Search JAT Search tips

JAT Bulletin 182, May 2000

This article examines the different types of translation a patent translator is likely to encounter and explores how to deal with them.

When translating patents, it is important to keep two things in mind:

• What is the purpose of the translation?
• Who will read the translation?

1. Translation for Filing

This term generally refers to the translation of patent specifications for the purpose of filing them in a foreign country. [1] Translation for filing does not have to be "literal" translation as required of other kinds of patent translation. In particular, the applicant is allowed to do a number of things, for example:

• correct errors that appear in the original
• make changes
• add text
• delete text
• combine two or more original applications into one translated application (in fact, this is frequently done by Japanese applicants)

Let us now return for a moment to the question of who the most likely and important readers of our translation are:

• Patent office examiners
• Judge/jury in court proceedings
• Potential licensees

Judge/Jury: Once a patent has been granted, it may land in court, where it will be scrutinized by judges and juries. Neither can be assumed to have a technical background worth mentioning (and to reflect this fact, the jury is sometimes referred to as "the twelve street sweepers" by some U.S. attorneys). The possibility of a patent being examined by a jury in the U.S. (there are no juries in Japan) has been reduced greatly since the introduction of so-called Markman hearings, but that still leaves the judge. Judges are trained in legal matters, but most certainly not in the field of the invention. Then there is also the "annoyment" factor. Many judges abhor patent cases, because it means that they have to dig through lengthy, dull specifications and other technical documents. Specifications that are lengthier, duller and even more convoluted than the average patent specification will not win any goodwill points with those judges (or the examiner for that matter).

Potential licensees: A truly successful patent is one that earns money. Besides winning an infringement action, this can be achieved by licensing the patent to third parties. Needless to say, the chances for this are greatly enhanced if the specification is easy to read and understand.

In short, patent specifications are ideally written in a manner that even a layman can understand them. There are several ways, a translator can contribute to the "understandability" of the specification, and thus add value to the patent:

• If there is a choice between two termini, choose the simpler one.
• If the sentences of the original are too long, make them shorter.
• If the original is convoluted, untangle it.
• If the original contains redundancies that make the translation difficult to read, delete them.
• If the original contains factual errors, correct them (or at least point them out to the author)

Patent translation for filing is always a teamwork effort, and close cooperation between the translator and the author of the original (i.e., the benrishi or the inventor) is necessary. Often, the author is not aware of avoidable problems that may occur down the road (e.g. due to different rules and case law in different countries), and it is one of the jobs of the patent translator to point them out and provide alternatives.

But, the reader may ask, can all this still be called translation? Isn't this rewriting? And is it really the job of the translator to do this?

Call it what you want - somebody has to do it. And in the patent business, it is generally accepted to be the translator's job - in fact, the translator is usually the only one with enough (linguistic) understanding to do this.

Back to top

2. "Literal" Translation

• Translations for entering the national phase of a PCT application (i.e. PCT translations)
• Translations first filed in a foreign language
• Translations of evidence in court proceedings
• Information disclosure statements (IDSes)

Unfortunately, the term "literal" translation is very unclear, and to the author's knowledge, there are no court decisions that shed any light on the question what a "literal" translation is and what isn't. This can be the cause of some interesting and sometimes vexing problems for the translator:

• What if the original Japanese doesn't contain a subject? Can we add one?
• Is it OK to split long sentences?
• Is it OK to add paragraph breaks?
• What about kanji-henkan errors?
• Is it OK to resolve ambiguities in the original?
• Can we make information that is implicit in the original explicit in the translation?
• How closely and literally must the translation remain to the words of the original when the meaning is best expressed using words not found explicitly in the original?
• To what extent can the translator augment the words to faithfully express the meaning?

When the translation should make the applicant "look good" (e.g. in PCT translations), the author of this article does not shy away from adding subjects or other implicit information, or resolving ambiguities where the intended meaning can be deduced with reasonable certainty, and still calls the result a translation that is "true to the original". However, it is not advisable to add paragraph breaks, because this may make the impression that text has been added to the original, and may lead to problems later in the application process.

Translations first filed in a foreign language: Some patent offices offer the possibility of filing an application in a foreign language. For example, the USPTO allows filing of an application in Japanese, whereafter an English translation has to be submitted within several months.

Information disclosure statements (IDSes): In the U.S., the applicant has the duty to submit to the USPTO all information that may be relevant to the application. Obviously, if this information is in a foreign language, it has to be translated into English. The examiner can - and will - use the IDSes filed by the applicant against the application. Thus, if the examiner believes that the information in the IDS anticipates the invention claimed by the applicant, the corresponding claims will be rejected. This means, it is not in the best interest of the applicant (i.e. our client) that the information in the IDS is coherent and easy to understand. Consequently, if the original consists of long, convoluted sentences that are barely comprehensible, there is no reason why the translation should be any better. Our client's interests are best served if the translation is comprised of equally long, convoluted, and barely comprehensible sentences, because that reduces the chances that they will be held against our client's application.

Back to top

3. Summarizing and other translations for the client's information

4. Example

【請求項１】　多孔質セラミック基板上に、多孔質電極層、プロトン伝導性固体電解質蒸着層、多孔質電極層が順次積層されて構成されたセンサ素子を備えてなることを特徴とする薄膜式ガスセンサ。

1. A thin-film gas sensor, comprising:
   a porous ceramic substrate; and
   a sensor element comprising a first porous electrode layer, a proton-conducting solid-electrolyte vapor-deposition layer, and a second porous electrode layer, which are layered in that order on the porous ceramic substrate.

Comments:
The gas sensor includes two main element, namely a substrate and a sensor element arranged on the substrate, which is clarified in the translation for filing.

The original Japanese recites two "porous electrode layers" and refers to these two elements by the same name. Under U.S. practice, this is not proper (rule of thumb: different elements have to be given different names), and filing a claim as in the "literal translation" with the USPTO will most likely result in a rejection of that claim. It is the patent translator's job to anticipate such problems and to solve them (e.g. adding "first" and "second").

Including the phrase "characterized in that" in the "literal" version is largely a matter of taste. It is the author's personal choice to add these words in "literal" and PCT translations for the sake of consistency, i.e. to have corresponding English for every Japanese phrase. However, as the phrase 「ことを特徴とする」 is present in virtually all Japanese patent claims, it is largely of a formalistic nature and adds only deadwood to the translation, so it should not be included in the translation for filing.

---[1] Recall from the first article in this series that after filing a first application for a patent (a so-called "priority application") it is possible to file applications in other countries within one year from the filing date of the priority application and maintain the benefit of the filing date.

Contents | Bulletins

Top | Home | Feedback