A potential customer inquired a few days ago whether my cost estimate for translating a patent application from French to English would include also charging for words such as “and”, “a”, and “the”. I always feel like responding by saying “only if you want me to include these words in my translation”, but I never actually do that. It is not a good idea to antagonize potential customers.
So let’s consider his question. If the translation cost is based on the word count, is it reasonable to charge the same amount also for simple words like conjunctions and articles?
The word count (or character count, or line count or page count) is a handy quantifier because it can be easily verified. That is why most translators base their cost on one of these quantifiers. But even when they charge by the word, translators are not really charging for the translated words – they are basically charging for their time. If I know how long it will take me to translate 2,000, 3,000 or 4,000 words, I also know how much I will be making per hour.
Some people say that it would be more fair if we could charge by the hour. I would not mind doing that, but I basically have to charge by the word because most translators charge by the word, or by the project – most translators also have a minimum charge for short translations, and some will simply ask for a flat fee for a given project.
What our clients may not realize is that the handy per word quantifier hides a lot of freebies that most translators generally offer to their customers, i.e. a lot of additional work for which we do not charge extra fees.
I just finished translating a fairly long chemical patent from Japanese to English. It was mostly straightforward work for me because I must have translated thousands of patents like this one. Although I charged my lower, non-rush rate, which is reserved for customers who give me enough time to work on their project, I was still making about 600 hundred dollars a day without working too hard.
So I am not complaining. But translating is not just about words. It is about making sense of things. Words are the most visible ingredients of what goes into every translation, but there are many other ingredients in every translation.
Here are few examples of some of these other ingredients that I simply threw in for free in that particular patent translation.
1. Long chemical terms count as 1 word
How many words is “2-(meth)acrylamido-propyltrimethoxysilane, 3-(meth)acrylamido-butyltrimethoxysilane, 4-(meth) acrylamido-hexyltrimethoxysilane”? It looks like about 20 words, but Microsoft Word counts it as 4 words. Who am I to argue with the wise and almighty Microsoft Word? I just accept the MS Word count, which would be in this case the same as the count of these 4 words:”the, “a”, “and”, and “not”.
2. Chemical patents have a lot of formulas which may throw off formatting
These formulas must be scanned and pasted into the text. It generally does not take a long time to do that, but it does take time. And when there several formulas on the same page, it may throw off the formatting, and once something goes wrong with the formatting, I may have to waste a lot of time trying to fix it. This is a problem especially with Japanese which takes up less space on a page because 2 Japanese characters on average correspond to 1 English word.
Because most people don’t charge an additional fee for scanning and formatting of chemical formulas, I don’t dare to add an additional charge either.
3. Chemical patents often include very long and very complicated tables
It is much easier to fit a few Japanese characters into the rows and columns of a table in Japanese than to fit the corresponding English words in a table in English. For example, “重合度” (jūgōdo) means “degree of polymerization” in English. The word “polymerization” alone is about 1.5 times the size of all the three Japanese characters in this chemical term. Recreating a complicated table that was originally in Japanese in English can be a nightmare, especially since MS Word always for some reason sadistically messes up the final formatting of complicated tables.
This particular chemical patent had one complicated, full-page Japanese table in “landscape” orientation in which the words ended up being nonsensically truncated in English by MS Word. After 3 unsuccessful attempts to recreate the table in MS Word, I created it in WordPerfect, which did a much better job, but because I had a file in MS Word, I printed it out and scanned and pasted it into the text as a graphic file.
The table was very important as it was the main proof of the claims of the patent, so I had to keep the same format in English. Because I inserted a graphic file to make sure that everything in the table will be instantly understandable, I did not charge anything for several hundred words contained in that table. Plus, of course, I could not charge anything for the three unsuccessful attempts to create the table in landscape format in MS Word either before WordPerfect saved my life once again.
4. Foreign words transliterated into Japanese may be very difficult to ascertain
There are several reasons for this. When foreign words are transliterated into Japanese, they often become unrecognizable unless you know precisely what they mean. For example, “sexual harassment” becomes “seku hara“, “power harassment” becomes “pawah hara“, etc. Well, the meaning of “seku hara” and “pawah hara” is obvious enough, but the meaning of technical terms adopted from foreign languages and then butchered beyond recognition courtesy of the Japanese language may be less obvious, especially since I don’t necessarily know from which language the term was originally imported. The foreign word could have been in German, or French, or Dutch or another language rather than English.
These foreign words transliterated into one of the Japanese alphabets called katakana can be found relatively easily with a search on the Internet … except when they are misspelled in Japanese. And they are often misspelled in Japanese patents because Japanese chemists and patent agents don’t particularly care about the correct spelling of foreign words in Japanese. For example, I remember a Hitachi patent agent who was consistently misspelling the word “analog”. It should have been written as “anarogu” in Japanese, but the esteemed benrishi was instead writing it as “anaguro“.
Patent agents do this kind of thing all the time. If the foreign word was originally for instance a Dutch or German name, for example of a Dutch or German (or French, or Russian or Polish?) chemist who invented a new method in chemistry, it takes forever to find out what was the original word if it is misspelled in Japanese.
5. Every language and every subject has its own unique challenges
Every translation from every language has its own challenges, and the more different the language is from English or the language into which one translates, the more challenging the translation is likely to be. Overcoming these challenges is also what makes translating so interesting.
Johann Sebastian Bach once famously said:”It is very easy to play any musical instrument. All you have to do is touch the right key at the right time and the instrument will play itself.”
The same is true also about translation. It is very easy to translate anything into another language. All you have to do is know which keys to touch on the computer keyboard to formulate the right words in another language, just like J.S. Bach knew exactly which keys to touch on an organ’s keyboard to play one of his fugues.
This knowledge is what translators are being paid for, not the simple or complicated words that they are typing while translating.
I can understand it when a client asks a question such as the one in the title of my post today. But when a translation agency insists on discounts for “fuzzy matches” or “full matches” in repeated words or repetitive passages determined by means of software, that makes me really mad. I would never work for a translation agency that does that.
Although discounts may be in order in case of repetitive documents, for example updates of printer manuals or communication software, these discounts should be determined ahead of time based on an agreement between a translator and the client, not based on software controlled by the agency. The translator must be in control at all times, not some software in the hands of a dishonest broker.
Insisting that the ingenious principle of so called “fuzzy matches” or “full matches” should be applied to all translations is a criminal concept which is known in law as “theft of labor”. This kind of fuzzy thinking is very different from a naive question of a customer who is asking whether a translator should also charge for words such as “and”, “a”, and “the”.