My favourite way of learning languages is by reading – and no, you are not surprised.
And of course it may take some time before one is proficient enough to really read – but I find that, once I have the hang of basic grammar, one great way to reach that level of proficiency is what I like to call the Rosetta Method: reading the same text in a language I know, and then in the language I’m learning. I love the way it makes me see the language in action, while at the same time forcing me to work out the different cogs-and-wheels of different languages…
Of course, allowances must be made for translation, and perhaps it is better to begin with non-literary texts, where less freedom is used. Travel guides, for instance, or museum websites, or even the less strictly technical parts of user’s manuals… Once you begin to look around, you’ll find a vast choice.
Personally, I have a penchant for opera programmes. Back when I was a young and avid opera goer, I used to devote intermissions and long train rides to the comparative reading of opera synopses and director’s notes, in an effort to better my German and Spanish. I must say, it rather worked. In the end I dropped my study of German when I no longer needed it, because I never liked the language to begin with – but with Spanish, I was soon enough able to upgrade to historical fiction and non-fiction, so… a success.
Again, allowances must be made for translation. As a rule of thumb, the more important the theatre/museum/institution/company, the better the chances of a good translations – but of course it ain’t necessarily so, and… well, things happen.
Once, on my way back from Rome, having run out of German and Spanish translations, I went on to read the synopsis of Verdi’s Don Carlo in English and French. And I found that whoever had done the two translations (the same person, I assume) had got their verbs mixed up somewhere between to bless and blesser – French for to wound. Their confusion resulted in having Rodrigo di Posa blessed to death. Considering that poor Rodrigo’s murder is ordered by the Holy Inquisition, and in that particular production the actual murder was carried off by a robed and hooded monk, the translation ended up being slightly Monty-Pythonesque. I confess that I burst laughing on the crowded train, and have never been able to look at the murder of Posa quite in the same way again.
All this to show the inherent weakness of the Rosetta Method – but I still say that, when used with some little caution, it is most definitely worth a try.