“Can’t people just use Google Translate?”
If you’re a professional translator or work in the translation industry, you’re used to hearing this question whenever you explain your job to someone. It’s an annoying question, but not an unfair one—for many people, Google Translate may be adequate for their translation needs. If you don’t travel abroad, don’t live or work in a big city, or don’t do international business, maybe all you need to translate is a word or two here and there, probably just our of curiosity. But if you’ve ever tried to use Google Translate to try communicate with someone in another language, or tried to interpret something from someone else who used it to contact you, you probably realized pretty quickly that Google Translate just won’t cut it.
Google Translate is a service that provides what is called machine translation. As you might expect from the name, the translation is done entirely by computer, with no human intervention. Machines are very good at simple yes/no logic—binary code is the whole foundation they’re built on after all, which has either an on (1) or off (0) state. A machine could easily tell you that the Spanish word for orange is naranja. With enough time, a machine could easily do a search and replace and swap out words in one language with another—however, this is not how translation works.
Languages weren’t thought out, planned, and designed; they evolved naturally over thousands of years. This means there’s no way to simply “crack” them like a code. Computers are not yet at the level where they can contextualize words in sentences in a meaningful way. Therefore, if you asked the computer to translate “I am eating an orange,” it would have to estimate whether you meant orange the fruit or orange the color—and though this example is simple, maybe the context isn’t part of the same sentence. Maybe it needs to understand the sentence before to know which form of the word “orange” you were referring to. Or the paragraph before. Or maybe the context isn’t in the words itself, but maybe it’s part of an overall theme, or depends on the original author, or is part of a pun, or if the request is made in a certain tone—these are all things that humans do easily, but subconsciously, and are way beyond the scope of current computer technology!
To put into perspective the current power of computers to parse human language, there is a concept in computer science called the Turing Test, which is a sort of game that gauges the accuracy of a computer to emulate a human by convincing actual humans that the computer’s responses are not generated by computer, but by another human. To date, the best results have only had a 33% success rate—and this is only for communication within a single language!
Here’s a simpler test you can do from your computer or phone right now: go to Google Translate and type in any simple sentence you want and translate it into another language. Now copy the translation and paste it back over the original text and translate it back into English. Did you get your original text back exactly the way you wrote it?
If you value your text enough to spend the time writing with proper spelling, grammar, capitalization, validate the information, and have proofread, then the only way to ensure that your thoughts are conveyed equally accurately in another language is to have it translated by a professional, human translator.
Personally, I also sometimes use google translate to learn some words in languages like French, German mostly, but that is only from English to French or German and vice versa, however, I can’t access its translations in my native language though it’s a rare language, but you can find human translators who can translate from English to my native language. Therefore, you can’t rely on google translate if you want to know something in a rare language, yet you can find human translators to do that for you