This paper reviews the article "The Perspective of Machine Translation and Horizons of the Future," examining arguments for and against machine translation as a replacement for human translators. It discusses the roots of machine translation in cryptography, the limitations of word-for-word software translation, and specific systems such as Systran, ACENTINUS, and Natural Language Processing (NLP). The paper argues that while machine translation offers practical value for businesses and organizations operating in a globalized economy — particularly for speed and cost savings — it cannot fully replicate the nuanced judgment of skilled human translators, who will remain essential at minimum as proofreaders and quality reviewers.
The paper models source-based argumentative writing: it summarizes a primary source article section by section, then evaluates each claim against counterarguments and contextual evidence. This technique — paraphrase, quote, evaluate — is a foundational academic writing skill, and the paper demonstrates it cleanly by maintaining a clear thread between the source's position and the writer's own critical voice.
The paper opens with a broad hook about Google Translate before narrowing to the specific article under review. It then moves sequentially through the source's main arguments: historical origins, the pro-automation case, the anti-automation case, a specialized use case (ACENTINUS/EU), and a taxonomy of translation software types. Each section builds on the previous one, culminating in a conclusion that the technology is a complement to — not a substitute for — human expertise.
Almost everyone is familiar with the convenient Google feature that allows for instantaneous translation of foreign words. This automated, or "machine," translation is a practical way to read websites in different languages. No longer does a reader need to know someone who speaks a foreign language or hire a professional translator — the translation is provided quickly and easily by software. However, for many professional translators, there is a fear that this mechanized process will render their profession obsolete. The article The Perspective of Machine Translation and Horizons of the Future argues that such fears are unfounded. There is a useful function that machine translation can perform — one that will enhance current translation capabilities for businesses, individuals, and other organizations — even if it is not a perfect replacement for human translation.
The article begins by noting the vital need for translation today, given the increasing globalization of the economy. Despite the facilitation of communication that globalization enables, business documents are often lengthy and complex. Hiring a translator to handle these documents can be expensive for small businesses and large businesses alike, resulting in costly delays and loss of income. Machine translation — accomplished via software — is far faster than human translation.
"The roots of machine translation are in cryptography; translating messages and then decoding them character by character. The first language translation was word for word and, as we all know, language cannot always be translated word for word without some very embarrassing mistakes."1 A machine cannot understand the subtleties of language, so there will always be a need for human translators, even if only as proofreaders. As with the introduction of any new technology, fears abound that it will be used to eliminate jobs from skilled human beings who have devoted many years to learning their craft. However, most of the existing — and relatively scant — literature on the subject does not support this thesis. Even when human beings conduct literal translations, the results are frequently awkward due to inappropriate wording and a poor ear for colloquialisms, and most existing software can only render language in an equally literal fashion.
Supporters of the ability of machine translation to completely replace human intelligence point to the program Systran, which has been used to translate entire newspapers from Spanish into Catalan without the assistance of proofreaders. However, this discrete example is not truly transferable to other forms of translation as a whole. First, Spanish and Catalan are relatively similar languages, and the ease of translating between them is not comparable to, for example, translating Russian into Chinese or vice versa. Furthermore, newspapers are relatively standard, plainly written documents that are not analogous to complex legal and business texts, and they lack the subtleties and complexities of literary writing.
However sophisticated these methods become, they are merely ways to make machines a bit less prone to error and more useful. They cannot replace the feel for language possessed by a skilled human translator. Machine translation is best understood as a powerful complement to human expertise — one that expands access to cross-language communication in a globalized world, while leaving the finer work of nuanced, accurate translation firmly in human hands.
You’re 55% through this paper. Sign up to read the remaining 3 sections.
Sign Up Now — Instant Access Already a member? Log inAlways verify citation format against your institution’s current style guide requirements.