A Conversation on Editing

Photo by Flickr user Daniel Pietzsch. CC BY-NC 2.0

Editing is a balance of many considerations. Photo by Flickr user Daniel Pietzsch. CC BY-NC 2.0

Following discussions with newsroom editors in September, some of us rolled up our sleeves and went to work on revamping our guide on decentralized publishing, which explains the process for editing and publishing stories first in a non-English language, then translating and sub-editing them in the English-language newsroom.

Decentralized publishing has always been tricky. The process has many steps and works across two languages, and changes found necessary in the English-language newsroom can sometimes send a discouraging message to authors, whose work is already published. We have to find a happy medium between respecting the author, editor and translator's work and making sure the post is written in a way that supports Global Voices’ mission.

Even when it doesn't involve other languages or translations, editing can difficult. We have to strike a balance between many considerations. We need to respect the work our wonderful GV community does, respect our readers from around the world, and respect our mission of building bridges. We do our best, but there is always room for improvement and conversation.

While working on the guide, Francophone Africa editor Lova Rakotomalala, Latin America community manager Laura Vidal and news editor Lauren Finch had an exchange that touched at the very heart of this editorial juggle, and we thought it might be beneficial to let the community in on the discussion. Read the back-and-forth (which has been slightly edited so that it makes sense taken out of its original context) below.

Hello Lauren,

Thank you for this guide.

Here is a hypothetical situation: one author pitches a story in a non-English language. It has been edited and agreed upon with the language editor and published on that site. The post is then translated and sub-edited for GV in English. A few editorial changes were made in the structure of the article (paragraphs orders were modified). It will be quite difficult for the original author to accept the changes to the original article considering that it was already worked on and edited the first time around and that it was already published.

Wouldn't it make more sense to let the original posts stay as the author and editor originally published (not including factual corrections or links that are necessary)?

The strain that such changes impose on the relationship between language editors and authors is not inconsequential. One ought to also take into account the challenges that GV has when it comes to recruiting authors and imposing a certain style on their writing. We ought to be careful to not add additional barriers to their writing. We also ought to consider that correcting original posts basically double the amount of work for the editor. In effect the editor would have: 1) edited the original story, 2) identified a translator quickly, 3) edited and checked the translation for quality, 4) applied the corrections from sub-editors, 5) applied the changes to the original article, 6) explained to the author (and the translator) why their original work were changed.

I realize there is no perfect solution here, but I am hoping that we are not losing the point of view of editors and authors in our efforts to get Lingua sites as close a translation as possible of the English site.


Hi Lova,

Thanks so much for the comments!

Lova, you’ve touched upon the very trickiness that is inherent in decentralized publishing that we cannot solve fully until we can one day afford to contract sub-editors in all our languages. The idea that major changes made to the English need to be reflected back in the original isn’t new; that’s in the original guide. I tried to expand this a little bit in this guide to make it clear that not every change needs to be reflected back:

“Changes to headlines, excerpts and certain stylistic elements in the English translation do not have to be reflected in the original story post-publication unless it is a matter of factual accuracy, given that different languages have different conventions.”

We have to somehow find a balance between keeping our translators happy by making sure their translation matches the original as much as possible; our authors happy by not making too many changes to the translation after it’s already been published; our editors happy by respecting their judgment; and keep up our editorial standards to 1) meet our mission by writing for a global audience, 2) which in turn keeps our funders happy.

You asked, “Wouldn’t it make more sense to let the original posts stay as the author and editor originally published?” That all depends. In your hypothetical scenario, it’s a bit hard to say for certain because it depends on what the paragraphs were and why they were moved. For example, say it was an important context paragraph buried at the end of an 1,000 word article. Perhaps that is an afterthought for a French-speaking audience, but for a non-French speaking audience, it’s a risky gamble to assume they’ll have the patience to be confused for 1,000 words until finally reaching that paragraph (and then possibly with that new information in mind, have to go back and re-read the article to truly understand it). We’d do a much better job of communicating the author’s message, of building that bridge, if we brought that up to the top, but that might not be evident until it has reached translation and sub-editing stage.

Edits like this (changing paragraph order) would be made in situations when, for whatever reason, the placement of that particular paragraph seems distracting or incongruent in its current location, and could be detrimental to the reading experience in English. This could be an English thing and work perfectly fine in French. Among all the things we have to keep in mind in the decentralized publishing process, we have to remain a bit flexible to accommodate styles across different languages while at the same time being true to the original. That’s a tall order sometimes.

In sub-editing, our philosophy isn’t to make changes willy nilly, and even when we do make a change it remains open to conversation. I have a hard time envisioning a scenario in which a sub-editor wouldn’t work with the editor and author who expressed that they didn’t agree with the flipping around of a few paragraphs, for taste or other reasons, because we’re very much aware of the strain this can put on editors and authors. Perhaps part of the issue is us as sub-editors haven’t done a good enough job explaining our reasoning in the edits we make, and I do know in the past we haven’t made it clear enough that sub-editing should be always be a conversation the same way editing is a conversation, and if an editor or author 100% disagrees with a change, they should absolutely say so.

Another bit might be that we need to manage the expectations of our authors more clearly – let them know beforehand what will happen to their post, that publication in French might not be the end of their writing road since we work across so many languages.

One way to avoid such scenarios is to do what GV Bangla or GV German does whenever a post isn’t super time-sensitive — work on the original non-English, translate it without publishing, sub-edit the translation, negotiate any changes in that moment to make both versions match, and publish simultaneously, instead of publishing immediately the non-English post. It regularly works rather well with those two teams.

What do you think? That’s a novel I just wrote. Is there any specific change you’d like added to this guide? Perhaps more emphasis on editing philosophies or the challenges of keeping so many people in mind during the process…?

I appreciate it, Lova. Thank you!


Thanks Lauren for the helpful clarifications.

I think you touched on the right solution for non-time-sensitive posts.



Hi all,

First of all. This summary of the process is very effective and clear. I like it.

Now, some more ideas on the question of decentralised publishing and sub-editing:

-In the case of the LatAm team (where 98% of our publishing is decentralised) the question around editing and sub-editing has shown the need of a more fluid communication between editors and sub-editors, although it has improved significantly in the last months. Now, to be able to make this process more fluid, we need to work even further. There are agreements that need to be reached in terms of style, content and context, so these aspects don’t take too much time to negotiate when a post is being worked on.

I wonder if a good solution for this is to build bridges between different editors and sub-editors to discuss regularly and reach some sort of general agreement so we can all be more or less on the same page when working with our authors.

-About content and context: I agree and support the idea of being attentive to context. In our case, it has been a challenge. Negotiation has been constantly needed. Since Latin America is made of many countries that speak Spanish and whose written language can work well across borders, there’s a sort of idea going around (specially when authors are publishing first in Spanish) that Spanish-speaking audiences will surely understand contexts happening in other Spanish-speaking countries.

I believe this has improved significantly, but we have had to meet halfway many times in terms of style and context. There’s a moment in which context and style start interacting with each other. There are moments in which the need for context can break with the style that the author intended at the beginning. Sometimes, context makes the post longer and less attractive for the author. Some sub-eds might prefer to sacrifice the style of the author to make things clear to the reader. I think this is an important point, and it’s something we can hardly solve with a guide. What has made things improve here has been constant exchange and open communication.

The order of paragraphs as an example is a great one. It shows the difference between changing style and working on context. There are moments in which changing the order of ideas will make the post work in English, but doing it in the original language might break the rhythm the author was aiming for, and might not be crucial to understand the main idea better.

This tells me that it’s important to understand that audiences should be thought of as divided by both language and region. Of course, there will be elements familiar to people speaking the “same” language, but this is far from being the case 100% of the times. French-speaking audiences from Senegal might not necessarily get the context of a story happening in Belgium. This, sometimes, has been a difficult point to convey to authors. Making authors understand the sub-ed process is a must, because it’s not about them losing their way of explaining things, but reaching a sort of common ground in which their story can go far. From what I have seen, this happens differently from author to author.

Decentralised publishing has opened a huge door in many ways, and I think these interactions, though complex (and sometimes complicated) can help us reach a flexible and more “global” style, and adopt/adapt styles that can work for our stories, even if sometimes these may differ from the traditional journalistic styles we know and like.

Phew! Thanks for reading all this! I hope it’s helpful!



Join the conversation

Authors, please log in »


  • Please treat others with respect. Comments containing hate speech, obscenity, and personal attacks will not be approved.