Let’s be honest: who among us has never turned to an automatic translation tool to convert a sentence from one language to another?
It’s no secret that we rely on machine translation tools in our daily lives. Think about social media platforms like Instagram, which automatically translates our friends’ captions for posts or stories, or Youtube, where subtitles are generated to help us understand the content of a video. We also use these tools for school assignments in a foreign language to save time, or at work, relying on computers to translate official documents. In short, automatic translation has become an integral part of our everyday life. And it’s not surprising, considering it has revolutionised the translation industry, making languages and cultures from all around the world accessible to everyone in just a few seconds. It has even led us to believe that we no longer need to learn a foreign language to communicate, understand, and write it effectively. But is this truly the reality?
Lost in translation
Undoubtedly, computer-based translation systems have simplified language interactions, giving us access to multilingual information sources, and speeding up the translation and interpretation process. However, I bet that at least once in your life, you’ve come across sentence that was auto-translated in a bizarre or even nonsensical way. This tells us that automatic tools can’t replace humans’ abilities.
The challenge of idioms
Idioms are expressions with a figurative, non-literal meanings that are not deducible by looking at the individual words.
Here’s an example:
It’s raining cats and dogs.
Of course, this idiom doesn’t actually mean that cats and dogs are falling from the sky. It means that it’s raining very heavily.
Translating idioms poses problems for machine-based tools, as they tend to translate phrases word-by-word incorrectly.
Imagine chatting with your new online Italian friend that asks about the weather, and you answer that canids and felines are falling from the sky… embarrassing, isn’t it?
Synonyms: which one should I pick?
Languages often have rich vocabularies, meaning that they are loaded with synonyms. This can be a real challenge for automatic translators, which must pick the right equivalent from among numerous words.
Take, for instance, the adjectives sick and ill. When translated into Italian, they both end up as are both translated into malato (for males) or malata (for females). But guess what? Human translators can spot the difference in meaning and select the appropriate translation.
Sorry machine translation tools but dealing with synonyms isn’t your forte.
Don’t joke with me
If you think translating a joke or conveying ironic messages in another language is an easy task, think again. Humour, for instance, is deeply rooted in culture and relies on linguistic tools, such as wordplays, puns, and cultural references to get a smile out of the readers.
In translations, sometimes it gets so challenging to preserve the twist of humour in a different language that human translators might just throw in the towel and try to explain why it’s funny. But once you have to explain a joke, the funny part is pretty much over.
And if reproducing the humorous effect is tough for humans, let alone a computer. Automatic tools often provide a literal translation of ironic phrases, completely missing the mark. Instead of a laughter, they leave readers scratching their heads in front of nonsensical phrases.
The final verdict
But let’s get down to serious business: is it still worth learning a foreign language?
Well, even though technology has been constantly improving, by reading the examples above you can easily understand that automatic translation is far from being accurate and shouldn’t always being trusted. However, this doesn’t mean that it shouldn’t be used when translating a text. Quite the opposite: it can be a helping hand to communicate effectively in another language, given its constant refinement and speed. Personally, I believe it’s important to realise that all these tools are a double-edged sword and have their limits. While they help us save time, they can mistranslate phrases.
References
https://www.sciencedirect.com/science/article/pii/S187704281300253X
https://mt-archive.net/00/IJT-2001-Hutchins.pdf
https://www.researchgate.net/publication/360572568_CAN_MACHINE_TRANSLATIONS_TRANSLATE_HUMOROUS_TEXTS
https://knowledge-centre-interpretation.education.ec.europa.eu/en/news/translating-humour-blessing-and-curse
I really enjoyed hearing your thoughts on this topic! I agree that the cultural aspects of language such as humour or idioms are very difficult to pick up on for machine translators, and I believe there is something really beautiful about the fact that a bit of human touch is still needed to breathe life into language. Also, there is nothing quite like the connection you make with a person by speaking their language, and leaving it to machines would rob humanity of an important tool for community-building.
Loved this post! I definitely agree that yes, despite how advanced digital translation tools have come, they still and probably will always have their faults. To take it even farther, I enjoy quite a few pieces of media in languages that I don’t personally speak so I have to rely on translated subtitles. Whether these subtitles are done through machine or manual, I don’t fully know, though I doubt technology isn’t somehow involved. Regardless, I often do deep dives into the actual original meanings posted by fans who speak the native language of the media to better understand what the translation fails to fully convey in the English language. As you say, many things such as idioms just aren’t easy to transfer over. It’s honestly inspired me to slowly start learning the actual language. So, in a way, translation tools shortgivings can definitely serve as an incentive.
As a translator student, I find your blog very interesting, I think this is a hot topic nowadays that we don’t discuss that much in real life. I believe automatic translation will never be better than human translation, but it’s true that we, translators, use them a lot to make our work easier.
There are many ways to work translating, can be translating from the beginning just you, common sense, and a dictionary; or you and the internet, it’s a great tool to help us work faster than without it. We don’t help ourselves only from automatic translation but also from urban online dictionaries and sayings that people quote on their papers.
It is undeniable that translation software is becoming more and more powerful. As a non-native English speaker student, I really benefit from the various conveniences brought by translation software. But as mentioned in the article, this is a double-edged sword. Language will regenerate, and people are constantly creating expressions that belong to our era. The innovation of language also makes translation software always lag behind. current.
I really liked your post because I have always been fascinated by the idea of learning a new language! I do agree that a machine translator can be very useful when traveling but I believe that its accessibility allows for many people to establish in new places and not feel the need to learn the language. I am guilty of this as since I moved to the Netherlands, I have become a constant user of the Google photo translator and do think that it has impacted my ability to soak in the language that now surrounds me. Nevertheless, there are things that translators can not do (as you say) which is that as they are embedded in culture, it tends to be a one-sided thing. The user understands what it says but -probably- can not verbally communicate. Sometimes I just want to go to the bakery and have a conversation in Dutch with the owner, and that, well that can Google not do…