Google Translate is free, and its algorithm is increasingly sophisticated. In a world of increasingly multinational business connections, entity professionals need more and more the ability to translate important documents into foreign languages for the benefit of non-Anglophone investors, subsidiaries and so on. But especially when it comes to Google Translate and similar free translation services, often, you get what you pay for.
In this post, we outline some reasons automated translation is, at least at this stage, unsuitable for the needs of most multinational businesses. Consequently, a central part of any entity management system that operates across linguistic barriers is — and should continue to be — paid, professional translation services.
Language Is an Immensely Complex Human Creation
There is a whole academic discipline — linguistics — and legions of scientists who, since the earliest days of multinational encounters, have been pushing forward the understanding of the roots of human languages; how they change, adapt and grow; the links between them; and the ways some of their structures overlap while others diverge wildly from each other. We won’t even try to introduce such a broad topic here.
If we think about what the greatest testaments to human accomplishment are, the development of language far outranks such monuments as the Egyptian pyramids or Angkor Wat in Cambodia for its infinite adaptability in the creation of meaning. Anyone who’s sat down to learn another language, even one that’s genetically close to English (Swedish, for example) or a common second language in countries such as the United States (like Spanish), knows that translating even basic meanings can be a time-consuming and unproductive chore that can never be perfected completely.
The simple fact is, automated language-learning software, even software as advanced as Google’s neural machine, can’t imitate the vast complexity and variability of the thought patterns that real humans produce when we process language. While it may not lead to results as embarrassing as some of the examples listed above, it’s likely that a great deal of meaning will be lost.
Automated Services Do Well Translating Into English — Not So Much From it
Most entity management professionals resorting to Google Translate and similar services are looking for a cost-saving way that allows them to share important documents and ideas with colleagues, investors and employees of a multinational enterprise who have an incomplete or nonexistent understanding of English — a requirement for business when the international system has English as a lingua franca but many of its local and national components still have not mastered it.
So mostly what we ask from Google Translate in these circumstances is accurate translation from an English original into, for example, the Mandarin used by a subcontractor in Taiwan or the German in which a merger deal in the EU is being conducted.
Problem: Google Translate’s algorithm was designed by people whose first (and sometimes only) language was English, and its program bears many of the marks of the monolingualism that’s distressingly common in the Anglophone world. In particular, this means that while it sometimes translates quite well from another language into English, it has more trouble translating from English into languages where its source code is not as knowledgeable.
Google Translate produces results that are incredibly diverse in quality depending on the languages involved and the circumstances under which a text has been produced. The reliance of the algorithm on professionally translated EU documents can produce decent results when you’re dealing with English and one of the major European languages like French, Spanish or German. Also, with the growing interconnections between Chinese firms and the developed world, programmers have been compelled to develop more sophisticated methods of translating to and from Mandarin. But its ability to deal with dozens of other languages has yet to catch up, and there is no timetable to achieve uniform quality.
Business Requires a Certain Understanding to Convey Full Meaning
The language used by any enterprise when it conducts business is deeply influenced by jargon that is industry-, or sometimes even organization-, specific. Jargon relies on a deeper level of meaning fostered by development that’s shared within organizations and among individual sectors.
Business jargon can be difficult to understand by an outsider because it is so context-specific. Automatic translation isn’t equipped to pick up on this deeper meaning and to find adequate expression for it in the target language.
Translating the specific idioms and expressions of a certain organization or industry requires an expert — a human one — who is fluent in both languages and familiar with the history and practices of the specific industry they’re working in for the translation to be adequate to the tasks it’s meant to perform. This is a specific specialty that is much more involved and expensive than copy-pasting a block of text into Google. But if you aren’t reaching toward this standard, it’s likely to cost your organization much more over the long run.
The Security Concerns With Google Translate
Language issues aside, Google Translate comes with other risks. These risks deal with privacy, which is a great concern for most, if not all, businesses. As a legal professional responsible for managing a global company’s entities, using Google Translation puts those legal documents in the hands of an organization that is very well known for data mining. Through openly putting legal documents through their algorithm you risk the opportunity to actually keep your information private and while you still maintain control of your information, you have also given Google the opportunity to read and scan any materials that have gone through their system. The same goes for all other Google Products, so if you want to ensure the privacy of your documents, you should consider hiring an individual who you can put on an NDA so as not to expose any pertinent information on the Internet where hackers can easily find this information and spread it around with malicious intent.
Accurate translation is an increasingly important part of multinational entity management. Blueprint OneWorld’s platform specializes in meeting the basic needs entity management professionals have in an increasingly complex and interconnected world. Please call or email us today to discuss our solutions.