Justified True Belief

Mapping the Landscape of Good Reasons for the Truth of Christianity

Christian Evidences

Evidence & Arguments for Christian Theism

Divinity of Christ

Biblical Evidence for High-Christology: Jesus is God

TBD

Old Testament Criticism

Evidence for the Reliability of the OT Bible

Common Objections

Objection Analyses to Christian Theism

World Religions

Critical Analyses of Non-Christian Religions

Philosophy

Phileō Sophia - to Love Wisdom.

Epistemology

Study of Knowledge

Evidentialism

(●) Evidentialism holds that a belief is justified when it fits the believer's evidence. + Ruth comes home from work and finds muddy paw prints running from the back door to the couch. The back door is ajar, and the dog is asleep on the cushions with mud drying on his legs. Ruth does not see him go out, and she does not see him come back. But she believes, without a moment's hesitation, that he got out into the yard and came back in, and anyone standing in her kitchen would say she is right to believe it. Her neighbor, who has heard nothing about any of this, believes the dog spent a quiet afternoon indoors. He has no reason to think so. He simply likes to think well of the dog. Both of them hold a belief about the dog, and only one of them is being reasonable. The difference between them has a name: Ruth's belief fits her evidence, and her neighbor's belief fits nothing at all. - Evidence is what a person has to go on in deciding whether something is true: what they see and hear, what they remember, what they already have good reason to believe, and what they can see to be true by reflection alone, like two plus two making four. - Justification is the good standing a belief needs, besides being true, to count as knowledge. A justified belief is one it is reasonable to hold, given where the believer stands. - Evidentialism is the view that justification is fit with evidence. A belief is justified for a person when, and to the degree that, it fits the evidence that person has. Earl Conee and Richard Feldman gave the view its modern statement in 1985. On their account a person can take one of three attitudes toward a claim. They can believe it, disbelieve it, or suspend judgment, which means holding off either way. Whichever of the three best fits the person's evidence at the moment is the one that is justified for them. - Think of a weather forecaster looking over the morning's data. Depending on what it shows, the right call is "rain," "no rain," or "too early to say," and it is the data that settles which call is reasonable. Two more ideas complete the picture. - First, having good evidence is not the same as believing because of it. Picture a man with excellent evidence that his business partner is honest, who trusts her only because a fortune cookie told him to. He has the evidence, but he is not using it. Conee and Feldman call a belief held on the basis of the evidence that supports it well-founded, and his is not. - Second, on Conee and Feldman's account evidence consists of things in the believer's mind: experiences, memories, and other beliefs. That makes evidentialism a form of internalism, the view that justification depends only on what lies within the believer's own mind. But it says something narrower than internalism alone, because not everything in the mind justifies. Hopes, fears, and wishes are in the mind too, and on evidentialism none of them justifies anything.

(●) Evidentialism has deep roots among Christians and their critics alike. + The idea is old, and both sides of the argument over God have claimed it. - John Locke, in a chapter he added to his Essay Concerning Human Understanding in 1700, wrote that one unerring mark of a person who loves truth in earnest is "the not entertaining any proposition with greater assurance than the proofs it is built upon will warrant." Locke was a Christian, and he had already defended the faith on those very terms in The Reasonableness of Christianity (1695). - David Hume put the same idea in a single line in his essay on miracles (1748): "A wise man, therefore, proportions his belief to the evidence." - W. K. Clifford turned it into a moral rule in "The Ethics of Belief" (1877): "It is wrong always, everywhere, and for anyone, to believe anything upon insufficient evidence." His example was a shipowner who quiets his doubts about an old ship, sends it to sea full of emigrants, and collects the insurance when it sinks. Clifford said the man would have been guilty even if the ship had come home safe, because he had no right to his comfortable belief. From there the story divides. - Critics of Christianity turned the rule against faith. If belief in God needs evidence, they said, and there is not enough of it, then believing is irrational. Philosophers call this the evidentialist objection to belief in God. - William James answered Clifford in "The Will to Believe" (1896), arguing that some beliefs may rightly be held beyond the evidence (D1). - Alvin Plantinga and the Reformed epistemologists answered the objection by denying its starting point. Belief in God, they argued, can be reasonable without being inferred from evidence or argument. It can arise directly from experience, the way belief in other minds does (D3). • Epistemology is the branch of philosophy that studies knowledge and reasonable belief: what it takes, for instance, for a jury's verdict or a doctor's diagnosis to be well grounded. The Reformed epistemologists are Christian philosophers of knowledge, named for the Reformed (Calvinist) tradition that Plantinga and several of his colleagues came from. - A long line of Christian thinkers took the other road and accepted the challenge. Joseph Butler (1736) and William Paley (1794) set out the evidence for the faith at length. Richard Swinburne and Timothy and Lydia McGrew stand in that line today. The McGrews' version is the one defended in (C2), chosen for its rigor: it measures the fit between belief and evidence by probability, and it holds that the rules of that measure are fixed by reason (P3).

(●) Probabilistic evidentialism measures justification by what the evidence makes probable. + On the McGrews' view, a belief is justified to the degree that the believer's evidence makes it probable. - Probability, here, is a measure of how strongly a body of evidence supports a claim. It runs from 0 (ruled out) to 1 (certain), and 0.5 means the evidence leaves the claim a coin toss. - The measure is taken on the believer's total evidence: everything they have that bears on the question, not just the part they like best. - The rules that say what evidence supports what are not matters of taste. The McGrews hold that they are analytic, true because of what their words mean, and a priori, knowable by reflection alone. Suppose nine of the ten tickets in a hat are winners, and you know nothing else. Then the ticket you draw is probably a winner, and nobody needs to survey raffles to see it. - On Timothy McGrew's account, evidence rests in the end on foundations that are certain. Some are beliefs that simply take in one's own present experience, such as noticing an ache and thinking "this is how it feels." Others are truths seen just by understanding them, such as two plus two making four. Everything else is reached by inference from these, that is, by reasoning from what is already known to what it supports. - This picture of knowledge as a building on bedrock is called classical foundationalism. It says where evidence begins, while evidentialism says what evidence does. Timothy McGrew holds both, but the case for evidentialism does not need the question of foundations settled first. That makes probabilistic evidentialism different from Clifford's in one important way. Clifford talked about duty and guilt, about what a person has no right to believe. Probabilistic evidentialism talks about fit. A belief that fits the evidence is justified whether or not anyone had a duty to believe it, in the same way that the answer to a sum is correct whether or not anyone was told to work it out. The argument for this view is deductive. Think of its premises as a locked chain: if each link holds, the conclusion cannot slip out. Each premise need only be more plausible than its denial, not certain beyond all doubt.

(P1) Justification is how well a belief is supported as true from the believer's point of view. + This premise says what justification is: how good a bet on the truth a belief is, given what the believer can see. "Supported as true" means backed by reasons to think it true, and "from the believer's point of view" means using only what the believer has to go on. (1) To believe something is to take it as true. - Belief is not hoping, imagining, or pretending. To believe it will rain is to take "it will rain" to be true. That is why someone who says "It is raining, but I don't believe it is" sounds so confused. - So whatever makes a belief a good one has to be its standing with respect to truth. A belief can be comforting, popular, or useful and still be a bad belief, in the way a map can be pleasant to look at and still be a bad map. (2) A reward for believing does not make a belief any more likely. - Suppose a billionaire offers you a million dollars to believe, by noon tomorrow, that the number of stars in the Milky Way is even. You now have an excellent reason to want that belief. What you do not have is any reason at all to think it is true. - Try it, and you will find you cannot do it. However generous the offer, it never touches the one question it would need to touch, which is whether the number really is even. - Philosophers call this transparency. When a person asks "Should I believe this?", the question turns at once into "Is this true?", and only considerations that bear on truth get a hearing. (3) A link to truth the believer cannot see cannot guide them. - A belief could be connected to truth in ways the believer has no access to, through some hidden mechanism that happens to work. But the question "Should I believe this?" is asked from the inside, and it has to be answered with what the asker actually has. - Think of a hiker whose compass points true only when there is no hidden magnet nearby. Today, as it happens, there is none, so the needle is right. But she has no way of checking for magnets, and so its accuracy is real but useless to her as a guide. That is the internalist point: what justifies a belief has to be something the believer can reach. (4) The better the support, the better justified the belief. - Support for truth comes in degrees, and justification rises and falls with it. A belief with a little support behind it is a little justified. A belief with overwhelming support is justified overwhelmingly. - Think of two card players. One holds a hand that wins nine times in ten, the other a hand that wins six times in ten. Both may be right to bet, but nobody thinks their bets are equally good, and the difference between them is nothing but the difference in support. - Nothing else is left over to add. Once you know how well a belief is supported as true from where the believer stands, there is no further question of how justified it is. (Whether the believer holds it because of that support is a different question, about how the belief is held, and (C1) takes it up.) Therefore, a belief is justified exactly as well as it is supported as true, seen from where the believer stands.

(P2) Only the believer's evidence supports a belief as true from the believer's point of view. + This premise says that evidence covers everything a person has that bears on truth, and that nothing else they have bears on it at all. (1) Evidence includes everything a person has that bears on truth. - It is a generous category. Sight, hearing, memory, the testimony of people one has reason to trust, the plain seeing that a sum is right, and the experience of God's presence in prayer all count as evidence, insofar as they are present to the mind and bear on the question. - Ruth's evidence is of exactly this kind. The paw prints, her memory of leaving the door latched, and her knowledge of what dogs do in mud all count. None of them is an argument, and she does not need one. (2) Benefits and wishes do not bear on truth. - Comfort, belonging, the hope of reward, and the fear of what others will think are real reasons, of course. But they are reasons to want a belief, not reasons to think it true. - A man holding a lottery ticket wants very much to believe he will win. His wanting is in his mind, and it is strong. It still does not change the odds by a single ticket. (3) Facts the believer has no access to are not part of their point of view. - Whether a person's eyes are working properly, or whether a source is reliable, matters a great deal to whether they get things right. But until those facts show up somewhere in the person's experience or reasoning, they do not bear on truth from where the believer stands. - Think of a juror. Whether her verdict is reasonable depends on the evidence laid before the court, not on a fact that never came out at trial. (4) Much of this premise is a definition. - Once evidence is understood as whatever a person has that bears on truth, the "only" in (P2) follows at once, the way "every bachelor is unmarried" follows once you know what "bachelor" means. There is no trick in that. But it does mean the definition, on its own, settles nothing. - The real weight rests on what falls outside the definition (wishes and benefits, and facts out of the believer's view) and on (P1), so that is where the disputes lie. One is whether anything besides support for truth, such as the stakes of a choice or the loyalty love asks, can add to justification (D1, D2, D8, D11). The other is whether a given experience, such as a sense of God's presence, really bears on truth and so counts as evidence (D3). Therefore, the believer's evidence, and nothing else, supports a belief as true from the believer's point of view.

(C1) Therefore, a belief is justified to the degree the believer's evidence supports it. + A belief is justified to the degree the believer's evidence supports it, and that is the core of evidentialism. It comes from putting the two premises together: justification is how well a belief is supported as true from the believer's point of view (P1), and only the believer's evidence supplies that support (P2). The step is a substitution of equals, a valid form: when two things are each the same as a third, whatever is true of one is true of the other. A cook who knows the recipe calls for a cup of milk, and knows that a cup is eight ounces, already knows the recipe calls for eight ounces. (1) A justified belief is not always a well-founded one. - (C1) is about justification in the first of two senses: what a person's evidence entitles them to believe. It does not say they believe it for that reason. - Go back to the man who trusts his honest business partner because of a fortune cookie. His belief is justified in this first sense, since his evidence supports it. It is not well-founded, because he holds it on the wrong basis. Evidentialism needs both ideas, and it keeps them apart. That leaves one question open, which is how support is measured, and (P3) answers it.

(P3) Evidence supports a belief to the degree the believer's total evidence makes it probable. + This premise says evidence comes in degrees, probability measures those degrees, and the measure is taken on all of a person's evidence. (1) Evidence comes in degrees. - One muddy paw print near the door is some evidence the dog went out. A trail of prints to the couch, an open door, and a muddy dog asleep on the cushions are far more. - A detective's case works the same way. It grows stronger with each clue and weaker when one turns out to have been planted, and support that can grow and shrink like that is a matter of degree, not of yes or no. (2) Probability is the measure of those degrees. - A piece of evidence raises the probability of a claim when the evidence is more to be expected if the claim is true than if it is false. Muddy prints are just what you would expect if the dog went out, and very surprising if he did not, which is why they count so heavily. - This is the heart of Bayes' theorem, named for Thomas Bayes, an eighteenth-century minister and mathematician. The strength of a piece of evidence depends on how much more likely it is on one explanation than on its rival. - Nobody needs the formula to reason this way. A mechanic who hears a particular knock and thinks "that is far more likely a worn bearing than a loose heat shield" is doing it already. (3) Probability is the consistent way to measure support. - Why probability, and not some other scale? Two classic results answer that. The first is the Dutch book argument, named for a set of bets that guarantees a loss. Suppose a woman is 60 percent sure it will rain tomorrow and also 60 percent sure it will not, which breaks the rule that the two must add up to 100. By her own lights, 60 cents is a fair price for a ticket that pays a dollar if it rains, and 60 cents is fair for one that pays a dollar if it does not. Sell her both, and she pays $1.20 to collect exactly $1 whatever the weather. - The second result shows that any way of grading support that keeps a few plain rules of consistency (for instance, that the grade a claim gets fixes the grade its denial gets) can be converted onto the probability scale, the way a Fahrenheit reading converts to Celsius. - Both results rest on assumptions philosophers still debate, such as whether support always comes in exact amounts. What they show is that a rival scale would have to give up some piece of consistency. (4) Probability must be judged on the total evidence. - Picture a doctor who looks only at the one test result that fits her favorite diagnosis and ignores the three that do not. She is not following the evidence. She is choosing it. - Philosophers call this the requirement of total evidence: a claim's support is measured against everything the believer has that bears on it. (5) The rules of support are fixed by reason, not by mood. - On probabilistic evidentialism, the principles that link evidence to probability are analytic and a priori, like the ticket principle (nine winners in ten tickets makes a win probable). So the probability a body of evidence gives a claim is not a matter of how the believer happens to feel about it. - That is what makes it possible to criticize a belief, with real meaning, for going beyond the evidence or falling short of it. Therefore, evidence supports a belief to the degree that the believer's total evidence makes it probable.

(C2) Therefore, a belief is justified to the degree the believer's total evidence makes it probable. + A belief is justified to the degree the believer's total evidence makes it probable. That is evidentialism in its probabilistic form, and it follows by the same substitution of equals as before: a belief is justified to the degree the believer's evidence supports it (C1), and evidence supports a belief to the degree the total evidence makes it probable (P3). If the premises hold, the conclusion follows necessarily. (1) Evidentialism does not require an argument for everything. - A child who sees her mother walk into the room is justified in believing her mother is there. The experience is her evidence, and most justified beliefs rest on nothing more elaborate than that. (2) Evidentialism does not require working out numbers. - The probabilities are a measure of support, not a procedure anyone has to carry out. A good forecaster says "seventy percent" without writing down an equation, and a careful reader of the news weighs sources without a calculator. (3) Evidentialism suits a faith that gives reasons. - The New Testament presents itself as testimony to be weighed. Luke writes his Gospel so that Theophilus "may have certainty concerning the things you have been taught" (Luke 1:4), and he offers that firm assurance through eyewitness testimony he had carefully investigated (Luke 1:1-3). On this view, that certainty is high probability built on good evidence. - Jesus "presented himself alive to them after his suffering by many proofs" (Acts 1:3). And when Paul preached in Berea, his hearers "received the word with all eagerness, examining the Scriptures daily to see if these things were so" (Acts 17:11). - On evidentialism, the question "Is my faith reasonable?" has an answer a believer can look at and share: the evidence they actually have. Therefore, if (P1), (P2) and (P3) hold, evidentialism in its probabilistic form follows: a belief is justified to the degree that everything the believer has to go on makes it probable. See also: Philosophy / Epistemology: Internalism Philosophy / Epistemology: Classical Foundationalism Philosophy / Logic: Inductive Arguments Philosophical Theology / Prolegomena: Meta-Apologetics: Evidentialist

On evidentialism and justification: Adams, Robert Merrihew. "Kierkegaard's Arguments against Objective Reasoning in Religion." The Monist 60 (1977): 228-243. Reprinted in The Virtue of Faith and Other Essays in Philosophical Theology. New York: Oxford University Press, 1987. Carnap, Rudolf. Logical Foundations of Probability. Chicago: University of Chicago Press, 1950. Conee, Earl, and Richard Feldman. "Evidentialism." Philosophical Studies 48 (1985): 15-34. Conee, Earl, and Richard Feldman. "Internalism Defended." American Philosophical Quarterly 38 (2001): 1-18. Conee, Earl, and Richard Feldman. Evidentialism: Essays in Epistemology. Oxford: Clarendon Press, 2004. Cox, Richard T. "Probability, Frequency and Reasonable Expectation." American Journal of Physics 14 (1946): 1-13. de Finetti, Bruno. "La prévision: ses lois logiques, ses sources subjectives." Annales de l'Institut Henri Poincaré 7 (1937): 1-68. Feldman, Richard. "Epistemic Obligations." Philosophical Perspectives 2 (1988): 235-256. Feldman, Richard. "Reasonable Religious Disagreements." In Philosophers without Gods, edited by Louise Antony, 194-214. New York: Oxford University Press, 2007. Keynes, John Maynard. A Treatise on Probability. London: Macmillan, 1921. McGrew, Timothy J. The Foundations of Knowledge. Lanham, MD: Rowman & Littlefield, 1995. McGrew, Timothy, and Lydia McGrew. Internalism and Epistemology: The Architecture of Reason. London: Routledge, 2007. McGrew, Timothy, and Lydia McGrew. "The Argument from Miracles: A Cumulative Case for the Resurrection of Jesus of Nazareth." In The Blackwell Companion to Natural Theology, edited by William Lane Craig and J. P. Moreland, 593-662. Oxford: Wiley-Blackwell, 2009. McGrew, Timothy. "Miracles." In The Stanford Encyclopedia of Philosophy, edited by Edward N. Zalta. 2010; revised 2019; revised with Robert Larmer, 2024. Ramsey, Frank P. "Truth and Probability" (1926). In The Foundations of Mathematics and Other Logical Essays, edited by R. B. Braithwaite, 156-198. London: Kegan Paul, 1931. Shah, Nishi. "How Truth Governs Belief." Philosophical Review 112 (2003): 447-482. White, Roger. "Epistemic Permissiveness." Philosophical Perspectives 19 (2005): 445-459. Williams, Bernard. "Deciding to Believe." In Problems of the Self, 136-151. Cambridge: Cambridge University Press, 1973. Critics and rival views engaged: DeRose, Keith. "Contextualism and Knowledge Attributions." Philosophy and Phenomenological Research 52 (1992): 913-929. Fantl, Jeremy, and Matthew McGrath. "Evidence, Pragmatics, and Justification." Philosophical Review 111 (2002): 67-94. Fantl, Jeremy, and Matthew McGrath. Knowledge in an Uncertain World. Oxford: Oxford University Press, 2009. Howson, Colin, and Peter Urbach. Scientific Reasoning: The Bayesian Approach. 3rd ed. Chicago: Open Court, 2006. James, William. "The Will to Believe." In The Will to Believe and Other Essays in Popular Philosophy. New York: Longmans, Green, 1897. Keller, Simon. "Friendship and Belief." Philosophical Papers 33 (2004): 329-351. Kelly, Thomas. "Evidence Can Be Permissive." In Contemporary Debates in Epistemology, 2nd ed., edited by Matthias Steup, John Turri, and Ernest Sosa. Malden, MA: Wiley-Blackwell, 2014, 298-313. Kelly, Thomas. "Peer Disagreement and Higher-Order Evidence." In Disagreement, edited by Richard Feldman and Ted A. Warfield, 111-174. Oxford: Oxford University Press, 2010. Kierkegaard, Søren [Johannes Climacus, pseud.]. Concluding Unscientific Postscript to Philosophical Fragments. 1846. Plantinga, Alvin. "Reason and Belief in God." In Faith and Rationality: Reason and Belief in God, edited by Alvin Plantinga and Nicholas Wolterstorff, 16-93. Notre Dame, IN: University of Notre Dame Press, 1983. Plantinga, Alvin. "The Foundations of Theism: A Reply." Faith and Philosophy 3 (1986): 298-313. Plantinga, Alvin. Warranted Christian Belief. New York: Oxford University Press, 2000. Rosen, Gideon. "Nominalism, Naturalism, Epistemic Relativism." Philosophical Perspectives 15 (2001): 69-91. Sagan, Carl. Cosmos: A Personal Voyage. Episode 12, "Encyclopaedia Galactica." Television series. PBS, 1980. Stanley, Jason. Knowledge and Practical Interests. Oxford: Oxford University Press, 2005. Stroud, Sarah. "Epistemic Partiality in Friendship." Ethics 116 (2006): 498-524. van Fraassen, Bas C. Laws and Symmetry. Oxford: Clarendon Press, 1989. Historical sources: Butler, Joseph. The Analogy of Religion, Natural and Revealed. 1736. Clifford, W. K. "The Ethics of Belief." Contemporary Review 29 (1877): 289-309. Hume, David. An Enquiry Concerning Human Understanding. 1748. Section 10, "Of Miracles." Locke, John. An Essay Concerning Human Understanding. 1690. Book 4, chapter 19, "Of Enthusiasm," added in the 4th edition, 1700. Locke, John. The Reasonableness of Christianity. 1695. Paley, William. A View of the Evidences of Christianity. 1794. Pascal, Blaise. Pensées. 1670. Biblical citations: Luke 1:1-4; John 20:25; John 20:29; John 20:30-31; Acts 1:3; Acts 17:11; Acts 17:26-27; Romans 8:16; 1 Corinthians 13:7; 1 Corinthians 15:14; Hebrews 11:1. Scripture quotations are from the English Standard Version.
+ William James: when a question is live, forced and momentous, and evidence cannot settle it, we may rightly believe beyond the evidence. Waiting for proof risks missing the truth altogether.
1. James's case deserves a hearing. The strongest form of this objection is James's own. He sorted choices between beliefs by three marks: - A choice is live when both options are real possibilities for you. For most readers today, whether to worship Zeus is not a live choice. - It is forced when there is no way to stay out of it. Someone offered a job that must be accepted by Friday cannot avoid answering, because silence on Friday is a no. - It is momentous when it is unique and the stakes are high. James's example was an invitation to join an expedition to the North Pole, a chance that would probably never come again. Where a choice like that "cannot by its nature be decided on intellectual grounds," he wrote, "our passional nature not only lawfully may, but must, decide." By our passional nature he meant our hopes, fears and will, as opposed to our reasoning. Refusing to believe until the evidence is in is itself a choice, and it gives up the chance of being right in order to avoid the chance of being wrong. This is a serious argument by a serious philosopher, and many Christians have found it moving. 2. Trust that helps make itself true is evidence in disguise. James's best example is friendship. Suppose I wonder, "Do you like me or not?" Whether you do may depend, he said, on whether I meet you halfway. Here my believing it helps make it true. - In a case like that, my willingness to trust is itself part of what makes the outcome likely, and an evidentialist can count it. The belief fits the evidence once the evidence includes my own part in the story. - God's existence, though, does not depend on anyone's believing in it, so that form of the argument does not carry over. 3. Seeking God is different from believing ahead of the evidence. James saw that reply coming and gave a subtler religious case. If God is personal, he argued, the evidence for God may come only to those who meet him halfway. Think of someone at a dinner party who demands a guarantee before every friendly gesture and will not take anyone's word for anything. He goes home without a single new friend. In the same way, a skeptic who holds out in what James called "snarling logicality," waiting to be compelled, might shut himself out of the very acquaintance with God that would settle the question. - This deserves granting. If God is personal, the way a person seeks may affect what evidence they ever receive. Paul told the philosophers of Athens that God placed people where they would seek him and perhaps find him (Acts 17:26-27). - But what follows is a reason to seek: to pray, to read, to put oneself where the evidence can arrive. That is wise action under uncertainty, and an evidentialist welcomes it. - When the evidence comes, belief fits it. What does not follow is that a person should believe now, beyond what they have, in order to get the evidence later. A man who wants to know whether a stranger is trustworthy can extend her a small trust and watch what happens. He does not have to decide in advance that she is honest. 4. James gives a reason to act, not a reason to think something true. James is right that we often have to act before the evidence is complete. A doctor in an emergency treats the most likely condition, and a person may pray, attend church, and read the Gospels while still weighing the evidence. All of that is reasonable. What none of it does is make the claim more probable than the evidence makes it. - Think of a firefighter who must decide in seconds whether to enter a burning house. He rightly acts on the best evidence he has, and he does not have to pretend he is sure the roof will hold. 5. Christianity does not need James's exemption. James assumed that the question of God cannot be decided on intellectual grounds, and that is the very thing evidentialist Christians deny. The case for God's existence and for the resurrection of Jesus is open to inspection, and the apostles presented it as testimony to be weighed (C2). A faith that steps beyond the evidence by choice has given up the strongest thing it has.
+ Pascal's wager shows that believing in God can be rational even without strong evidence, because the stakes are infinite. The practical case for believing outweighs the evidential one.
1. The wager is not evidence that God exists. Blaise Pascal's argument runs like this. If you believe in God and God exists, you gain everything. If you believe and God does not exist, you lose little. So if the evidence leaves the question open, the wise bet is on God. Whatever its strength as a piece of reasoning about choices, it does not claim to show that God exists. It weighs the value of believing, not the probability of the claim, which makes it a reason to seek rather than evidence. 2. Pascal knew belief cannot simply be chosen. Pascal had advice for the person persuaded by the wager but still unable to believe: act as believers do, take holy water, have Masses said, and in time belief would come. He told that person to work, not by adding proofs of God, but by lessening the passions that held him back. - The first lesson is one evidentialism shares. The wager by itself does not produce belief, because it does not bear on truth (P1). - The second needs care. If habit only clears away the fears and appetites that keep a person from seeing what the evidence already shows, evidentialism has no quarrel with it. Anyone who has sat on a jury knows that prejudice can hide the plain meaning of the facts. - But suppose habit were to produce a belief the evidence does not support. Then evidentialism says that belief would not be justified, however it was acquired, and read that way, Pascal and the evidentialist really do disagree. 3. Pascal also gave evidence. The wager fragment is only one note among hundreds in his Pensées ("Thoughts"), the notes for a defense of Christianity that he left unfinished at his death. Many of the others set out evidence for the faith: prophecy, the miracles, the character of Christ, the witness of the apostles. Pascal did not think the wager could stand in for those. It was addressed to a person convinced that reason could not settle the question. 4. Two kinds of "should" are in play. "Should I bet on God?" is a practical question, about what to do. "Should I believe God exists?" is a question about what is true, and only evidence answers it. Run the two together and you get the illusion that a good bet is a good reason for belief. - Think of a traveler who packs an umbrella on a day the forecast calls probably dry, because a soaked suit before an interview would be a disaster. Packing it is a wise decision. It does not make rain one bit more likely.
+ Reformed epistemology: belief in God can be properly basic, reasonable without being inferred from evidence or argument. Evidentialism would condemn the millions of believers who have never weighed an argument.
1. Reformed epistemology is a serious view held by fellow believers. In the 1980s Alvin Plantinga and the Reformed epistemologists argued that belief in God can be properly basic: reasonable without being inferred from other beliefs. Your belief that other people have minds is like that. It is reasonable, though you have never proved it. They argued, with real force, that the evidentialist objection to belief in God held faith to a standard nobody applies to memory, perception, or other minds. Christians who hold this view are brothers and sisters, and the disagreement is about the theory of knowledge, not about the faith. 2. Evidence is not the same as argument. Plantinga never said basic belief in God is groundless. He said it arises from experience: gazing at the starry heavens, feeling guilt or gratitude, reading Scripture. And on evidentialism as Conee and Feldman state it, experiences are evidence. So the quarrel is narrower than it first looks, because evidence does not mean argument. A believer's experience of God's presence in prayer, the testimony of people they have good reason to trust, the beauty and order of the world, and what Paul calls the Spirit bearing witness "with our spirit" (Romans 8:16) are all evidence, insofar as they are present to the believer's mind (P2). Most believers have a great deal of it, though they would never call it that. - Think of a woman who has never studied the history of her country. She still has good evidence that it exists: maps, coins, newspapers, neighbors. She simply has not organized it into an argument. 3. Basic beliefs still answer to evidence. Plantinga himself grants that a properly basic belief can be defeated, that is, overturned by contrary evidence. It survives only if the defeater is answered, whether by argument or, as he holds, by the sheer strength of the belief itself. So even on his view, evidence matters to whether belief in God stays reasonable. - Think of your memory of locking the front door. You believe it without any argument, but if you come home to find the door standing open, the memory alone no longer settles the matter. The evidentialist adds only one thing. Whatever makes the belief reasonable in the first place is also something the believer has to go on, such as the experience itself. It is not a fact about her faculties (the mental equipment, such as memory and perception, that produces beliefs) which she cannot see. 4. The disagreement is about where reasons start. Plantinga argued that classical foundationalism, the theory behind the old evidentialist objection, refutes itself, because no one can show by its own standards that it is true. Timothy McGrew holds classical foundationalism all the same, and its defenders have a reply: the view is the conclusion of an argument from premises seen to be true, so it passes its own test. - On that view, the foundation is the believer's certain awareness of her own experience: the sense of being forgiven, the stillness of God's presence in prayer. The belief that it is God she is meeting is then reasoned to from that experience, as its best explanation, the same way belief in the world outside rests on how that world looks and sounds. - So the two sides agree that such experiences count. They differ over whether belief in God is the foundation itself or the first thing built on it. - You might think that makes every believer a philosopher. It does not. Reasoning from experience to its best explanation is something people do all day without naming it, the way a mother hears a crash in the kitchen and knows the cat is on the counter (D6). - (C2) itself holds whichever way the question of foundations is settled, since it says how evidence supports a belief, not where evidence begins. See also: Philosophy / Epistemology: Warrant Philosophy / Epistemology: Classical Foundationalism Philosophical Theology / Prolegomena: Meta-Apologetics: Reformed
+ Faith that rests on evidence is not faith. Jesus blessed "those who have not seen and yet have believed," and Hebrews calls faith "the conviction of things not seen."
1. "Not seen" does not mean "without evidence." Jesus spoke those words to Thomas, who had refused to believe until he could touch the wounds. Thomas was not short of evidence. The other disciples had told him they had seen the Lord (John 20:25), and he would not accept their word. "Blessed are those who have not seen and yet have believed" (John 20:29) is a blessing on people who would believe without seeing Jesus with their own eyes. It is not a blessing on people who believe without reason. The very next verses say so: "Now Jesus did many other signs in the presence of the disciples, which are not written in this book; but these are written so that you may believe" (John 20:30-31). John wrote down the evidence precisely so that people who had not seen could believe. 2. Hebrews speaks of trust in someone with a track record. "Now faith is the assurance of things hoped for, the conviction of things not seen" (Hebrews 11:1). The chapter then lists people who acted on what God had told them, often for years before they saw the outcome. Their faith was trust in a God who had shown himself faithful, and "not seen" refers to what was still future or invisible, not to an absence of grounds. - Think of a man who boards a plane to a city he has never seen. He believes it is there, and he has excellent reasons: maps, the airline's record, the people who have been. 3. Trust is reasonable when its object is trustworthy. Faith in the New Testament is personal trust in God and in Christ. Evidentialism is a claim about belief that something is true, and trust in a person rests on beliefs of exactly that kind: that the person exists, that they have promised, that they keep their word. Evidence for those things does not compete with faith. It is what makes faith something better than a leap. - Think of a patient going under for surgery. Her trust in the surgeon is personal, and it rests on things she believes about him: his training, his record, what her own doctor told her. 4. Evidence short of certainty can carry a whole life. The strongest form of this objection is Søren Kierkegaard's, set out under the pen name Johannes Climacus, and the worry is real. Historical study yields only an approximation, and it never finishes: there is always another manuscript to collate and another scholar to answer. A faith that waited on the scholars would wait forever. Worse, faith as he saw it needs risk, the passion of holding fast to what cannot be made objectively certain, and a faith measured out by probability would be lukewarm, ninety percent committed. - The worry about waiting proves too much, which is to say that if it worked here it would work everywhere. Every great commitment in life, from marriage to emigration, rests on evidence short of certainty and open to further inquiry, and no one thinks it worse for that. Evidence in hand can be enough before inquiry ends. - The worry about lukewarmness confuses two things. Evidentialism proportions belief to evidence. It does not ask anyone to proportion commitment. A bride who is very confident, though not certain, that her husband is a good man does not marry him ninety percent. She marries him, and her whole life goes with the vow. - Nor does evidence take away the risk Kierkegaard prized. "Very confident" is not "certain," and the bride still stakes everything on a man whose heart she cannot see into. A probable truth, embraced wholeheartedly, is a fit object of trust.
+ Clifford's principle refutes itself. There is not sufficient evidence for the claim that we must never believe on insufficient evidence, so by its own standard no one should believe it.
1. The objection hits Clifford's moral rule, not evidentialism. Clifford stated a moral rule about guilt and duty, and he stated it in sweeping terms: wrong always, everywhere, for anyone. That rule overreaches, and a careful evidentialist need not defend all of it. Evidentialism as stated in (C2) is a claim about what justification is, not a rule about when people sin. 2. Evidentialism is itself supported by evidence. The case for evidentialism is the argument for it: the bribe that cannot buy a belief (P1), the lottery ticket whose owner's wishes do not change the odds (P2), and the doctor who ignores the inconvenient test results (P3). A reader can weigh every one of those. So evidentialism does not exempt itself from its own test. It is believable exactly to the degree those considerations are good ones. 3. Some of its support is knowable by reflection. On the McGrews' view the basic principles of evidence are analytic and a priori. A person can see, just by understanding the words, that nine winners in ten tickets makes a win probable. Seeing a truth of that kind is evidence of the most basic sort, so a principle known that way is not believed without evidence.
+ Evidentialism is impossibly demanding. No one tracks their total evidence or works out probabilities, so on this view almost no one is justified in anything.
1. Justification is a standard, not a procedure. Evidentialism says what makes a belief justified. It does not say a person must consciously run through their evidence before believing anything, and a belief can fit the evidence without the believer stopping to check that it does. - Think of grammar. A sentence is grammatical if it follows the rules, whether or not the speaker could state them, and most people speak grammatically all day without ever opening a textbook. 2. Most beliefs are easily justified. Ruth's belief about the dog, your belief that you are reading, a child's belief that her mother is home: these fit the evidence so obviously that no reckoning is needed. Hard cases are rare. And in the hard cases we do slow down and weigh things, just as evidentialism says we should. 3. Degrees of justification fit ordinary life. Because justification comes in degrees, a person who has weighed things roughly but sensibly is justified to a reasonable degree, even if a careful statistician could do better. Ordinary reasoners approximate, and their approximations are often very good. - Think of an experienced carpenter who judges a board's length by eye. He has no tape measure in hand, and he is seldom off by much.
+ Probabilities are subjective. Two people starting from different assumptions will read the same evidence differently, so "what the evidence makes probable" is only a matter of opinion.
1. Subjective Bayesianism is a serious view. The strongest form of this objection comes from subjective Bayesianism, the view that probabilities are degrees of confidence a person happens to have, constrained only by the rules of probability. On that view a person's prior, their starting confidence before the evidence comes in, is up to them, and two people can reason flawlessly from the same evidence to different conclusions. Many philosophers of science hold this view, and it has real attractions. The subjectivist also has history on his side. The older attempt to fix every prior by a formal rule ran into well-known trouble. The obvious rule, the principle of indifference, says to give equal probability to possibilities you have no reason to tell apart. The trouble is that it gives conflicting answers depending on how the possibilities are described. - Think of a factory that makes cubes with sides somewhere between 0 and 2 centimeters. Spread your confidence evenly over the side lengths, and the chance that a cube's side is under 1 centimeter is one half. Spread it evenly over the volumes instead, from 0 to 8 cubic centimeters, and the chance of the very same thing is one eighth. 2. On the McGrews' view the rules of evidence are not personal. The McGrews hold that the principles connecting evidence to probability are analytic and a priori, as the ticket principle is. On that view the support that evidence gives a claim is a fact about the evidence and the claim, in the way that the sum of a column of figures is a fact about the figures, whoever adds them up. John Maynard Keynes and Rudolf Carnap held views of the same family, called logical theories of probability. - The ticket principle is a case where the possibilities really are alike: ten tickets, each as likely as the next. The cube factory is a case where almost nothing is known. Keynes knew puzzles of this kind, and his answer was to restrict the principle to possibilities like the tickets, not to abandon it. Real questions usually come with background knowledge that breaks the tie, the way learning how the factory sets its machines would. - Whether every starting probability can be fixed by reason, and not only the clear cases, is the most disputed part of the view and the hardest to defend. 3. Evidence tends to swamp starting points. Even on the subjective view, as evidence accumulates, people who started far apart are driven toward agreement, provided neither started out absolutely certain and both agree how much more to be expected each piece of evidence is on one answer than on the other. This is called the swamping of the priors, because the evidence washes out the starting points. A juror inclined to trust the defendant and a juror inclined to suspect him will converge once the fingerprints, the receipt, and the security footage are in. - The convergence is a long-run result, though. On a single question with a fixed body of evidence, such as a disputed event in ancient history, starting points can still matter a great deal. Even there, the force of the evidence can be argued on its own terms, by asking how many times more to be expected it is if the claim is true than if it is false. A large enough ratio will bring round to the claim anyone who did not start out all but sure it was false. That is why the view finally rests on the objectivity of support, not on swamping alone. 4. Subjectivism does not escape the question. Subjective Bayesians can still fault a person who contradicts himself, or who refuses to update, that is, to revise his confidence when new evidence comes in. What they cannot fault is a strange starting point, faithfully followed. Imagine a man who begins all but certain that the moon landings were staged, and updates flawlessly on every photograph and testimony afterward. He may end still convinced, and on pure subjectivism he has made no mistake. Few people, subjectivists included, accept that verdict in practice, and the reason they do not is a reason to think the support evidence gives is not merely a matter of taste.
+ Pragmatic encroachment: a person with little at stake can know the bank is open Saturday, while someone with the same evidence and a mortgage payment riding on it cannot. So justification depends on stakes, not evidence alone.
1. The bank case is real and deserves an answer. Imagine two people who drove past the bank last Saturday and saw it open. On Friday afternoon each decides to deposit a check tomorrow instead of waiting in line today. For one, nothing much depends on it. For the other, a bounced mortgage payment does. Many people judge that the first can rightly say "I know the bank will be open," while the second should go in today and check. Jason Stanley, and Jeremy Fantl and Matthew McGrath, argue that what you are justified in believing, or know, shifts with the practical stakes. The view is called pragmatic encroachment, because practical matters (what you stand to win or lose) push into, or "encroach" on, questions about knowledge. Its defenders do not say the second customer has less evidence. They grant that the evidence is the same. Their argument runs through a link between knowing and acting: if you are justified in believing something, you are entitled to act on it. The second customer is not entitled to act as though the bank will be open, and so, they conclude, she is not justified in believing it, though her evidence matches the first customer's exactly. 2. Probability stays fixed while the threshold for action moves. On probabilistic evidentialism, both people have the same evidence, and so the same probability that the bank will be open, say 0.95. What differs is whether that probability is high enough to act on, given what is riding on the choice. A five percent chance of a bounced mortgage is a risk worth avoiding. A five percent chance of a second trip is not. - Think of a weather forecast of ninety percent sunshine. It is the same forecast for the family planning a picnic and for the couple planning an outdoor wedding, and the couple rents a tent anyway. 3. Both customers are justified to the same degree. The probabilistic view measures justification in degrees (C2), and on that measure both customers are justified to exactly the same degree. The link between knowing and acting is aimed at something else: whether a person may flatly say "I know" and act without hedging. - Whether that flat, all-or-nothing standing shifts with the stakes is a live debate, and (C2) does not need to settle it. Let the bar for flat-out belief, or for the word "know," rise when a mortgage is on the line. The support the evidence gives is still the same for both. 4. The mortgage does not bear on the bank's hours. What the rival view does deny is that support for truth is all there is to justification. On its account, two people with the same evidence, reasoning the same way, differ in whether they are justified about the bank, and the difference comes from a fact (the size of a payment) that has nothing to do with whether the bank is open. The evidentialist answer is that the stakes change when it is wise to stop checking and act. Whatever they do to the bar for "I know" (point 3), they do not change how well anything supports the belief. Support is what (P1) identifies with justification, and a mortgage no more moves it than a bribe does. - Think of two people reading the same thermometer. One is deciding whether to wear a jacket, the other whether a newborn needs a heavier blanket for the night. The second should check twice, but the thermometer does not become less trustworthy because her stakes are higher.
+ Reasonable people with the same evidence disagree all the time, about politics, philosophy and religion. So evidence cannot fix one right answer, and evidentialism is wrong to suggest it does.
1. Usually the evidence is not the same. Two people who disagree about God, or about a court case, almost never share exactly the same evidence. They have had different experiences, heard different testimony, and read different things. Much disagreement is simply evidence that has not yet been shared. 2. Exactly the same evidence leaves little room. Richard Feldman argues for what he calls uniqueness: a single body of evidence supports a single attitude, so two people with exactly the same evidence cannot both be fully reasonable in reaching opposite conclusions. Roger White adds that if the evidence allowed either conclusion, which one a person held would come down to something other than the evidence, such as temperament or upbringing. Others, called permissivists, deny uniqueness, and Gideon Rosen and Thomas Kelly are among them. They argue that reasonable people can weigh the same evidence by different standards, as when one scientist prizes a simple theory more than her colleague does. Kelly draws on James: one person may care more about avoiding error and another more about believing truths, and the same evidence can then leave the first rightly holding off and the second rightly believing. - The debate is live, and it matters here, because (C2) speaks of the degree the evidence supports. If permissivism is right, the evidence fixes a range rather than a single point, and a belief is justified to the degree that range allows. - Even permissivists usually allow only modest slack, not room for any belief at all. - Think of two accountants auditing the same books. They might reasonably differ on how to classify a borderline expense. They cannot both be right if one says the company made a profit and the other says it went bankrupt. 3. Disagreement is itself evidence. Learning that a thoughtful person disagrees with you is a piece of evidence, and an evidentialist takes it into account. Sometimes it should lower your confidence. Sometimes, on reflection, it does not, because you can see where the other person's reasoning goes wrong. Either way, the response is governed by evidence. - Think of two hikers who disagree about which fork leads back to camp. If she has walked this trail many times, her disagreement should give you pause. If you can see she is holding the map upside down, it need not.
+ "Extraordinary claims require extraordinary evidence." By evidentialism's own standard, a resurrection is the most extraordinary claim there is, so no testimony could ever justify believing it.
1. The slogan contains a truth. The line was made famous by Carl Sagan, and it has an ancestor in Hume's essay on miracles. Put in terms of probability, its truth is this: a claim with a low starting probability needs evidence that is much more to be expected if the claim is true than if it is false. Evidentialism agrees. 2. Evidence can be strong enough. How strong evidence can be is a matter of measurement, not of slogans. Testimony from several independent witnesses who had no motive to invent their story, who were in a position to know, and whose accounts fit what else we know, can be extraordinarily strong. Each witness multiplies the weight of the others, provided they are independent in the sense that matters: none got the story from another or from a shared source, so that the only thing linking their reports is the event itself. - Think of a jury hearing five witnesses who did not know each other, each of whom saw the same unusual car at the scene. Coincidence would have to be very great to explain their agreement. - Timothy and Lydia McGrew have set out the evidence for the resurrection in exactly these terms, measuring how much more to be expected the testimony is if Jesus rose than if he did not. 3. The objection is an application of evidentialism. The challenge is not a rival to evidentialism at all, which is why evidentialism welcomes it. The Christian evidentialist accepts its terms and asks only that the evidence be weighed, all of it, rather than ruled out in advance. Paul staked the faith on a historical claim (1 Corinthians 15:14), and the apostles invited inspection. See also: Philosophy / Logic: Inductive Arguments CE / Resurrection: Maximal Data Method
+ A loving friend or spouse should believe the best of the one they love, even against the evidence. Evidentialism makes love calculating and disloyal.
1. The objection names something real. Sarah Stroud and Simon Keller argue that good friendship involves epistemic partiality. ("Epistemic" means having to do with knowledge and belief.) Friends interpret each other generously and are slow to believe the worst. And Paul writes that love "bears all things, believes all things, hopes all things, endures all things" (1 Corinthians 13:7). 2. Love changes how we look, not what the evidence shows. Much of what partiality asks for is evidentialist already. A good friend looks harder for an innocent explanation, asks before accusing, and gives weight to years of good character, which is real evidence. - Think of a wife who hears a rumor about her husband. She has thirty years of his faithfulness to weigh against one anonymous story. Believing him is not loyalty against the evidence. It is what the evidence supports. 3. Believing against clear evidence is not a kindness. A parent who refuses to believe, against all evidence, that a child is addicted does the child no favor. Love that will not see cannot help. Paul's phrase is about love's refusal to give up on people, its readiness to hope and trust. It is not a command to believe what the evidence shows is false. 4. Love shapes action where evidence is unclear. Where evidence is evenly balanced, love rightly decides how we act: we extend trust, give the benefit of the doubt, and stay. That is a choice about what to do, and it is one of the best choices people make (D2).
+ Evidentialism hands the atheist the verdict. Most believers hold their faith with far more confidence than their evidence supports, so by evidentialism's own standard most Christian belief is unjustified.
1. The standard binds believers too. The strongest form of this objection grants everything argued for evidentialism and then turns it around. Very few Christians have weighed the case for their faith, many say they are certain, and on (C2) confidence beyond the evidence is a fault. Richard Feldman, himself an evidentialist, sharpened the point. A believer who knows that thoughtful people who have weighed the same case reject it should often suspend judgment, he argued (D9 answers that part). This should be granted as far as it goes. Evidentialism has no exemption for Christians, and it should not have one. - Think of a sports fan who is sure every spring that his team will win the championship. His loyalty is fine. His certainty is not something the standings support, and nobody is being unkind in pointing that out. 2. Accepting the rule does not settle the verdict. Evidentialism says what justification is. It says nothing yet about where the evidence on any particular question points. Whether the evidence for God and for the resurrection of Jesus is strong is a question about the world and its history, and it has to be answered there. - Think of two lawyers who agree on the rules of evidence before a trial. Agreeing on the rules decides nothing about the verdict. It only makes sure the verdict, when it comes, rests on what was shown in court. 3. Ordinary believers have more evidence than the objection counts. A believer's total evidence includes far more than formal arguments: answered prayer and the sense of God's presence, the testimony of people she has good reason to trust, the order of the world, the changed lives she has watched up close, and whatever she knows of the history behind the Gospels. When several independent lines (linked only through the truth of the matter, as in D10) each fit Christianity better than its denial, their support compounds, so that the whole is far stronger than any one line. - Think of a woman who is confident her husband loves her. She has no single proof. What she has is ten thousand small things, each of which would be odd if he did not. - How much all of that adds up to differs from one believer to another, and some believers do hold more confidence than they have grounds for. 4. Confidence is not the same as commitment. Many believers who say they are "sure" are describing the wholeness of their commitment, not claiming that every doubt has been ruled out. A person can be wholly committed on strong evidence short of certainty (D4). - Where a believer's confidence has outrun her evidence, evidentialism prescribes what the Bereans did when Paul first preached to them: look at the evidence (C2). See also: Philosophy / Logic: Cumulative Case Arguments CE / Resurrection: Maximal Data Method Philosophical Theology / Prolegomena: Meta-Apologetics: Evidentialist

Metaphysics: Free Will

Whether Our Choices Are Up to Us

Public Theology

Theological Analyses of Societal Issues

Theology of the Family

Where Faith and Family Intersect

Biographies

Notable Works & Great Quotes from Key Figures

Contemporary Voices

1960 AD – present
Playlist
Justified True Belief