I see a lot of competitions going on hiring translators and interpreters for the European instutions and that makes me think about the issue.
The E.U. administration should provide a table with two cells: how much does the European administration spend on translating and interpreting, and how much does it spend on the research and administrative work.
I'm pretty sure at some point in the near future the language related expenses will surpass the real-work expenses.
It seems obvious, to me, at least, that a true E.U. community cannot exist without a common language.
Those who object to the idea of a common E.U. language will bring the argument that such a choice will produce a uniform E.U. , and I would reply that there should have been concerns already when the global businesses were sweeping the local flavors of everything, it's already too late to object to this blanketing. I can also add that international scientific institutes are using a common language and you don't find many people there thinking uniformly in scientific matters. I invite the ones who still disagree to sell their car, end their telephone, TV and internet connections, and never buy a plane ticket again: then we'll have plenty of diversity, a diversity which will happen to be ignored, or be visited only over the weekends.
So we should use a common language in the E.U. schools .
Now comes the more difficult part: which language? Each E.U. country with a population larger than 30 millions hopes it will be their language to be chosen as the common one. In this matter we should forget about the concept of nation which, I agree, was useful to start wars and make people kill around in its name, but now it's a brake in communication.
Pragmatic reasons should be at the base of such a choice: how many people are talking that and that language? Chinese, hindi, english, spanish are the first four languages in use today globally. The most popular language used on Internet is english for obvious reasons but that may change with the numbers of PCs connected in the non-english speaking countries, so an argument using the "main language on the internet" statistic is weak.
As europeans, it will be easier for the new generations to commonly learn one of the old european languages (english, spanish, french, german, italian) or to learn an entirely new one, like esperanto. The global use of these languages select english and spanish and discard french, german and italian.
So E.U. is left with three options for the common language: english, spanish or esperanto.
Esperanto might be more difficult to put into effective use, but if taught early in schools, as a comoon E.U. alternative language, it might win in the long term. This way, the national pride of any E.U. country stands unaffected. But preserving the national pride of all the E.U. countries might kill the intention of having a common language: the peoples will still use their own language, and use esperanto only for contact with E.U. foreigners, which will end up in the situation we are today: the majority of germans, french, spaniards barely know a foreign language (e.g. english), even if taught in schools. One can feel that on the street.
So esperanto might be a failure because it's fresh new for the most. It only relies on early schools and local politics to be promoted to a successful use. That might be costly, although not as much as the current translating/interpreting practices in the E.U. administration.
So let's return to english and spanish. It's easier and cheaper for either of them to become the common E.U. language. The adoption of english might seems advantageous today because its the "language of science" and "language of the internet". But none of these two are real arguments: 1. the few global publishers of today who adopted english language as a standard might dissapear sooner than you think and the scientific publishing might get distributed sooner than I think. 2. the language of the Internet goes with the users of Internet, and it might be possible that the majority of these users to become spanish or portuguese, not english, globally, as their infrastructure grows.
There is also a cultural argument against english: the habits of the native english language users are associated with the english language in a fuzzy way. As these habits resemble arrogance more than human compassion, the english might lose sympathy if nothing in global politics changes.
E.U. can wait until english installs as a de facto common language, but this waiting costs a lot. And the result of a natural evolution might still be spanish.
So we're left with spanish. The advantages of promoting it as a common main or second language for all the E.U. countries might bring some fun with the novelty itself, and the fact that many E.U. languages have a lot of latin in them can only help. This way, E.U. might also gain a feeling of community identity (as opposed to the dangers of being confused with U.S. in terms of style).
The conclusion: start teaching spanish in all the primary schools in E.U. as the main national language and start using it in the public administration, teach english as the second language and keep the rest for the arts/history classes. Make some room for chinese, hindi or bengali classes.