You’re considering hiring a translator or agency to help you with translation, but you’re new to the industry. Isn’t translation just a matter of finding the right word in the new language for each word in the current language? Because then all you need is a free online translator, right? Not really, and we can explain why.
What Is Machine Translation?
In the translation industry, “machine translation” means using software instead of humans to translate words. When you’re on your favorite social media platform and see a small “translate” option under a post written in a language other than your profile language, that’s machine translation. It has come a long way from its early days, but it still requires a human touch to work with.
Can’t Machine Translation Just Look Up, Well, Translations?
Your foreign language studies, even if they were sparse, likely focused on memorizing lists of words:
All you had to do was memorize these lists, know where to put the words in a sentence, and voilà! You could speak in tongues.
Can’t computers do that? As we say in German, “jein.” (In English, that would be “yo,” which doesn’t mean the same thing. See how confusing this can quickly become?) What that German word—a combination of ja and nein—means is “yes and no.”
Yes, you can look up the words, but words by themselves don’t convey meaning, so no, you can’t immediately translate them. If I say, “keyboard,” and nothing else, not even with my body, you won’t know what I mean. Do I mean an input device for a computer? A musical instrument? A piece of playwood in a garage that has keys hanging on it?
We need sentences (and usually other factors) to convey meaning, e.g., saying “My keyboard is broken” to an IT team member at work most likely does not refer to your musical instrument standing in your living room at home.
(However, if that IT team member loves music, then it actually could. But do you see what’s happening here? We’re looking at the context of that statement, something machine translation still needs to perfect.)
Machine translators can’t detect context unless you give it to them, and even then, it’s impossible to train them to see all contexts. The human brain continues to surprise science with its capabilities, and we can’t technologically reproduce those yet.
One Pesky Word That Can Change It All
Prepositions are used in English to connect nouns to verbs and nouns to nouns. For example, these sentences are identical except for the preposition:
- I ran into the house.
- I ran over the house.
- I ran around the house.
- Run the washer with the dryer.
- Run the washer after the dryer.
- Run the washer around the dryer.
Indeed, two of those examples may have made you laugh, because they’re impossible, like running the washer around the dryer. But hopefully you now see the role prepositions play in English.
So, can’t machine translation do that? Often, yes, but it depends on the meaning of that sentence. For example, “They’re out to lunch if they think that will work” was translated by a well-known machine translator into German to mean that literally, i.e., “They’re going out for food at about noon if they think that will work.” It didn’t understand the implied context and return, “They’re crazy if they think that will work.”
You might think that the solution is to add in the translation “to be out to lunch” means “to be crazy.” But if the machine translator doesn’t understand the proper context, it may return that crazy translation when the literal one is intended.
Prepositions may be (usually) small words, but they’re big players when it comes to understanding an English sentence. Humans use many clues to give us context that MT can’t pick up on yet. Machine translation is getting better, but it’s still miles from perfect.
Languages Are Built Differently
Did you know that not all languages have prepositions? They may change a noun or verb to include the meaning we give to prepositions in English. Other languages don’t differentiate between the sentences “I’m working” and “I work” because they don’t conjugate (change the form of) verbs to account for an event that is being carried out right now and one that is carried out over a longer period of time or is a habit.
This is another reason why you can’t translate one word for another. German only has “I work” and requires other signals to help an English speaker understand the time component.
What’s the Solution?
A good machine translator will of course pick up many of the differences we’ve discussed, but not all of them. That’s because computers, despite all the incredible advances in AI, still can’t process context the way a human can.
The assumption that everyone in the world speaks English also doesn’t hold any water: most people don’t. Estimates range from about 1.1 billion to 2 billion speakers, including native speakers and those who speak it as a second or foreign language. Translation is the only solution if you expect your company to have any strong influence globally and have any hope of ultimately conquering any global markets.
In some cases, human-only translation is the best option. But in the usually straightforward language of business and technology, machine translation can help save time and money.
But you always need a human involved who knows both languages exceptionally well.
That’s because effective and fast translation requires a solution that lets standard translations be carried out automatically while still allowing for a human to confirm the context and approve the translation. This is the best of both worlds and a solution you should expect from the translation agency you choose.
There are time-saving ways to translate the documents, applications, and multimedia that your company needs done to reach a wider audience. At text&form, we use a variety of translation tools to complete your translations accurately without resorting to word-for-word lookups and time-consuming, human-only translation processes.
About the Author: