What if someone told you that there was a recent event of global significance that involved people named Meyvezer and Pak’yao. Would you guess that it was related to, say, North Korean nuclear ambitions, or Chinese naval aggression in the Sea of Japan? Or are you clever enough to see that it involved two guys standing in a boxing ring in Las Vegas?
Here is a recent headline from the BBC Russian Language Service:It is frequently amusing (at least to some people) to see what happens when a well-known personal name is “round-tripped” (i.e., transliterated from its native written form to a different writing system, then re-transliterated back to its native script). And that is how Mayweather became Meyvezer and Pacquiao became Pak’yao, once these names were converted first into their equivalent forms in Cyrillic (the writing system for Russian), and then re-converted from Cyrillic back into the Roman alphabet, the native writing system in which they originated.
So what does this have to do with name-screening and name-searching, especially in the context of AML/KYC/CIP requirements? Quite a bit, as it turns out…
At the very heart of any meaningful name-screening operation, there is a process that compares one person’s name (viz., the intending participant in a business transaction) with a list of names (e.g., OFAC/SDN, PEP List). There is a great deal of guidance provided by cognizant governmental authorities concerning the acceptable sources for the name to be screened, but little, if any, comparable information about the names on the lists that must be searched. This lack of symmetry can be the source for significant risk in the name-screening process, because there is no way to know what methods and systems were used to render the Romanized forms of names on the list that originated in non-Roman writing systems, such as Arabic, Russian or Chinese.
For example, a careful review of the Islamic names present in the current OFAC SDN list shows a variety of transliteration styles, evidenced by such things as the inclusion or exclusion of the ‘ayn (usually represented in Romanized name as an apostrophe), the segmentation and spelling of the common Islamic name components عبد and ال, which appear spelled as ABD-AL, ABDEL, ABDUL, ABDUR, sometimes attached to the following name and sometimes apart from it. Despite the helpful inclusion of numerous alternative Romanizations in the OFAC listings in recent times, there is no way to know which of the Romanized name-forms might stem from an “official” identity document, and which is from an informal or popular source.
A further wrinkle is introduced by important differences in the Romanization styles used for Islamic names in the Maghreb (Western) countries of the Middle East, most of which were under French colonial rule at one time or another. This style, developed understandably for use by French speakers, represents the male name حسان (commonly rendered as HASSAN or HASSAN) in the forms HACÈNE or HACENE, and the family-name الزرقاوي (commonly rendered as ZARQAWI) as ZARKAOUI. Both Levant (Eastern) and Maghreb (Western) Romanization styles for Islamic names are found in the current OFAC SDN list.
Back to the pugilists. If it is not possible to know the provenance of the Romanized names on the OFAC SDN list, it is also very difficult indeed to know if the customer name that is being screened, however CIP-compliant the source for that customer name may be, will be matched when it should be. If, by chance, the Filipino name PACQUIAO had been collected from a source that was originally in Russian, it might appear in the list as PAK’YAO, just as MAYWEATHER might instead appear as MEYVEZER. While this situation seems quite unlikely for personages who have spent their entire lives living in the English-speaking world, it is certainly possible for people whose names were originally spoken and written in Arabic, then later brought into the Roman alphabet by a variety of methods.