טכניון מכון טכנולוגי לישראל
הטכניון מכון טכנולוגי לישראל - בית הספר ללימודי מוסמכים  
M.Sc Thesis
M.Sc StudentBerlin Elena
SubjectLanguage Disenfranchisement and the Cost of Translation: The
Case of the Expanding European Union
DepartmentDepartment of Industrial Engineering and Management
Supervisors Dr. Michael Ben Gad
Dr. Benjamin Bental
Full Thesis textFull thesis text - English Version


Abstract

Over the past 35 years economics has also been used to study a variety of languages problems.

The EU is founded on the principle of ‘unity in diversity’: diversity of cultures - and languages. Nonetheless, a large number of official languages imposes substantial financial, communication and legal burdens on the EU.

Ginsburgh, Ortuno-Ortin and Weber consider a model of a society where individuals are distinguished on the basis of their language characteristics, and this society has to select a set of languages to be used for official proposes. Researchers have used dichotomous and Dyen methods to determine the optimal sets of official languages for the EU15 (i.e. a union including 15 countries). Results suggest that no matter what index is chosen, the best solution contains three languages: English, French and German.

The goals of our research were to extend Ginsburgh, Ortuno-Ortin, Weber work to the EU29 (i.e. a union including 27 countries plus Turkey and Croatia), and further - to a potential EU35 (i.e. a union including 35 countries) and then to extend  the basic model by introducing language distances as an argument in  the cost function. Using data about population and native language of each country of EU35 (i.e. a union including 35 countries) and survey on language proficiency for the EU29 (i.e. a union including 27 countries plus Turkey and Croatia) that was conducted by the INRA  we've tested our major hypothesis that the number of languages chosen for the optimal set will  not be always monotonic in the degree of disenfranchisement. 

It is important to note that despite on the public opinion that English is the most common language in EU it is appeared in the first best choice only for some data sets in Dyen and dichotomous case.