In the second of a three-volume modular approach, Plantinga continues his investigation into one of the most fundamental questions of epistemology since Plato: what precisely is a warrant? Before something is held to be true, it must be justified or warranted, and yet the contours of a warrant are not fully clear. Plantinga explores the basic parameters. Cognition must be functioning properly (and functioning for the cognitive environment for which it was de-signed), the original plan of design must a good one, its relevant modules aimed at truth, and the objective probability of a belief being true (depending on all these underlying factors) must be high. In rejecting an internalist and naturalist view, Plantinga turns to theism. If warrants are to be understood, they need to be found in a naturalistic epistemology that is set in the context of supernatural theism.

KEY WORDS: Epistemology, Warrant, Theism, Internalism, Foundationalism, Naturalism, Descartes, Perception, Purpose, Knowledge: self, way of memory, of others, Belief, Reliability

For Plantinga, Plato’s analysis of the contours of knowledge ‘set the agenda for Western epistemology’. As the title intimates, the author’s focalization is on one aspect of knowledge which differentiates it from belief: this quality or quantity is identified as warrant. Warrant is that which distinguishes knowledge from mere belief. Before something is held to be true, it must be justified or warranted. Unfortunately, warrant is an elusive creature. Attempts have failed to define what analysis precisely is and warrant is far more complex. Plantinga hears the reliabilist siren song. Ordinarily, one assumes cognitive faculties are functioning properly, that is, they are reliable; they do produce and would produce truth under normal conditions. The author, though, recognizes a fallacy.

Plantinga thus begins his search by examining the need for epistemic faculties that function properly and how, nevertheless, cognitive systems can go astray. Even when in order, the resulting belief may still lack warrant. What is then needed is a design plan. A comprehensive development of design is the dominant aspect of his text. Plantinga lays out the basic parameters around a warrant. Cognition must be functioning properly (and functioning for the cognitive environment for which it was designed), the original plan of design must be a good one, its relevant modules aimed at truth, and the objective probability of a belief being true (depending on all these underlying factors) must be high. Only then can there be a discussion of reliability.

Foremost, the system must be designed to have a proper function, free of malfunction. Normality, in the nonstatistical sense, is the foundation. But the design plan must be such ‘that our faculties are highly responsive to circumstances’. The design must naturally be stable but also able to change. This means it is a reactional, anticipative one that troubleshoots and repairs in the event of a malfunction. Even the ability to admit mistakes and correct them presupposes a system designed to allow introspection. All this applies analogously rather than univocally. This is how Plantinga distinguishes between what he calls the design plan and the responsive maximum plan. Perceptual experience plays a key role in responding effectively. As Plantinga says, perception roughly constitutes knowledge if certain conditions are met. ‘Perceptual beliefs taken in the basic way, have or may have very high degrees of warrant.’ The author is not convinced. He believes that belief, truth and justification are not sufficient for knowledge: ‘[I]nternalist accounts of warrants are fundamentally wanting.’ Plantinga finds the answer in purpose.

Throughout Plantinga’s philosophical analysis, there is a small voice that only is more fully heard at the end of the book: theism. Caught in a Gordian philosophical entanglement, the author resorts to deus ex machina to find a solution.

From a theistic perspective, ‘God set out to create rational creatures with that astonishingly subtle and articulate battery of cognitive faculties.’ More specifically to the discussion, Plantinga is convinced that in this natural world there is knowledge; there are warrants which are used by natural organisms which have systems that are designed well and function properly. Such an organism can make propositions that stand in causal relations. A purely naturalistic explanation, though, cannot capture the nature of proper function. Naturalism is not false but merely irrational. If there is to be doubt, then there can be deception. The idea that ‘there are natural phenomena of which we can gain proper understanding only by the way of such notions as purpose and function . . . is a felicitous fiction’. Indeed, the way to be a naturalist in epistemology is to be a supernaturalist in ontology.

The traditional theist, on the other hand, has a set of stable beliefs. ‘He has no corresponding reason for doubting that it is a purpose of our cognitive systems to produce true beliefs.’ Descartes can’t be followed, as he started from a condition of doubt. Doubt is the first step on the slippery slope. Doubt about the efficacy of cognition leads to a suspicion of truth, leading to a doubt of God’s veracity. The theist has ‘nothing impelling him in the direction of such scepticism in the first place, no elements of his noetic system points in that direction.’ Furthermore, ‘there is a rejection of the belief that our cognitive faculties have the apprehension of truth as their purpose and for the most part fulfil that purpose.’ Plantinga appears to hold to a view popular among medieval philosophers: ‘[P]ropositions are divine thoughts, properties divine concepts, and sets divine collections.’ In this way, the causal objection to a priori knowledge ‘can be easily sidestepped’. A naturalistic explanation is a good beginning, but to function most effectively, an account of warrant needs to found in a naturalistic epistemology that is set in the context of supernatural theism./div>