Translation plagiarism is relatively a new term. It means that individuals copy the works of others as their own. Even when you take the written content and translate it from one language into another, it would also be a translation plagiarism.
Translation plagiarism has become one of the major problems for automated plagiarism checking tools. All of these services use some kind of fingerprinting, designed for detecting plagiarism. In comparison with it, translation plagiarism is very imprecise due to various ways of possible translations, which makes it difficult to detect for plagiarism. In spite of this, all these tools have been in great demand for many years already.
In fact, translation plagiarism existed long ago before Internet has gained worldwide fame. But with a speedy Web development, writing something has become so easy. Nowadays creating your own blog and website is more than possible. There are so many people, who produce content, which in most cases is known as unauthorized.
If someone sees an article in French and likes it, he can easily translate it into English and post on his website. It’s quite a common scenario on the online scene today. There are so many websites, which take news and articles from other languages, translate them into their own and post on the website. Are they doing right?
Why Does Plagiarism Happen?
According to the empirical review study from 1941 till 2005, the results of using translation plagiarism are the following: from 20% of students up to 75% are engaged in some form of plagiarism. One thing is clear: an opportunity to use all possible technology tools today provides them with more opportunity to misbehave while creating content. In some way, the contemporary students are spoiled with all that material, available on the Web.
Plagiarism detection is the process of including plagiarism elements within a text. Due to the development of Internet, copying other works has become very easy. The majority of translation plagiarism is found in academic writing. In fact, it can be see almost in any other field of writing, including art designs, scientific papers, source codes, novels etc.
Moreover, plagiarism detection does not fully detect plagiarism, but only copies. The methods would vary, but the results would be always the same: these systems are looking for matching phrases, and if there are any, they look deeper to see more matching between the texts.
This system is very efficient, as it helps automated plagiarism checkers looking through a vast amount of content for any possible similarities. The disadvantage of the tool is that it always relies on exact matching phrases. If you change a number of words, fooling plagiarism checkers becomes so easy.
It’s very difficult to find high-quality translation tools. Getting a good translation is not so easy; it requires a lot of time and efforts. In the majority of cases, people, who use plagiarism detection, will be caught by their teachers and instructors, who would notice changes in student’s writing. In fact, humans are one of the best tools in detecting this kind of plagiarism. Of course, it doesn’t mean that all plagiarism detecting tools are useless.
As a matter of a fact, in order to avoid being caught with plagiarism, one should change 1 of every 3 words in the written content. A lot of plagiarism detecting software developers confirm that plagiarism of the majority of English articles has become so vast, that their customers often ask them for any possible means to detect plagiarism in translation. That is a rather interesting and promising statement. At the same time, how can such system guarantee high quality results? To achieve that, the first step would be to create huge multi-language database with all types of academic papers. When the most challenging part would be creating original plagiat-checking script that would be able to work with each and every online vocabulary, so that it excludes every possible translation format. How real does this sound? Would it even be ever possible?
3 Problems in Detecting Translation Plagiarism:
- There is no one right and exact way of translating a word. There are so many nuances concerning this, but automated systems may see the meaning of one word as completely different words.
- Each language has its own grammar structure: a word-for-word translation is not possible, when it comes to grammar. Even languages in one family may have different grammar structure.
- Automatic translation system is not effective, when we take an English text, translate into another language and do it back again – we will get truly a hilarious result. These systems work quite well to understand the sense, but not detecting exact matches.
How to Defeat Automated Plagiarism Detection?
There are so many predatory publishers and unethical authors, who know a lot of tricks how to make it more difficult for automated system detecting plagiarism. For example, PDF files are usually made of several layers. There is a visual layer and the text layer. It’s possible altering the unseen text by changing all the letters into mojibake. So, when you use Ctrl+C and then paste it into the Notepad, you’ll get only garbage characters. Because of this, automated detection tools are not able to read the text. Moreover, this also makes it difficult finding the authors papers by the search engine in Google.
Back Translation – a New Form of Translated Plagiarism?
Back-translation is a new approach of many students, especially international. Using particular tools, they can take a text in English and translate it into another language, then ‘retranslate‘ again into the original language, thus hiding their cheating. To avoid the problem is not so easy, but possible. The teachers should give students such kind of tasks for them to be unable to use back-translation.
Translated plagiarism remains one of the most difficult areas to deal with. According to 2011 test results, there were no systems available to cope effectively with the issue of translation plagiarism. It’s believed it would remain a big problem in the nearest future.
Everything is explained by a low ability of automated systems to detect translated plagiarism. Of course, the issue wouldn’t become rampant and some other methods of detection will certainly appear soon.
Also published on Medium.