Sobering FactsSome unseen and rarely spoken-about realities of the translation industry
As a result of content editing and consolidation, you might have been redirected to this page from one or more other pages.
(July 16, 2022, updated October 19, 2023)
This article contains facts which might be disturbing to some readers. If it does not, that is truly disturbing.
We have a FAQ (frequent asked questions) page that answers questions that have actually been asked of us. Here are some additional sobering facts about the translation "industry."
These facts are known to most translators and all operators of translation companies, including true translation companies and mere brokers. They might come as a shock, however, to those outside the translation tent.
Most Japanese-to-English translation (and particularly translation sold outside of Japan) is sold by companies that have no translation capability themselves.
These entities, positioning themselves as translation companies, are best characterized as translation brokers and are in the business of purchasing and reselling translations.
Translation brokers almost never have any significant number of translators and sometimes no translators in their employ.
This is just the way it is with most translation companies other than small, specialized operations, rendering almost any larger company a translation broker rather than a true translation company. The brokers purchase translations from translators or from other brokers and resell them to clients, which sometimes apparently think that the broker has had a material role in the execution of the translations they sell. Such actual participation in the translation process is rare.
Translation brokers claiming to have many thousands of "vetted" translators are simply lying.
It might be comforting to believe such claims, but the comfort is accompanied by the risks of ignoring common sense. Translation brokers do not have the ability to vet thousands of translators. Since they almost never have Japanese-to-English translators in their employ, they would need to outsource even the vetting task. They are highly unlikely to have vetted (or even to "have") thousands of Japanese-to-English translators.
What these claims mean at best (and even that "best" is rare) is that they know the email addresses of that many translators and perhaps have gotten them to agree to work for a rate that makes their business model succeed. At "worst" (which is probably more common) the claims are simply lies.
Japanese-to-English translations ordered from major translation brokers in the US will these days very often be done in China.
This is an indisputable fact that we have verified numerous times. In depositions, we often see discovery document that have clearly been translated from Japanese-to-English by translators in China. They are quickly identifiable as such, even though the translation broker often hides or falsifies the name of the translator. There are some characteristic errors, Shibboleths if you will, that reveal the origin, and sometimes (but it is rare) the broker just discloses the name of the translator in China.
Accuracy certification letters provided by major translation brokers are often meaningless and sometimes fraudulent.
They are often meaningless because they are very commonly signed by someone who is not the translator of the document and often by someone who is not even a translator and does not have the ability to judge the quality of a translation. We have verified this to be true.
Part of the reason for the above-noted signing by an unrelated person is that translation brokers, which usually do not add value to translations but merely broker (purchase/resell) them, cannot afford to disclose the identity of the actual translator to the client.
With regard to fraud, in doing deposition interpreting we have on more than one occasion seen a translated document presented to a deponent that had a certification letter with the name of a translator who had nothing to do with the translation. We know this because we were able to verify the fraud during the deposition by immediately emailing the purported translator while we were conducting the deposition and verifying that he had nothing to do with the translation. That translation was sold by a major US translation broker known for selling translations to law firms. The translator indicated ("They're doing it again") that the agency involved was a repeated offender, having used his name without permission in the past.
The name of the translator who created a translation you have purchased is often not known or knowable even to the translation broker who sold you the translation.
Even if the translation broker does not fear the client learning the name of the translator, if they are following the common business model that can include subcontracting to yet another broker, the broker might themselves be unable to learn the name of the translator. And if the sub-broker is in China, something very common these days, there are potential problems that extend beyond quality and accountability.
Official quality certifications of translation brokers are by their very nature not an assurance of quality.
There are several types of QA certifications, such as offered by ISO, that translation brokers sometimes boast of. As far as we can tell, none of these go beyond recognizing that the translation broker has reported to the certifying body what their administrative procedures are for translations they sell and reported that they follow their procedure. They have checked the correct boxes on the application form. The qualifications do not specify substantive metrics for translation quality, and these qualifications give assurances of nothing more than that the broker knew how to fill out the application forms correctly.
There is an ongoing shift by translation agencies from using cheap Japanese-to-English translators in places like China to using even cheaper machine translation inhouse and then having it repaired by a human. Many agencies are now boldly offerring machine translation as a service, with varying degrees of post-editing. The problem is that the translation might be only as good as the post-editor, and people capable of doing good post-editing are often not required to resort to post-editing work, since they can translate. The entire machine translation post-editing process is opaque, placing a burden of due diligence on the client in selecting a vendor. That burden is unbearable by some clients, and the results can be much less than optimal. Additionally, some translation brokers are using machine translation for hte rough translations without disclosing it, in the hope that they will not be found out. Care is required.