ISSN: 1579-9794

Hikma 23(2) (2024), 1 - 8

SIN-WAI, CHAN. ROUTLEDGE ENCYCLOPEDIA OF TRANSLATION

TECHNOLOGY, 2ND EDITION. OXON/NEW YORK, ROUTLEDGE, 2023,

834 PP., 978-0-367-76736-5

Acknowledging that translation technology has brought about

“fundamental changes and additional dimensions to all aspects of the

contemporary world of translation” (p. xxxi), this encyclopaedia aims to serve

as a key reference of fundamental concepts related to translation technology

and of the research undertaken in this field. Given the fast pace at which

technology evolves and the numerous developments involving the use of

artificial intelligence (AI) in translation since the beginning of the 2020s, the

publication of this second edition, which follows a first edition dating back to

2014, comes with updated content and five new chapters, making it therefore

particularly relevant and topical.

With a total of 47 chapters or entries, this collective volume is

particularly welcome considering the scarcity of general and non-research

focused monographs on translation technology. Austermühl (2001), Bowker

(2002) and Quah (2006) became key reference works on translation

technology in the 2000s and the early 2010s, yet we have had to wait until

recently to read a similar publication, such as the textbook authored by

Rothwell et al. (2023). Recent contributions of a more general nature are less

generic and more focused on specific technologies, scenarios and types of

translation (see, for example, Bolaños García-Escribano, 2025). Given that

many of the aspects covered in works on translation technology will inevitably

become dated, in current works the focus is often placed on key concepts and

technologies, and how they have been implemented in representative tools

(e.g. authors might choose a tool like memoQ to explore concepts related to

translation memory (TM) tools).

This is also the approach taken in Sin-wai’s edited work, especially in

Part 1, where general issues and topics within translation technology are

discussed. While there is a strong emphasis on machine translation (MT) in

this part, with 8 out of 15 chapters being devoted to this type of translation

technology specifically, sizeable attention is also paid to the wider notion of

computer-aided translation (CAT) and its training. Although at times it is not

clear whether MT is seen as a type of CAT or rather as a technology that

pursues full automation of the translation process, this part attempts to clarify

the terminological confusion surrounding the very concept of CAT, including

detailed definitions of key notions. Readers will also find compelling reflections

on how the different types of translation technology and resources needed to

operate these are related to each other. Some chapters delve into the concept

2 Rocío Baños

Hikma 23(2) (2024), 1 - 8

of localisation and localisation tools within translation technology, with the

focus often being placed on software localisation and, to a lesser extent, on

web localisation, thus missing the opportunity to introduce media localisation,

a field nevertheless explored in part 3. Despite not having been updated since

the first edition, Chapter 3 is of particular interest. Authored by Garcia, this

entry distinguishes between classic and more current CAT systems and

discusses interesting concepts with a focus on TM and terminology

management.

Within Part 1, one of the most valuable chapters is Bowker’s, which is

articulated around key questions in relation to translation training, such as why

translators need to learn about translation technologies, which types of tools

are relevant to practitioners and which ones should be included in the

curriculum. Bowker provides an updated overview not only of key challenges

in translation training but also of current tools and technologies in professional

environments. When compared to the first edition, it becomes clear that this

chapter has been revamped to include topical aspects such as the need to

consider ergonomic design or to address ethical issues as regards the use of

translation technology in the translation classroom, as well as the role of

neural MT (NMT). Despite some obvious overlaps, a new and welcome

addition is the chapter entitled ‘Teaching translation technology’, authored by

Chan and Shuttleworth, which complements Bowker’s chapter by discussing

different approaches and methodologies to teach translation technology. In

doing so, the authors refer to up-to-date empirical studies, with a refreshing

section devoted to the effectiveness of translation technology teaching.

A further and important addition to the second edition in Part 1 is the

chapter on AI in translation authored by Xueting and Chengze. Together with

chapter 47 on deep learning and translation technology (see below), this

chapter bears the responsibility of providing an overview of new developments

in translation technology, exploring key notions and implementations of deep

learning, machine learning and AI in translation. Although their chapter seems

to focus widely on NMT, Xueting and Chengze also illustrate how AI has

previously been used in other MT models or approaches that have now been

superseded by NMT. Previous MT models are covered in detail in other

sections of the encyclopaedia such as rule-based MT (RBMT) (see chapter

10), example-based MT (EBMT) (see chapter 7), or statistical MT (SMT) (see

chapter 11). AI, understood by the authors as a “field of studies that enables

computers and machines to mimic the perception, learning, problem-solving,

and decision-making capabilities of humans” (p. 280), can be implemented in

translation-related tasks in many ways and not solely to achieve full

automation of the translation process. For example, AI can be used to

automate or at least streamline specific stages of the process (e.g. the

Reseñas 3

Hikma 23(2) (2024), 1 - 8

documentation undertaken by translators or the quality control by reviewers),

as well as for translation-related tasks such as automatic speech recognition

(ASR). However, the focus of this chapter is mostly on MT. Other applications

are nevertheless discussed in chapter 47, which places greater emphasis on

deep learning as a form of machine learning, “which is in turn an approach to

artificial intelligence” (p. 797).

Given the predominance of NMT in current times, the relevance of

some of the chapters on MT included in this part may appear limited in scope.

This is particularly true for approaches that have not been consolidated (see

chapter 9 on pragmatics-based MT). However, these discussions are indeed

essential for readers to understand how the field of MT has evolved over time

(see for example Hutchins’s chapter on the history of research and

applications of MT) and why specific approaches became more prominent

within specific time periods, contexts and environments.

Context and environments gain prominence in Part 2 of this

encyclopaedia, which illustrates “the development and application of

translation technology in different social and cultural situations and at different

levels of technological advancement” (p. xxxv) in diverse countries and

regions. Although the rationale for choosing some regions over others has not

been clearly argued and some continents are not represented (i.e. South

America), praise should be given to the editor’s attempt to provide a global

perspective and showcase development in regions not often reported in the

mainstream literature on translation studies in general and translation

technology in particular, with a wide and diverse coverage of Asian regions.

In this section of the encyclopaedia, authors have adopted diverse

approaches, with some chapters focusing on research and others paying

attention to professional practices. For instance, the reflections on how

technology has evolved in Canada based on Macklovitch’s personal

experience as a translator at the Canadian Translation Bureau are particularly

insightful, showing how some of the then-new tools such as word processors

and terminology banks have hitherto maximised translators’ efficiency. His

arguments on the potential of automatic speech recognition (ASR) are

particularly compelling. Through a series of consultation sessions undertaken

in 2011, the author was surprised to find out that many employees of the

federal translation bureau still used a Dictaphone to dictate their texts, thus

highlighting the potential that ASR could have if it were to be integrated in their

translation workflow. Yet, Macklovitch argues that ASR performance is still not

suitable, both because “word error rates remain too high to allow automatic

dictation to be cost effective” (p. 325), especially in languages other than

English, and because ASR systems are not yet satisfactorily integrated with

the tools often used by translators (particularly TM systems). Although some

4 Rocío Baños

Hikma 23(2) (2024), 1 - 8

of these issues remain in the 2020s and performance is likely to vary between

languages, the situation has improved considerably (see Zapata et al., 2023)

thanks to some of the developments on deep learning and speech translation,

further discussed in chapter 47 of this encyclopaedia.

In chapter 22, Chung-ling also attempts to reflect the views of

professionals, but in a more systematic manner, even undertaking an

empirical investigation to report on the application and teaching of translation

technology in Taiwan. In addition to identifying key trends from the data

gathered, the author proposes concrete and useful suggestions to improve

the situation and address the weaknesses identified, such as offering joint

lectures and on-the-job training on translation technology or providing

government incentives to boost the growth of the localisation industry in

Taiwan. Despite being specific to Taiwan, many of these suggestions arise

from in-depth reflections and are thereby applicable to other contexts, thus

making this chapter a highly valuable contribution.

The remaining chapters from this section also uncover many other

region-specific peculiarities such as the paradox of languages being under

threat in a global economy such as the UK, where translation technology has

certainly thrived in the past (see chapter 24 by Declercq), or DeCamp’s

portrayal, in chapter 25, of MT as an obsession in the US, “not only in

universities and development labs but also in the general public” (p. 423). In

addition to referring to how MT has been portrayed in influential books,

television shows and films (namely, the universal translator in Star Trek or C-

3PO in Star Wars), DeCamp reflects on how the use of technology was

affected by socio-historical events throughout history (such as 9/11 or the

Covid-19 pandemic), offering an incredibly informative timeline of the history

of translation technology in the US at the end of her chapter. Other chapters

focus on language-specific issues such as the challenges of working with

Japanese using translation technology in general and MT in particular (see

chapter 20 by Isahara) or the issues experienced in MT development for the

many languages spoken in South Africa due to resource scarcity (see chapter

21 by van Huyssteen et al.).

Far from providing an overview of all regions across the globe, which

would be an impossible endeavour for a publication of this nature, Part 2

succeeds in offering a snapshot of different practices in an array of contexts,

with varied restrictions and circumstances. These chapters are well suited for

a project of this scale, and the reader will therein find key information, including

similarities and differences between regions, as far as the development of

translation technology is concerned.

Reseñas 5

Hikma 23(2) (2024), 1 - 8

With 22 entries, three of which constitute new additions, Part 3

introduces specific topics in translation technology. Some of these refer to key

aspects or processes within natural language processing (NLP) such as part-

of-speech tagging, segmentation or information retrieval and text mining.

Some chapters, nevertheless, seem to be more generic, such as chapter 37

(localisation), chapter 38 (NLP), chapter 43 (subtitling) or chapter 47 (deep

learning and translation technology). Attention is also paid to key resources

for translation technology tools, such as corpora (chapter 31) or TM (chapter

45), as well as translation management systems (chapter 46). In addressing

key concepts and applications, this section brings to the fore phenomena and

practices that had been described as recent or future developments in

Chapter 1 by the editor of this encyclopaedia, and which include

crowdsourcing (see chapters 39 and 43, among others) or the prominence of

cloud-based translation technology (see chapters 33, 43 and 46).

Probably due to the significance that post-editing (PE) has acquired

over the past decade in the translation industry, considerable attention is

directed towards editing in translation technology in general (chapter 32),

editing in audiovisual translation (chapter 33), and MTPE (chapter 34). These

last two chapters are very relevant additions. Chapter 34 complements

chapter 32, which already includes a section on PE. In addition to providing

relevant definitions and an overview of guidelines and types of PE, as well as

a summary of research findings on aspects such as PE productivity and effort,

chapter 34 includes very interesting reflections and suggestions on the skills

needed to become a successful post-editor. Considering both the surge that

AVT has experienced in recent decades and the specificities of subtitling, the

inclusion of chapter 33 is a welcome gesture which complements existing

encyclopaedia entries on AVT and technology (see Baños, 2018 or Díaz-

Cintas and Massidda, 2019). As the authors argue, the complexities of editing

tasks – whether human or machine-generated – are often greater in AVT in

general, and subtitling in particular, as they involve both linguistic and

technical editing. Indeed, Bolaños García-Escribano and Declercq identify at

least 6 types of editing that are often performed in subtitling before, during or

after the translation process (namely, pre-editing, post-editing, revision,

proofreading, quality assurance or quality control, and post-quality control

viewing), as well as a supra-editing AVT-specific type of editing that can

appear at any point of the editing process (e.g. truncation).

Chapter 47, also a new chapter of this second edition, not only

illustrates the dramatic changes that deep learning has brought about to NLP

but also helps readers to understand how some of the technologies and

processes explored in Part 3 (e.g. MT, ASR, speech synthesis, sentiment

analysis or text classification, among others) have been affected by such

6 Rocío Baños

Hikma 23(2) (2024), 1 - 8

developments. The subsections on the impact of deep learning on

intersemiotic translation and translation memory are of great relevance, as is

the final discussion of trends that propel the advancement of deep learning for

translation and cross-lingual communication according to the authors. Large

language models, which are widely popular in the translation industry at the

time of writing, are briefly mentioned in this chapter, especially in relation to

those developed by Open AI (i.e. GPT). Xueting and Chengze refer to GPT-3

as both an “enormous” (p. 283) NLP processing system and a “recent

transformer model for text generation” (p. 300).

In addition to the breadth of topics covered, one of the main strengths

of this publication is that it capitalises and reflects on the work of many

scholars and specialists who work on and with translation technology in varied

settings and environments. Among them are authors who once undertook

pioneering research in this field and have thus become household names

immediately associated with translation technology (such as Bowker,

Hutchins or Melby, to mention but a few).

While some chapters may appear too technical and therefore targeted

at expert users or readers already familiar with a specific technology (see, for

example, chapter 26 or 35), many others succeed in providing a clear, easy-

to-understand overview of key concepts for a wider audience while also

introducing more technical aspects for more experienced readers. An example

of this style is the chapter authored by Forcada (chapter 8), on free/open-

source MT, which provides very clear and useful explanations of how different

MT approaches work in addition to exploring the differences between

commercial, free/open-source MT and non-free, open-source MT. Despite

including complex technical information, the chapter authored by Sánchez-

Martínez (chapter 40) masterfully explains how part of speech (PoS) taggers

work, introducing them as programmes that attempt “to assign the correct PoS

tag or lexical category to all words of a given text” (p. 702). This chapter also

explores how PoS taggers are widely used in NLP applications such as MT,

with a focus on RBMT and SMT.

As regards coverage, readers might be surprised not to find specific

chapters on either NMT or speech recognition. While such chapters would

have been very welcome, NMT is widely covered in chapters 15 and 47, as

the editor rightly argues in the preface to this second edition. A substantial

section on speech recognition is also included in chapter 42 on speech

translation.

As regards organisation of content, readers might find that the

distribution of the chapters can be slightly confusing. Indeed, although the

volume has been divided into three parts, chapters are not always presented

Reseñas 7

Hikma 23(2) (2024), 1 - 8

following a specific logic. For instance, Part 3 seems to have been organised

largely alphabetically, according to the name of topic covered, with the

exception of the new chapters, resulting in some inconsistencies. Additional

inconsistencies might also be found in the presentation of information in this

second edition and while many could likely have been addressed through

more detailed proofreading and editing, they are also to be expected in a work

of this extent and scope.

As explained in the editor’s preface, entries act as independent

chapters and authors have clearly been granted substantial freedom to

approach their work as they considered appropriate, which brings a novel

angle and diversity to this book. Nevertheless, at times a more systematic

approach would be desirable in a reference work of this kind. For instance,

readers would probably appreciate finding an introduction and conclusion in

each entry, and perhaps a section with useful references and resources

(which has been included on occasion, but not methodically). As already

mentioned, there seems to be a slight imbalance as regards which of the

entries included in the first edition have been updated. While some have been

very carefully revised (see, for example, chapters 4, 29 or 46), others have

not. Consequently, some technologies are described as needing further

research or development (e.g. ASR) and trends as being emerging instead of

well-established (i.e. crowdsourcing), thus overlooking or not reporting on

recent advancements. Considering the fast pace at which technology evolves

in this field, this fact detracts from the quality of the publication. Having said

that, the work involved in editing and reviewing an encyclopaedia of this

magnitude can be endless, so ensuring that contents remain relevant

throughout and are addressed appropriately as well as authored by renowned

authorities are in themselves enormous challenges that the editor has

overcome in a very successful manner. Overall, and despite the minor caveats

mentioned in this review, this encyclopaedia is to be regarded as a key

reference work for both students and scholars who would like to further their

understanding of both general and specific aspects of translation technology.

REFERENCES

Austermühl, F. (2001). Electronic tools for translators. St Jerome.

Baños, R. (2018). Technology and Audiovisual Translation. In C. Sin-wai

(Ed.), An Encyclopedia of Practical Translation and Interpreting (pp. 3–

30). Chinese University Press.

Bolaños García-Escribano, A. (2025). Practices, education and technology in

audiovisual translation. Routledge.

8 Rocío Baños

Hikma 23(2) (2024), 1 - 8

Bowker, L. (2002). Computer-aided translation technology: A practical

introduction. University of Ottawa Press.

Díaz-Cintas, J. and Massidda. S. (2019). Technological advances in

audiovisual translation (pp. 255-270). In M. O’Hagan (Ed.), The

Routledge Handbook of Translation and Technology. Routledge.

Quah, C. K. (2006). Translation and technology. Palgrave Macmillan.

Rothwell, A., Moorkens, J., Fernández-Parra, M., Drugan, J., & Austermühl,

F. (2023). Translation tools and technologies. Routledge.

Zapata, J., Secară, A., & Ciobanu, D. (2023). Past, present and future of

speech technologies in translation — life beyond the keyboard. In J.

Moorkens & V. Sosoni (Eds.), Proceedings of the 44th Translating and

the Computer (TC44) Conference (pp. 16–25). Tradulex.

[ROCÍO BAÑOS]