Is the well trained musical ear:

a. sensitive to tones of a definite frequency OR

b. sensitive to pitch zones around the given sound

The question of pureness in violin playing does not seem so straighforward.  First there is the problem of non-diatonic intonation which means that C sharp and D Flat are not enharmonic: sharp tones sound higher, flat ones lower according to their direction and we have to produce them accordingly.  The other thing to consider is that the accuracy of a given note is likewise modified by the place it occupies in the context of a chord.  So (b) must be the answer. 

If you listen to a recording from 3 different players, is it possible that the difference in intonation between the 3 is so great that it is impossible to include them into a common system?  I would say yes. 

So "pure intonation" seems to be defined by several conditions:  the place of the given note in the context of melody and harmony as well as the function is occupies in this place.  But even more important is the zonal nature of the performers hearing.  This allows you to deviate (distinctive voice of the artist) so that the same note may be accurate in various senses.