Sir Andras Schiff's BWV 797 Recording

[serito]

Hey Friends,
I have come across Sir Andras Schiff's Decca Recording of Bach's Two and Three Part Inventions (1985). For me the notable piece is BWV 797, where he deviates from the written music quite often and extensively. The deviations are in bars 19, 21, 44 to 48 and 52. I was wondering if anyone could help me understand if those are acceptable as ornamentations and if it would be okay to copy his version for a recital. And in general why would Schiff change the piece like that?

It is a rare case however where a Gould recording (Sony 1957 live recording) is closer to the written music than Schiff. I recommend a listen to both versions even if you can't answer my question!

[quantum]

Improvised ornamentation is part of baroque performance practice.  Adding to, modifying or even selectively omitting the written ornamentation is appropriate. 

Both the Schiff and Gould recordings you mention are well within the scope of informed performance practice. 

As you build up your experience listening to baroque performance, you will start to notice certain patterns in the music where performers may improvise their own ornamentation.  It is like the difference between learning a language from a textbook, vs learning a language by speaking to locals. 

I would recommend becoming familiar with the Bach table of ornamentation.  In your practice sessions, try to apply these ornamentations to baroque music, in a manner that suits the flow and character of the piece. 

Schiff has many masterclasses available on Youtube.  I suggest you watch a few to gain insight into his approach. 

The least musical thing you could do is to play the score verbatim.  That's not making music, that's reading a score.  Music is not made when a performer reads a score, music is made when a performer uses insight to bring the printed dots on paper to life.