifteen example sentences struggle to.

As the test direction. Here's one that surprises people: your reviews should often start from your language. If you've read the notebook post, you know why: testing from L1 (or from meaning) to English forces production, which is the skill you actually need in conversation. The learner who covers the English column and retrieves make a decision from their own language's prompt is doing the most valuable exercise in this entire blog, and it's translation, formally speaking. Nobody tell the commandment.

As an error detector. Knowing how your language differs from English is a superpower. If you know your language says the equivalent of "close the light," you can put turn off the light on a watchlist. Teachers who share their students' first language fix errors twice as fast for exactly this reason; you can be that teacher for yourself.

As emotional bedrock. Planning something difficult to say, checking you've understood something important, processing what you're learning: doing this partly in your own language isn't weakness. It's how a whole adult mind participates in the project instead of a beginner-sized fragment of one.

03 · Where Translation Sabotages You (The Actual Sins)

Where Translation Sabotages You (The Actual Sins)

Live sentence assembly. The big one. Composing the sentence in your language first, then converting it word by word while someone waits. It's slow (two languages' work for one sentence), it produces structural transfer errors by design, and it locks you at a permanent speed limit, because conversion can never run at conversation speed. If "stop translating" means anything useful, it means stop this.

One-word equivalences. Writing get = [one word in your language] and considering the matter closed. Words aren't coins with fixed exchange rates; they're territories with different borders. Your language's word for make probably covers some of English's do territory, which is exactly why that pair torments you. The fix you already know: translate chunks, not words. Make a decision = [your language's whole expression]. Chunk-to-chunk exchange rates are honest. Word-to-word ones are counterfeit.

Translating idioms and collocations piece by piece. Your language's beautiful expression, rendered literally into English, produces either comedy or confusion, occasionally both. The traffic also runs the other way: dismantling heavy rain into its parts to check them against your language teaches you nothing, because collocations aren't built from logic in any language. They're learned whole, in both directions.

04 · So What Does "Thinking in English" Actually Mean?

So What Does "Thinking in English" Actually Mean?

Let's rescue this phrase, because it describes something real, just not what learners imagine. It doesn't mean your inner monologue switches nationality, or that your first language gets evicted from your own head (it won't, and that's fine; even lifelong bilinguals don't fully evict anything).

It means this: for the situations you've practised, retrieval replaces conversion. When someone asks how your weekend was, you don't consult your language, translate, and convert. The chunk "Pretty quiet, actually. What about you?" just arrives, whole, because it was learned whole and retrieved often. That's "thinking in English": not a mystical state, but a growing collection of situations where the English comes pre-assembled. And notice how you get there: not by banning your first language, but by learning chunks and retrieving them until they outrun the conversion route. Every post on this blog about chunks and spaced retrieval has secretly been a post about this. The commandment says suppress the old road. The working method says build a faster one, and let the old road quietly lose its traffic.

05 · Common Mistakes

Common Mistakes

Mistake 1: Translation guilt. Feeling you've failed every time your language appears in your head. It will appear for years, decreasingly, and that's the normal curve of every successful learner who ever lived. Guilt adds nothing except a reason to quit.

Mistake 2: The other extreme: everything through translation, forever. Reading English by converting each sentence, speaking by composing in L1 first. If this is your only mode at intermediate level, the speed limit will define your ceiling. The exit is chunk retrieval, built gradually, one situation at a time.

Mistake 3: Bilingual flashcards with naked words. The tool is fine; the grain size is wrong. One word to one word breeds false equivalences. One chunk to one natural expression in your language: that card is honest and useful.

Mistake 4: Using translation to check your own English word by word. "Does this sentence exist in my language? Yes? Then it's correct." That test approves "say me" and "an advice" all day long. The correct test is "have I met this chunk in English?", which is a memory question, not a translation question.

06 · Try This Today

Try This Today

An honest audit of your own machinery, two minutes. Say something mildly complex in English, out loud, about your plans for next week. As you speak, just notice: which parts arrived whole (retrieval), and which parts you built by converting from your language (assembly)? Don't judge the ratio. Just meet it. Every learner has both; the ratio is what changes with practice.

Then take one sentence that came out via conversion, find out how English actually says it (the natural chunk, not the literal build), and put that chunk into your reviews, tested from your language to English, out loud. You've just used translation twice, correctly, in both of its legitimate jobs: once as a detector, once as a retrieval prompt. That's the whole peace treaty, in miniature.

07 · The Peace Treaty

The Peace Treaty

Back to the commandment, and the guilt it installed. Here's the replacement, suitable for the poster: your first language built your mind, and it isn't leaving, and it shouldn't. Use it for what it's brilliant at, reaching meanings and testing retrieval and spotting the traps it set for you. Refuse it the one job it's terrible at, which is assembling English sentences live, on its own blueprints, at a speed it can never reach.

Translation was never the enemy. Word-by-word assembly was, and the way past assembly was never suppression. It was chunks, retrieved until they win the race. Your language gets to stay for the whole journey. It just doesn't get to drive.

WordCat is built around exactly this treaty: chunks tested from meaning to English, so retrieval outgrows conversion one situation at a time. Try it at wordcat.app.

Keep reading:

• False Friends That Will Embarrass You (Eventually)