Risk Analysis in Foreign Language Litigation: Choosing the Right Vendor
Author: Remu Ogaki
In the context of foreign language eDiscovery, I was once directly asked a question—why would I choose anyone but the cheapest vendor? This was a question I loved, because it gets at the heart of a central issue that many people are afraid to ask. Or worse, don’t think to question.
Why does expertise matter in the first place? Can’t I just use an AI plugin to translate everything and hire an English language team of contractors that would be significantly less expensive?
The framework I like to use to discuss this is simple: risk. How much risk does it make sense for your client to take on?
At the simplest level, risk value1
(Probability of Occurrence) x (Consequence) = Risk Value
For example, lets say someone wants to skiing but there’s a chance you will break your leg.
(10% chance of breaking leg) x ($3000 hospital bill) = $300 Risk Value
9 out of 10 times, you may get through without having suffered any harm, but the one time the bad event “hits,” you get charged with $3000, so your average risk value is $300.
Take it a step forward, if you could pay to eliminate or reduce that risk for less money than it costs to do so (a safety harness that costs $100 for example), it’s worth it. $100 eliminates $300 of harm on average.
So, how does this all relate to foreign language eDiscovery?
The biggest risk litigators take in simply going with the cheapest foreign language review vendor, or relying on an AI translation tool instead of a team of experts is the risk of missing
important evidence, or of ceding legally significant points of dispute without even realizing they exist.
This goes beyond simply the issue of accuracy in translation—if accuracy is
A few real life examples I’ve encountered in my career.
During an FCPA (international bribery) investigation, we were asked to look into whether any sales people were taking government officials to dinner and inflating the number of dinner participants to skirt rules on maximum allowable per-person costs. While reviewing receipts, we noticed that some of the restaurants listed simply did not sound like restaurants. A few google searches later, we realized we had a receipt from a brothel.
Another example was where nigiri, a term for sushi, was used in an antitrust conspiracy as a code-word for “rigged bidding” due to its similarity to golf-slang in Japanese for making a bet on the golf course.
Yet another: in a case alleging false public statements by corporate executives during an SEC investigation, an executive we were representing told a subordinate to make amendments on 表 (hyou). The problem was that Hyou can mean “table” as in a table of numbers. One legitimate and perfectly accurate translation would be “make changes to the table of figures.”
However, Hyou can also and just as legitimately be translated to “on the surface.” An equally valid translation was “make changes to the numbers on the surface”—i.e. a direct order to falsify the public-facing numbers.
Which translation was more accurate was unclear in context, thus either could be chosen and defended as an accurate translation. We recommended utilizing the translation that did not incriminate our client.
These are instances where cultural knowledge proved essential to identifying critical evidence, and they are not uncommon. Even an accurate translation of these events could produce significant legal risks, because accuracy was not where legal risks emerged.
It was in understanding the legal importance of these linguistic and translation quirks that identified these possibilities as important.
Which circles back to the issue of risk.
When selecting an eDiscovery vendor, particularly in foreign languages and cultures, it is important to be aware of what the risk value or risk exposure the client is being subjected to in vendor selection.
(Probability of Occurrence) x (Consequence) = Risk Value
For consequence, what is the monetary value of what is at stake for the company? What are the potential damages?
Probability of occurrence is harder to determine. If the legal issue is factually extremely simple with little chance for ambiguity, the probability may be lower. The more factually complex the issue, or the more incentive for the evidence to be intentionally obscured or hidden, the greater the risks of material harm to the client by utilizing low-cost solutions.
Even if the chances a high quality vendor discovering a material issue that the low-quality vendor does not is only a 10% increase, if the potential damages to the client are $5M:
10% x $5,000,000 = $500,000
This is the risk value benefit that the high-quality vendor would bring to the client. In other words, if the price difference between the high-quality vendor and low-quality vendor is less than $500k, the switch would be well worth it to the client.
For these reasons, when assisting a client in selecting an eDiscovery vendor in a foreign language case, a litigator should approach the issue holistically.
- How much is at stake for the client?
- How complex are the underlying factual issues?
- Is there any reason for obscuring or intentionally hiding evidence?
Instead of simply selecting the lowest cost option for vendors, this risk analysis should play a role in determining what type and quality of vendor best serves the interests of the client.
A machine translation of the evidence, reviewed by a minimally managed team of low-cost English language contractors will almost always provide the lowest cost solution to eDiscovery. It is important to understand the tradeoffs in litigation risk such a solution presents to the client and to make an informed decision.
About the Author
Remu Ogaki,
Counsel; Foreign Language Review Services
Remu has well over a decade of experience managing large teams of foreign language attorneys, paralegals and translators in the US and Japan. Follow him on LinkedIn here.
