15 September 2016

OSASCOMP in the news

OSASCOMP: Applied analysis is by far the most popular post on this blog.
Today OSASCOMP made, if not the front page of the Guardian, at least Tim Dowling's
Reference and languages Notebook under the heading
Order force: the old grammar rule we all obey without realising.
followed by:
I had no idea there was a specific order for adjectives until I read a viral post.
The viral post focussed on the following quote:

from  a book called The Elements of Eloquence by Mark Forsyth.


The paragraph concerned the order of adjectives – if you’re using more than one adjective before a noun, they are subject to a certain hierarchy. You know it’s proper to say “silly old fool” and wrong to say “old silly fool”, but you might never have thought about why – or if you did you probably imagined it was just some time-honoured convention you picked up by rote. But it isn’t. There’s a rule.

Unlike many laws of grammar or syntax, this one is virtually inviolable, even in informal speech. You simply can’t say My Greek Fat Big Wedding, or leather walking brown boots. 

In a piece for the BBC, ... Mark Forsyth examines a rare exception to the adjectival hierarchy: the Big Bad Wolf. Bad is opinion, and should therefore come first. However, as Forsyth points out, this phrase is too busy obeying another rule I’d never heard of: the rule of ablaut reduplication.

Other examples of the rule in action include chit-chat, singsong, flipflop and hip-hop. When you shift vowel sounds for effect this way, the vowels always follow a specific order: I, then A, then O.

If you’re a native speaker, the hierarchy is ingrained in you.

With moderation

Broadly I agree with everything both Dowling and Forsyth say and am very pleased to see that the topic is winning increased attention as it is often especially important to into-English translators and more particularly to those who translate technical journalism. I'm also pleased to have learned about the exception attributed to ablaut reduplication.

I would, however, like to make a couple of qualifying statements:
  1. The statement "If you’re a native speaker, the hierarchy is ingrained in you" sometimes fails when the OSASCOMP categories are difficult to assign or the subject matter is beyond the writer's comfort zone (as occurs often when good into-English translators tackle subjects they don't fully understand). On other occasions, OSASCOMP-attentive readers will observe that even good writers do occasionally slip up for whatever reason.
  2. For technical writers and translators, OSASCOMP is difficult to apply (a) because the categories are difficult or impossible to assign, or (b) because the challenge is to combine one or more adjectives with other types of qualifiers. This applies especially to noun qualifiers and things like prices or values combined with units of measurement as qualifiers.
Does '12-metre' before 'boat' or 'US$4-million' before 'ship' follow the rule as if it were a 'size' adjective?
I will follow this up with examples of (a) OSASCOMP slips by mother-tongue writers and journalists and (b) OSASCOMP challenges encountered in technical communication and the translation of technical journalism.

English should have a word for ...

Here are two suggestions from The Atlantic's What Concept Most Needs a Word in the English Language? column:
We need a word for the mental suffering that results from someone else’s misuse of a word or phrase in one’s presence, the distress being magnified by an abiding sense of politeness that precludes correcting the other person—coupled with an intensifying melancholy about the confused changes that so many words are undergoing as a result of mass indifference to linguistic tradition. I suggest wordschmerz.
Bryan A. Garner, author, Garner’s Modern English Usage 
The fallacy of attributing every unfortunate outcome to deliberate intent,and neglecting the possibility of incompetence, unplanned by-products, or entropy.
Steven Pinker, author, The Blank Slate

09 September 2016

Workflow for low-tech .doc or .docx to .idml, #2

Since posting Workflow for low-tech .doc or .docx to .idml in July, I have an opportunity to put my own suggested workflow into practice.

It's certainly a pity we didn't follow step 1 to the letter.
After laying out far too many Word documents in InDesign, we noticed that the non-breaking hyphens (aka NBHs or hard hyphens) had apparently disappeared in InDesign.

McKee (@MacKeyComp), author of Moving Text From Word to InDesignexplained that, when importing a Word document into InDesign, NBHs are not recognising as such. Indeed, in his testing, they are treated as Bookmarks. If you open the Bookmarks Panel, you’ll likely find Bookmarks with names like “_GoBack” in place of where the NBH was entered in Word. You can’t search for Bookmarks in InDesign’s Find/Change, but you can use the Bookmarks Panel to go right to it and enter it manually. It helps if you have Type > Show Hidden Characters on in InDesign.

Jamie also reminded me that, had I wanted to, I could also have made changes to the Word documents already drafted with NBHs all over the place using Word’s Find and Replace to search for “^~”.

The next step was to ask Stanislav Okhvat, the developer of TransTools, about batch processing Word files. He's planning on adding this capability to his product, but it's not available yet.
He suggested two options:

  • a freeware Word add-in by Funduc Software (see here) for one replacement at a time in multiple files, and
  • for multiple replacements according to a replacement list, try Batch Replacer for MS Word, a shareware product that neither Stanislav nor I have tested to date, but which looks promising.


Jamie's finding (Word NBHs become bookmarks) only applies, apparently to certain combinations of hardware and software on the different computers.
Tests I conducted showed that on other occasions Word NBHs simply disappear into the ether leaving the InDesign document with the hyphenated elements contiguous.
Such is life!

Anyone out there got any workarounds?

01 September 2016

Le FMI juge incompréhensible le langage du gouvernement français

Thanks to Cath Cellier-Smart ‏@Smart_Translate  for flagging the following article in Le Figaro économie

One extended quote says it all:
Ces deux expressions (Crédit d'Impôt pour la Compétitivité et l'Emploi et Pacte de Responsabilité et de Solidarité) figurent bien sûr nommément dans le rapport d'une soixantaine de pages du FMI, mais elles n'ont pas été traduites en anglais. Et pourtant il s'agit de mots simples, «crédit, impôt, compétitivité, responsabilité, etc.». Leur équivalent anglais ne pose aucune difficulté. Il n'y a vraiment aucune complexité ni finesse littéraire qui serait impossible à retranscrire dans une autre langue. Mais les économistes de Washington ont préféré renoncer. Et on les comprend assez bien.
A quoi bon aurait servi de traduire «Pacte de responsabilité et de solidarité» par «Solidarity and Responsability Pact»? Le lecteur anglophone n'aurait pas mieux saisi ce que cela signifie dans sa langue, car cela ne veut en réalité rien dire. C'est de la langue de bois, ampoulée et opaque, qui ne se comprend pas mieux d'ailleurs en français qu'en anglais. Que veut dire «responsabilité», «solidarité» dans un tel accouplement? Il faut deviner les intentions, plus ou moins démagogiques, de leurs auteurs qui sont derrière. Mais à la lettre, de tels assemblages de mots n'ont aucun sens. De même la mélasse sémantique que renferme le CICE est incompréhensible et donc intraduisible.
Writing to impress rather than express I'd say!
Wonderful, however, that Le Figaro wrote this up in this way and agreed in full with the IMF's views.

For more information or to download the report entitled France: Selected Issues; IMF Country Report No. 16/228; June 24, 2016, go here.

26 August 2016

St Jerome and transcreation

In July 2016, theology student Lydia Lange's doctoral thesis was published in book form in German by De Gruyter.
Lange, Lydia: Die Juditfigur in der Vulgata. Eine theologische Studie zur lateinischen Bibel. Verlag De Gruyter, 2016. 456 pages, 129,95 Euro. ISBN 978-3-11-048823-4
Her supervisor was Prof. Barbara Schmitz, Head of the Chair of Old Testament Studies and Biblical-Oriental Languages at the University of Würzburg.

The University's online press release, complete with some interesting Germanisms, is entitled 

Lydia Lange (32), Faculty of Catholic Theology, 
Julius-Maximilians-Universität (JMU) Würzburg in Bavaria, Germany
in the library of the chair of Old Testament Studies and Biblical-Oriental Languages

"The Book of Judith is one of the more recent works of the Old Testament; it was written down for the first time about 100 b. C., in Greek"
"Jerome emphasizes Judith's chastity - but the Greek original and the other Latin translations do not mention that word at all",
Jerome may also have used a strategy that is commonly used in theater and literature up to this day: recasting an old story for the present.
The implications for theology, sacred book translation and translation tout court are considerable, to say the least.

20 August 2016

The parlance of pilots

The parlance of pilots is, quite simply, an amazing piece. You can read it on the aeon.co site. It was written by Mark Vanhoenacker, a senior first officer with British Airways, the author of Skyfaring: A Journey with a Pilot (2015) and a regular contributor to The New York Times.

A few quotes:
It’s hard to imagine a system more in need of a common language. And that language is English (or English-derived Aeroese).

I like the word atmosphere, for no other reason than because we so rarely think of the air as a sphere, one that floats just above and envelops the heavier world of land and water.
Each Boeing or Airbus airplane is sold with an entire library of associated technical manuals. They’re all in English, too (a particularly remarkable fact for Airbus, which is headquartered in France). A small number of airlines might translate these manuals – an expensive endeavour, and a never-ending one, as the manuals are frequently updated.
The same is generally true of the checklists that pilots read to one another at key points in a flight. These checklists – a sheet of laminated paper or, increasingly, a display on a computer screen – are a simple but critical component of flight safety (see ‘The Checklist’(2007) by Atul Gawande in The New Yorker).

It’s also worth noting that Airbus and Boeing have distinct dialects of Aeroese. For example, there’s a system on airliners that we can regard as a sort of cruise control (though that’s not a perfect analogy). On Boeings, it’s the autothrottle; on Airbus jets, it’s theautothrust.

Another defining (and simplifying) feature of radio Aeroese is its small vocabulary. Indeed, in addition to callsigns and numbers, the words regularly used in everyday air-traffic communications probably amount to only a few dozen. Even these few words are subject to usage and pronunciation rules designed to correct the shortcomings of quotidian English.
For example, we’re instructed to pronounce three as ‘TREE’ and nine as ‘NINER’, and 25,000 as ‘two-five thousand’ (more specifically, ‘TOO FIFE TOUSAND’), not ‘twenty-five thousand’, because experience has shown that these modified pronunciations are less likely to be misunderstood.
For the technical communicator, translator or journalist, the precision of Mark Vanhoenacker's technical language combined with his sense of style makes every sentence a joy to read. If you don't believe me, just listen to the version on the aeon.co page, courtesy of curio.io. This is the sort of flow that technical writers and translators should aim for. Writing worth listening to is the name of the game.

19 August 2016

Beyond curly quotes

Brit Bitch Berlin is one hell of a name for a blog and The Short and Curlies is one hell of a name for a post ... but the lady (Ms Galina Green) writes both well and wisely.

The Short and Curlies is on:
Quotation marks, double inverted commas, speech marks, guillemets, goose feet, citation marks, duck feet, smart quotes, curly quotes, dumb quotes, whatever you want to call them, we really have to master them in the languages we work in – there’s no goose stepping around it. Oh and then there are the scare quotes and – my favourites – air quotes.
and it's fascinating.

At one point she writes:
Just for the record, the only place you should be using dumb quotes is when you’re coding. Which is probably never.
On this I beg to differ.
The other place is when compiling glossaries and doing terminological work. Why? Because once your efforts have been indexed by an indexing engine, some search engines are liable, when launching a search containing quotation marks or an apostrophe will sometimes miss the curly kind, but never the dumb ones. At least that's been my experience to date.

Her post on  is also very good.

01 August 2016

Preparing Word files for InDesign

As discussed in Workflow for low-tech .doc or .docx to .idml, this workflow requires that the translator carefully format Word files containing text to be imported into InDesign. Today I want to say a little more about adding hard spaces (aka NBSPs) to the said files to reduce the workload when proofreading documents laid out using InDesign.

Thanks to some excellent support by Stanislav Okhvat the developer of TransTools, I now have this excellent bag of tricks properly configured. I have now configured the suite's NBSP Checker to automate the otherwise fastidious task of inserting NBSPs.

Stanislav's procedure to configure this TransTools function is described here.
It took me a while to get my head around a couple of key steps, but it was well worth the effort.

Thanks to Workflow for low-tech .doc or .docx to .idml and TransTools I can now insert hard spaces automatically, more systematically and far more quickly.

For me, the procedure needs to go just one step further. Stansilav, perhaps you could now apply your proven skills to develop a Hard hyphen checker similar to your excellent NBSP checker.

A few hours later:
... And you know what, Stanislav just informed me that he's actually working on a hard hyphen and dash checker. He says it may take a while to finalise, but I, for one, am looking forward to the results.

26 July 2016

Who needs to go from .doc or .docx to .idml?

Following my earlier posts on workflows involving the delivery of translations as .doc or .docx files with a view to producing end products laid out using InDesign, some readers have asked me to explain the sorts of projects I refer to.

First, it's true that many translation projects involving InDesign come from clients who want the translator to produce an end product in the target language from a source document already laid out using InDesign. For the corresponding workflow, see Workflow for .idml to .idml.

Today's question is, however, Who needs to go from .doc or .docx to .idml?
Well, there are translation customers in France and elsewhere who publish some documents in one or more target languages but not in the source language.
Indeed, when you think about it, this situation is potentially applicable to any organisation marketing or promoting products or services in foreign-language markets differently from the way these things are done on its home market.

More specifically, some of my clients draft technical journalism articles in French that are tailored specifically to promote French products and services in English to international target audiences using arguments, writing styles and publication formats that are different from those used in France. In the case of end documents laid out using InDesign, the source articles are drafted over a period of several weeks, translated and laid out progressively as the articles are made available, then published as an online and/or printed magazine as the last step of a long process.

The French-mother-tongue technical journalists draft their articles using Microsoft Word; the translators draft their translations using Word; and the page layout team does the layout using InDesign. The completed magazine is published in English, but not in French.

15 July 2016

Workflow for low-tech .doc or .docx to .idml

Note: the Caveats listed in Workflow for .idml to .idml apply.

Case 3: Source language text mostly unformatted in Microsoft Word, but target language version to be laid out using Adobe InDesign.

The following workflow is based on a tip provided by Jamie McKee (@MacKeyComp) in the course of private exchange. Jamie McKee (@MacKeyComp) is the author of a high recommended post entitled Moving Text From Word to InDesign.

This workflow is for translators and graphics layout teams working for translation clients aiming to produce a target language documents laid out using Adobe InDesign from mostly unformatted source language documents in Microsoft Word. More specifically, this workflow is for cases where that described in Really smart workflow for .docx to .idml is not feasible for whatever reason.

Proposed workflow
1) Run a test by compiling a short Word file using only Word's "Normal,n" and containing local character attributes (bolditalics, etc.), non-breaking spaces (NBSPs), non-breaking hyphens, curly apostrophes, curly quotes and so forth. Ask the layout team to import the text into InDesign using the Place command (in the French version of Adobe InDesign this is called Importer) and to check that the said NBSPs) non-breaking hyphens, curly apostrophes and curly quotes are preserved.
2) If the test fails, find out why and repeat until successful. It may, for instance, be necessary to use a different style or styles in the Word document.
3) Ask the authors and translators to build their Microsoft Word files using only the Word style or styles successfully tested in step 2 and local character attributes (for bold, italics, etc.). From the outset authors and translators should use non-breaking spaces (NBSPs), aka hard spaces, non-breaking hyphens, curly apostrophes, curly quotes, and any other relevant tricks to ensure that layout and proofreading progress as smoothly and quickly as possible.
4) When the translation work proper has been completed, the translators should run Stanislav Okhvat's TransTools to ensure the consistent use of apostrophes, quotation marks, non-breaking spaces (NBSPs) and the like.
5) The translator(s) now forward their .doc or .docx files to the layout team. Be sure to ask the layout team to import all text exclusively using the Place command (in the French version of Adobe InDesign this is called Importer) as this is the only way to preserve local character attributes (bolditalics, etc.), non-breaking spaces (NBSPs), non-breaking hyphens, curly apostrophes, curly quotes and so forth.

Really smart workflow for .docx to .idml

Note: the Caveats listed in Workflow for .idml to .idml apply.

Case 2: Source language text in Microsoft Word formatted in strict compliance with a comprehensive Word template and target language version to be laid out using Adobe InDesign.

The following workflow is based on a post entitled Moving Text From Word to InDesign by Jamie McKee (@MacKeyComp). Although a keen user of Word styles, I have yet to have the privilege of working with a client, or a client's graphics layout team, that appreciated their benefits.

This workflow will only work well if the translation client, the translator(s) and the translation client's graphics layout team share in-depth knowledge and a commitment to continuous improvement in the use of both Microsoft Word and Adobe InDesign.

Proposed workflow
1) Ask the layout team to create their InDesign Template as detailed in Moving Text From Word to InDesign and taking into account the recommendations in 8 Steps to Optimize InDesign Files for Translation.
2) Ask the layout team to export the said InDesign Template to Word so that authors and translators work consistently with the same set of styles.
3) Ask all authors and translators to exercise extreme discipline in the use of the Word template and local character attributes. From the outset authors and translators should use non-breaking spaces (NBSPs), aka hard spaces, non-breaking hyphens, and any other relevant tricks to ensure that layout and proofreading progress as smoothly and quickly as possible.
4) When the translation work proper has been completed, the translators should run Stanislav Okhvat's TransTools to ensure the consistent use of apostrophes, quotation marks, non-breaking spaces (NBSPs) and the like.
5) The layout team should now be able to import the Word file (including its template) directly into Adobe InDesign as detailed in Moving Text From Word to InDesign with maximum efficiency.
6) Ask the layout team to fine-tune the layout then generate pdfs to send to the translators and others for proofreading. Again, this step should now be far more efficient than using less sophisticated workflows.

This workflow calls for a lot of preparatory work, understanding, commitment and rigour. As a result, it may not be feasible in the case of a one-off project. In the case of regular projects using the same authors, translators and layout team, it should, however, prove a real time saver.

Workflow for .idml to .idml

This is the first in a series of posts concerning translation projects — and more particularly technical journalism translation projects — where the target-language text is laid out using Adobe InDesign.

Caveats: I am writing this primarily for translators who do not use Adobe InDesign, but have customers that do. I write as a translator who has often found this sort of project challenging; also as one who has no first-hand knowledge of Adobe InDesign or any other desktop layout software. If, as a result, my explanations need correction or refinement, please feel free to comment accordingly. I further write as a translator who has no first-hand knowledge of any of the TenTs mentioned here (sorry if that comes as a surprise ...), so, again, if my explanations need correction or refinement, please feel free to comment accordingly.

Case 1: Source language text already fully laid out using Adobe InDesign and translation client wants the target language version laid out in (approximately) the same way with Adobe InDesign.
Note that this is not a type of project that I have encountered personally.

Proposed workflow: Use a state-of-the-art translation environment tool, or TenT — also, but less precisely, known as a computer-assisted translation (CAT) tool or a translation memory (TM) tool. For memoQ, see Kilgray’s Language Terminal. For Trados Studio 2015, see here.
For more on TenTs, see Jost Zetzsche's Tool Box Journal (basic subscription is free), or The Translator's Tool Box: A Computer Primer for Translators Version 12.
The best workflow would appear to be that suggested by Jost on p346:
Kilgray’s Language Terminal has changed the way translators can work with InDesign files. One of the various features of Language Terminal is the ability to upload InDesign IDML files of any version to a server, which converts these files to a memoQ-specific version of XLIFF. 
More from Stanislav Okhvat, developer of TransTools:
Kilgray’s Language Terminal also supports Adobe InDesign INDD files (which are otherwise not readable by anything other than InDesign). It does this by converting them to IDML behind the scenes thereby allowing translators to download a PDF preview of an INDD + MQXLZ package for translation in memoQ. This package can be analysed by any ZIP archiver in order to extract the XLIFF file and translate using any other another TenT tool. By uploading the MQXLZ file back into Language Terminal, one can then download the translated IDML.
Also, Memsource Cloud now integrates with products by several new technology partners including Frontlab offering similar capabilities: besides IDML, one can upload INDD to Memsource Cloud (which will be converted to IDML behind the scenes), translate it and download the translated IDML file.
This workflow works better if the layout team is familiar with translators' needs and more particularly if they prepare the layout specifically for translation. One of many possible links on this is called:
8 Steps to Optimize InDesign Files for Translation.

Note that lots can go wrong, so time must be allowed for corrective work by the layout team.

As 8 Steps to Optimize InDesign Files for Translation says under the subheading Intelligent use of white space in InDesign document layout:
The biggest challenge in designing InDesign document templates for multilingual projects is creating page layout that will accommodate post-translation text expansion. It is not easy to create a source English document that has enough of white space or "breathing room" around text elements. Many languages (e.g. Russian, German, Italian, Latin American Spanish) can expand the line count by as much as 35%.
Language expansion is further magnified by narrow containers, e.g. side notes in the margin, table cells, indented text or boxed cautions and warnings. In addition, some eastern languages (e.g. Arabic, Farsi, Urdu) display text "right-to-left"; this requires right alignment and right-to-left layout modification. Text expansion and text direction require a flexible layout designed by professionals who understand the challenges of multilingual desktop publishing and graphic localization. Your translation company can be a great help in this regard.
In the case of French to English, and many other language pairs, the problem is not so much language expansion, but language contraction.

06 July 2016

dtSearch + IntelliWebSearch, latest

As mentioned in my previous post on dtSearch + IntelliWebSearch on 12 June 2015,
dtSearch can instantly search terabytes of text across a desktop computer, making it ideal for translators, terminologists, LSPs and wordworkers in general. 

For translators and terminologists working on personal computers running Microsoft Windows, the combination of dtSearch Desktop and dtSearch offers remarkable benefits.

Those using IWSv3 can invoke a local dtSearch Desktop search by following the procedure detailed at 
How to set up a local dictionary in IntelliWebSearch
using these settings:

Notes=Change path according to your set-up (you may also add {Enter} to the end of Finish).
Quotes Off=Yes
Pluses Off=Yes

If you have both IWSv3 and IWSv5, you can set up the above settings for IWSv3 then import them into v5.
If you only have v5, follow the information and links above.
If all else fails, you can contact me or Michael Farrell and ask for the IWSv5_dtSearch.ini which you can then import into IWSv5.


Here's a find for into-European Portuguese translators researching not only current terminology but also terminological changes since the first web sites in PT-Eu were put on line.

If you're interested, take a look at Arquivo.pt.

The man behind the project is Daniel Gomes. His Facebook page is here. The Arquivo's Facebook page is here.

28 June 2016

Terminological consistency #5: Metonyms et al.

I wish I had read Matthew Kushinka's excellent article À Propos: Synonyms in French under the ATA French Language Division's À Propos rubric before starting this series of posts on Terminological consistency.

Matthew found the terms and examples that my pieces lacked. The good news, I think, is that we are thinking about the same issue from parallel perspectives.

A metonym is a figure of speech in which a person or thing is called by another name rather than its own. (Think about how many times you’ve seen l’Elysée used to refer to the French government.)
I was pleased to see that Matthew's best example of multiple stand-ins in French is drawn from a piece of science writing for the general public, a text type that I see a form of technical journalism:
Mais le comble ? In an article in Science & Vie magazine by French science writer Lise Barnéoud titled Vers la fin des grands arbres, les grands arbres are referred to in almost twenty different ways: as doyens de la nature, maîtres de l’espace et du temps, rois des forêts, and titans ligneux, to name a few. (You can read my post 18 Ways to Say ‘Large Trees’ in French for the other phrasal synonyms that she uses.)
Matthew goes on to say:
Unfortunately, I don’t have any data on “synonym density” between French and English. (Corpus linguists, consider that an idea for your next academic paper!) But I suspect that the French use synonyms, metonyms, and other lexical stand-ins more frequently than Americans.
Like Matthew, I suspect that the French make greater use of metonyms and other lexical stand-ins than English-mother-tongue writers of the text types under discussion. And I too would like to read some serious corpus-based research confirming or disproving this hypothesis.
And if anyone wants to take the concept even further, it would be fascinating to see languages ranked according to this criterion.

18 Ways to Say ‘Large Trees’ in French includes literal translations of the 18 stand-ins for les grands arbres used in Lise Barnéoud's Vers la fin des grands arbres.

When it comes to translating text types that may make greater use of metonyms and lexical stand-ins in either the source or target language (journalism, political discourse or management speak are three text types that come quickly to mind), I believe that today's theories of terminology, not to mention the supposed benefits of CAT/TenT termbases, leave much to be desired.
CAT: computer-assisted translation
TenT: translation environment tool

23 June 2016

Terminological consistency #4: A different approach to meet different challenges

Below I explain how I applied the experience, analysis and observations recorded in the last three posts — Terminological consistency #1, Terminological consistency #2 and Terminological consistency #3 — to the task of compiling my French-English glossary of naval technology.


The broad context in which I spent 15 years translating corporate literature and technical journalism concerning the French naval defence industry can be summarised as follows:
  • Most documents can be broadly classified as technical journalism as described in Why 'technical journalism'?TJ and term variability and Defence is different.
  • Most documents were written by engineers, in-house communicators or other journalists writing in French journalistic styles and completely unaware of the benefits of technical communication guidelines, beginning with terminological consistency. On the other hand, I am bi-cultural enough to readily acknowledge that most mother-tongue readers of French naturally object strongly to repetition and probably also find it off-putting. Conclusion: While terminological consistency would make life easier for the Fr-To-En translator it is very unlikely to improve this type of text for the designated readership.
  • Application of the method described in Value propositionTranslation by emulation, take #1 and Translation by emulation, take #2 meant that I typically aimed to emulate the style and devices used by mother-tongue contributors to publications like IHS Jane's Naval Forces.
  • Terminologically, Jane's Naval Forces and most of the other technical and defence journalism that I have read over the years are characterised by a combination of (a) some rigorous standardised terminology, (b) some precise terms and designations established by product manufacturers, programme authorities and the like, (c) extensive use of acronyms and abbreviations that typically appear once in long form early in each article, (d) extensive use of metonyms, synonyms, generics, and other lexical stand-ins and the like, and, last but not least, (e) a constantly evolving sprinkling of jargon. The last three categories are used, naturally enough, to avoid repetition and to add sparkle and variety.
Such is the essence of the translation challenges posed by technical journalism and such is the essence of the highly personal structure and organisation adopted in compiling my French-English glossary of naval technology (aka #NavTechGloss).

Metronyms in #NavTechGloss

  • Balardgone (= Pentagone à la française / État-Major des armées / Ministère de la Défense et État-Major des armées / Centre de Planification et de Conduite des Opérations / CPCO)
  • bateau gris (= navire de guerre)
  • bateau noir (= sous-marin)
  • Hôtel de Brienne (= Ministère de la Défense) 
  • Hôtel de la Marine (= État-Major de la marine)
  • La Royale (= la Marine française)
  • rue Royale (deprecated) (= État-Major de la marine)
  • rue Saint-Dominique (deprecated) (= Ministère de la Défense)
For more on metronyms, see Terminological consistency #5.

Terminological consistency #3: Observing the French

In this post I summarise some observations concerning terminological consistency as an English-language technical communicator and an into-English translator who put two children through the French education system (primary, secondary and tertiary levels) and has spent 40 years working with and for French engineers employed in the aerospace, remote sensing and naval defence industries.

Some personal observations of the French education system and its graduates since say 1975:

  • Literature was consistently held up as the sole model for nearly all types of writing. Indeed I seldom if ever heard anyone discuss the importance of simple, direct language or consistent terminology.
  • Many, many French people clearly remember and, as a result, remain enslaved to, the dictates of primary- and secondary-level teachers concerning the importance of non-repetition when writing essays. Then, because they were not also taught that other writing styles demand different techniques, they applied this rule throughout their lives to anything and everything they are required to write.
  • The memory I refer to was remarkably vivid because it often took the form of red circles around any repeated word or expression on any given page joined by a red line overwritten with a big red R for répétition.
  • I was amazed how often this image of the big red R came up when explaining why my translations featured greater terminological consistency, hence more repetitions than the original. My clients often asked me to use more synonyms as they saw terminological consistency (aka repetition) as poor writing style.

Terminological consistency #2: Some broad generalisations

Before explaining my thoughts on the limitations of vendor claims on terminological consistency allow me to record here some generalisations that (a) I hold to be broadly valid, and (b) provide a framework for my analysis.

On the relative importance of technical communication and literature in different languages

  • In most languages, the status of technical communication is low to virtually non-existent compared to that of literature.
  • The small number of languages in which technical communication enjoys relatively high status compared to literature include English, German, the Scandinavian languages and Russian.
  • Western European languages in which technical communication has low to non-existent status compared to literature include the Romance languages, with French showing early signs of a little change.
  • Languages in which literature enjoys very high status and technical communication has low to non-existent status generally — please correct me if you think I'm wrong — consider terminological consistency as either a low or irrelevant priority in most writing styles.
  • Broadly speaking, where literature enjoys very high status, I believe that this correlates with cultural emphasis on the importance of non-repetition. Similarly, where technical communication enjoys relatively high status, I believe that this correlates with at least a degree of cultural awareness of the importance of terminological consistency in a broader range of writing styles

On the importance of terminological consistency in translation

  • As a first approximation, the terminological consistency of a translation should reflect that of the original. If the original is inconsistent, the translation should probably follow suit. Corollaries: (1) Term mining remains critical, but work on termbases is not likely to be cost-effective. (2) The benefits of translation memories and termbases approach zero.
  • All of the claims made by language service providers (LSPs) and translation software vendors regarding terminological consistency are 100% valid where the original was produced by competent technical communicators writing in their mother tongue in rigorous compliance with a comprehensive termbase.
  • The situation just described is far more common when translating from languages in which technical communication enjoys relatively high status (e.g. English, German, the Scandinavian languages and Russian).
  • When translating into a language in which technical communication enjoys relatively high status, greater terminological consistency in the target language may represent significant added value, particularly in writing styles where the target readership can be assumed to be familiar with good technical communication. Corollary: This often implies N equivalents in the target language for M terms in the source language where N is significantly smaller than M.
  • When translating into a language in which the status of technical communication is low to compared to that of literature, it may be wise to submit a trial batch of translated material to the client for analysis. There are writing styles in these languages where excessive repetition — the inevitable corollary of rigorous terminological consistency — may be viewed as a shortcoming.

Terminological consistency #1: Vendor claims

Below I have compiled a handful of claims by translation and language software vendors on the importance of terminological consistency.

SDL Trados Studio

SDL offers a downloadable brochure entitled The importance of corporate terminology. On page 8 it says:
By defining a concept and assigning one allowed term per language to it, a company can make sure that everybody involved in the communication process is clear and consistent, no matter which department, subject area or which text type is concerned.
 The SDL MultiTerm page says:
SDL MultiTerm is SDL's terminology management tool used worldwide by content owners, project managers, reviewers and translators to ensure consistency in terminology across all content types and languages. It can be used as a standalone desktop tool to create terminology databases and glossaries or with SDL Trados Studio to improve overall translation quality and efficiency.
Elsewhere we read:
A key finding from the TTI survey is that nearly half of translation rework is caused by terminology inconsistencies


The page headed memoQ benefits for individual translators says:
The bigger the project, the harder it is to ensure consistency. Terminology and recurring phrases must be applied consistently. memoQ helps you achieve this ...


Referring to an Acrolinx case history, we learn that:
Introducing the Acrolinx platform also helped raise in-house awareness of the importance of consistent terminology. This, in turn, simplified translations into other languages and allowed the company to speak with one voice.
With the exception of SDL's naive statement about "one allowed word per language" for a given concept, I would agree entirely with all of these claims if only they were made with a few reservations. These reservations will be discussed in my next two or three posts.

21 June 2016

A glossary worth its salt

Allison Wright author of a That elusive pair of jeans at wrightonthebutton.com has posted an insightful review of A French-English glossary of naval defence under the heading A glossary well worth its salt.

Allison makes a number of, as I said, insightful observations including the following:
... this glossary provides a cogent argument for building something worthwhile for your own practical reference within a specific subject field. Such glossaries are far more dynamic than is initially apparent.
In many texts, including those where the proportion of ‘technical’ content is relatively high, consistency is not always desirable, and this is where a glossary which lists two or three target options next to one source head word is much more helpful than anything one’s TM might contain.
There is more than one way to skin a CAT, as the saying goes, and in this case, a glossary ranks high on my list of priorities.

09 June 2016

Defense technical journalism: Noun forms vs verb forms

Many technical journalists, communicators and translators show a tendency to use noun forms in preference to verb forms, for example 'acquisition' in preference to 'acquire'. This gives rise to much discussion. Today, I do not wish to add to the theoretical discussion, but merely to quote some examples then add a few brief comments.

My examples are drawn from an article dated entitled Navy May Back Away From Advanced Arresting Gear for Ford Carriers by Sam LaGrone, a seasoned technical journalist who has covered legislation, acquisition and operations for the Sea Services. The article was published on the U.S. Naval Institute's USNI News website on 24 May. (Hence U.S. spelling and original punctuation and capitalization.)

The article discusses the General Atomics-built Advanced Arresting Gear under development for the U.S. Navy's next-generation Gerald R. Ford-class (CVN-78) aircraft carriers.

An artist’s conception of an installed Advanced Arresting Gear (AAG) on a U.S. carrier. General Atomics Image

The quotations I wish to focus on are:
“The Advanced Arresting Gear has become a model for how not to do acquisition of needed technology,” a senior Navy official told USNI News on Tuesday.
Question: Why 'how not to do acquisition' rather than, say 'how not to acquire'?
1)  While most style guides suggest that verbs are preferable to nouns, this general rule is often flouted by technical communicators, particularly when the noun carries connotations that the cognate verb does not.
2)  When understood as shorthand for 'defense acquisition', the noun brings to mind university courses, guidebooks and more. [The Defense Acquisition University provides mandatory, assignment specific, and continuing education courses for military and civilian personnel. DAU guidebooks here.] The verb 'acquire' carries none of this mass of information, know-how, procedures, constraints, etc.

If anyone out there in the blogosphere knows of a a more thorough treatment of this topic, please let me know.
The promise of the AAG and the Electromagnetic Aircraft Launch System (EMALS) on the other end of Ford was to allow the carrier to launch and recover aircraft that weren't built to the high tolerances of the current arrested landing and catapult systems and expand the types of aircraft that can make an arrested land on a carrier. [sic]
Question: Why 'aircraft that can make an arrested landing'?
1)  First, I assume that the author meant to write 'aircraft that can make an arrested landing'. 
2)  This example is grammatically different from the one above. Here the challenge is to find a verb form combing 'to land' and arresting gear. ... But now, having isolated the problem, I'd suggest that the author might have done better to write 'expand the types of aircraft that can land on carriers using arresting gear'.


While there are clearly cases where technical journalism demands certain noun forms rather then the cognate verb forms, frequent recourse to noun forms in preference to verb forms can result in poor style.

08 June 2016

Numbers in boxes: Grammatical and other tricks

This post is about layout, editorial and other tricks used in general and technical journalism to present numbers in information boxes, side columns and the like.

Here are a couple of examples from the FT's The Big Read on 6 June. The article, by Harriet Agnew, Miles Johnson and Patrick Jenkins, is entitled Inside McKinsey’s private hedge fund:

Would anyone like to describe the grammatical liberties taken here in order to achieve the designed graphic design/layout?
While the liberties taken here will probably pain many a grammar maven, the visual impact is impressive, particularly, IMHO, for scan reading, or for scanning to decide where to focus one's gaze more intently.

31 May 2016

Dashing through and other punctuation points, take #1

From Lingua Franca (Language and writing in academe.)

Ben Yagoda on My Favorite Shibboleth.

In language, a shibboleth is a usage that members of a certain group engage in not for meaning or elegance but in order to recognize each other (and exclude everybody else). Sometimes it reflects the state of the language decades or centuries before; other times it doesn’t have even that justification.
What are your shibboleths, and, more importantly, are you at least aware of them?
Mine include em-dashes with spaces (see below), minimal capitalization, and, when writing for specialist readerships, the uninhibited use of the sorts of acronyms and abbreviations that my readers use in their own writings.

Here's Ben again:
As it happens, the Quartz article included my own personal favorite shibboleth: “Farther refers to physical distance, while further describes the degree or extent of an action or situation. ‘I can’t run any farther,’ but ‘I have nothing further to say.’” True to form, M-WDEU demurs, pointing out that further and farther are historically the same word, and that both have been used by the best writers in both contexts for centuries. (However, the most recent citation for farther-meaning-additional is Edith Wharton, 1920: “He became aware that Mr. Jackson was clearing his throat preparatory to farther revelations.”)

Anne Curzan on Dashing Through.

Once you start using the dash in your writing, it can be hard to stop. I’m talking about the em-dash here — that punctuation mark that is so helpful at linking phrases and clauses that don’t seem well served by a comma, semi-colon, or colon.
Most style guides provide a good amount of leeway in terms of how the dash can function — it can function like a colon (as it did right there), parentheses (as it did in the first sentence of this paragraph), or a comma (as it did in the second sentence of this post).
The dash has a certain flair to it in its informality and its versatility. It makes a parenthetical a bit more prominent — a bit less parenthetical — than parentheses. It adds more sentential importance to an additional thought or an afterthought than a comma can do.
Note that Anne uses em-dashes with spaces — one of my shibboleths.

Lucy Ferriss on Language Shrapnel or how to present obscenities in formal writing. 

30 May 2016

Terminology in journalism

In “Why Terminology matters in Journalism” Jessica Mariani (a Study Visitor at Termcoord and PhD Candidate in English Language and Translation at the University of Verona) informs us:
  • Reuters, one of the leading news agencies on a global scale, was given birth (sic) as a translation agency in 1851. 
  • Bielsa and Bassnett report in their book Translation in Global News (2008), “news agencies are effectively vast translating organizations with the technology and skills required for the production of fast and accurate translations”.
Thank you Jessica.

04 May 2016

Mary Norris clarifies "mic" vs "mike"

In her TED Talk entitled Mary Norris: The nit-picking glory of The New Yorker's Comma Queen, Mary clarifies "mic" vs "mike" as follows:
The music industry spells it "mic" because that's how it's spelled on the equipment. ... New Yorker style for "microphone" in its abbreviated form is "mike."
I found this both interesting and instructive.

01 May 2016

Australian poet Les Murray was once a technical translator

The latest issue of the Paris Review includes, in the Interviews section, an interview (by Dennis O'Driscoll) with Australian poet Les Murray. The article is entitled simply The Art of Poetry No. 89. The link is to a scanned and OCRed version (including a number of OCR typos) of an article published on 1 April 2005. Recommended reading. A fascinating article about a fascinating and extraordinarily gifted man who is also an Australian National Living Treasure.

We learn that Les as a young man, and before going on to much greater things:
Later, as a student at Sydney University, he concentrated less on prescribed reading lists than on a serendipitous exploration of the university's library stock, and was, he says, blase about the examination system. Nonetheless, his brilliance as a linguist won him a position as a scientific and technical translator at the Australian National University in Canberra.
Later, in response to the question: What aspects of your life were most crucial to your development as a poet? Les says:
I was a freak, but happily my freakishness was in language—not, say, in classifying antique crankshafts. We seem to get a word-freak once or twice a century in the Murray family. Sir James Murray of the Oxford English Dictionary was my cousin, for example. When I'd argue points in the OED with my Russian fellow-translator at the National University in Canberra, I'd tell him we Murrays owned the damn language! Being some other kind of freak has its attractions, mind you. I envy painting its impasto and sheer color-play, how it's not held in by that stubborn insuspendable lexicality that words have. I get out into nonsense as far as I can. Lord knows, though never for nihilist ends . . . There's also the wonderful advantage of music and painting and sculpture, that they don't have to be translated.
Here's a particularly perceptive comment about rural Australians:
Probably half of longer-established Australian rural families have some Aboriginal admixture; and yet most are still in denial about it, dead scared of it, even as educated town folk start gingerly to yearn after that connection. An immense common property of black and white rural folk is what we've been learning to call "country," an intense connection with one's home region as a resource not just of survival but of the spirit. That has probably saved my life, more than once.
On Ezra Pound:
I realize now that poor Pound was mainly a man like Howard Hughes, one who slid so gradually into insanity that people were slow to detect it. He was a resourceful translator of languages he half knew.
I have no real superstitions, except that if I ever start doing a lot of translating, it will signify that poetry has left me. It's a pity, because I'm not a bad translator ...
We then moved to Canberra, where I was a translator of scholarly material for the Australian National University for four years.
The Labor Party's minister of justice, Michael Tate, asked me to revamp the old Oath of Allegiance.

(Under God) from this time forward
I am part of the Australian people.
I share their democracy and freedom.
I obey their laws.
I will never despise their customs or their faith
and I expect Australia to be loyal to me.
"Under God" was optional; and by "their faith" I meant whatever tradition sustained them and underwrote their own good faith. The minister's public servants liked my text down to that line, but were scared stiff of the last line. Here was a social revolution in a paragraph; The people made themselves into citizens and immediately demanded loyalty from the nation, as if the premise of democracy were actually true and the people were sovereign. The contrast with subjecthood under a mighty crown could not be more absolute, or sudden. The first part of the pledge survived in the recasting. The citizen still makes the essential move and constitutes himself as such. But the last two lines fell away, as I expected, and every epithet was doubled to give an appearance of weight. It was turned into legalese, in fact, and the rhythm was gone along with the daring.

28 April 2016

DCNS subs for RAN

This post concerns the Australian government's recent announcement that it has selected DCNS of France as the preferred international partner for the design of its next-generation submarines. I write from the viewpoint of a long-time provider of language services to French industry and more particularly the French naval defence industry.

My comments — and more particularly the congratulations — are made relative to the naval defence industry, not to broader political, philosophical and moral issues. Call it a "cop out" if you will. I may return later to questions like Can one justify the existence of the armaments industry and spending a slice of one's career in its service?

On 27 April Les Echos published a dossier entitled Sous-marins australiens : tout comprendre sur le "contrat du siècle" .
On 28 April Mer & Marine published a dossier comprising Australie : Ambitions navales et Eldorado industriel and Sous-marins australiens : Un partenariat stratégique et historique avec la France.

Thoughts on this huge contract

  1. Congratulation to DCNS, DCNS Australia, the negotiators and the teams that put together the bid, from engineers to technical writers, translators, support staff and more. All must have done an exceedingly good job.
  2. Congratulation to the Australian government on making a major defence decision which, at this stage, looks as though it was genuinely based on sound technical analysis. This is significant given that many defence procurement decisions leave more room for doubt.
  3. Very encouraging to read in Future submarine announcement from the RAN's media room that senior USN officers contributed to the bid analysis process thereby guaranteeing, presumably, that (at least at this stage and this level), the idea of integrating a US-designed combat system with a French-designed submarine, complete with systems and sensors, has been thoroughly explored by all concerned from the outset.
  4. Also very encouraging to read that the Australian government's “decision was driven by DCNS’s ability to best meet all of the Australian government’s requirements. These included superior sensor performance and stealth characteristics, as well as range and endurance similar to the Collins-class submarine. The government’s considerations also included cost, schedule, program execution, through-life support and Australian industry involvement.” (Source: Future submarine announcement). That's a very impressive statement.
  5. Congratulations to the US DoD for (presumably) giving the all clear for the integration of a US-designed combat system with a French-designed submarine, complete with systems and sensors.
  6. Congratulation to the French Ministry of Defence for (presumably) giving the all clear for the export, and in many cases the transfer, of technologies previously considered sovereign know-how, including the state-of-the-art stealth hull, the ultra-quiet propulsion system (not least the pump-jet propulsor) and advanced shipbuilding expertise.
  7. Very encouraging to read in Sous-marins australiens : Un partenariat stratégique et historique avec la France that
    Le franco-australien Ross McInnes, ancien de Thales et président du Conseil d’administration de Safran et de la Chambre de commerce franco-australienne, est notamment présenté comme l'un des artisans de la victoire. (...) Ross McInnes symbolise aussi l’union de l’ « Equipe de France », comme aime l’appeler Hervé Guillou.
Despite the fact that Mr McInnes's background is mostly in corporate finance and as a CFO he is likely to be well aware of the importance of technico-cultural issues in ensuring the success of this sort of project, including the differences between the technical communication cultures of client and prime contractor.


  1. Shortfin Barracuda Block 1A: great name!
  2. As a technical writer and translator I would love to know how much attention the bid evaluation teams paid to technical writing, presentation and related aspects. I would also like to know how much effort DCNS put into these aspects of bid preparation. While these considerations are obviously orders of magnitude less important than confidence, cost, schedule, program execution, through-life support and local industry involvement, among yet other factors, they have been known to contribute to or, on occasions, compromise the impact of some bids.

On technical writing

I would now like to make some admittedly sweeping statements based on my personal experience in France as a technical writer and translator in the hope that may be of interest or possibly even direct benefit to some of the people about to embark upon several months of negotiations.
  1. Despite some notable exceptions, technical writing is less appreciated, less understood and less expert in France than in English-speaking countries like Australia.
  2. In France much technical writing is done by engineers often without the assistance of trained technical writers.
  3. Very few naval engineers in France have English as their first language.
  4. Some French engineers draft directly in English, their second language. Others draft in French then have their documents translated into English.
  5. The language service providers that win the contracts to translate "technical manuals" are typically selected primarily on the basis of price.
  6. These companies typically subcontract to freelance translators who only earn a proportion of the amount billed to the client and often too little to attract the very best, much less to keep them motivated to work to the highest standards.
  7. What are the RAN's expectations in these areas and will DCNS and its subcontractors be able to deliver?
I like to think the both client and prime contractor will explore these issues during the contract negotiation phase, that the prime contractor will rise to the challenge and that the client, the RAN, will be happy with with all aspects of this massive project, including the admittedly secondary issue of technical communication.

26 April 2016

Challenge for MT freaks

From here:
The word with the most meanings in English is the verb 'set', with 430 senses listed in the Second Edition of the Oxford English Dictionary, published in 1989. The word commands the longest entry in the dictionary at 60,000 words, or 326,000 characters.
That's an excellent piece of information for any who thinks they want to get involved in machine translation (MT). Challenge begin with sentences containing different uses of the word 'set' then use your MT system to translate them into N languages, then check the results carefully. Repeat the process until you achieve at least accuracy.

Congratulation DCNS

Malcolm Turbull, the Australian Prime Minister, announced today in Adelaide, that the next generation of 12 submarines will be constructed in Adelaide, with DCNS of France selected as the preferred international partner for the design.

The FT put it this way:
DCNS of France wins A$50bn Australia submarines contract

Le Monde put it this way:
Sous-marins vendus par DCNS à l’Australie : les coulisses d’un contrat « historique »Le constructeur naval militaire français a été retenu par l’Australie pour la construction de douze sous-marins. 4 000 employés de Cherbourg, Nantes et Lorient vont y travailler pendant six ans.
This is a momentous decision for Australia and the RAN and an extraordinary achievement for DCNS. My congratulations to all concerned and to all my clients and acquaintances at DCNS.

I worked for DCNS's head office in Paris for 15 years or more as the leading freelance supplier of French-to-English translations. It was in the course of some 20 years' work for DCNS and other members of the French naval shipbuilding industry that I compiled 
'A French English Glossary of Naval Technology'

Earlier today I tweeted:
#DCNS to supply #AusNavy subs (#NavyDaily ). #t9n jobs ahead. Get #NavTechGloss (https://goo.gl/hyHhXj )

22 April 2016

The go-to page on transcreation?

On 18 April English-to-Italian translator Laura Cattaneo posted an excellent piece under the heading
Trans-form your communication with trans-creation.

This may well become a, or even the, go-to page for anyone seeking a concise explanation in English of transcreation.

One very minor point: I think I would have stated the infographic with "Also known as" or "Aka" rather than "A.K.A." which, to me, is typographically a bit clunky.

01 April 2016

Paris on April fool's day 2016

Paris métro operator RATP really lightened up today.

Web page Métro - RER : la RATP renomme 13 stations (in French).


Best (IMHO):

Origine de l'expression « poisson d'avril ».

Origin of the expression "April fools' day".

From "classification is culture" to terminology

Listening to anthropologist and award-winning Financial Times journalist Gillian Tett discussing her book The Silo Effect, the translator and terminologist in me was struck right between the eyes by the expressions:
classification is culture and culture is classification.
(I was listening to another Big Ideas programme from Australia's ABC Radio National, this one called 
The silo effect and how to break down the barriers.)

Googling for the expression classification is culture without quotation marks (the link is to the very first returned by Google) led me to "culture classificiation" complete with a definition of "ontology" which was nice since I started this blog aiming to link Tett's expressions to the cultural component of terminology.

The "culture classificiation" page is part of the web site dedicated to the Virtual Terrain Project which links to "the fields of CAD, GIS, visual simulation, surveying and remote sensing", GIS and remote sensing being two fields in which I translated hundreds of pages in the 1980s and 90s and translator to French space agency CNES and satellite imagery subsidiary Spot Image, now part of Airbus Defence and Space.

Clearing this rambling calls for more work ...

30 March 2016

Aramaic in Britain

Today, listening to a podcase of a Big Ideas programme from Australia's ABC Radio National entitled Mary Beard's ancient Rome (audio download, but not transcript) I was amazed to learn that there is a tombstone in northern England dating from Roman times with an inscription that is partly in Latin and partly in Aramaic. The woman in the tomb was apparently from southern England while her husband was a speaker of Aramaic from Palmyra in what is now Syria.

More on this remarkable tombstone here.

Like professor Beard, I find this utterly amazing. One wonders if there were any other speakers of Aramaic in the area at the time to give the man some mother-tongue company. Again like professor Beard, one also cannot help wondering what language the couple used in their daily life.

Professor Beard's book is called SPQR: A History of Ancient Rome (link is to a review in The Atlantic).


SPQR (Senatus Populusque Romanus)                                                SPQR
     The Senate and People of Rome                      on a modern-day manhole cover in Rome.