That's an excellent question.
No one is claiming that tracking applications are a panacea or a silver bullet, neither the privacy community nor public health authorities. However, it is quite conceivable that tracking applications could contribute to the solution and help identify people who have COVID-19. More traditional methods could then be used to contact them and provide them with advice on how to stay home.
Applications will therefore not necessarily be the solution as such, but they can contribute to it. This is a context where we are facing a completely new phenomenon.
Privacy principles say that you have to be able to demonstrate the effectiveness of the measure, but that does not mean that you have to demonstrate absolutely that the measure is going to be absolutely effective. This is perhaps, once again, where we could have a flexible and contextual approach.
What we're looking for is a scientific basis. We're not trying to prove the effectiveness of the measure beyond doubt. It's about demonstrating that it can be effective with other measures.