The legal industry is assumed to be a traditional and organized field, owing its beginnings in the ancient past as laws and jurisdictions are created alongside civilizations and societies. Alongside it is the development of legal translation, as the different parties involved in agreements and negotiations needed to understand the terms and decisions that shape families in the small scale, and entire empires at the larger level. Limited by the technology of their time, the legal translation of the past mainly depended on human language experts and mechanical writing tools.
However, in modern times, the world of legal translation is undergoing a significant transformation, thanks to the advent of Machine Translation Post-Editing (MTPE). This innovative approach melds the rapidity and efficiency of machine translation with the nuanced understanding of human editors, offering a compelling alternative to traditional translation methods.
How do these translation approaches compare to each other now? Does Machine Translation Post-Editing (MTPE) have something to offer in the table of legal translations?
Let’s find out.
Legal translation has a history of more than 3,000 years. In a paper by Jianmin Zhao, it’s mentioned that the rules of translating legal text were codified during the times of Roman Empire, wherein East Roman Emperor Justinian gave the directive of a word-for-word translation from Greek to Latin in order to “preserve the letters of the law.” This kickstarted the beginnings of legal translation into a very literal mindset, causing obscure, awkward and difficult translations in the years and centuries to come.
The shift to translating words in context happens much later, during the Middle Ages, when France became a dominant international power. With French as the then lingua franca, a proposal was made for this “new form” of literal translation.
However, it’s not until the 19th century that translators made the conscious decision to make better-quality translations, making efforts to follow the rules of the target language. Human translators have long been the cornerstone of this field, ensuring precision and contextual accuracy. And the process, while thorough, was often time-consuming and labor-intensive.
The rise of digital technology marks a significant shift in the translation landscape. This introduces machine translation systems that were in the same vein as when legal translations were first structured: simple and straightforward, offering literal translations without considering context or idiomatic expressions.
However, with advancements in artificial intelligence and natural language processing, machine translation has evolved remarkably, leading to the development of MTPE. From the basics, Machine Translation Post-Editing (MTPE) uses computer software to translate text (either written or spoken) from one language to another, and a human checks the output after.
But legal translation has become a meticulous process, requiring extensive knowledge of legal terminology and the judicial system. In practice, creating a machine that can translate with the same finesse and nuance that humans have developed in the thousands of years of legal translation history has proven tricky, if not outright impossible.
That, however, doesn’t mean that MPTE does not have its benefits in the legal field.
MTPE has found a unique application in legal translation. One of its more highlightable benefits is the ability to streamline workflows while maintaining high accuracy levels. Legal documents like contracts, patents, or routine litigation documents are particularly effective for initial machine translation, as these types of formats often contain repetitive phrases and standard formats. These parts of the legal process also benefit from the speed of machine translation, with human oversight during and after the translation process ensuring legal accuracy, context, and cultural appropriateness.
Additionally, when legal professionals involve only humans in the translation process, they need to wait for the completed translated material before making their next moves during the case. This waiting period can be wasteful for both clients and firms. The raw machine translation, while not suitable for official or public legal documents, still can be valuable for internal use. It provides a quick overall understanding of content, aiding in decision-making processes or preliminary document reviews.
With the addition of a human linguist, the necessary guidelines when it comes to document style and format are maintained. This helps especially when documents and evidence are required to be submitted to the courts. These completed, post-edited documents can also serve as references for future clients who may ask for similar instances during their cases, or for other editors who are working on legal documents for the first time.
When comparing traditional methods with MTPE, three factors stand out: efficiency, accuracy, and cost.
MTPE significantly accelerates the translation process, an essential factor in time-sensitive legal matters. Deadlines are met strictly, and for the justice system that looks into hundreds and thousands of cases daily, MPTE can min-max productivity for legal professionals.
While traditional methods have been synonymous with high accuracy, MTPE has rapidly closed this gap, offering comparable precision. As discussed above, there are still situations where a complete human team is still required to get the best possible result. But with machine translation lending a hand, legal professionals can maintain consistent accuracy in a shorter time frame.
MTPE is generally more cost-effective, reducing the time and resources needed for translation. Human translators have to be paid for their skills and expertise, which some machine translation systems can already replicate, albeit in a lower standard of quality. For businesses and clients who may not have the resources to support a complete translation or localization team, MPTE can be an adequate replacement.
Despite the efficiency of MTPE, certain situations still demand exclusively human translation. Unedited machine translated documents have been banned in some states in the U.S., such as New Mexico for example, for formal purposes like court proceedings or exhibits. In Canada, the Immigration and Refugee Board requires legal documentation to be submitted alongside a certificate or an affidavit that swears to the translation’s accuracy and the translators’ language proficiency. Additionally, some certificates cannot typically be translated through machine translation, such as birth, death, and marriage certificates.
Complex legal arguments, documents with substantial cultural nuances, and cases where the stakes are exceptionally high, such as high-profile litigation, are also areas where human expertise remains irreplaceable. As Global Voices notes in its article, the documents presented at court can mean the difference between justice and injustice in a legal case.
There are a few ways to resolve these concerns:
When it comes to MT, the main problem at times isn’t the software; it’s the original text. A stray comma or word can affect the meaning of the entire text section. Before starting the machine translation process, proofread the source text first.
There are many options in the market for either free or paid options, so consider choosing a software that can be algorithm-trained with legal-specific datasets.
Despite human translators having domain expertise, style, format, and layout might differ from location to the legal firms handling such cases. Have a copy of such guidelines readily available for reference, to ensure that positing is seamless, collaborative and adjusted for any sudden deadlines.
Machine translation for legal translations should be “good and fast enough” for everyone involved, including the courts.
Set expectations on what errors to correct, on which parts of the raw translations should be used, and which sections human translators should focus on. If necessary, add certifications for every finished document to assure readers of accuracy and quality.
Once MT has become an established part of the legal translation process, check to make sure that the quality of the raw translations remain consistent, and assess how much legal translation post-editing is being done for each document.
MTPE is proof that history does tend to repeat itself, in this case, marrying the best of machine efficiency and human expertise.As technology continues to evolve, the reliance on MTPE is likely to grow, reshaping the landscape of legal translation.
In time, the future of legal translation is one that the history books might experience for the first time: human expertise augmented, not replaced, by technological advancements, leading to more efficient, accurate, and cost-effective translation solutions.
Raphaella Funelas is a creative writer who graduated from the University of the Philippines Diliman with a Bachelor of Arts degree in English Studies, specializing in Language. She likes learning about anything new in any field, and has pursued that interest through a writing career. She always has an ear on the ground for any exciting topics, and an enthusiasm to share any newfound knowledge through her words.