What Happens When a Crowd Translates the Bible?

Crowdsourcing ScripturesIn the past, Bible translation was typically the work of small, closed groups of experts. Hindsight now reveals some significant weaknesses with this approach to translation.

A team of Bible translators would draft Scripture, and then bring the draft to the community. Through rounds of testing, they wanted to find out: Is the translation clear? Does it accurately convey the meaning of the original texts? And does it sound natural?

But because the draft was already semi-complete, community members could only participate in small ways, suggesting incremental changes to the draft here and there. Once the translation was finished, it was handed to the community for their use.

Crowdsource the Translation

But what if the translation process was opened up to full community participation from the start? Simply stated, the notion behind “crowdsourcing” is that a small group of experts can be smart about some things, but they can’t know everything. They also tend to think alike, so their ability to keep learning is rather limited. The group needs more minds.

Wisdom of crowdsCrowdsourcing is what Wikipedia does. When you open a process to a larger community of interested people, you end up with a far greater aggregation of diverse knowledge, insights, experience, and even areas of expertise that exceed the intelligence of the small, closed group of experts.

Diversity is the key: the larger the crowd (the community of participants), the smarter it becomes (see The Wisdom of Crowds).

Astounding Results From CrowdSeed Project

This year, The Seed Company, in collaboration with our partners in South Asia, conducted an experiment to see what would happen when a Bible translation project is opened to full community participation from the start.

The results were astounding. Within months, over 3,000 people participated via their own custom-designed Web site where the translation work resides. About 78 people were confirmed by the community as quality drafters. Over 100,000 votes were cast, answering essentially the same questions: Is the translation clear? Does it accurately convey the meaning of the original texts? And does it sound natural?

All segments of the community participated. Significantly, women and youth were able to participate, adding their perspectives which are typically missing because of cultural constraints. Non-literate people were able to participate because the people chose to work in groups. People from seven regions, across denominational boundaries, worked together with surprising unity and harmony. And most importantly of all, they view the translation work as their own from the very start, and it is already making an impact in their community in ways we could not have guessed.

Next Step: Confirming Accuracy of the Translation

Now, we’re working with experienced consultants to confirm accuracy of the translation. It should be much easier given the rich and intelligent involvement of the crowd.

If you were invited to participate in a new, crowdsourced revision of an English translation, or in whatever language you read, would you join? Why or why not?

Further Reading:

Which Book of the Bible Do You Translate First?
The Seed Company Earns The Mission Exchange’s 2011 eXcelerate Award for Innovation

Crowd photos in images courtesy of Bayhaus and Anirudh Koul on Flikr.

18 Comments to What Happens When a Crowd Translates the Bible?

  1. December 15, 2011 at 10:07 am | Permalink

    I’m a big believer in the power of crowdsourcing provided that the crowd in question has the ability to do the task and that they do not have a bias that impacts the quality of their work. Those Wikipedia pages are best that draw a huge and diverse crowd. Bias can creep in on smaller-audience pages that are more controversial. I’ll be interested to see how this particular project works out. I think this one has some significant potential. It will be interesting to see how the accuracy works. I think this would work for languages where there is a large and diverse and interested Christian community.

  2. December 15, 2011 at 12:30 pm | Permalink

    Thanks Gilles for sharing this. We enjoyed hearing you speak a few weeks ago at WCIU and ever since then “crowd sourcing” has been suggested several times as a way to accomplish just about every task that comes up! Seriously though, I appreciated particularly the emphasis on giving marginalized groups such as women and youths a voice through this process. We look forward to hearing about the end results of this translation!

  3. December 16, 2011 at 7:42 am | Permalink

    Thanks for the update, Gilles! Great to see the high level of interest from the community when they were given “ownership” of the project through their active participation. I expect this will increasingly become the preferred model for translation, for many reasons, including improved quality and efficiency. I look forward to finding out more about the technology you are using to enable the crowd-sourcing, as well as the license you are using to manage the IPR of the contributors. Keep up the good work – we’ll be praying for good success.

  4. December 16, 2011 at 9:55 am | Permalink

    Having lived in this region, this is so fabulous to hear! Additionally, this is very culturally relevant and significant to these people. Yahoo!

  5. Gilles's Gravatar Gilles
    December 16, 2011 at 11:13 am | Permalink

    Thank you all for your great comments. Justin makes a good point about how group bias can creep in and eventually dominate while trying to force a perspective. But what we’ve seen in the Ganbi experiment so far is that ensuring diversity from regions, cultural strata, denominations, educational levels (all of the rich diversity) seems to prevent one group from taking control. We still have much to learn, but we’re very encouraged by what we’ve seen so far.

  6. Jim's Gravatar Jim
    December 17, 2011 at 10:01 am | Permalink

    Sounds interesting except for a few small points. Unless the crowd involved is proficient in Greek, Hebrew, and Aramaic, the original languages of the Bible then the “translation” is not a translation, but simply a paraphrase which would still have the bias of whatever experts did the original translation that they read. Believing as I do that the Bible is God’s inspired word, I guess the Holy Spirit could influence this “crowd” that did the work. I still believe for my study I will read those translations which learned men have spent years of work at, making sure they are translating God’s original intent with the original writers.

  7. Gilles's Gravatar Gilles
    December 17, 2011 at 11:32 am | Permalink

    People like pastors, theologians, and even those good at doing Bible studies still need to be part of the crowd. The notion of crowd sourcing means increasing diversity instead of limiting it, as Bible translation has typically been practiced. These trained people can compare the work against the original languages and inform the crowd on what they find. The members of the crowd grow in understanding and intelligence. That’s what we are testing though this experiment, at least.

    Thanks for your comment Jim!

    • Pat Tiller's Gravatar Pat Tiller
      December 18, 2011 at 7:22 pm | Permalink

      This seems very promising. As I see it, you need linguists, philologists, and historians to provide the scholarly input on which most of the translation work is based. Most of them may not even be part of the “crowd” but would provide source material for the crowd. Then you would need to have people knowledgeable in both the source and target languages who have access to the scholarly source material who would propose initial drafts and provide some quality control. You also need readers, hearers, pastors, and theologians to evaluate the usability and propriety of the proposed translation for the target community and to suggest alternatives. The ability to get them all involved early on seems very positive as long as there are agreed-upon quality control procedures to bring things back when the translation starts to become inaccurate or misleading.
      What I think has been missing (at least in the English world) is the involvement of people like speech-writers, who know how to express ideas accurately while avoiding stilted language on the one hand or bland language on the other. Until that last few hundred years, the Bible was read aloud for an audience as a sort of “performance” far more than it was read by individuals for private study or edification. In my opinion translations of the Bible should also be suitable for public “performance” reading, especially in cultures were literacy is not near 100%. Every culture has people who are able to give or write a stirring speech and who know how to use language not only to impart information but also to stir emotions or move people to action. These kinds of people should be sought out and included in the “crowd.”
      How is it working in your experiment? Are there people who take on (or are assigned) various roles? You say that the next step is to get the consultants involved. Would it have been better to have them actually join the “crowd” and provide more immediate feedback? Or does that inhibit the community’s participation?

  8. Gilles's Gravatar Gilles
    December 19, 2011 at 12:11 pm | Permalink

    Pat, regarding your questions, we intentionally did not begin by instituting too much structure and formal guidance, since, according to crowdsourcing strategies, that usually inhibits wider community participation from the start. The idea is to see how the community builds and grows, and what sorts of expertice is added. Also, we didn’t want to assume certain Western practices were the default way to work. Instead, it’s important to allow the community to determine how they prefer to work together. Ideally, consultants would be part of the community crowd from the start, and that may be the case in subsequent test projects, but the goal is to see community members develop as consultants through this learning process, too. Crowdsourcing is a more interative than linear process.

    Thanks for your comments.

    • Pat Tiller's Gravatar Pat Tiller
      December 20, 2011 at 7:55 pm | Permalink

      OK. I think I made some wrong assumptions. So now I’m guessing that the “crowd” includes people who are familiar with the source language, whatever that might be, as well as the target language, so that they can work together to come up with their own draft. The “expert” review comes later. Is that closer to what you are trying to do?

  9. Gilles's Gravatar Gilles
    December 21, 2011 at 11:37 am | Permalink

    Pat, your description is correct. As I mentioned before, ideally, people skilled in exegesis are also part of the community crowd during the process. They guide and inform the crowd as they work together, learning and ever improving. However, with this experiment, we wanted to see how the community participants would shape things first, so the consultant review comes last.

  10. Karl Dahlfred's Gravatar Karl Dahlfred
    February 24, 2012 at 4:09 pm | Permalink

    I’d never heard of that before. Good way to get more input but I’d be nervous if qualified Bible translators didn’t have final say. But still, wouldn’t those who know Greek & Hebrew need to do initial work & present it to people? Only small number of people could translate from the original, although this would be a good way to get more feedback and thus produce a better translation. In some ways though, crowdsourcing seems like the same as old way but there are just more people giving feedback on the initial draft

  11. Gilles's Gravatar Gilles
    February 24, 2012 at 4:40 pm | Permalink

    Karl, you do understand this correctly, People who know how to reference the original languages to confirm accuracy indeed need to be part of the crowd. The idea behind crowdsourcing is “increasing” the types of skills, wisdom, insights, etc., that ideally can produce a better translation. This was not possible in former times, but now that so many people, even in poor and remote places, have access to the Web, it is possible.

    Gilles Gravelle

  12. July 16, 2012 at 6:57 am | Permalink

    A recent article by a professional, non-biblical translator notes that people competent in both the source and target languages have a 70% failure rate when they try to do professional translation. http://thebigwave.it/what-clients-want/why-we-still-need-more-good-translators/

    He goes on to identify other kinds of knowledge that are needed to succeed at translation. Crowdsourcing appears to be a way to bring to bear those other kinds of knowledge.

    Also, many professional, non-biblical translators use Internet forums where they can post issues to other same-pair translators, and/or they use software that can show them how a word or phrase has been translated in the past. Crowdsourcing appears to be another way to accomplish the same need for a translator, or a small team of translators, to get wider input on options.

    Another point to note is that many modern translations into major languages are done with very large teams, as was the KJV. The large teams often include people who specialize in one aspect of the translation such as proofreading or (English) style. They work alongside the Hebrew and Greek scholars. Having a team of 50 or 80 working on every minority language translation is not feasible. It appears that crowdsourcing may be a way of extending the team so that minority languages can benefit from the multiplicity of skills used in modern major-language translations. (Or it might even give the major-language translations a new tool.)

    So, for me, crowdsourcing is new as a technique, but the idea of using many types of skills in translation, including Bible translation, is not new. Rather, it is part of a long tradition.

  13. September 29, 2014 at 9:39 pm | Permalink

    Everything is very open with a really clear description of the issues.
    It was definitely informative. Your website is useful.

    Thanks for sharing!

  14. September 30, 2014 at 8:01 pm | Permalink

    Often, the item is printed with a motivational message that the company
    has been using in their advertising. What you choose
    to give as corporate gifts will depend on who you’re giving
    them to and your industry (average life time value of
    a client not what’s customary in your industry; don’t copy your competitors,
    it’s rarely a good idea). You can get all sorts of gifts like coffee mug, pen,
    pen stand, mobile stand, wall clock and many other things.

  15. October 6, 2014 at 8:00 am | Permalink

    Often, the item is printed with a motivational message that the company
    has been using in their advertising. In every gathering they will go to, they can put on a quite show that he is someone responsible that mean business.
    It also calls attention to the entire staff
    that the employee has been rewarded.

  16. Chuck's Gravatar Chuck
    May 8, 2015 at 6:39 am | Permalink

    Of course, the statement “the larger the crowd (the community of participants), the smarter it becomes,” is not always the truth. Or rather, though they crowd may feel they become smarter, meaning more entrenched in their confident opinion, their opinion may not be objectively true. Hence the need for divine Revelation! The research, experience and disciple of the few may be more accurate than the received assumptions of the crowd, and this is as true in Biblical studies as it is in translation. As helpful as crowd-sourcing may be in certain communities for improving naturalness, presentation, distribution and overall ownership, it is no substitute for the guidance of fully qualified consultants, with experience in the original languages, linguistics, as well as the language and culture of the community itself.

    Of course, a very small percentage of the bibleless people groups will have a crowd of vernacular literate people with sufficient experience in biblical exegesis (including the original languages) within their own community so as to outweigh (or transform) the preconceptions of the majority without careful structuring. How would they have acquired such knowledge without a vernacular translation? If they already have such a high level of knowledge of the Bible, is the translation even needed? In most communities needing original translation, (rather than revisions of existing translations) crowd-sourcing doesn’t sound very viable (or the “crowd” will be so small it could just look like a translation team!)

    However, in those places where crowd-sourcing is viable, perhaps for revisions of older or overly literal translations, I would recommend the model not be Wikipedia, which has aggregated a huge number of pages, but with plenty of inaccuracy and bias and vitriol and hatred among the contributors and editors. (And don’t even think that Christians can’t be just as nasty about their opinions! The crowd-sourcing process itself could become an opportunity for factions to try to out edit each other as often happens on Wikipedia.) Besides Jimmy Wales, there was another founder of Wikipedia, Larry Spranger. He left Wikipedia, stating that it was “broken beyond repair” and cited not only bias and inaccuracy, but an “anti-elitist” attitude. He felt Wikipedia could have been much better had there been a more welcoming attitude toward experts. He has since founded Citizendium, which has no where near the name recognition or number of pages that Wikipedia does, but requires contributors to use their real names and be accountable for their contributions, to submit to structured peer review, and which has recognized experts in the field as editors. This sounds like a better model for using crowd-sourcing for translation work that the unharmonious, unstable and unreliable world of Wikipedia.

    In the noble desire effort to widen involvement, we ought not to discount the importance of experts, even if they come from outside of the community or even the country. Truth is not the product of human consensus, and the accurate transmission of God’s Revelation to humanity ought not to be subject to the rules of either Western democracy or postmodern epistemology. Let’s not do the Jesus Seminar with minority Bible translation.

  1. By on February 22, 2012 at 9:07 am
  2. By on March 19, 2012 at 9:33 am
  3. By on July 3, 2012 at 6:30 am
  4. By on October 24, 2012 at 9:55 am

Leave a Reply

You can use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>