The Common European Framework of Reference for Languages (now CEFR) (2001) has defined plurilingualism as the multifaceted practice of an individual’s languages. The linguistic imbalance means that each competency is not well placed in different closed drawers, but that competencies are decomposed and they “interact” with each other. Moreover, with the publication of the CEFR, a great stereotype has broken down, although it still persists in some: the fall of the myth of perfect bilingualism. The imbalance in skills does not mean the fall or instability of the person, but on the contrary, it is a matter of seeing this plural as an additional wealth for the identity of the individual (CEFR, 2001). However, an obscure shadow follows and glue to the skin of this metaphor of the imbalance and prevents it from truly developing in language learning. The imbalance then fluctuates between wealth, thanks and by the CEFR, and disability, through its purely negative disnotation. How does this wealth or disability translate into a bilingual context? The Valle d’Aosta [now VDA] is a region in northern Italy characterised by its Franco-Italian institutional bilingualism. The school system provides bi-/multilingual education throughout school, from nursery school to upper secondary school (the upper secondary school). Since 1998, with the reform of the Esame di Stato (the baccalaureate) in Italy, the French tests have been redesigned in VDA and the fourth test (quarta prova) [now 4th test] in French has been introduced in order to offer a test to the Esame di Stato included in this continuum. However, this territory is currently divided between the ideology of bilingualism as a perfect addition of two languages and the reality that this perfect bilingualism does not exist and that all stakeholders would benefit from designing an unbalanced plurilingual competence in the context of multilingual education. The ideological aspiration of perfection has therefore had an impact on the education system. The 4th French VDA test has a very high taxonomic level. The French test is the certified copy of the Italian test, which is, for its part, national; but this calm raises many doubts. The questions on which the investigation was based were therefore as follows: 1. given that the system gives them the same tests, do the learners have an unbalanced or balanced perception of their skills in French and Italian? 2. does the VDA present a situation of perfect bilingualism or unbalanced plurilinguism? 3. between entering and leaving the secondary school, in French and Italian, is there an increase, stability or decrease of SEP? These questions show that it is not an assessment of the level of skills or success, but of the perception of skills. There would have been no need for an assessment of skills, as research in the territory clearly shows asymmetric skills. Nevertheless, the linguistic ideology in VDA and the implicit behind a symmetrical assessment at the Esame di Stato may suggest that learners truly perceive symmetry in their skills. Consequently, there would be no concern to sit symmetrical tests between the two languages. The aim of this Communication is to present the results of an empiric-deductive survey considered qualitative, by its interviewing and exploratory nature (Blanchet, 2000) on skills perceptions. It was carried out in six schools with very heterogeneous training courses, i.e. out of 346 pupils. The assessment of skills perception in Italian and French was carried out in the first and last year of secondary school. General results and some variables will be presented in order to highlight this VDA paradox and to show that strong skills, such as oral understanding, are not valued in the certification assessment of secondary education, while lower skills, such as written production, are highly evaluated and penalised.