Before summarizing Warrant: The Current Debate (henceforth WCD), it is helpful to understand, in broad outline, Plantinga’s Warrant trilogy as a whole. In WCD, Plantinga surveys various naturalistic versions of warrant and, by examining scenarios in which the conditions for warrant posited by a given theory of warrant are met but knowledge is still lacking, teases out what the missing ingredients are. In the next volume (Warrant and Proper Function) Plantinga fleshes out his proposed definition of warrant and examines its adequacy by applying it to a baker’s dozen of our cognitive faculties (including memory, perception, and testimony). Along the way he notes that several of the aspects of his version of warrant fit better with theism than with naturalism. In the final volume (Warranted Christian Belief) Plantinga examines the role of warrant in theistic belief in general and Christian belief in particular.
In the introduction to WCD Plantinga lays his cards on the table: he is an externalist in epistemology. All kinds of subtle qualifications to the definitions of internalism and externalism can be found in Plantinga’s work, but the basic idea is that an internalist with respect to warrant is concerned with how things go with an individual ‘downstream from experience’. For example, an internalist may be most concerned not with what is going on in the environment external to the individual, but what is internal and (to some extent) under the control or present to the awareness of the individual. An externalist, as might be expected, places more emphasis on states of affairs which are external to the individual or outside the control or awareness of the individual. (Again, these are imprecise terms. At various places throughout WCD, Plantinga encouragingly notes the ambiguities and the need for multiple examples before a given definition in this area becomes clear). And, of course, there are multiple varieties and subvarieties of both internalism and externalism.
Plantinga notes that many modern versions of interalism exist, but that in the last few decades interest in various types of externalism has been growing. However, he thinks that this is not a discovery but rather a re-discovery.
“We may therefore see present-day externalists as calling us back to our first epistemological love, after a brief and ill-starred fling with the seductive siren of internalism. In this book and its sequels, I hope to heed that call." (p. v of Kindle version)
Plantinga further notes that there is widespread agreement on a couple of things that knowledge presupposes: a belief that is true. There is also widespread agreement that more is required, however. For example, if Plantinga believes that he will win the Nobel Prize in question, and by happenstance that turns out to happen, it does not follow that he knew the belief. Rather, it seems more like a belief that turned out, by accident, to be true. This does seem intuitively correct. But then the question becomes: what property or properties (henceforth warrant) are such that, when added to true belief, yield knowledge?
The approach Plantinga takes is to examine various naturalistic versions of warrant and see if, when their conditions are met, knowledge results. He is especially interested in determining if the conditions of a given model are (a) necessary and/or (b) sufficient for warrant. If knowledge can occur without meeting a model’s conditions, then that model is not necessary for warrant. If the model’s conditions can be met and knowledge not result, then the model is not sufficient. If both come to pass, the model is neither necessary nor sufficient.
Broadly speaking, the models Plantinga examines can be thought of as ranging from internalist to quais-internalist to externalist. As things become more externalist, he finds the models to be closer and closer to satisfactory—although even the fully externalist models he evaluates are less than fully satisfactory, as will be seen.
The first variety of internalism Plantinga tackles is deontologism. The basic idea behind deontologism is that we have what might be called epistemic duties. There are ways we ought to go about forming our opinions. For example, lazily adopting a belief without attempting to truly evaluate evidence for or against it would, on many construals, be a violation of our epistemic duties. So the idea is if one behaves in a dutiful way, one is not to be blamed. To introduce yet another term, one is justified, not blameworthy, has met one’s epistemic duties. The problem, as Plantinga points out, is that it is fully possible that someone be epistemically dutiful—or even epistemically supererogatory acts—and still not obtain knowledge. 
To show that deontologism cannot be the correct account of warrant, Plantinga conducts two thought experiments. First he describes an individual Paul who has a brain lesion. The brain lesion causes Paul, upon hearing a church bell, to form a nearly irresistible impulse to form the belief that he is seeing an orange object. He doesn’t know about this defect in his cognitive faculties, and he is nonculpable in this matter. He is dutiful in forming his beliefs: maybe he is unusually dutiful. But he still forms a false belief upon being presented with a church-bell sound. Hence he does not have knowledge, and he also (by definition) does not possess warrant. So the deontological model is not sufficient for knowledge. Nor, Plantinga argues, is it necessary. Suppose that in another possible world Paul has a brain lesion which causes him to believe (for opaque reasons of their own) our overlords from Alpha Centauri dislike Paul thinking “I see something red”. His brain lesion also causes him to further believe that, upon learning that Paul is thinking “I see something red”, the overlords will punish him by bringing it about that the majority of his other beliefs are false. Given that it is an epistemic duty to hold mostly true beliefs, it becomes Paul’s duty to avoid thinking “I see something red”. But, given human nature, this means that Paul must accomplish the (almost) impossible: upon being ‘appeared to redly’ (that is, when seeing a red bus or a stop sign) Paul must *not* form the belief or thought “I am seeing something red”. However, after a few hours of this, Paul gets tired of the effort. “Heck with this”, he thinks. Thus, he violates his epistemic duty to avoid having mostly false beliefs…and acquiesces in the belief, upon seeing a red bus, that “I see something red”. But in this case Paul has *not* been epistemically dutiful, but has nonetheless *gained* knowledge. So the deontological model is neither sufficient nor necessary for knowledge.
Plantinga then turns to a version of internalism known as coherentism, which basically says that a belief gains warrant by its cohering (not contradicting) other beliefs we hold. Plantinga again demonstrates that the coherentist account of warrant is not sufficient by constructing thought experiments to demonstrate that it is neither sufficient nor necessary for knowledge (and hence inadequate as accounts of warrant). He shows that it is not sufficient for knowledge by describing what he calls the’epistemically inflexible climber’. Suppose Paul (poor Paul!) is standing at the top of a mountain. He sees the sunlight, snow, trees, and hears an eagle in the background. Paul is then exposed to a burst of cosmic radiation. His beliefs are now ‘fixed’ in his mind. His friend helps him climb back down—but Paul is still being appeared to ‘mountainly’. His first-person perspective is mistaken; he thinks he is still on the mountain top. His beliefs, presumably, are coherent. There are no contradictions in his beliefs. Still, he is mistaken: his belief that “I presently hear an eagle” is incorrect, his belief “I am on a mountaintop” is incorrect, and so on. So coherentism is not sufficient for knowledge. Nor, however, is it necessary.
"I am an arboreal expert giving a lecture on trees; among other things I claim that no oak trees grow in the state of Washington. Naturally I believe that I have never seen an oak tree in any part of that state. I suddenly notice you in the audience. seeing you jogs my memory: I seem to remember an occasion on which you and I noticed a particularly luxurious oak flourishing on the campus of Western Washington University...At the moment when it seems to me that I do so remember, the proposition that I have seen an oak tree in Washington has warrant for me, despite the fact that it does not cohere with my noetic structure...the change that is called for, of course, is not that of rejecting or trying to reject the memory belief in question; what is called for is revising the rest of my noetic structure in such a way that it is coherent with the belief in question. Coherentism, therefore, is neither necessary nor sufficient for warrant. [p 83, Kindle version]
After dealing with other versions of coherentism and quasi-internalism, Plantinga turns at last to three versions of reliabilism. Reliabilism, Plantinga thinks, is closer to the necessary and sufficient account of warrant he has been seeking because it is externalist. He claims that reliabilism is externalist because
On externalist views, warrant-making properties are such properties as being produced by a reliable belief producing mechanism, or standing in a causal chain appropriately involving the subject of belief, or standing in probabilistic relation R to certain other relevant propositions; and none of these properties is one to which we have the relevant kind of special access. (pp. 5-6, Kindle version). As an example of how Plantinga demonstrates that reliabilism (in the three versions examined by Plantinga) fails, consider his critique of Dretskian reliabilism. Dretske proposes various ‘principles’ which, he thinks, are necessary and sufficient for warrant (and hence for knowledge). Plantinga, as before, uses thought experiments to demonstrate that Dretskian reliabilism is not.
Consider Dretske’s principle D5:
(D5) K knows that s is F if and only if K believes that s is F and there is a state of affairs is being G such that (1) is being G causes K to believe that s is F and (2) P((s is F)l(r's being G & k)) = 1 and P((s is F)lk) < 1. (p. 195 of Kindle version)
In non-philosopher speak, what D5 says is as follows. First, I can only know X if (a) I believe X (2) there is some state of affairs G that causes me to believe X (3) the probability that X is true is, given G’s obtaining plus my background knowledge, is 1.00 (that is, is certain), and (4) the probability of X being true is, on my background knowledge alone (without G obtaining) is less than one.
Less formally still, I can know X only if something (G) causes me to believe X, and if that something (G) makes X certain when X would not otherwise be (if ~G).
To see what is wrong with this principle, Plantinga presents what he calls the Case of the Epistemically Serendipitous Lesion. Suppose, he says, that Paul (poor Paul!) is suffering from a brain lesion. The brain lesion causes widespread havoc in Paul’s noetic structure. It generates many false beliefs, but also causes the true belief in Paul that “I have a brain tumor”. This satisfies all of the above conditions. Paul (1) believes that he has a brain tumor because (2) of a certain state of affairs (he does indeed have a brain tumor)
(3) the probability of him having a brain tumor given the occurrence that he has a brain tumor (plus his background knowledge) is, quite obviously enough, 1 and (4) the probability of him having a brain tumor on his background knowledge alone, minus G (him having a brain tumor) is quite obviously less than 1. But surely Paul does not know that he has a brain tumor. And again, as with all the other examples, what is lacking is the idea of proper function. Deontologism, coherentism, reliabilism all fall prey to the same lack.
Plantinga finishes the book by summarizing his conclusions and adding a few further conditions which he thinks a necessary and sufficient account of warrant must include. For example, he notes that not only must we consider the notion of proper function, but also the idea of cognitive faculties being ‘aimed at truth’. For example, it might be part of our cognitive faculties that, in certain circumstances, we form (properly) ‘wishful thinking’. Say I get a terrifying cancer diagnosis. Then it might be part of the proper function of one or more of my cognitive faculties to believe (in the teeth of all evidence) that I will survive. Again, it seems as if I do not *know* that I will survive—even if, perhaps, I in fact wind up doing so. He also adds a brief discussion of ‘appropriate cognitive environment’. Our sensory faculties are such that (whether designed by God, evolution, or both) they work fairly well in the conditions here on earth. There is no guarantee that they would continue to so work on Mars, however. And, of course, proper function implies a design plan. And the design plan must be a good one. If we were designed by an inferior engineer (say, one of Hume’s infant or decrepit deities) then the design plan might be so inferior that we would never (or hardly ever) obtain knowledge and therefore would also not have warrant. The stage is now set for an extended discussion and defense of this view of warrant, as well as Plantinga’s contention that theism is a more comfortable fit with this view of warrant than is naturalism.
Apologetics 315 Book Reviewer Latter Day Inkling is a U.S.-based research psychologist for the military. He is especially interested in epistemology and natural theology.
 Consisting of Warrant: The Current Debate, Warrant and Proper Function, and Warranted Christian Belief
 For example, Plantinga even explores at least two different versions of internalism proposed at different times by Roderick Chisholm. Rather than discussing at length all the different subvarieties, I will focus on those which most clearly, in my estimation, highlight problems with the broader type of internalism or externalism to which a specific subvariety belongs.
 Plantinga traces the main lines of deontologism-as well as foundationalism, of which he has much more to say in later volumes—to Locke and Descartes. The modern apostle of deontologism covered in WCD is Roderick Chisholm. It should be noted that although Plantinga criticizes (at length) Chisholm’s account of warrant, his respect for Chisholm’s work is evident.
 It is important to note how deontologism is bound up with an internalist epistemology. We can only be held responsible (it can only be our duty) to perform certain acts if we have access to the relevant facts about those acts. If some event X is beyond my knowledge or control, how could I be held responsible in any way for the fact that X happened?
 I.e., this has occurred through no fault of his own.
 Again, Plantinga’s fair-minded approach is evident. Although he (obviously) ultimately rejects coherentist models of warrant, he also defends them against a particular criticism which is often propounded by individuals with whom Plantinga is, on other grounds, broadly sympathetic. The criticism he rejects involves the charge of circular reasoning. Plantinga thinks, rather, that the coherentist is more fairly construed as seeing coherence as the sole basis for warrant of a belief.
 Another way of putting this point is as follows: something can be coherent but not true; something that is incoherent is necessarily false.
 See footnote 4 for why this is externalist rather than internalist.
 Dretskian reliabilism, per Plantinga, is concerned with the probability of a particular belief. Goldmanian reliabilism, by contrast, is concerned with what we might call the ‘overall hit rate’ of our beliefs: the ratio of true vs. false beliefs generated by our cognitive faculties. But similar problems (plus others) plague the Goldmanian version as well.