Zoë Mullan is the Editor-in-Chief of the open access journal, The Lancet Global Health. She is an Ex-Officio Board Member of the Consortium of Universities for Global Health; an International Advisory Board member of Sun-Yat Sen Global Health Institute, Guangzhou, China; and a Scientific Advisory Board member of the Centre for International Health Protection at the Robert Koch Institute, Berlin, Germany.

Between 2013 and 2017 she was a Council Member and Trustee of the Committee on Publication Ethics. She trained in Biochemistry at the University of Bath, UK, before joining the publishing industry in 1997 as a Scientific Information Officer with CABI. She moved to The Lancet in 1999, where she has worked since, variously as a technical editor, section editor, and founding editor of The Lancet Global Health.

In this episode, Zoë and Jo talk about the importance of multilingualism in Global Health research and how the editorial team behind The Lancet Global Health is facilitating bilingual research article submissions.

To see all episodes, please go to our CONVERSATIONS page.

TRANSCRIPT

Jo: Welcome back to our podcast at Access 2 Perspectives conversations. And welcome, Zoe Mullan, to another episode where we will be talking about translations. It’s a great pleasure having you here. 

Zoe: Thanks very much, Jo. It’s great to be here. 

Jo: So Zoe works for The Lancet who recently announced that they established translations or that they invite translations of research summaries for that matter, and are in the process, or have established a process to standardize translational work to make research articles more accessible to bilingual and multilingual research audiences and readers. So maybe getting into that topic, would you mind talking a little bit about how this decision by The Lancet editorial team came about and how it’s being adopted by the researchers and authors? 

Zoe: Sure, yeah. So it all started, I think, with my colleague Anne Rocker, who’s no longer with us, but she was a senior editor back in 2018, I think we started this. And she is a French speaker. Her native language is French. And so she had a strong interest in multilingualism and she made the point very strongly that as an open access journal, The Lancet Global Health was opening up accessibility to authors, researchers and indeed the general public on a financial level. So we have no barriers to reading in terms of being able to see the work that we publish. There are no subscription barriers, for example. But that wasn’t the only barrier. It wasn’t the only access barrier. There is a vast number of people in the world who do not speak English at all or do not speak English as a first language. And so, in effect, we were still failing to reach a great number of potential readers just through publishing solely in English. So that was where it came from. It was the idea that we wanted to be open, we wanted to be accessible, but we hadn’t managed to break down all the barriers to accessibility, just the financial ones. So she was the person who really kind of got this onto our radar. And once we started to talk about it internally, it was seen as a no-brainer in a way, really. Of course, just publishing in English is a barrier to access, so we kind of took it from there. 

Jo: Okay, so basically having a team colleague in house who knows of the challenges and is very much aware because of this case, her Francophone background and was it difficult to convince the team and also the management to take the next steps? Because I assume there’s a lot of organizational extra work that’s coming your way and with that, of course, an easy decision to acknowledge. It makes sense, but then on the operational level, it’s a whole extra effort, basically on top of the already ongoing tasks and deliverables. 

Zoe: Yes, that’s right. Absolutely. Those were the main internal barriers, I would say are the kind of logistical ones, the work related ones. We’d kind of toyed with this idea a little bit in the past, how could we start to translate more content more regularly? But it is an expensive process to take on linguists who know the subject matter and which languages you choose and how do you work that additional step into a streamlined workflow that doesn’t hold up the publication process too long. So we’d always found that there were too many ifs and buts and difficulties to get it going. But this time we aim to come up with a workflow that basically disrupts the existing workflows as little as possible and at virtually no cost. We actually had no budget to put this into place, so our aim was to find something extremely cheap and non disruptive and the best way we decided we could do that was by actually having it as an option that the authors took on themselves. So we decided to offer authors the opportunity, once the manuscript had been accepted, to translate the abstract into whichever language they felt most appropriate to the audience. And that could be more than one language and they could be very unusual languages. So we’ve had translations into Zulu and Indonesian as well as the French and Spanish and Portuguese main languages and we post them as additional files alongside the main manuscript, so they’re not incorporated into the body of the manuscript once it’s published. They are supplementary files, so there are some limitations. It’s not an ideal situation. We are exploring how we can make them more visible at the moment. So that’s the main ongoing challenge. But there’s certainly been quite a good uptake. I would say about 25% of accepted manuscripts get translated into one language or another or several, so we see that as a pretty good vote of confidence in the project.

Jo:  It’s like a quarter of the submissions. It’s pretty good. And do you assign a supplement? Do the translations also get assigned their own DOIs? 

Zoe: No, they’re part of the same manuscript, so they’re under the same DOI. So the main problem with that is that they’re not searchable. So if you wanted to do a search for the manuscript in the language of your choice, you couldn’t find it in PubMed or a Google search at the moment because as I say, it’s not embedded into the HTML of the manuscript, it’s only a supplementary PDF. So that’s the main issue with it. But I must say there has been some great feedback from readers and from the authors. 

It’s almost a signal of intent as much as anything, I think that this is an important thing and this is the first step towards making it something. 

Jo:  It’s a very dramatic step. Also, like I said, it’s low budget, is affordable for a publisher or journal or other editorial team with a limited budget. It incentivizes the authors also who probably have an urge to make their research up as available in various languages to provide some sort of infrastructure to make that possible. And yeah, as I said, it’s certainly a step in the right direction, opening up to a vast majority but many researchers are not capable of reading or speaking English so well or at all. Even though it’s not machine searchable as such just yet. It’s possible to send along the translations to point out that there is a translation available. But is that in your team? Is that basically on the to do list? I heard what you said earlier, it’s like one step and we’re learning from the community also as whatever the infrastructure allows for how to further improve. Right, so it’s work in progress, so to say, also amongst your team. 

Zoe: Yeah, that’s right. I think the next step is to make them more visible. At the moment it’s a little bit difficult to find manuscripts that do have a translation. It’s not immediately obvious. So at the moment we’re thinking about how we can flag that up more prominently for readers to see once they come across the English version. Please note there is also a French or Portuguese or Aramaic version available. So that’s the next step and the slightly larger step after that is to try and somehow make them searchable. But that is very infrastructure dependent really. Yeah, and somewhat might not be able to happen until we move the whole of elsevier onto a different platform, for example, which is obviously a big ask on something that isn’t going to happen just for our small translation project. But it’s something that a lot of the team, because we have a very multilingual team across the Lancet, there’s always going to be somebody driving this internally, I think. And now we’ve established it externally and started to sort of promote it a bit more that will hopefully hold us to account as well that sort of keep going with it and make it more usable in the future. 

Jo: Yeah, that’s great. And what do you count as a translation? Is it only if really the whole article is being translated? I’m asking because I find this is like a huge extra effort to ask for the researchers and normally it would require a whole league of experts, translators, science journalists, researchers who are knowledgeable of the other language, but probably the practicing researchers who have been working on the project, but not necessarily only. And then because with Africarxiv, a project we’re coordinating, we’re currently doing such a project where we thought we would translate the research articles and then came to learn that it’s actually more feasible to create a lay summary and translate that into traditional African languages just for the fact that some only spoken language, they’re actually written languages, the ones that chosen. But many traditional languages do not have such vast technical terminology. So we actually have to also define certain works and terminologies or terms and also depending on what we want to achieve with making the research available in various languages, it’s also maybe even more approachable and accessible to have it in a summary style instead of an emirate structure research article style.

Do you see different kinds of translational styles being submitted or co-stubmitted as supplements?

Zoe:  So this is strictly research abstracts only. 

Jo: Okay, that’s good. 

Zoe:  Again, that makes it more doable because we have to build it into the workflow such that, because all our research manuscripts are quite heavily edited post acceptance by our inhouse team, we want the translated version to reflect the edited English version. So it means the translation has to come quite late down the workflow after English editing. So there’s not really time, I don’t think, to ask an author to translate the whole article after it’s been through the internal editing process. So we just asked them to do the summary. And I think the reasons you gave there for just doing the summary are actually quite relevant to us as well because it may be the people who have less technical knowledge in the first place who want to be reading the translation. And you’re right, that actually because English is the sort of lingua franca of science and has been for a long time, there are pieces of terminology that just don’t exist in languages other than English. So I think it makes sense for us just to do the abstract for the time being at least on this kind of regular workflow project. Having said that, we do actually offer or we accommodate translations of entire manuscripts or reports if the authors want to do it. So, for example, we published a commission on financing primary health care recently which is a sort of 25,000 words document and it was published on our website. And after the fact the author said well, we have a little bit of budget left to pay a translator to translate the whole thing into French. So in that instance, we just enabled the authors to have the translated version styled into our word processing kind of style and to upload it onto the website, the platform with the English version. So, although that isn’t something we offer as a sort of regular workflow, I think if we want to be as open as possible to any author who has the means and the inclination to translate more than just the abstract. 

Jo: Yeah, I was also in a working group that was led by cross rep where we discussed mostly preprints and then also as part of preprints like translational works or translations and summaries or summaries as translations, no translated summaries. And there we agreed there were a few preprint community preprint service and platform leaders and coordinators were present and we all agreed that translations and also summaries of translations, as we do with declarative science and I presume also what you do with Lancet, should be able to get their own DOIs, just like you said, to make them searchable and discoverable. And I was personally hesitant before, but the translational work also comes with some sort of interpretation because in a different language you always bring the experience and the cultural heritage of the language group into the research project to some degree more or less like suddenly more in cultural studies and sociology and maybe less so in bioscience and modern science now with gene technology and those things. And therefore there’s basically also a level of interpretation and that basically then comes almost like a review sort of thing which probably deserves its own DOI to make it count as its own research output in a way because there’s also so much contextual knowledge from the language group going into the research and vice versa. Like learning from the research presented in English. Making that accessible to the other language group. But it’s certainly not established and this is not to criticize again, but rather to encourage further like also your team and other teams, maybe some other editors are listening to establish workflows where that is possible, that your eyes can be assigned, but also connected obviously to the original article that is being either translated in full or as a summary.

Zoe: Can I just make a quick point on that? I think that’s really interesting actually that the point you make and I think it probably is much more relevant for kind of the social science type of work. I think at the Lancet we would want to see the translation as a faithful duplicate, if you like, in a different language from the English version. And that’s why we only get the item translated after it’s already been edited in house. So it’s almost a carbon copy, but in a different language, if you like. So I think in that case having a separate DOI would be slightly misleading for us because it would suggest that it was a different version, whereas in fact we want it to be the same version, just accessible in a different language. So yeah, it might depend on the subject area and I think for biomedicine, like you say, it might not be quite as relevant. It’s a really interesting area because I think book translation, for example, if someone’s translated a piece of fiction that’s almost like a new book in itself because there’s so much input from the translator into the writing of a novel which is as beautiful prose wise in a different language as it was from the original, that’s almost like a new work in itself. But I think certainly for biomedicine we’d want it to be a faithful copy rather than a different one.

Jo: Yeah, a very good point and the discussion is certainly not turning over. Even for that, I also think not to be misunderstood, like I think the level of interpretation is minimal, like less than 1% or less than zero 1% really where cultural heritage comes in through the language. But if you look at it from a different perspective, if a non English speaking author or English as a second language author writes about their research which they have conducted in their own mother tongue a lot of the contextual knowledge might get lost. And again, also here, certainly more in virtual studies and sociology or social sciences. But then it’s just like I just generally want to make the statement that language carries cultural context, and with that, also some sort of traditional knowledge and contextual knowledge which might get lost to a certain degree if you strip it up, if you translate to another language we’re not very firm with as a submitting author, and also even for English speakers. And if you publish in English and you think, oh, it has to be technical English, but some aspects of every research project, I would guess, still have some cultural dimension which might get lost if the research offer is presented in a purely technical manner. But that’s becoming a little bit philosophical. I don’t know if that’s too. I think it’s a concept.

Zoe: Yeah, it’s a tricky one, isn’t it? But I think it’s definitely worth discussing a greater length at some point. But it’s probably not within the scope of our little project just at the minute. 

Jo:  What is your personal take on the translational work? Was there like an eye opener in the process where you realized, oh, yeah, that actually makes sense? I mean, it’s obvious we can all agree it makes sense to open scholarly publishing opti multilingualism of some degree, but do you also have a personal story that comes along where you felt, okay, we as a team, we were not able to facilitate one particular project to reach wider where you were personally involved with, which made the extra effort worthwhile?

Zoe: I was surprised. I think. Even myself at the range of languages that were taken up when we launched this. And I think slightly humbled by just how keen authors were to translate into what are to me quite unusual languages. But they are so important because they are done in a community that speaks that language principally. And not English and not even perhaps the official language of the country. Which might be something other than English. But it’s yet another more regional or tribal language that is not rare. But you wouldn’t think to see it in the pages of a global journal. Something where the script is very unusual.

We are in our work in global health, trying to do a lot of decolonizing work where we make some effort to make sure that research projects that we publish have been done in an equitable way. So if we see some research that’s been submitted, and it’s been done in a community, in a low income country in Africa, for example, and we know that there’s been a lot of input from researchers on the ground, from health workers on the ground, from people collecting data on the ground. And yet the authorship on the manuscript is sort of 100% North America and Europe. We’re now starting to say, no, that research isn’t going to be published with us. 

Jo: Yeah, I saw that announcement was so well taken up, like, in my community, which are all very much concerned with exactly that issue that you highlight. Like, can you just say a few more words as to what brought your team to make that announcement? Because that’s really a game changer. And it’s also perfectly lined with let me just briefly add that, like, with our work with Africarxiv, we tried from the beginning to have the same approach primarily. Like, Africarxiv has an open access portal and preprint repository with a focus on research coming out of Africa by African scholars primarily, and also non African scholars who study African topics. And for the latter, we want to encourage increasingly. So I want to sensitize that exactly what you just said wouldn’t happen as much anymore in the future or at all. If there’s any research being done by non African scholars, they would be requested to seek collaborative research projects with African scholars, scholars as partners, as project partners, and then submit as co authors and co contributors. We still see a lot of research out there. Also what we kind of trace and can see on Petrol Studies and African Studies on Africa without any involvement of African scholars. So what was the behind the scenes discussion that led to that announcement at the Lancet? 

Zoe: Yeah, I think it was a long time in the coming, and I think there have been lots of very powerfully written pieces. We published a collection earlier this year called What’s Wrong With Global Health? And we invited scholars from lower middle income countries principally to write to us about what they saw with the issues, with researching global health specifically. And so that actually allowing the people who kind of live global health to speak out and say what the issues were from their perspective, I think was very illuminating for us and led us more firmly down this path to the point where we made some of these statements. So I think when we’re talking about global health particularly and making research aims and objectives stated in projects, they’re about the community. We need to keep that community in mind all the way through and whether that’s involving researchers, health workers, community members from that community, not just the writing of the report, but the design of it from the start. We don’t want to be having somebody in a high income country invent a research question back in their ivory tower in London and then sort of come in and say, right, we’re doing this project, and would you like to put your name on it? Because then it looks slightly better so it’s proper involvement of individuals from the community in the research design as well as the process and also the writing and then through to the dissemination of it. I think that’s where the language fits in. If you are seriously concerned principally about the community that you’re working in and not just your own kind of CV or publication profile, then you should be making efforts to translate it literally to the community and to disseminate it and truly try and make a difference there. 

Jo: Absolutely. Often the complaints I hear indirectly are that when research in communities is being, like knowledge is being extracted but never brought back to the community and sometimes even the researchers often want to come back but then there’s no time in our budget to do so but if it’s properly planned for. As I said before the project even starts and to think about the potential stakeholders who should also participate in the project design to reasonable and feasible expense because they usually come without any payment or some of the stakeholders in the region don’t get paid for the extra work and the interviews and the collection. But they should be compensated one way or the other. And actually in many cases there should actually be a budget also to pay these people. But as an editorial team you can do so much and that’s what you did right to raise awareness to that necessity and best practice. But when submissions are coming in, the work is already done like the project is done and now it’s being presented and submitted to you or do you see that you have an influence maybe for the future project of these authors to sensitize for other projects? 

Zoe: Yeah, I mean we have rejected manuscripts simply because the authorship was not reflective of the work that was evidently done on a couple of occasions. Well, hopefully that’s led to future projects being better thought through in terms of equity. 

 But in a couple of instances it has actually led to a resubmission to us with input from some of the people who should have been on the authorship to begin with. So it doesn’t get away from the fact that perhaps the research question wasn’t entirely put together with the community involvement but certainly the writing of it to a certain extent that can be adjusted after the project has been done. For instance, this goes back to cultural interpretation. So you may have the same findings from your research, you’ve got the same numbers in table three, but when you’re writing your discussion, your interpretation, that may be very different if you’ve got somebody who’s actually working and living in that community looking at those data and thinking well, what does that actually mean? And it might be something quite different from what the Northern partners had actually interpreted it as meaning. So on a couple of occasions we have actually had resubmissions in which the authorship has changed, the conclusions have been rewritten with the input of those on the ground. So yeah, I think that’s potentially a good outcome and hopefully we’ll lead more thoughtful project planning from the beginning and the next time around. 

Jo: And just to close the circle back to our original topic, because this is obviously also a topic that’s very important to both of us and also Access 2 Perspectives. I had a previous episode on helicopter research which is basically the term for what we described earlier. He is a Kenyan scholar and he said even as Kenyans, like even us, we do the same mistakes nowadays, like working with, in his case, fisherman. It’s an extra effort, an extra thought process. But his team also makes a lot of effort to educate but also learn from the fishermen about fisheries and fisheries for his fisheries research and to unlearn making assumptions from the ivory tower perspective about how fishery should be or not be and how it can be improved or not. Because the fishermen have the actual experience on the ground and how they can breed the fish best. What the obstacles are, what to beware of, weather conditions, what not, water quality. Where the researchers can then come in with their science and knowledge about the ecosystem and in collaboration actually have both stakeholders and also other stakeholders learn from the exercise of working together in a project. Looking towards the translational aspect, a question to come back to our original topic of translations would be have some of these decolonized articles also had translations into our respective local languages just by coincidence? And then I was also going to come to another question like with the credit taxonomy or does your team also look at the credit taxonomy and where do you draw the line? And who would still be considered as a co contributor or A.K.A co author? Can they also be non scholars or especially when it comes to global health, said there is medical personnel who might contribute to the research project. Would these also be accepted or be encouraged to be co contributors along the credit taxonomy ideology? That make sense? 

Zoe: Yeah. So a few questions. I think I’m not going to be able to point to a specific article that was where the authorship was sort of changed and then we translated it, but I’m sure it must have happened. But in terms of the second part of your question, the Lancet, it doesn’t actually use the credit taxonomy as such. I think there was a concern that having something like that makes it a little bit too easy for authors to sort of assign authorship on almost like a checkbox basis. Oh, I think we could probably shoehorn so and so into that particular line in the taxonomy. So we feel that a more free text description of a contribution is a slightly better way to do it. So we actually prompt authors to write in their own words what each author contributed to the paper and we find that tends to give a more realistic description of what those contributions actually were. And of course we refer authors to the ICMJ authorship criteria. They should be having those in mind when they’re assigning authorship. But we tend not to get involved with saying who can and can’t be an author. And that certainly means that you don’t have to have an MD or a PhD to be an author on a Lancet journal. So I think anyone who has made a substantial contribution to the design, the analysis, the writing, whoever they may be, should qualify to be an author. And if you can describe honestly what that person has contributed, they in your mind count as an author, then that’s what we go with. We’re not particularly prescriptive on who should be an author and who shouldn’t. That’s very much up to the team themselves. But we do ask for a definition of what they’ve done. 

Jo: Okay, well, yeah, makes a lot of sense. And I personally see credit taxonomy more like a guideline.

And also data collection is essential to the success of a research project and that can be done by an undergraduate student or an assistant. And then the question is, do these people need the acknowledgement through co authorship or is it for, is this also career boosting for them? So does that provide some sort of compensation for the effort? And otherwise do they get paid for the work? And even if they get paid, can they make use of that acknowledgement in that form? Or is there another more feasible compensation mechanism? But still, I think it’s also a matter of research integrity to list the people who contribute to the success of the results of a research project one way or the other and acknowledge the acknowledgement section. I’ve always found the funny part like okay, we thank these people for contributing to our work and why are they not co – authors? It’s also very random where people draw the line here. And I think it’s also a cultural thing, like a disciplined cultural thing of having acknowledgements or not and what’s enough contribution to make you a co contributor. But I think it’s never very random. Long story to discuss in depth. 

Zoe: Could be. Yeah, I think as someone who’s just collected data, I think probably doesn’t constitute full authorship, but I think anyone who has collected data should be given the opportunity to get involved in the rest of it. If they should have the time and the compensation potentially. And if they don’t, then I think they should be acknowledged. Because of course authorship has great responsibilities as well. So if you’re signing as an author, you have to remember that this person also takes responsibility for the integrity of the whole manuscript. So if this person has just done a little bit of data collection, they might not actually want that huge responsibility. So. They should be given the option. As I say, I think to get involved at any level and whatever they choose as their level of involvement should be acknowledged in the appropriate way, be that authorship or acknowledgment. 

Jo: Yeah, from that perspective, it makes a lot of sense. Again, I also felt like at some point when I was invited for co-authorship as an undergraduate student, I felt both honored but also overwhelmed. Like, what? But I only did so little. And how is that even, like, intellectual contribution to the paper? But I think it’s also part of the process in the academic career trajectory and growing into becoming a scholarly author and contributor. Cool. Great. Thank you so much for sharing all these insights from the Lancet Global Health team. 

Is there anything that you would still like to share with the audience, with the listeners? Any concluding remarks or something that came to mind which we haven’t still talked about and you think belongs to this topic? 

Zoe: I think we could probably talk for hours about this topic and it’s things that radiate off from that topic. But no, I think I’ve covered everything I wanted to talk about in this podcast. Thank you. And I would just say if anyone is listening and wants to contribute, give us feedback on our program, then I’d be really open to that. Because our projects are ongoing, we’re still polishing it and it would be good to hear what people think of it. 

Jo: Yeah. I’ve also learned a few new aspects of what to consider when it comes to translations.

Yeah, there’s certainly more to explore and to learn from each other as we tap into the aspects of digitally facilitating translational works in scholarly results and presentation there and grow the multilingual scholarly community. 

Zoe: Absolutely.

Jo: Thanks again and yeah, welcome again where we can deep dive on any other of the other topics that we’ve briefly touched upon. There’s plenty more to talk about indeed. 

Zoe: Thanks very much for having me.