Note on Patent Claim Usage: "Characterized in that"

[Warren Smith]

I have a client who is very insistent that a claim translation that does not include "characterized in that" is a serious translation error. I did some research on this topic, and thought I would pass it on to my colleagues. 

In a search of all patents issued in 2017 at the USPTO, I discovered the following:

Patents issued: 351405

Patents including "characterized" in the claims: 5812

This is 1.7% of the total. (So I guess it is not *wrong* to use "characterized in that," but just rather unusual.)

This surprised me though:

Patents issued based on Japanese priority: 46632

Patents issued based on Japanese priority and including "characterized" in the claims: 311

This is only 0.6% of the total -- despite virtually all Japanese patent having the term 特徴とする in the claims. 

I guess Japanese patent translators are pretty well trained!

A caveat to this is that if the translation is for European consumption, perhaps "characterized" is correct -- I found an article online discussing the differences between US and European patents, and this difference was highlighted. Wikipedia on "Patent Claims" also discusses how "characterized in that" is correct for Europe, and not for the US, as quoted below.

Just thought I would pass it on, for general interest to my colleagues. 

Warren Smith

European Patent Convention[edit]

Main article: Claims under the European Patent Convention

Under the European Patent Convention (EPC), a claim must define the matter for which the protection is sought in terms of technical features.^[8] These technical features can be either structural (e.g. a nail, a rivet) or functional (e.g. fastening means).^[9]

Regarding the structure of a claim, under the European Patent Convention, what is called the "preamble" is different from the meaning the "preamble" has under U.S. patent law. In an independent claim in Europe, the preamble is everything which precedes the expression "characterized in that" or "characterized by" in a claim written according to the so-called "two-part form", and therefore everything which is regarded as known in combination within one prior art document, namely the closest prior art document.^[10] For this reason, in Europe, the preamble of a claim is sometimes also called "pre-characterizing portion".^[11]

United States[edit]

A claim may include the following parts:

·                 A preamble that recites the class of the invention, and optionally its primary properties, purpose, or field: "An apparatus..." "A therapeutic method for treating cancer..." "A composition having an affinity for protein X..." This preamble may also reference another claim and refine it, e.g., "The method of claim 1..." (See "dependent claim" below.).

·                 A "transitional" phrase that characterizes the elements that follow. The phrases "comprising", "containing" and "including" are most often used and (under some patent laws, specifically US) preferable, as it means "having at least the following elements..." and are therefore open (inclusive) and do not exclude additional limitations. The phrases "consisting of" and "consisting essentially of" are (under some patent laws, specifically US) more limiting, as they mean "having all and only" or "virtually only" and are therefore closed (exclusive). In the US, the phrase "consisting of" excludes additional limitations, while the phrase "consisting essentially of" excludes additional limitations that would "materially affect the basic and novel characteristic(s) of the claimed invention".^[12]

[Matthew Schlecht]

On Tue, May 8, 2018 at 8:59 AM, Warren Smith <drwo...@gmail.com> wrote:

I have a client who is very insistent that a claim translation that does not include "characterized in that" is a serious translation error. I did some research on this topic, and thought I would pass it on to my colleagues. 

In a search of all patents issued in 2017 at the USPTO, I discovered the following:

Patents issued: 351405

Patents including "characterized" in the claims: 5812

This is 1.7% of the total. (So I guess it is not *wrong* to use "characterized in that," but just rather unusual.)

 

     I believe the results of your analysis, but must say that I am surprised that the usage of the phrase "characterized in that..." is so meager.

     Early on I would put "wherein..." where the "characterized in that..." might otherwise appear.

     Then, I had a chance to pose this question to James Phillips (formerly of Horsefrog, now Director of the PCT Translation Division at WIPO), and his recommendation was to use the wording "characterized in that...", advice that I have followed ever since.

     Most of the "for filing" clients that I have insist that the translation must be as literal as possible, which also points toward rendering "特徴とする" as " characterized in that".

     I imagine that the ultimate arbiter of style will always be client preference, but I must still remark that I am surprised by your finding.

     Might it be that there are variations not counted in your analysis, such as "characterized in having..." or "characterized in being...", or even "characterized in comprising..."?

Matthew Schlecht, PhD
Word Alchemy Translation
Newark, DE, USA
wordalchemytranslation.com

[Carl Sullivan]

Matthew,

Interesting research; have wondered about this. The patent translation guidelines put out by the JLD years ago mentioned that this is optional; I would agree that it is at about that ratio of use, or a bit more, both in Japanese and Korean. James saying to use it is significant. I use it myself, but have allowed translators their own style when editing and I don't correct it. 

Carl

--
You received this message because you are subscribed to the Google Groups "Honyaku E<>J translation list" group.
To unsubscribe from this group and stop receiving emails from it, send an email to honyaku+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

--

Masae and Carl Sullivan 

MasaCa Translation Services, LLC 

CDR Carl Sullivan, USN Retired

P.O.Box 145

Manti, UT. 84642 

Cell: 801-369-6288 

[Jon Johanning]

> On May 8, 2018, at 11:11 AM, Matthew Schlecht <matthew.f...@gmail.com> wrote:
>
> Most of the "for filing" clients that I have insist that the translation must be as literal as possible, which also points toward rendering "特徴とする" as " characterized in that”.

Either for filing or for information, I think “characterized” is best, because it is what’s in the source.

Jon Johanning
Japanese-to-English translation
jjoha...@igc.org
www.jcjtrans.com

[Warren Smith]

Matthew conjectured:

     Might it be that there are variations not counted in your analysis, such as "characterized in having..." or "characterized in being...", or even "characterized in comprising..."?

My search was just for "characterized," so it picked up these cases. 

Here are my search strings and the results returned (copied and pasted from the USPTO search page):

ISD/2017: 351405 patents

ISD/2017 AND ACLM/characterized: 5812 patents. 

(ISD/2017 AND PRIR/JP): 46632 patents.

((ISD/2017 AND PRIR/JP) AND ACLM/characterized): 311 patents. 

I am having some severe problems coping with this client. I try to be flexible, but in the 34 years I've been doing this I have become somewhat calcified in my idea of what is "right" and what isn't. If one of you were to press me on changing my choices, I would give your opinion a lot of weight, but this particular client seems pretty clueless in general. For example, he objected to my translation of エンジン回転数 as "engine RPM," and we ended up going with his "the engine's number of revolutions," instead. I have a very difficult time bending that far....

[Warren Smith]

Either for filing or for information, I think “characterized” is best, because it is what’s in the source.

Jon Johanning

I find this very interesting – three of my most respected patent translation colleagues have taken this position, quite contrary to my own. Your opinions, of course, are extremely worthy of consideration, and must not be dismissed lightly. 

Warren Smith

[Matthew Schlecht]

On Tue, May 8, 2018 at 12:04 PM, Warren Smith <drwo...@gmail.com> wrote:

Matthew conjectured:

     Might it be that there are variations not counted in your analysis, such as "characterized in having..." or "characterized in being...", or even "characterized in comprising..."?

My search was just for "characterized," so it picked up these cases. 

Here are my search strings and the results returned (copied and pasted from the USPTO search page):

ISD/2017: 351405 patents

ISD/2017 AND ACLM/characterized: 5812 patents. 

(ISD/2017 AND PRIR/JP): 46632 patents.

((ISD/2017 AND PRIR/JP) AND ACLM/characterized): 311 patents.

     Thanks for the clarification.  I must be outside the mainstream on this issue, then.

 

I am having some severe problems coping with this client. I try to be flexible, but in the 34 years I've been doing this I have become somewhat calcified in my idea of what is "right" and what isn't. If one of you were to press me on changing my choices, I would give your opinion a lot of weight, but this particular client seems pretty clueless in general. For example, he objected to my translation of エンジン回転数 as "engine RPM," and we ended up going with his "the engine's number of revolutions," instead. I have a very difficult time bending that far....

     That's just dumb!

     It all depends on how much hassle you wish to endure, though.

     One client who insists on "as literal as possible" translations frequently sends jobs in which the claims are pretranslated (I guess, for legacy purposes) and I must use the terminology established therein.  In the one I am doing today, what I usually express as "wt%" was pretranslated as "per cent by weight", with percent as two words.  I know that if I press the point, that will require several iterations of communication and the end result will still be the instruction to follow the example of the pretranslated claims.  The only chance of making headway on such things, in my experience, is that the suspect phrasing must not only be dumb but also either too ambiguous or clearly wrong.

     So, I suspend judgment and complete the task as ordered, and I'll just add a note concerning this oddity.

[Kirill Sereda]

“Percent” as two words (per cent): I believe it’s the preferred spelling in British English.

 

Kirill Sereda

[Matthew Schlecht]

On Tue, May 8, 2018 at 12:31 PM, Kirill Sereda <kvse...@att.net> wrote:

“Percent” as two words (per cent): I believe it’s the preferred spelling in British English.

     I just checked, and you're right!
     The two-word form is preferred for EN(UK), which makes sense because that's the target language in this case.

     Learn something every day!

     I still think it would make more sense and be less wordy to put this as "wt%", though.

[Herman]

On 08/05/18 05:59, Warren Smith wrote:

I have a client who is very insistent that a claim translation that does not include "characterized in that" is a serious translation error. I did some research on this topic, and thought I would pass it on to my colleagues. 

In a search of all patents issued in 2017 at the USPTO, I discovered the following:

Patents issued: 351405

Patents including "characterized" in the claims: 5812

This is 1.7% of the total. (So I guess it is not *wrong* to use "characterized in that," but just rather unusual.)

This surprised me though:

Patents issued based on Japanese priority: 46632

Patents issued based on Japanese priority and including "characterized" in the claims: 311

This is only 0.6% of the total -- despite virtually all Japanese patent having the term 特徴とする in the claims. 

I guess Japanese patent translators are pretty well trained!

To non-rigorously test the theory, I searched USPTO using the search string "ISD/2017 AND PRIR/JP", and then looked at the Japanese priority application for 10 patents selected randomly from near the top of the list.

Of these, 5 did not contain 特徴 in the claims at all.

Of the other 5 that did contain 特徴, for 4 of the patents, the corresponding US claims were not a translation of the Japanese claims (as determined based on the fact that the number of claims was different between the two documents).

So your theory held for 1 of the 10 documents in question.

Thus, I would provisionally conclude that the reason US patents with a Japanese priority do not contain "characterized" in the claims is mainly that the US claims are not direct translations of the Japanese claims and/or because the Japanese claims themselves do not contain "characterized".

Herman Kahn

[Herman]

On 08/05/18 05:59, Warren Smith wrote:

I have a client who is very insistent that a claim translation that does not include "characterized in that" is a serious translation error. I did some research on this topic, and thought I would pass it on to my colleagues. 

In a search of all patents issued in 2017 at the USPTO, I discovered the following:

Patents issued: 351405

Patents including "characterized" in the claims: 5812

This is 1.7% of the total. (So I guess it is not *wrong* to use "characterized in that," but just rather unusual.)

This surprised me though:

Patents issued based on Japanese priority: 46632

Patents issued based on Japanese priority and including "characterized" in the claims: 311

This is only 0.6% of the total -- despite virtually all Japanese patent having the term 特徴とする in the claims. 

I guess Japanese patent translators are pretty well trained!

[Herman]

If "for filing" means that the translation will be filed as-is, then
using "characterized" is probably not advisable because it is not the
common US patent usage.

However, if "for filing" means that the translation is going to be
further manipulated by a US patent attorney, then, unless you have been
instructed otherwise, using "characterized" would be generally
appropriate, because in this case, your task is to produce a maximally
accurate translation of the source document to be used as a starting
point for generating the US application. Not using "characterized" for
特徴とする in this case would tend to imply that the original did not contain
the term, thus violating the underlying expectation that the translation
accurately reflects the source.

Herman Kahn

[Warren Smith]

So your theory held for 1 of the 10 documents in question.

Herman Kahn

Wow. This is very interesting. Thank you.

While in my mind  "特徴とする" still means "wherein" (so that translating it as "wherein" in a literal translation, just like translating 回転数 as "rotational speed" or "RPM" is every bit as much of a "literal translation" as "revolution number," and "many" is a "literal" translation for 沢山...), it just might be that "correct practice" is for me, as a translator exercising my function as a translator, to translate this a "characterized in that," where the statistics seen at the USPTO is because the attorneys, exercising their attorney function, then rewrite to a form that is more suitable for US filing. While I have never had complaints about this in the past (aside from this particular client, for which I have done work, indirectly, in the past), it could just be that my clients have been unaware of the source text). (That being said, I have also had attorney clients who are NJS, and they have not complained, so perhaps this too is not the full explanation...)

On the other hand, can any argument be made that there is a difference in substance between the two forms? If there is no conceivable difference in substance, then why would anybody argue that local conventions should not be followed (with "wherein" for use in the US, and "characterized in that" for use in Europe)? 

At any rate, thanks, Herman, for the interesting observation.

Warren Smith

[Matthew Schlecht]

On Tue, May 8, 2018 at 2:57 PM, Herman <sl...@lmi.net> wrote:

If "for filing" means that the translation will be filed as-is, then using "characterized" is probably not advisable because it is not the common US patent usage.

However, if "for filing" means that the translation is going to be further manipulated by a US patent attorney, then, unless you have been instructed otherwise, using "characterized" would be generally appropriate, because in this case, your task is to produce a maximally accurate translation of the source document to be used as a starting point for generating the US application. Not using "characterized" for 特徴とする in this case would tend to imply that the original did not contain the term, thus violating the underlying expectation that the translation accurately reflects the source.

     That is an excellent assessment.

     I would hope that attorney end clients would use such highly literal translations (and whatever notes we supply along with the translations) as starting material, and not simply slap on a cover page with their names on it.  Otherwise, what are _their_ clients paying them such big bucks for?

     Only lawyers or highly qualified associates would have the know-how and experience to fine-tune an application for a particular jurisdiction, and make choices about inclusions, deletions, or modifications that will be viewed as consistent with current patent filing practice and case law.

[Warren Smith]

     I would hope that attorney end clients would use such highly literal translations (and whatever notes we supply along with the translations) as starting material, and not simply slap on a cover page with their names on it.  Otherwise, what are _their_ clients paying them such big bucks for?

Matthew Schlecht, PhD

Yes! I once had the rather shocking experience of discovering that my rather marked-up, commented version (with concerns highlighted, intralinear comments, etc.) had been submitted to the USPTO without any modification whatsoever!  The client was then somewhat putout that it bounced (had an OA). Now I am very careful to state, very noticeably, in the cover email that there are issues highlighted for the "executive decision" of the attorney. 

Warren Smith

[John Stroman]

What we're discussing here is an Improvement claim or "Jepson-type"​ claim that "recites some or all of the elements of a known article, process, composition, or combination in the preamble to the claim, includes a transition that states "wherein the improvement comprises" or "the combination o=with the old article, etc., of," and recites in the body of the claim only the new or modified elements...i.e., recites the improvement." (quotation from Section 57 of Landis On Mechanics of Patent Claim Drafting by Robert Faber, 3rd edition, 1990). For many years the Landis book has been considered the most authoritative work on drafting of patent claims in the United States.

Further on, Landis states: In European country applications, the transition phrase is usually translated into English as something like "characterized in that" or "characterized by comprising." The "characterized" form is acceptable in the United Sates although not very commonly used. This language is frequently objected to by examiners, but the claim form can be retained through using a more typical Jepson claim language transition phrase. Except when the "characterized" form is used, the body of the claim starts with the word "comprises" or "comprising" as in normal United States claims...

Basically, "wherein" or "comprises" or "comprising" is preferred in the US, but in European claims an expression usually translated as "characterized in that" serves the same purpose.

In my experience, I've seen all of these usages, and they all perform the same function in the structure of the claim. I normally use "wherein" because I am an American based in the US. If the application is intended for EPO or WIPO, I might check with the client or arbitrarily use "characterized in that" if the Japanese uses 特徴とする.

John Stroman

 

[Herman]

On 08/05/18 12:01, Warren Smith wrote:
>
> While in my mind  "特徴とする" still means "wherein" (so that
> translating it as "wherein" in a literal translation, just like
> translating 回転数 as "rotational speed" or "RPM" is every bit as much
> of a "literal translation" as "revolution number," and "many" is a
> "literal" translation for 沢山...), it just might be that "correct
> practice" is for me, as a translator exercising my function as a
> translator, to translate this a "characterized in that," where the
> statistics seen at the USPTO is because the attorneys, exercising their
> attorney function, then rewrite to a form that is more suitable for US
> filing. While I have never had complaints about this in the past (aside
> from this particular client, for which I have done work, indirectly, in
> the past), it could just be that my clients have been unaware of the
> source text). (That being said, I have also had attorney clients who are
> NJS, and they have not complained, so perhaps this too is not the full
> explanation...)
>
> On the other hand, can any argument be made that there is a difference
> in substance between the two forms? If there is no conceivable
> difference in substance, then why would anybody argue that local
> conventions should not be followed (with "wherein" for use in the US,
> and "characterized in that" for use in Europe)?

If "wherein" is used instead of "characterized in that" in the
translation of a patent claim, where the context makes clear that
"wherein" is essentially synonymous to "characterized in that", then
beyond hair-splitting, I don't think that it could be substantively
argued that this is a wrong translation.

On the other hand, in the case of patent-related documents involved in
prosecution, litigation or other patent-related discussion, where the
clarifying context may be absent, rendering 「特徴とする」 as "wherein" could
become somewhat problematic. For example, if 「上面が平坦であることを特徴とする装置」　and
「上面が平坦である装置」 are both rendered as "a device wherein the top surface
is flat", it would conceal the fact that in the first case, "the top
surface is flat" is explicitly stated to be a distinguishing feature,
which is not necessarily true of the second case. So if this
hypothetical discussion were to be citing a patent application in which
you translated 「上面が平坦であることを特徴とする装置」　as "a device wherein
the top surface is flat", you may be forced to change it now to "a
device characterized in that the top surface is flat". In short, an
asymmetry would result between the Japanese and English, hindering the
ability to freely translate back and forth between the two without
something being lost in translation, thus leading to a situation where
the reader may ultimately have to refer to the source document in order
to fully understand what is being said. So the desirable characteristic
that a translation should "stand on its own" would be somewhat compromised.


Herman Kahn

[Warren Smith]

....thus leading to a situation where

the reader may ultimately have to refer to the source document in order
to fully understand what is being said. So the desirable characteristic
that a translation should "stand on its own" would be somewhat compromised.


Herman Kahn 

Well argued, sir! 

But your argument applies to use of the term 特徴 outside of the claims. Outside of the claims (and parallel claim language in the Specification), my go-to translation for 特徴 is "distinguishing feature," and I heartily agree that "wherein" would be incorrect. (Perhaps my favoring "distinguishing feature" is just a reflection of an irrational aversion I have developed over the years to "characterized." This is kind of like how I refuse to use "realize" for 実現する, favoring "achieve" instead...)

Warren Smith