In Getty Gettier Gettiest, I wrote about how the Gettier problems that I have seen don't present a problem to the notion of justified true belief, because they all involve fooling the subject into false believing something specific, having that subject infer something general from the false belief and then having the general inference proved to be true.
I want to expand a little in this article, having discussed the issue over at CraigLand in terms of "warrant" (more of "warrant" a bit later).
It's generally understood that to have knowledge, one must have justified true belief. In other words, in order to know something, you must have sufficiently good reasons for believing something that is true. What is built into this model, but apparently overlooked, is the idea that you can have good reasons for believing something that is not true. So, you can have justified false belief. Let's revisit the scenario of the faux-barns as an explanation (slightly complicated per Goldman 1976, as taken from EIP).
The fake barns (Goldman 1976). Henry is driving in the countryside, looking at objects in fields. He sees what looks exactly like a barn. Accordingly, he thinks that he is seeing a barn. Now, that is indeed what he is doing. But what he does not realize is that the neighborhood contains many fake barns - mere barn facades that look like real barns when viewed from the road. And if he had been looking at one of them, he would have been deceived into believing that he was seeing a barn. Luckily, he was not doing this. Consequently, his belief is justified and true.
So what we have here is an observer, Henry, whose eyes have received a configuration of photons which is consistent with having seen a barn. The problem is that he could have received a functionally identical configuration of photons without having seen a barn, if he had been looking at a fake barn, rather than a real one.
So the question is, given that he could just as easily as been deceived as having been correctly informed about a barn in the field, does this constitute knowledge? My instinct tells me that it does. Henry has a good reason to believe that there's a barn there, there is a barn there and he believes that there's a barn there.
Those arguing that there is a problem will raise a few other related scenarios.
First, Henry looks into another field and receives a configuration of photons that is consistent with there being a barn in that field and as a consequence believes that there is a barn in the field. However, it's not a barn, it's a faux-barn. I don't think that this introduces a problem. Henry merely has a justified false belief; he's made a completely understandable error and this does not constitute knowledge.
Second, say that when looking into the first field, Henry is actually aware that in the region it is reasonably common for fields to have fake barns in them. He sees what looks like a barn, it is actually a barn, but now he does not know that there is a barn there. This seems strange. He's been given more information about the situation but the apparent effect of that is to make less knowledge available to him. The reduction of available knowledge, however, is not real - the total knowledge has increased but is differently distributed. Henry now has knowledge about other things that look like barns in fields and as a consequence of having that knowledge, he now will not form a belief that what he sees is necessarily a barn. Instead, he's forced to believe something more general, that what he is seeing is something that looks like a barn - which constitutes knowledge because it's a justified true belief. However, this more generalised knowledge will apply when he looks in other fields. Note that without this information about fake barns, Henry would otherwise have held false beliefs about barns when he saw faux-barns and thus had no knowledge at all.
Third, say that Henry is looking in a third field, not knowing about fake barns despite looking right at one. Hidden behind that barn, and thus not seen by him, is a real barn. This is a scenario that I covered in Getty Gettier Gettiest and, per that article, my position is that Henry is being fooled into believing that there is a barn in the field. He might actually be right in the general claim that there is "a barn" in the field, but believing that to be so is not justified by the information that is available to him. He could know that he is looking at something that looks like a barn. If he could justify believing that he is looking at a fake barn, then he could know that there is a fake barn in the field. But there is no relevant justification for him to believe that there is an actual barn in the field. The fact that there actually is a barn in the field is only "serendipitously" true.
Now, in Getty Gettier Gettiest, I raised the issue of sliding between specifics and generalities. Henry saw what he thought to be a specific barn and in order for there to be a problem with justified true belief, it would have to be valid to make an inference from a specific false belief to a general true belief. That's not logically sound.
At CraigLand, that was not precisely the approach that I took. Instead I pointed out that for justified true belief to constitute knowledge, the justification involved must be relevant and that the belief cannot be merely serendipitously true. In other words, the truth of the belief cannot be decoupled from the justification for the belief.
While the invalidity of inferring a general true belief from a specific false belief covers all the Gettier problems that I am currently aware of, I think that relevance of justification and absence of serendipity may be key in addressing a more generalised Gettier-like challenge to the notion of justified true belief as a basis for knowledge.
Now, I did say that there would be more on "warrant". In the discussion I was having at CraigLand, the definition of warrant was:
the quality or property that turns mere true belief into knowledge
This definition is very close to how Plantinga introduces it in Warranted Christian Belief (I have elsewhere challenged what Plantinga thinks warrant is (as opposed to what it is defined as)).
Think about it. Warrant is being considered almost like an alchemical agent. Not only that, it would appear that "warrant" only pertains to true beliefs. Could you have a warranted false belief, if "warrant" is the "the quality or property that turns mere true belief into knowledge"? Plantinga's theory is a dual plus theory of knowledge, masquerading as a tripartite theory of knowledge.
And, even worse, it's basically a theological approach to knowledge as one of the WLC fans at CraigLand wrote "I would say it has to be an actual interface with God". Plantinga himself writes about how warrant is related to some sort of "design plan" - a cognitive system "working the way it ought to" (ie as designed).
Anyway, part of the motivation to look towards something like warrant is because some people think that the justified true belief theory of knowledge is somehow threatened by the Gettier problem. Once it's realised that the Gettier problem doesn't affect justified true belief theory and that, at best, warrant is just a limited form of (relevant) justification for (non-serendiptiously) true belief, the justification for even considering "warrant" quickly dissipates.