For Mīmāṣakas, a non-defeated belief counts as knowledge as long as the opposite is proven. This means that according to Mīmāṃsakas, for the Veda, the absence of defeating conditions is in itself equivalent to its truth. 
This, however, does not amount to its truth from the point of view of a theory which considers only justified true belief as knowledge. Incidentally, the Mīmāṃsā’s refusal to distinguish between justified belief and knowledge offers a way out of a difficulty found in every account of linguistic communication as an instrument of knowledge, i.e. the problem of how we can understand false utterances (see Chakrabarti 1986, Matilal 1990:61-8, Mohanty 1992:253-5, Ganeri 1999:18-25). Roughly, the problem lies in how we can understand that there is a snake in the next room after hearing the sentence “there is a snake in the next room” although there is no snake in the next room. Linguistic communication is an instrument of knowledge, but the belief that there is a snake in the next room cannot amount to knowledge. How can this content be possibly conveyed? In order to justify that we understand false sentences, Indian theories of linguistic communication as an instrument of knowledge would need a (preceding) status of non-committed awareness of the meaning, claim the authors listed above.
However, this is not needed in the case of Mīmāṃsā. Mīmāṃsakas would describe this situation by saying that our initial knowledge of the presence of a snake in the next room is later defeated as soon as we see that there is no snake there.
Comments and discussions are welcome. Be sure you are making a point and contributing to the discussion.
This is an interesting contrast. From the sabda pramana point of view you have the conveying of information about a possible experience. This is a valid means of knowledge which prepares you and is therefore a rational procedure. It is a generalised conceptual view of the total encounter with a reality. The other view of justified true belief as knowledge comes from the actual experience. This seems an artificially attenuated account which is empiricist and takes each experience one at a time. However, sabda covers the experience as well by allowing that the default assumption of validity (with apta qualifications) can be altered by contrary experience.
Yes, Michael, śabda is always falsifiable by particular experiences proving the opposite. But so is pratyakṣa, which can also be falsified by further sense-experiences. It cannot be invalidated by the Veda only because the Veda does not deal with the same contents.