December 1, 1999
What’s in a Page?
by Lynne E. Riggs
FAQ corner
Setting Your Units of Charge
What’s in a page? 250 English words, 350 words, 200 words, 25 lines, 2000 characters, 400 characters, 200 characters, 1200 characters . . .
All of these, and others, are commonly encountered “standard” pages, by one measure or other. Translators, editors, proofreaders, rewriters, book designers, typists/keyboarders—wordsmiths of all kinds—consort with several of these in doing professional work in Japan.
Client: “We have 10 pages of text that needs proofreading (kōsei). How much do you charge and how soon can you finish?”
Proofreader: “What are your ‘pages’ like?”
“We have something we want translated by Monday. It’s four A4 pages’ worth.”
Translator: “How many characters on a ‘page’”?
“What is your rate per page?”
Translator: “That depends on the ‘page.’”
One of the questions asked most frequently by those entering the editing, rewriting, proofreading, typing, and translating fields is how to calculate the “pages” or other units proper to their kind of work. While the “page” may be the unit most immediately graspable to the client, it turns out to be a most elusive unit. A recent thread of discussion on SWET-L, SWET’s electronic mailing list, brought out some of the varied issues the subject of the “page” elicits among experienced professional. Here we compile some conventional knowledge of SWET members, and draw on comments from the SWET-L thread. Thanks to Miriam Bloom, Daniel Day, Martin Ecott, David Eunice, Maynard Hogg, Hugh Miller, Wayne Root, Dennis Schneider, Fred Uleman for their contributions.
In mediating the counting gap between client and professional, a number of factors come into play. The very first thing to be established with the client is what the unit of charge for professional services will be: hour, page (and what unit of page), English (alphabetic) character, Japanese character, or English word. Below are some of the typical units of charge and ways of counting “pages” and “words.”
The English word
The English word is a common unit of charge for translation, editing, proofreading, and rewriting. Where once words were counted by hand by various methods, today most computers and word processors do the job. Some people charge by the total number of words in a document; others use a fixed word unit as their standard of charge.
The 250-word page
This unit is part of the legacy of the typewriter era, particularly in the United States, where the 250-word, double-spaced pica-typewritten 8 1/2 x 11-in. page was the unit for high school compositions, university research papers, and expository writing class essays. A page with that quantity had good-sized margins and plenty of space between the lines for corrections and comments. Publishers liked this format too, which gives the editor’s red pencil plenty of room to work. With that long tradition, it is still around as a unit. The words may be counted by the computer, but for editing, rewriting, and sometimes translating, the unit is still often 250 words as a “page.”
The 280–350-word page
In Japan, where the standard manuscript page is A4 size, 25 lines on a page leaves a lot of wasted paper. A full page of text with 3-cm. left/right margins and 2.5-cm. top and bottom margins in a font like Palatino contains about 330 words. This format is preferred for print-outs of running text to be read by authors, clients, editors, and proofreaders, regardless of the unit in which professional services are charged.
The English (alphabetic) character
For some texts, such as those consisting of short captions, questionnaire responses, diagrams, “pages” defy regular counting. One way of calculating the quantity is to count the characters in the entire text, and charge per character. If it is easier to transform the number of characters into word units, an average “word”—say 5 characters—can be determined. It can also be used as the basis for a “page” unit.
The Japanese character
In J-E translation, one of the most common units of charge is the number of Japanese characters. The genkō-yōshi (grid-lined printed sheets used for writing manuscripts) provided a “page” with a 200- or 400-character standard, and writers used one square for each character and each item of punctuation. Today, even though the total number of characters in a word-processed Japanese text can be counted easily by computer, the 400-character unit is often retained as a familiar unit.
The line of text
For convenience, such as when translating titles of articles, names of works of art or craft in an exhibition, captions for photographs or illustrations, it may be useful to set a “per line” charge, defining a line as, for example, an average of 10 words, and setting the price in accordance with an amount that reflects the time taken, the expertise assumed, or the difficulty of the work.
Kinds of work and units of charge
Writers
For writers, the most common unit is the word. Commissions are generally made for a piece of desired length: a 100-word paragraph on kusamochi, a 2,000-word essay on Tanabata, or 5,000 words on the politics of Japanese ODA. Some magazines pay by the printed page (their printed page).
Editors, proofreaders, and rewriters
Perhaps because work in these areas is closely involved with publishing, it is often paid by the page, although the “page” may be defined differently by the client or by the freelancer. Most often the unit of charge is the 250-word page or the page of the particular publication, after typesetting and layout, with adjustments made for non-textual areas.
Translators
J-E translators of various types are paid in a number of ways: Here are some of the more common units of charge:
- Literature, by 400-character genkō-yōshi page
- Non-fiction (essays, journalism, scholarly papers), 400-character genkō-yōshi page
- Medical papers: by the final English 250-word page or by the word
- Manuals: by number of words in resulting translation or by number of characters of original Japanese
- Consumer/marketing reports: by number of English words
- Website texts: by characters of original Japanese or by number of resulting English words
- Government documents: by number of characters, as counted digitally by client or as calculated by translator; some translators charge by number of English words of resulting translation
E-J translation is usually calculated in terms of the number of Japanese characters in the resulting translation. This was easily calculated in pre-computer days when translations were written on genkō-yōshi sheets. Today, computers count the number of characters in a document.
Who counts?
In some cases the client will have counted the number of words or the number of characters (English or Japanese) digitally or by other means, and know the amount of work to be done at the outset. (It is a good idea to see the text physically before taking on a job. A third-hand fax of a manuscript in long-hand or a word-processed manuscript by a poorly trained keyboarder who couldn’t really read the author’s handwriting—resulting in many typos—is a sure formula for slowing down the translator.)
Often the client has no experience and asks the translator, editor, or proofreader how he/she will charge. In any case, it is up to the professional to know what unit to use for particular kinds of work and what rate to charge. It is always a good idea to check the count received from a client.
If your count of a manuscript is the basis for billing, be sure your procedure is agreeable to the client. There are basically two ways of counting: (1) digitally, as the computer will do it, which gives the strict number of Japanese characters, the number of English words, number of 1-byte characters, etc., and (2) the looser approach, following the old genkō-yōshi principles, in which every line of text counted, even with a few characters in the line, as a full line.
Advice for the page-wary: Excerpts from the SWET-L Thread
The important thing before accepting the job is to get it clear what the unit is and how much it is going to cost. Any talk of page size elasticity from an agent is totally disingenuous.
Client education is one of the things that we neglect at our peril. I try to maintain flexibility, but I am unwilling to work with people who think that I can be squeezed. Like most people who work with other people’s words, I care about what I do and I don’t like being infected with a couldn’t-care-less state of mind by middle merchants.
One thing that you have to educate clients about is that rates which allow editors (or writers or translators) to make a decent living from a 40-hour work week are reasonable. . . . be assured that Japanese businesses know all about unit sizes. You have to stand your ground in what can be a tedious bargaining ritual, but that is what a lot of business is all about.
There is a “standard” Japanese page that is 400 characters. This is a 400-ji genkō yōshi. And you might be able to work this into the explanation by saying that your 200-word (or 230-word, or whatever) page is equivalent to a 400-ji genkō yōshi and that’s why you have made it your standard.
. . . And if you count the total and divide, you don’t even need to mention “page.” Just tell your clients it is so much per 1,000 per words or portion thereof. (Using 1,000 rather than 200 as the base is probably easier for them to understand, and it gets you away from the “page” idea.)
. . . it sounds as if the easiest thing to do may be to not even bother looking for a translation for “standard page” but to simply ask them to use A4 paper for convenience, to double-space with 25-30 mm margins so you’ll have room to make changes, and to use 12-point so your eyes won’t rebel. Specify things so it comes out to 220-250 words per page. Then just charge them by how many pieces of paper they send.
If the average English word is 5 letters plus space (USA Today readability, not medical papers), 250 words is 1,500 keystrokes without the carriage returns and punctuation that make it readable.
I second [the] suggestion about using 200 words as your “page equivalent” yardstick—especially since it fits so well with the “divide by 2” upper bound for converting 400-ji genkō yōshi input text to English output. For starters, most of you are no doubt using word processors that conspire against fitting your output into the Procrustean bed of the mythical 230-word page: 65 fixed-width characters per line, 23 double-spaced lines per page.
. . . “What about short jobs?” I say: consider the total production-line cost of setting up before (switching your mind from the last job, digging out your notes, researching new bits, thinking up a file name
and sending off the finished product . . . Moral: Deep down, you’re charging for your time, so set a minimum charge to cover all the non-translation time required.
As a freelancer, I use the character-counting system to work out standard lines, 55 characters (including spaces) per standard line, rounded up to the nearest line and multiplied by my line rate (varying according to the nature of the text, my relation to the customer, etc.)
. . . All other “consulting work” (copywriting, revision, proofreading, typing in eastern European languages, tracking down appropriate fonts, etc.) is charged on an hourly basis.
I have had contact with some freelance translators who actually get paid by the hour for all translation work. That’s fine if the customer is prepared to accept you are worth so much an hour, but my experience is that in most cases I would lose out like this because generally having a reasonable rate enables me to compensate for more time-consuming texts with the dead easy ones.
You can try the international word count. That’s one of the tools translators used before they had word processors that would count words. It was originally devised on typewritten work that was not right-justified. Count the longest line on any page, including spaces, and divide that figure by 5. Then count all the lines in the text omitting any line that is less than half a page in length. Then multiply that figure by the figure divided by 5. It will give you a count that is fairly close to what you would get if you counted every word, and the longer the text the more accurate it becomes.
(Wayne Root, California)
From Newsletter No. 87 (December 1999)