Book Review
The book by Chiew Kin Quah is entitled “Translation and Technology” and, as the title implies, devoted to the utilization of computerized tools in translation. The author acknowledges the development of English as the universal language of business and, simultaneously, the globalization that requires boundary-spanning all over the world. As a consequence, the author addresses the issues of technology usage and enlists occupation-related issues in translation to provide a full informative overview of contemporary trends in the field. The book is, thus, primarily oriented for students since it mainly discusses the development processes of CAT tools. The book omits extensive formulae, which makes it comprehensive for those non-professionally interested in the field.
The author has gathered and analyzed an extensive body of data to complete the work, although it is by no means confined to machine translation as such. Rather, the book deploys a multi-perspective approach analyzing machine-aided human translation and the same process in reverse.
The book consists of six chapters that are quite concise in length. The first chapter provides a discussion of terms to most optimally classify the tools used on the field. The second one briefs out the fundamentals of Translation Studies as an academic discipline and outlines linguistic theories used in machine-aided translation. Chapter three reviews the systems of machine translations in the diversity of their designs and provides a concise history of their evolution over the decades. Chapter four is aimed at the description of standards for data exchange via CAT tools and also discusses the commercial CAT existing in 2006. The fifth chapter concentrates on various methods of translation evaluation and contains a theoretical framework.
The final chapter summarizes the recent tendencies in the field and maps the possible ways of further development. The author also uses addenda to refer to throughout the work. Some information on a Web-based translation platform is also given.
The author organizes the book for the readers’ convenience. As said, the author refrains from using extensive formulae, relying mostly on descriptive writing and structuring the text for the readers’ convenience. The terms and abbreviations used are deciphered to simplify the reader’s advancement through the text further still. The writing style is simple yet more than attentive to the details, as in popular scientific literature. Overall, albeit the short-length format, the work is detailed, and the writing makes it appropriately informative to a professional audience as well as for the public.
From the Translational Studies perspectives, the author offers a thorough exploration of the association between the theory and tools. Particularly, she states that linguistic formalism has played a more significant role in the formation of machine-aided translation than the theoretic aspects of the field. Several approaches to machine-aided translation are also reviewed, e.g., EBMT which is semantics-oriented. The scope of the work is, thus, quite extensive. The author indeed uses the term “newer” about theories that were relevant back in the 1960s but, among its kind, the work appears an elaborate endeavor.
The recent developments in the field of machine-aided translations might appear not so recent considering that the work by Quah was published a decade ago. On the other hand, the review of what was recent ten years ago provides a detailed retrospective insight. The author has correctly predicted the expansion of the Web which, in turn, will increase the demand for machine-aided translation. Also, she expanded upon the semantics as a means of improving machine translation as viewed by a corpus-based approach.
The book is concluded with a critical evaluation of translation tools discussed throughout the work and squared into tables. The characteristics of the tools include the extent to which they involve automation and how – and whether – they can be integrated with others. Also, the author mentions the optimal input for each of the tools, whether they are overly dependent on the language, etc. The assumption that rule-based machine translation systems are not algorithm-based appears rather obscure but it can be conceded that the author has her unconventional perception of this notion.
To conclude, the book “Translation and Technology” can be deemed a competent piece of writing. The convenient structure of the book deserves every appraisal, the language is smooth, and the text itself features illustrative graphs that enhance the unprepared readers’ comprehension as opposed to formulae. The scope of discussion that the book offers is quite extensive, and the author’s vast theoretical base is visible. Students of machine translation, as well as those interested in the field, will find this work a big help for their studies.
CAT evaluation
The machine translation tools Chiew Kin Quah mentions in the book are electronic dictionaries, corpora of languages, concordancers, and translation memory systems. The subject of CAT tools is mainly overviewed in terms of future development and models of evaluation.
Each of the non-human translation tools has its advantages and disadvantages. Electronic dictionaries, for one, are convenient and fast at processing the data. One can easily import the word of interest into the base and receive a rapid output. On the other hand, some of the online dictionary platforms deploy a “Wikipedia” approach which is to say that any user can adjust the terms as they believe is appropriate.
At that, a linguistic corpus appears more elaborate and precise. The corpora are CAT tools used for a wide variety of linguistic purposes and come in the form of texts that machines are able to read. The texts of the corpora contain representative information including hundreds of millions of linguistic units and their combinations. That, as well as their sufficiency in resolving complex linguistics-related issues, can be undoubtedly considered an advantage. On the other hand, is constituted by a variety of areas of human experience does not make corpora an ultimate problem-solution tool. The reason being that they are pre-designed in a way including subject fields with the linguistic means appropriate to each of them and consequently may prove inadequate for more specific fields of study. Concordancers are more optimized at that.
These are electronic tools that provide a user with a range of situations and ways in which particular words can be used. Due to the amount of functions these tools are capable of rendering they have been widely applied to translational studies as well as other fields. The outputs of the concordancers can serve as inputs to TM systems. These systems, also known as databases, are sets of lexical entries in which there is a positive association of lexical units of different languages. The translation memory systems are valued for their segmentation qualities. The text is divided into blocks, paragraphs, sentences, and utterances, and in the process of translation, the database is automatically searched for a matching segment. The match can be full or fuzzy, of which the applications inform the user.
The systems have their strong points, which are high speed and quality of translation, and easy and user-friendly control. They are fully customizable and come in many forms from the most basic to the most advanced. TM systems, thus, are a powerful technology that might be sufficient in cutting localization prices. From the other perspective, the usage of TM systems has to be seriously considered since they have proven to be most effective in translating the texts with much repetition.
The point that Chiew Kin Quah has about any CAT tool, and which seems rather justified, is that non-human translation has to be used in moderation, as an extra source of assistance. She states the necessity to exercise one’s brain and remain creative to produce translations that can be considered masterpieces in their field.