Monday, October 4, 2010

Translation and the Art of Motorcycle Maintenance, by Gary Smith

Having begun my translation work through engineering (nearly twenty years ago in an industrial research centre), I have often faced suspicion from linguists who wonder what on earth I’m doing "invading their territory". However, when faced with translations on nuclear reactors or catalytic converters these people grudgingly accept that I can translate texts they themselves won’t touch. That said, this suspicion is mutual; I have also faced it from engineers who wonder why they need a translator ("what do linguists know about bridge construction?"). These attitudes give rise to misconceptions such as the myth that "good grammar doesn’t matter in technical texts". It is a sobering thought that a misplaced comma in the emergency procedures for a nuclear power station or an aircraft maintenance manual could have catastrophic results. But who is "qualified" to translate such texts? A monolingual nuclear engineer? A scientifically illiterate linguist? My aim with this article is to reverse the negative attitude of certain linguists towards scientists "invading their turf" and encourage them to do likewise – "invade" the area of science, where there is good money to be made and you are needed, whatever some engineers may tell you.

Firstly, let me stress that people who master different languages are by definition intelligent people, and therefore perfectly capable of learning new subjects, especially if they decide to specialize. And the wide world of science has enormous possibilities for specializing in something you like (astronomy, the environment, marine biology, medicine, sustainable architecture, antique motorcycle maintenance…). The problem is that many who have only studied linguistics believe that this is an esoteric world of microbiology druids whose secretive terminology is accessible only to a chosen few. This is a strange attitude in people who boast of having a greater vocabulary than most.

This was brought home to me when a translator friend of mine saw a text I was working on about the isomerization of certain compounds. She burst out laughing, finding it incomprehensible. Where to look? Which bilingual specialized scientific dictionary to use? If you have an ordinary, medium-sized monolingual dictionary at hand, look up the word "isomer" now. (If not, check the Wiki). Go on, I’ll give you a few seconds while I wait… Found it? Surprised that such a word exists in any old run-of-the-mill dictionary? You should be – after all, you are a translator and should therefore find no surprises in such a dictionary. (Take note, by the way – for scientific translation you will need a monolingual dictionary. There is little point knowing that isomer in Spanish is isómero if you still have no idea what the word means in your own language). My point is, science is not so mysterious nor the texts so difficult as you may think at first. In fact, you’ll probably find that the average 15-year-old high school student is familiar with isomers, as you yourself probably were at that age before your linguistic specialization led you to forget half the things you learned at school. This is a common drawback in education in the developed world. Again, it works both ways – I once had an engineering client who didn’t know what an adjective is. This may explain why many believe that there are no Da Vincis or Aristotles around today, which is quite untrue – they just specialize now. I myself know a production engineer for a multinational glass manufacturer who studied fine arts, and a globetrotting troubleshooter for the cement industry with a degree in philosophy. As translators, we are condemned to be cultivated and always win at Trivial Pursuit – but there’s a piece of the pie to be won too if you wish to show just how cultivated you truly are. Science is also culture, and it changes the world and society far more than any poet, politician or preacher. And it therefore provides a seemingly infinite amount of texts to be translated.

To my mind, there are two very distinct kinds of text: scientific and technical. The former are usually of a standard layout – abstract, method, results, conclusions and bibliography, rounded off by effusive acknowledgements for the translator, preferably with your e-mail address. These are often written by academics with fairly good linguistic competence, to be published in international journals. Technical translations, on the other hand, can provide more problems and should be taken with some humor if you are not to end up banging your head against your keyboard and sobbing loudly. They are often written by private company technicians with scant regard for such trifles as commas and full stops/periods.

A classic example is provided by instructions manuals. One may well be mistaken for thinking that everybody reads instructions manuals in the same way. Nothing could be further from the truth; even cultural issues come into play. Let us take the example of the mobile/cell phone manual, a technical text which we are probably all familiar with and which some of you may even have actually bothered to read. In Mediterranean and Latin countries, the procedure after purchasing a mobile/cell phone is as follows:

1- Open box.

2- Throw away pesky bits of paper (i.e. instructions manual and guarantee).

3- Proceed to fiddle with buttons and moveable parts for 10 days until:

a) you realize you don’t know how the phone works.

b) you receive an electric shock.

c) the phone breaks.

4- Look for instructions manual (being chewed by cat/dog and/or at bottom of bin).

5. Discover phone is indeed broken and/or parts are missing and/or is in fact an electric razor.

6. Look for guarantee.

7. Discover 10-day trial period has now expired.

8. Curse manufacturer.

In colder climes, however, the procedure can be quite different:

1- Open box.

2- Proceed to read instructions manual step-by-step for 10 days.

3- Take phone out of cellophane. (Keep cellophane out of reach of children as instructed.)

4- See points 5 to 8 above.

The point is, instructions are read differently depending on who reads them, and this affects how they should be written. In warmer climes people often skip "boring bits" and prefer to discover the product for themselves, using the instructions only as a reference when they come across something they don’t understand. This implies avoiding the use of pronouns or references to other sections, even if you end up being repetitive. Technical texts are not meant to be works of art, but to be unequivocal and unambiguous. Indeed, many of the rules that should be applied to instructions manuals can be applied to legal texts too.

Apart from cultural preferences, there may also be other reasons for skipping points. Take emergency procedures. All companies have two possible emergency procedures in the event of a fire, for example. Let us call them plan ‘a’ and plan ‘b’. Plan ‘a’ involves following a well-rehearsed drill and assembling staff at a pre-arranged meeting point. Plan ‘b’ involves racing round like headless chickens, arms flailing, screaming for dear life. Unfortunately, when people’s lives are in imminent danger, it is plan ‘b’ that prevails.

So let us imagine a fire breaks out in a factory. Our hapless hero worker seizes an extinguisher and begins to read the instructions with understandably great haste:

1- Congratulations on purchasing your X-300 fire extinguisher, the top of the range bla bla bla.

A ball of flame flares up, searing the seat of their pants.

2- Ensure this apparatus is cleaned every two months according to European guideline 1:2002 F. Begin by carefully applying a damp cloth etc.

As the hair on the back of their neck begins to singe, they will obviously move quickly on to point 10:

10- Connect the former to the latter and turn in the same direction as in point 5.

At this point the worker wisely decides to wield the extinguisher as a hefty weapon with life-or-death ferocity, mercilessly beating a ruthless warpath through the pack of panic-stricken people blocking the emergency exit (indicated in diagram ‘f’ of emergency procedure appendix II). At least it came in handy.

Clearly, essential points should be brief and come first, points 1 and 2 should come later if at all in publicly visible emergency instructions, and full nouns should be used rather than "former" and "latter". Again, one should avoid references to other points that may have been omitted by the reader.

Such rules when writing technical texts and instructions can be found in "controlled languages" created artificially for this very purpose. These are used in multinational joint ventures where people from many countries are involved and where English is used as the common language but the employees’ grasp of this language is not perfect. One such example is ASD-STE100 Simplified Technical English. They include rules such as avoidance of the passive voice, keeping sentences short and always using an article. Such rules, while helping to prevent misunderstandings, inevitably lead to a repetitive style, which in turn leads to a high degree of translation memory fuzzy matches. This fact has not been lost on some companies that have even produced controlled languages with this in mind (Caterpillar and CLOUT).

One more tip on translating technical texts – find pictures! To fully understand the parts of a ship, the cross-section of a road or the workings of a machine, the "Images" search you can make in Google is priceless, not to mention picture dictionaries such as the Merriam-Webster online. And what better example of a picture dictionary than an encyclopedia? And what better example of an encyclopedia than Saint Wiki? Obviously, the Wikipedia should not be used as a reference for correct terminology, but as a learning tool it is unbeatable, especially if you need to understand scientific texts. But first perhaps it is necessary to explain just how to use it – and how not to.

The Wikipedia is a site that depends entirely on contributions – both in terms of financing and information. The articles are written by volunteers, which should obviously set alarm bells ringing as to its reliability as a dictionary. Sometimes you’ll find a symbol next to the language that has been defined as a "good article" (often in German), and the site applies a "verifiability" policy which means that all pages should cite sources whose reliability you can then check. So why use it at all? Well, its reliability is proportional to the simplicity of the subject matter, due to its democratic form of editing. For example, if you decide to write an article about kangaroos, saying that they are a kind of gigantic mollusk that lives in the Arctic Ocean, it will immediately be removed by Wikepedians with more authoritative knowledge than yours on the world’s fauna, not least because they have a "no original research" policy and the idea of huge polar mollusks is, well, quite original. If, however, you decide to give the world the benefit of your meager knowledge about erythropoiesis in hemopoietic tissue, there will be far fewer people qualified to dispute you if you affirm that it is caused, say, by ingesting huge Arctic gastropods (though you’ll have to provide prior research into the phenomenon, which could be tricky). In other words, the Wikipedia is the Sesame Street of Science. It’s where Grover explains isomers to you, and in this respect it does indeed work and can be an invaluable self-learning tool.

Another important point to note about the Wiki is that the texts in different languages are written by different authors who usually bear absolutely no relation to each other. In other words, the texts are different and this may even lead to slightly differing definitions.

One trick to help check the names of flora and fauna is to look for the Latin name which scientists have generally agreed on to find the corresponding common name in each respective language. Again, double-check with official sites other than the Wiki. It is a curious fact that in certain scientific respects Latin is still the most widely used language (and some may argue that it still is in a wider sense through its daughter languages). A similar tip applies to chemical nomenclature. Find the chemical notation for a compound and you can find its name in another language – but beware of isomers! The chemical notation can also change slightly from Asian to European languages. You can also look up the International Non-Proprietary Names (INN) for chemicals at the WHO, with their corresponding official translations, to avoid prosecution from companies that have patented a chemical product (similar considerations should be applied to technical translations).

While my intention with this article is to encourage translators to delve into science, I must obviously stress that you should not attempt to translate texts that you honestly do not understand. All I’m saying is that before you reject a job upon seeing the title alone, give yourself fifteen minutes to check it out first and see if it really is as difficult as you thought at first. You may surprise yourself; maybe even Grover can understand it. That said, obviously medical texts, for example, should be left to the experts. Nevertheless, if this subject interests you, make yourself an expert by studying official courses for translators on the matter. The same goes for perhaps the most difficult of all subjects to translate: mathematical texts. These also require sound understanding. By no means do I advise anybody to try to bluff their way through the translation of a text they if haven’t really understood why the Dirac delta is not strictly a function since it should have total integral zero. I’d like to see Grover try to explain that…

Copyright © and the author, 1999-2010. All rights reserved.


  1. Spot on, Gary. Thanks for writing.

  2. I just went through the process of convincing someone that she could do technical translating. If she had seen this beforehand, she would probably not have listened to me. Still, I enjoyed this.