Skip to content

A Quest for a Lingua Franca Approach

“Lingua franca is the real medium to express feelings to people around you, but the real feelings that comes from deepest core of heart have no language, but inscrutable countenance.”

-Ankur Srivastava, poet and writer

The lingua franca concept denotes a language that can be used to enable communication and understanding between people who speak different languages, thus serving as a linguistic bridge in connecting different cultures and regions.

Lingua franca, also called common or bridge language, is currently English around the world but other languages (such as Latin and Koine Greek) have been the accepted lingua franca during other historical times. Another strategy similar to that of lingua franca is a multi-lingual formula in which two or more languages are intentionally bundled together so that everyone become increasingly more familiar with all included languages. One can frequently see examples of this approach in the form of multi-lingual signage at international airports (or at least remember these examples from pre-COVID travels).

We need this shared language philosophy (not necessarily a lingua franca but a multiple-language formula) for clinicians and data scientists when they convene together for meetings or share a manuscript. The most important dividend from everyone learning each other’s vocabulary and concepts is a genuine appreciation for the other domain experts and a deeper respect for the other field’s knowledge. This is akin to any Spanish-speaking patient and family appreciating a healthcare worker who is speaking Spanish during a clinic visit.

Interestingly, as difficult as English is for new speakers, Chinese is considerably more challenging to learn. For that reason, even though more people speak Chinese than any other language, it is not likely to be adopted in the near future as a lingua franca for the rest of the world. Similarly, perhaps the sophisticated computer programming with its hundreds of obtuse lines of code or complicated multiple-worded medical jargon (such as sphenopalatine gangioneuralgia for ice cream headache) can be as difficult to learn as Chinese.

Two possible solutions for this communication challenge. First, humans can start an understanding of the respective vocabularies of the other domain beyond what they have heard in the media or on television shows. This can be an excellent reason for spending time in the other domain. Clinicians can learn some of the vocabulary and concepts of AI just as data scientists can try to appreciate more common medical terms and knowledge. For example, while “RPA” is right pulmonary artery for most clinicians, it is robotic process automation for most artificial intelligence-focused stakeholders.

Another solution may lie in natural language processing and its ability to explain unfamiliar terms in real-time just as it is translating languages. With the recent advent of GPT-3 and other advanced and sophisticated NLP transformer type tools, this possibility of lowering the barriers between domains will be even more realistic.

The Healthcare Data Conundrum (Part I)

“It’s the economy, stupid.”  James Carville, Bill Clinton’s strategist for 1992 presidential campaign   The above battle mantra during the presidential campaign of Bill Clinton

Show Buttons
Hide Buttons