Does It Makes Sense To Charge For Translations By The Word?

This is how I have been charging for my translations, with some exceptions, since I hung out I my shingle as a freelance translator in San Francisco in 1987.

But is it the best way to determine how much a translation should cost?

As far as I know, the most common method in the United States is to charge based on the number of words in the target language for translations into English. The cost of my translations is determined by the number of English words in the translated version of the original document.

There is a way for translators to abuse this arrangement. For example, I can translate the word “mochiron” in Japanese, or the word “selbstverständlich” in German to English as:

1. “obviously”, which is 1 word,

2. “of course”, which would be 2 words, or

3. “it goes without saying”, which would be 4 words.

It goes without saying that many translators will tend to favor the “it goes without saying” version in their translations to achieve a higher remuneration of their efforts. And there are translators who will shamelessly abuse the English language by trying to cram as many English words into their translations as possible, which is a technique also known in the industry as “word count padding”. But since you can’t keep padding the words in your translation without making it painfully obvious, even to a relatively unsophisticated customer, that you are simply doing it to increase the cost, people who use this technique are very likely to ultimately lose the customer.

Other methods for arriving at the cost of a translation are used in other countries.

When I used to work for Japanese translation agencies back in the nineties, these agencies counted the pages in the original language in a format called “genko yoshi“, which means something like “a standard page”, roughly corresponding to a double spaced page in English with about 250 words on one page. I am guessing that this method is still probably used in  Japan. This method kind of makes sense, especially since it is not clear what is a word in Japanese as there are no spaces between words in this language.

A similar method is used in German-speaking countries, except that as far as I know, most people there use the price per a standard line with 55 characters on one line and 25 lines on a double spaced page. Since there are so many compound words in “the awful German language” as Mark Twain described it, this method also makes sense. For example the term “Medikamentenabgabeeinrichtung” is only 1 word in  German, but the English translation, “medication dispensing device”, has 3 words.

German translators who translate foreign languages into German are thus at a significant disadvantage because the word count in German is by about 30% lower than for instance the word count in English. Some translators into German for this reason charge based on the word count in the original language, for example English.

Charging for translation based on the source rather than the target language thus also makes sense, especially in view of the fact that by using this method, one can give the client a precise price quote ahead of time.

PDF files in languages such as German, French, Russian or Japanese used to present a major problem for me because the word count had to be estimated page by page, which took a long time. But it is quite easy now to convert PDF files in many languages into MS Word with Adobe PDF Converter, which costs 20 dollars for a yearly subscription. This program can be used online from any computer by clicking on the “Convert to MS Word” option. This is very helpful also for estimated word counts for translations from Japanese because if the character count in the Japanese MS Word file shows 6 thousand words (which is completely incorrect regardless of how one would define the “word unit” in Japanese), this means that the translation will have about 3 thousand English words because 2 Japanese characters (hiragana, katakana or kanji) roughly correspond to 1 English word.

Different languages will have slightly different word count in MS Word due to different language structure or orthographic rules, etc. For instance, because Slavic languages have no article (except for Bulgarian), personal pronouns such “I” are often optional, and also because they use fewer prepositions – things like “from”, “to”, or “about” can be indicated not only with prepositions in Slavic languages, but also with endings of nouns which will then count as 1 word, they tend to have a slightly lower word count than English. For example, “He said it to me”, which is 5 words in English, would be “Řekl mi to” in Czech, which is only 3 words.

French, on the other hand, tends to have a slightly higher word count than English, partly due to different orthographic rules in French – “in this model;”, which is 3 words in English, will become “dans ce modèle ;” in French, which will be counted as 4 words due to the obligatory space in front of the semicolon.

Some translators say that it would be best if translators were paid like lawyers, by the hour and in increments of 15 minutes.

Perhaps it would be best. But it is never going to happen because the customers would then have no control over how many hours we charge. Slow translators would also then be making more money than their faster colleagues for the same number of pages, which would hardly be fair. And some would be padding hours instead of words into their bills.

Customers do put up with having to pay lawyers by the hour because they really don’t have much choice in the matter since this world is run by lawyers – look at the US Congress, for example – which is why the world is in such a bad shape, many would say.

But since translators are hardly in a position to dictate their preferred payment terms to their clients, a remuneration method that is based on the word count in European languages or even in Japanese, with modifications depending on the language, is in fact very rational, at least in the field of technical and patent translation – provided that both the translator and the customer understand the rules in advance and agree to them.

On the other hand, I also think that the word count method is not really fair if we are talking about translation of novels, or translation of slick advertising and seductive commercial propaganda and other interesting translation niches in which translators often must be much more creative than mere technical translators such as myself.

As to the perennial question of whether translators can call themselves “professionals” when they are paid based on how many words they can input as if they were file input clerks, I think that the following three points are more important than the exact method used to determine the cost of translation:

1. How much we are in fact able to earn, regardless of the method used to determine our remuneration, which depends more on who we work for than how the cost is calculated,

2. How much control we have over the methods used to calculate how much we earn (agreeing to lower compensation or no compensation at all for “fuzzy matches” and “full matches” means in my opinion that we no longer control anything), and

3. Whether we are able to say NO to ingenious schemes designed to turn translators into cheap administrative help that has no choice but to accept low remuneration, demeaning conditions of “Nondisclosure Agreements” and other excellent inventions of the Brave New World of Translation Industry.