Language Polishing Issues and Options
Over the fifteen years since 1990, the increase in papers submitted from non-native English speaking sources, particularly from China, has affected the pulse of scientific publishing. Editors are faced with the dilemma of research papers that they cannot accept due to an unsatisfactory level of English at the time of submission. At times even the value of the paper cannot properly be judged due to problems of language. It has become an all-too common predicament.

Should editors be responsible for the task of language polishing? Are non-native English authors to by-pass the English-language journal community? Are publishers to intervene? We spoke with Paul Evans, Elsevier’s Vice-President for Science and Technology China, and Jessica Rhys-Griffiths, Elsevier’s Science and Technology China Coordination Manager.

Current options
Click Language Editing on the Elsevier Author Gateway site and you will see links to International Science Editing (ISE) (www.internationalscienceediting.com) and Asian Science Editing (ASE)
(www.asiascienceediting.com). A little more reading and you will find that these are owned by the same Irish company. ASE provides language polishing at €0.024 per word. This is approximately 40% of the ISE rate of €0.06 per word for their premium language editing service which essentially provides copy-editing.

New service
There are more services in the pipeline to be added. “I felt the service on offer was not enough,” says Paul Evans. “Elsevier wants to continue to provide options and encourage authors to use them.”

The most recent option to emerge is SPI Publisher Services. Since January 2005, SPI has offered authors a pre-acceptance language polishing service, operating out of Manila in the Philippines, for €0.032 per word (www.prof-editing.com).

Author pays how?
One of the typically convoluted and daunting problems to be solved is that the author pays for pre-acceptance language polishing. That they should pay is recognized by authors without question. Very complicated problems arise, however, when they try to open their wallets. “Payment is a difficult issue and a real problem for Chinese authors,” says Jessica Rhys-Griffiths.

Of course, the methods of payment are all the ones we are used to: credit card, cheque, bank transfer, wire transfer, and stored value card (also called electronic wallet or local debit card). But China’s fragmented banking system does not have a centralised clearing structure. So cheques and stored value cards do not work for foreign payments. Even if their bank’s connection to an international clearing centre allows an author to pay by cheque, it can take up to eight weeks for the money to be deposited in the supplier’s bank account. Suppliers feel they simply cannot bankroll that delay.

So perhaps they can pay by credit card. But many Chinese scientists do not have credit cards. No problem, they can pay by bank transfer. But the average language polishing transaction will probably be somewhere between €60-€150 making the banking charge 30-50% of the total cost which is very costly for the author. In addition, our research uncovered that unless an author specifically instructs their bank to assume all charges for the transfer, the supplier must pay them, sometimes forcing a supplier to ask the author for reimbursement.

So perhaps they can pay by wire transfer. Our research uncovered an average €20 charge per wire transfer transaction, however, for a typical ASE payment (remembering that ASE currently charges the lowest rates to the author). So this is no easy solution for authors, either.

University pays
A possible solution to investigate is for the Chinese universities to pay accumulated invoices for all the services used by authors within their institutions at intervals, perhaps monthly. Rhys-Griffiths explains, “This would result in one set of transfer fees which would improve the marginal banking cost per author and provide greater security of payment for the supplier. This in turn would allow the supplier to waive any self-protective charges they may feel they would otherwise have to levy.”

This solution could take, however, considerable time to implement. “Barriers to this scheme may exist within Chinese institutions,” explains Evans. “It may take a long time to implement it across many institutions.” Also, because of the difficulties faced by western publishers operating in China, due to the Chinese government’s caution with regards to foreign investment in its information industry, it is extremely difficult to consider making a similar accumulated-invoice offering to authors at this time, although options are being investigated.

Pre-review filter
An appealing programme developed by Elsevier in China, the Chinese Editorial Scheme, is currently underway at Tsinghua University. This scheme provides a pre-acceptance review filter for papers being sent to Elsevier journals. At the moment the scheme commits 19 Tsinghua professors to pre-review papers on 20 journals. “Besides relieving overburdened editors from the task of language reviewing, CES has the added benefit of bringing journals and scientists into closer contact,” says Evans. “This is the first initiative of its kind and we hope to expand it to other top Chinese universities and research institutes in the coming months,” adds Rhys-Griffiths.

Integrating boards
“In addition to the efforts being made to provide pre-acceptance language polishing services to authors, journal editors are more and more seeking to populate their editorial boards with Asian members,” says Rhys-Griffiths. The idea is that an author may favour submission to a journal that appears sympathetic to non-native English scientists. With so much science and technology currently coming from Asia, this is a timely movement across journals that editors seem to have taken out of their own resourcefulness.

By-passing English
Of course, authors could seek publication solely in their native languages and hope to be translated later if their article were in some way spot-lit. Although English is currently the language of science, technology, and medicine the world over, will this trend remain consistent?

“In China, English is still very much seen as the lingua franca of advancement,” explains Evans. In fact, the Chinese government has recently decided that fully 20% of all course material at the university level must be in English. Clearly, this will have a lasting effect on the level of English authors attain. But it will take time for the effects to become evident. So, for the immediate future, the need for language polishing continues to exist.


Points to consider

• “(Chinese) universities are churning out vast numbers of engineers and scientists. And they're good.” (James Dyson, The Richard Dimbleby Lecture on BBC ONE, 9 December 2004)
  
  
• “From 1990 to 2002, the number of Chinese enrolled in higher education in China grew from two million to 16 million.” (Paul Wells in Maclean’s magazine, 7 March 2005)

• “US patent applications from the Asian countries of China, India, Singapore, South Korea, and Taiwan grew by 759% from 1989 to 2001. Patent applications from the US during the same period grew more slowly at 116% (though ... it should be mentioned that the Asian countries started out at a much lower base level). (A report from the Task Force on the Future of American Innovation, www.futureofinnovation.org, 16 February 2005)

• “Scientific papers from China are growing rapidly by volume and quality. It is forecast that by 2010 only the USA will produce more papers than China with Elsevier.” (Paul Evans, Elsevier’s Vice-President for Science and Technology China)

Please send responses to this article to EditorsUpdate@elsevier.com

----------------------------------------------------------------------------------------------------------------------------------------

26) Paul Evans, Vice President Science & Technology China, Elsevier

Options for language polishing. What you thought.
Responses received from Elsevier editors concerning language polishing are not unanimous in their choice of option but overwhelmingly support option B: “The publisher should facilitate language polishing, training for non-native English authors or work proactively with universities. These activities add value to the publishing process”. A number commented instead or additionally option B that “it is the author’s responsibility to submit papers in final and correct English” and in those few that did combine these options it was stressed that language polishing should occur pre-review.

Some editors emphasized the point that they do not have time to do intensive language polishing as part of the review process. Other editors (a small minority) said they normally did do a language edit themselves. In general the company expects its editors to focus attention on managing the scientific quality and the strategic development of the journal overall rather than being bogged down in language editing but it is understandable in some cases that editors feel the need to perform an edit for the work to be fit for purpose. Others, perhaps with less time available, have less compunction in returning the paper to an author to ensure the language is improved by them, preferably in association with a native or near native speaker.

The company has developed services which are advertised on the Author Gateway to allow the authors to pay for different service options for language polishing services from different independent providers. We are encouraging a choice to ensure the most competitive offerings and to help make the service as affordable as possible.

Other comments note that training is not a core competence for a publisher while linguistic editing is closer to a service we can offer. In China, where I am vice president for Elsevier Science & Technology, I am increasingly finding we are being asked to provide training possibilities for different people we deal with including authors to contribute as strongly as we can to the improvement of scientific communication. So although it may not be meaningful to expect us to enter into language training per se, we are likely to increase our activities with regards to showing people how to write a good paper, giving examples of common linguistic errors etc as well, in some cases, trying to increase understanding of the need for proper objectivity in the peer review process and focusing better on innovation in the development of new science e.g. with new methodologies rather than just repeated experiments.

25)  Chris Morfey, Editor, Journal of Sound & Vibration

In common with many scientific journals,  our journal publishes Discussion items, in which a reader contributes a discussion of a previous article in the Journal. My normal practice as Editor is to check the technical validity of any such contribution before acceptance, using external reviewers if necessary.

The question comes with the original authors' right to reply. What do other editors do if the reply sent by the corresponding author of the original article, intended for simultaneous publication with the Discussion item, contains serious technical errors?

Because of the risk of a prolonged but pointless exchange in the Journal, my preference is to subject the reply to external review (if its technical correctness is in doubt). Therefore in principle, the original author might be effectively barred from replying, by refusal to modify the reply in response to a critical review. What should then happen?

24)  Peter Holland, Editor, Physics Letters A
     
This is a long-standing problem. In the past it was solved to a large extent by detailed copy-editing of accepted papers. I became aware that this was apparently no longer being done when papers started appearing with ungrammatical titles.
     
Elsevier should consider providing an in-house pre-submission polishing service that is FREE to the user. We offer free reviewing of articles, which generally enhances their value, so polishing could be treated as an extension of that.
     
From a strategic point of view it would make commercial sense for Elsevier to become renowned for good quality English across all its journals. It would also provide clear evidence of service to the academic community. Recently, I have had several top US scientists refuse to act as referees because of the ongoing dispute over subscription charges. Part of Elsevier's answer in this debate is the value added to the papers published.
   
 
23)  Jim Cooling, Editor, Microprocessors and Microsystems

My experience gained in 10 years of editing MicPros leads me to believe that solution (b) is the best approach to the problem.

Solution (a) is naive (somewhat like saying that we won't have road accidents because people are told the rules of driving).  Solution (c) has a problem in that it demands a great deal from reviewers who are already providing their time and expertise for free.  Moreover, many reviewers, though technically expert, are not native English speakers.  

Whenever I've had a submission, which needed improvement in language, grammar, etc., I always suggested to authors that on the rewrite they have the material proof-read by a competent English speaker.  The final polishing is, I believe, the responsibility of Elsevier.

22)  Carlos Brebbia, Editor-in-Chief, EABE Journal

My opinion is B - ie the Publisher should facilitate language polishing, training for non-native authors, or work proactively with universities.....

21)  Peter Scheid, Editor-in-Chief, Respiratory Physiology & Neurobiology

Concerning language polishing, I agree that this remains to be the sole responsibility of the author. However, I very much welcome ways in which the authors can receieve help from professional services. Whether or not they can pay for it will have to be decided by themselves; sometimes authors have a colleague or friend with native English capabilities, who can provide the service free of charge.

I welcome any links on Elsevier's websites to professional services.

Incidentally, in my view, it is not only non-native speakers who require polishing. The standard of language is often quite poor for native speakers as well. However, the main problem is indeed the Far East, China in particular.

20) Takeji Hashimoto, Senior Editor, Polymer

B) is the best.

19) Belur V. Dasarathy, Editor-in-Chief, Information Fusion

A). It is the author's own responsibility to submit papers in final and correct English.
We wish this could be true. But in real world it is next to impossible to expect this to work in all cases or even in most cases. Personal experience supports this unfortunate state of affairs.

B) The publisher should facilitate language polishing, training for non-native English authors or work proactively with universities. These activities add value to the publishing process.
Would love to see this happen or some version of check for language problems before or during final formatting from the electronic version

C) Language polishing is part of the peer review process
It is too much to expect of the reviewers who are already volunteering their time for free. They can only point out examples of problems. This does not seem to help in many cases.The authors correct just the ones that were pointed out. Many do not even search for similar problems. Also, in quite a few instances we have reviewers who have the same problem. Indeed I have even come across some hilarious cases wherein the sentence criticizing the language itself has serious problems. (I hope I am not committing one such blunder here!)

Example - It seems that author does not know good English!
(poor language skills often leads to information confusion instead!)

18)  Keith Attenborough, Editor-in-Chief, Applied Acoustics

I tend towards C in the sense that when I have the time I do more a less a complete rewrite as part of my duties as Editor-in-Chief. On the other hand it is simply not feasible to do this with the increasing number of papers requiring such rewriting activity.  

In many cases the authors should be advised to seek local help before submitting.

17)  Gerald Farin, Editor-in-Chief, Computer Science & Engineering

Let me focus on one detail. The most common mistake is the omission of articles. I wouldn’t be surprised if there is some software, which specifically fixes that kind of mistake. If not, I don’t think there are more than a dozen rules governing the use of articles and developing a piece article-fixing software doesn’t sound too hard.


16)  Uriel Frisch, Editor,  Physica D

I choose option B without any hesitation.

15)  Fritz Spener  Co-Executive Editor, BBA - Molecular and Cell Biology of Lipids

Language polishing is part of the peer review process. Reviewers and/or editors must indicate the need of polishing, authors must follow suit as good as they can. The publishing house has  native-English speaking personnel who is doing the final job.

14)  Michael Schieber, Founding Editor, Journalof Crystal Growth

My answer is more towards B. The Authors may have important data, which is useful for the Community, and must be helped. The Peer-Reviewer can not dedicate the required time to correct the English, but if a very good new Software would be available, we could send it to the Reviewer and ask him to mark the required corrections according to the new Correction program.

I think that Elsevier should invest in such much better software for English Corrections. The Window 's English checking is very poor. In most cases, it does not even recognize elementary errors. Instead if they see some technical terms unfamiliar to them , they immediately underline it in red un-necessarily. So they do with non-English Family names. They will underline in red the generally used term Semiconductor, and ask instead Semi-Conductor. There are many examples of this kind, which do not help the Authors to make their papers more readable and correct.

Therefore, new software can and has to be developed by Elsevier.

13)  David Muir, Editor, Hydrometallurgy

B) The publisher should facilitate language polishing, training for non-native English authors or work proactively with universities. These activities add value to the publishing process.

By working pro-actively with Universities the publisher could encourage and pay postgraduate students to polish language of papers in their field and develop a better understanding themselves of the publishing standards required.

12)  Violet R. Syrotiuk Editor, Computer Networks

How should we organize language polishing and editing? Please choose answer A, B or C.
I believe this is a very important problem because the quality of written English is deteriorating.  I am happy to see Elsevier start to address the problem.  I was asked to choose between A, B, or C and perhaps B is the best choice.

A) . It is the author’s own responsibility to submit papers in final and correct English.
This is the default and clearly does not work. See as evidence many recently published articles by Elsevier and other publishers

B) The publisher should facilitate language polishing, training for non-native English authors or work proactively with universities. These activities add value to the publishing process.
This may help address the problem.  I believe publishers used to have editors that would work on the English of papers.  I favour reinstituting such editors to improve the quality of the journals.  However, the burden should not rest entirely on the publisher.  Certainly the universities should also provide help to authors to improve their writing skills.

C) Language polishing is part of the peer review process.
It is not the responsibility of the reviewer to rewrite the paper, just as it is not the responsibility of the reviewers to correct technical errors.

I believe the second option (B) is the right direction to proceed to address the serious problem of very poor written English.

11)  Giorgio Ranalli, Editor-in-chief, Journal of Geodynamics

A. It is the author's responsibility.

10)  Leon Chaitow, Editor, Journal of Bodywork and Movement Therapies

The piece in the current Editor's Update about foreign language polishing is interesting but does not deal with papers received where the English used requires more than just basic editing. This last year I have received papers written by individuals whose 1st language is either Greek, Spanish or Norwegian (and two from Singapore where Chinese was indeed the first language).All these papers had value and the struggle to get them to a usable standard took an inordinate amount of time.

Would a language polishing service for these papers have been appropriate from the Irish company?

9)  Leonard I. Wiebe, Editor, Applied Radiation and Isotopes

I strongly support option ‘B’.  Authors whose manuscripts need help can obviously not give themselves that help (option A).  My experience is also that when encouraged to do so (get help) they go to someone who has spent some time in an ‘English speaking’ country, but that individual, despite local perception, may be no better than the author.

Equally importantly, I find myself overwhelmed by Editor duties without polishing responsibilities (although I do this from time to time), so option C is ‘not on’.

8)  Beverly K. Philip, Co-Editor-in-Chief, Ambulatory Surgery

B) The publisher should facilitate language polishing, training for non-native English authors or work proactively with universities. These activities add value to the publishing process.

A DEFINITE value-added to the publishing process.

For the researcher and for the reviewer, we should emphasize the scientific contents of their work. Language skills should not be the barrier.

7)  Kent Voorhees, Co-Editor, Journal of Analytical and Applied Pyrolysis

It is hard enough to get reviewers without asking them to polish the English.  First, I believe it is the author's responsibility to provide as good as copy as possible.  If that is not good enough, then it is up to Elsevier to correct the English.

6)  John Lavery, Associate Editor, Journal of Mathematical Analysis and Applications

I vote for option B with emphasis on language editing/polishing rather than training.

 Language editing/polishing is time-consuming work (as indicated by the rates that are charged commercially). An associate editor that handles 20 manuscripts per year may have 10 manuscripts that need a lot of editing and others that need light editing. Assuming that only half of the manuscripts make it through the peer review process, this still means 5 manuscripts with heavy editing and 2 manuscripts with light editing, resulting in 50-100 hours of work per year added to the normal work load of the volunteer associate editor (a year = a nominal 2000 hours of work, although most of us wouldn't have the time to be associate editors or indeed do our "daytime jobs" if we limited ourselves to this small amount of hours of work per year).

The language editing/polishing should happen *before submission*. Main reason: The language-edited copy will likely contain technical errors that neither the author nor the language-editor are aware of and that get caught only when peer-reviewed. To get a language-editing rate as "low" as that charged by the commercial organizations (we may not consider those rates low, but they are, compared to the rates that are described later in this paragraph), one has to rely on 1) language editors who speak and write good (native) English but who have little experience in mathematics, 2) language editors who may not speak the native language of the author (and therefore may not be able to make correct guesses when there are ambiguities in the copy) and 3) non-interactive language editing (author sends in the copy, the editor corrects it without asking for clarifications from the author, final copy is sent back to author). As a result, one can expect the language-edited copy to have technical errors. (Just like technical translations done by people with plenty of dual-language experience but with little technical experience contain errors.) To get a well-written manuscript that is also scientifically correct, one must 1) use expert mathematicians (preferably not merely expert in mathematics but with a lot of experience in the area of the manuscript) who are also expert at writing and also know the native language of the speaker and 2) allow interactive questions between language editor and author. However, this option would increase the cost of language editing by a factor of 10 and would put it outside the realm of feasibility. What is feasible is to get the language straight first (perhaps at the cost of introducing technical errors) and then let the peer reviewers do their job, which includes identifying technical errors.

 Working proactively with universities or collections of universities is an option to using language-editing services. Another option is to work with the Chinese government and perhaps with other publishers to establish an organization inside China that can do the language editing.

 Training is a long-term issue that requires (additional) educational infrastructure that is not part of the main activities of Elsevier or of other organizations that do not have direct responsibility for education (in China or elsewhere). A short-term objective of having an author learn how to write the next manuscript better based on corrections to the current manuscript is fine and will generally have some moderate success. However, it is unrealistic to expect that the training experience gained from the corrections to one manuscript will be sufficient for the author to write future manuscripts stylistically well (unless the future manuscripts are nearly copies of the manuscript that was corrected).

5)  Merv Fingas, Editor, Journal of Hazardous Materials

I think that there should be a service available at cost (option B) -- however the major problem is that many authors do not know that their English is not acceptable to start with.

I suggest that a submission check list be made up and could include:

“Is your Mother tongue English?  If not has this paper been reviewed and edited for English by someone who's mother tongue is English?”

4)  Barry Lever, Editor-in-Chief, Coordination Chemistry Reviews

Not (A) - often the writer has no access to someone who could re-write his article in good English (but see comment below)

I do not see (B) as being particularly efficient or practical

As regards (C)- if you mean referees can do it - no, it is not reasonable to expect them to.

Our policy is to polish the articles in the editorial office, i.e. the editor does it. The only exception is when the article is written in such poor. English that it would need almost total revision- then it would be returnd to the author for improvement.

3)  Ronald Blanton, Co-Editor-in-Chief, Microbial Pathogenesis

Option A does not seem to be an option.  Otherwise I have a feeling that this may depend on the journal, the amount of staff and the volume of work.  For myself, I generally do the polishing.  There are other journals that have staff specifically for this and suggest that reviewers not waste their time on this.  Some reviewers will offer extensive corrections, others place the onus back on the author to find an English-speaker to edit.  However, it can be very hard for some authors, especially in developing countries to find a native speaker of English capable of reviewing a manuscript.  I imagine there is a point at which the language is so tortured that polishing would amount to re-writing.  I have not run into any like this yet that weren't already rejected on the science.  Native-speakers are not immune, but at least generally get the prepositions right.

2)  Jean-Pierre Burg, Co-Editor-in-Chief, Tectonophysics

Answer B is the safest to make sure the language is appropriate. Many reviewers are also non native English speakers and cannot do the job (which excludes answer C) and answer A is a dream that no author with language difficulty will fulfil.

1)  Claude Brezinski, Editor-in-Chief , Numerical Algorithms

I choose B.

Some papers are so badly written that they could not be sent to referees. This is author's responsibility.

But, once a paper has been accepted, the publishing company has to help author to polish his English. It is not the work of the referees or the editor.