Argument

The Testimony Argument

Intro

Think about how much of what you know you have never checked for yourself. Your birthday. That the earth goes round the sun. That there is a country called Japan. That your parents are your parents. Almost all of it came from someone telling you, a parent, a teacher, a book, a map. Without trusting what others say, no one could know much of anything.

That trust works because people mostly tell the truth. Researchers find that we naturally believe what we are told unless something makes us suspicious, and that most people tell few or no lies on a given day, while a small number of people tell most of the lies. Even a liar depends on this. A lie only works because the listener expects the truth. Lying is a parasite: it lives on honesty and cannot exist without it.

Why would the world be like that? If life is only a struggle to survive, deceiving others would often pay, and we might expect a world of liars, not a world where honesty is the normal setting. The Testimony Argument says a world where truth-telling is the default is exactly what you would expect if reality comes from a God who speaks and cannot lie, and who made human beings in His image to speak the truth to one another.

◎ See this page's graph and discover linked topics →

In full

The Testimony Argument is a likelihood argument from the epistemology and psychology of testimony. Its first premise, that most of what anyone knows depends on testimony and that relying on it is rational, is defended by Reid, Augustine and, at length, by Coady's critique of Hume's reductionism. Its second premise, that testimony works because truthfulness is the human default and norm, rests on three independent strands: communication research (Levine's truth-default theory; Serota, Levine and Boster's finding that most lies are told by a few prolific liars), philosophy of language (Lewis's account of language use as a convention of truthfulness and trust; Davidson's principle of charity) and moral philosophy (the parasitism of lying, from Augustine and Reid to Kant). Its third premise notes that history as a discipline, including the history of Christian origins (Bauckham), is the weighing of testimony, so knowledge of the past depends on the general reliability of testimony. The spine (P1 to P5) argues that a world in which trusting testimony is rational and lying is parasitic on a norm of truth-telling is more expected if the ground of reality is a truthful, speaking God who made speakers in His image than on naturalism, where deception is often adaptive and honesty must be secured by vigilance and reputation. The strongest rival, the epistemic-vigilance account of Sperber and colleagues, explains a good deal of the data and is answered in the defeater gate. The conclusion is confirmatory and modest.

Format note

This page is a braided case: a valid spine with three supports under each premise (evidence, a comparison of rival explanations and agreement across independent fields), then a knot against double counting, a ledger and a defeater gate. Five fields are braided. History depends on testimony weighed, so the discipline itself presupposes that people are mostly truthful. Language works only if truth-telling is the norm, since words could not mean anything if speakers used them at random. Psychology measures the truth-default and finds that most people are mostly honest. Philosophy shows why trusting testimony is rational and why lying feeds on truth. Theology names the ground: a God who speaks, cannot lie, and made human beings to bear true witness. Other arguments in this format are listed at Ris3n Arguments.

The spine

Validity. P1 to P3 establish the datum: human knowledge, language and history rest on a norm of truthfulness that is actually and widely kept. P4 says the datum is more expected on theism; P5 makes it confirmation. The form is a likelihood comparison closed by modus ponens.

Strength of the conclusion. Confirmatory and modest. Naturalism has a serious account of honest communication, so the ratio in the ledger is not large; the argument's distinct force is in the normativity and the parasitic structure of lying.

P1, Dependence

Inductive strand

1. Augustine counted the cost of refusing testimony. In Confessions VI.5, Augustine reflected on how many things he believed without having seen them, events in history, places he had never visited, what friends and physicians told him, and even who his parents were, and concluded that unless we believe such things we could do nothing at all in this life.
2. Reid made trust in testimony a first principle. Thomas Reid, An Inquiry into the Human Mind (1764), chapter 6, section 24, wrote that "the wise and beneficent Author of nature, who intended that we should be social creatures, and that we should receive the greatest and most important part of our knowledge by the information of others," implanted in us "two principles that tally with each other," a propensity to speak truth and a disposition to believe what others tell us. He called them "the principle of veracity" and "the principle of credulity."
3. Coady's systematic defence. C. A. J. Coady, Testimony: A Philosophical Study (Oxford, 1992), argued against Hume's reductionism that we cannot justify testimony by checking it against our own observations, because the observations we would need, of maps, reports, records and languages, themselves depend on testimony.
4. Children learn the world by testimony. Paul L. Harris and Melissa A. Koenig, "Trust in Testimony: How Children Learn About Science and Religion" (Child Development 77, 2006), showed that children accept testimony about things they cannot observe, such as germs and the shape of the earth, as well as about God and the afterlife.

Abductive strand

Convergence

Patristic philosophy (Augustine), Scottish common-sense philosophy (Reid), contemporary epistemology (Coady) and developmental psychology (Harris and Koenig) reach P1 independently.

Objections and rebuttals

1. "We should believe only what we can check." Rebuttal. Try it for a day. You cannot check your own birth date, most of science, or the meaning of the words you would use to state the rule. The rule refutes itself in practice, which is Coady's point.

P2, The truth norm

Inductive strand

1. People default to belief. Timothy R. Levine, "Truth-Default Theory (TDT): A Theory of Human Deception and Deception Detection" (Journal of Language and Social Psychology 33, 2014), and Duped (University of Alabama Press, 2020), argue that people operate on a default presumption that what others say is basically honest, and abandon it only when something triggers suspicion.
2. Most people are mostly honest. Kim B. Serota, Timothy R. Levine and Franklin J. Boster, "The Prevalence of Lying in America: Three Studies of Self-Reported Lies" (Human Communication Research 36, 2010), surveyed 1,000 American adults: about 60 percent reported telling no lies in the previous 24 hours, and almost half of all reported lies came from 5 percent of respondents. The data are self-reports, but the skewed pattern has been replicated.
3. Language is a convention of truthfulness and trust. David Lewis, "Languages and Language" (Minnesota Studies in the Philosophy of Science 7, 1975), proposed that a population uses a language when there is among them a convention of truthfulness and trust in it. Without such a convention, sentences would not have the meanings they have.
4. Interpretation presupposes truthfulness. Donald Davidson, "Radical Interpretation" (Dialectica 27, 1973), argued that we can interpret another's words at all only by assuming that most of what they say is true, the principle of charity.
5. Lying is parasitic. Reid observed that even "the greatest liars" speak truth far more often than they lie, and that lies are "comparatively few," so that "the authority of human testimony is only weakened by them, but not destroyed." Kant argued in the Groundwork of the Metaphysics of Morals (1785) that a lying promise, made universal, would destroy itself, since no one would believe any promise.

Abductive strand

Convergence

Communication science (Levine, Serota), philosophy of language (Lewis, Davidson), moral philosophy (Kant) and common-sense epistemology (Reid) reach P2 by unrelated methods. Lewis and Davidson were not arguing for theism.

Objections and rebuttals

1. "People lie all the time; studies say everyone lies daily." Rebuttal. Earlier diary studies reported averages of one or two lies a day, but Serota and colleagues showed that the average hides a skewed distribution: most people tell few or none, and a minority tells most. The norm holds; its breakers are few.
2. "Reid said children never lie, which is false." Rebuttal. Correct; Reid overstated. Victoria Talwar and Kang Lee, "Social and Cognitive Correlates of Children's Lying Behavior" (Child Development 79, 2008), found that most children aged three to eight who peeked at a forbidden toy later denied it, and that lying tracks developing understanding of other minds. The argument keeps Reid's structural point, that truth-telling is the default and lying requires effort and motive, and drops his claim about children.

P3, History

Inductive strand

1. History weighs witnesses. The historian's craft is the critical assessment of testimony: who saw what, how close they stood, what they had to gain, whether independent witnesses agree. Reid noted that "the character, the number, and the disinterestedness of witnesses" can give testimony great strength.
2. The gospels as testimony. Richard Bauckham, Jesus and the Eyewitnesses: The Gospels as Eyewitness Testimony (Eerdmans, 2006; 2nd ed. 2017), argued that the gospels embody the testimony of named eyewitnesses and should be read as testimony, not as anonymous community tradition. His thesis is debated among New Testament scholars. See Richard Bauckham and NT Authorship and Eyewitness Apologetics.
3. Scripture presents itself as witness. "And he that hath seen hath borne witness, and his witness is true: and he knoweth that he saith true, that ye also may believe" (John 19:35, ASV); "that which we have seen and heard declare we unto you also" (1 John 1:3, ASV).

Abductive strand

Convergence

Historiography, philosophy of testimony (Reid, Coady) and New Testament scholarship (Bauckham and his critics) agree that history runs on weighed testimony, whatever they conclude about particular texts.

Objections and rebuttals

1. "Testimony to miracles is a special case." Rebuttal. Hume argued so in "Of Miracles." That debate is treated at Argument from Miracles and David Hume. This argument needs only the general reliability of testimony, which Hume also relied on.

P4, The hypotheses

Inductive strand

1. God is truthful and speaks. "God is not a man, that he should lie" (Numbers 23:19, ASV); "God, who cannot lie" (Titus 1:2, ASV); "it is impossible for God to lie" (Hebrews 6:18, ASV).
2. Truthful witness is commanded. "Thou shalt not bear false witness against thy neighbor" (Exodus 20:16, ASV).
3. Lying is named as parasitic and alien. Jesus says the devil "standeth not in the truth, because there is no truth in him. When he speaketh a lie, he speaketh of his own" (John 8:44, ASV). On Scripture's account, the lie is a departure from a truth that was there first.
4. Human testimony points to God's. "If we receive the witness of men, the witness of God is greater" (1 John 5:9, ASV).
5. Naturalism has no norm at the root. On naturalism, communication evolved where it paid. Signalling theorists have long noted that animal signals can be manipulative, and Robert Trivers (The Folly of Fools, 2011) argued that deception, and even self-deception, are widespread products of selection. Honesty is then a fragile equilibrium, held by costs and vigilance.

Abductive strand

On theism, a truth norm is part of human nature because human beings are made in the image of a God who speaks truly (Imago Dei), and lying is a corruption that depends on the norm it violates. On naturalism, honesty must be explained as a stable strategy wherever it occurs, and the normative claim that lying is wrong, not merely risky, has no ground beyond the strategy. Naturalism predicts much honesty where reputations are at stake; it predicts less well the presumption of honesty toward strangers and the felt wrongness of lies no one could discover.

Convergence

Biblical theology (a truthful, speaking God), the common-sense tradition (Reid's explicitly theistic account) and evolutionary signalling theory (which locates the problem honesty must solve) meet at P4.

Objections and rebuttals

1. "Honesty evolved because it pays in the long run." Rebuttal. Often it does, and the argument concedes that this explains a good deal. It does not explain the strangers' default, the private honesty no one will check, or why lying is felt to be wrong rather than merely imprudent. See the defeater gate.

P5, The likelihood principle

Inductive strand

Evidence favours the hypothesis on which it is more expected; see Bayesian Probability.

Abductive strand

Its rival would forbid comparing explanations by fit, which every party to the debate does.

Convergence

Bayesians and likelihoodists agree on the comparative use.

Objections and rebuttals

1. "The prior matters more." Rebuttal. The ledger is prior-agnostic, and priors are argued in the Cumulative Case for Christian Theism.

The knot

The strands are independent: patristic and early modern philosophy (Augustine, Reid), analytic epistemology (Coady), communication research (Levine; Serota, Levine and Boster), developmental psychology (Harris and Koenig; Talwar and Lee), philosophy of language (Lewis, Davidson), moral philosophy (Kant), evolutionary theory (Trivers; Sperber and colleagues) and biblical scholarship (Bauckham). Two overlaps are counted once. First, Levine's truth-default theory and the Serota prevalence study come from one research group, so the ledger counts them as one family. Second, Lewis and Davidson make a related conceptual point, that meaning presupposes truthfulness, and count once. The history strand overlaps with The Record-Keeping Argument, which concerns physical records; this page concerns personal testimony, a different kind of trace.

The ledger

Prior-agnostic: the ledger asks only how expected each datum is under each hypothesis. The ratings are judgments argued above, not measurements.

Net. No row is decisive, and naturalism does well on the first three. The argument's distinctive force is in the last two rows: a norm of truth that binds where no reputation is at stake, and lies that feed on it as parasites feed on a host. The verdict is a modest confirmation that joins the Cumulative Case for Christian Theism.

The defeater gate

Epistemic vigilance. The strongest objection is the account of Dan Sperber, Fabrice Clément, Christophe Heintz, Olivier Mascaro, Hugo Mercier, Gloria Origgi and Deirdre Wilson, "Epistemic Vigilance" (Mind & Language 25, 2010). Humans depend heavily on communication, which exposes them to being misinformed. Selection therefore produced a suite of mechanisms that monitor the reliability of sources and the coherence of content. Because hearers are vigilant, lying is risky, and honest communication becomes a stable strategy. Levine himself presents the truth-default as adaptive for individuals and the species. On this view, a world of mostly honest testimony is just what evolution predicts, and no truthful God is needed.

The reply has three parts.

1. Grant the mechanism; it explains the equilibrium, not the norm. Vigilance explains why lying is often a bad bet. It does not explain why lying is wrong, why people feel guilt for a lie no one could discover, or why we hold that a speaker owes the truth to a stranger he will never see again. The argument's distinctive data are normative, and vigilance is a theory of prudence.
2. Vigilance presupposes the norm it polices. A vigilance system is tuned to detect departures from honesty against a background of mostly honest speech. It is a guard on a structure, and the structure, language as a convention of truthfulness and trust, is prior. That is the parasitism point in another form.
3. Theism includes the mechanism. A Creator who made social creatures to learn from one another would also equip them to detect liars in a fallen world. Reid already said that reason, as it matures, "learns to suspect testimony in some cases, and to disbelieve it in others." Vigilance is not a rival to design; it is part of it.

What the gate concedes: epistemic vigilance is a good scientific account of how honest communication is maintained, and it removes much of the surprise that a naive version of this argument would claim. What remains is the normative and parasitic shape of truthfulness, which fits a truthful Creator better than an equilibrium of interests.

Concessions

• Grant: honest communication pays in the long run. Now collect: "it pays" explains why lies are risky, not why they are wrong.
• Grant: people lie, and some lie a great deal. Now collect: the research shows the lies are concentrated in a few, and every lie depends on the honesty of the rest.
• Grant: Reid was wrong about children. Now collect: children who lie know they are breaking a rule, and they must learn to lie, while truth comes without training.
• Grant: Bauckham's thesis is debated. Now collect: the debate is conducted by weighing testimony, which assumes the very reliability the argument explains.

Cheatsheet

The 30-second version (read aloud). Almost everything you know, someone told you: your birthday, the shape of the earth, most of science. That only works because people mostly tell the truth. Researchers find we naturally believe what we are told, and that most people tell few or no lies on a given day, while a small number tell most of them. Even a liar depends on honesty; a lie only works because you expect the truth. In a world built only by survival, deception would often pay. A world where honesty is the default and lying is a parasite is exactly what you would expect if reality comes from a God who "cannot lie" and made us to speak the truth.

Fast facts:

1. Reid (1764): God implanted "the principle of veracity" and "the principle of credulity."
2. Coady (1992): testimony cannot be reduced to our own observations, which themselves depend on testimony.
3. Levine (2014): people operate on a truth-default.
4. Serota, Levine and Boster (2010): about 60 percent reported no lies in a day; almost half of lies came from 5 percent of people.
5. Lewis (1975): using a language is a convention of truthfulness and trust.

The closing line. A lie is a counterfeit, and counterfeits only work where real money is the rule.

Cross-examination

Round 1, How much do you know by testimony?

• Q: How do you know when and to whom you were born?
• If they say records: records are testimony written down. Could you know anything about the past without trusting someone?

Round 2, Is lying parasitic?

• Q: Could a lie work in a world where everyone lied half the time?
• If they say no: then lying depends on a norm of truth. Where does the norm come from?

Round 3, Risky or wrong?

• Q: If you could tell a lie that would never be discovered and would benefit you, would it still be wrong?
• If they say yes: then honesty is more than a strategy. What grounds it?
• If they say no: then the felt wrongness most people report is an illusion on your view, which is a cost.

Dodge table

Live-cite kit

Scripture (ASV):

• Numbers 23:19: God is not a man, that he should lie, Neither the son of man, that he should repent (Numbers 23:19, ASV)
• Titus 1:2: in hope of eternal life, which God, who cannot lie, promised before times eternal (Titus 1:2, ASV)
• Hebrews 6:18: that by two immutable things, in which it is impossible for God to lie, we may have a strong encouragement (Hebrews 6:18, ASV)
• John 19:35: And he that hath seen hath borne witness, and his witness is true: and he knoweth that he saith true, that ye also may believe. (John 19:35, ASV)
• 1 John 5:9: If we receive the witness of men, the witness of God is greater: for the witness of God is this, that he hath borne witness concerning his Son. (1 John 5:9, ASV)
• John 8:44: When he speaketh a lie, he speaketh of his own: for he is a liar, and the father thereof. (John 8:44, ASV)

Scholarly:

• Thomas Reid, An Inquiry into the Human Mind (1764), ch. 6, sec. 24.
• C. A. J. Coady, Testimony: A Philosophical Study (Oxford, 1992).
• Timothy R. Levine, Journal of Language and Social Psychology 33 (2014): truth-default theory.
• Dan Sperber et al., "Epistemic Vigilance," Mind & Language 25 (2010): the strongest rival.

Aphorism: "Truth is the host. Every lie is a guest that cannot live without it."

Connection to Scripture

• Numbers 23:19, Titus 1:2 and Hebrews 6:18, the God who cannot lie.
• Exodus 20:16, the command against false witness, the norm of testimony.
• John 8:44, the lie as something foreign to truth, spoken "of his own."
• John 19:35 and 1 John 1:1-3, the gospel offered as eyewitness testimony.
• 1 John 5:9, human testimony received, and God's testimony greater.

Scholarship

Published academic works this page draws on, each matched to a publisher, index or library catalog record.

• Paul L Harris and Melissa A Koenig, "Trust in Testimony: How Children Learn About Science and Religion," Child Development 77 (2006). doi:10.1111/j.1467-8624.2006.00886.x PubMed 16686784
• Timothy R. Levine, "Truth-Default Theory (TDT): A Theory of Human Deception and Deception Detection," Journal of Language and Social Psychology 33 (2014). doi:10.1177/0261927x14535916
• Timothy R. Levine, Duped (University of Alabama Press, 2019). Open Library
• Timothy R. Levine, Duped: Truth-Default Theory and the Social Science of Lying and Deception (University of Alabama Press, 2019). doi:10.2307/jj.30297464
• Thomas Reid, An inquiry into the human mind, on the principles of common sense (Alexander Ewing, 1764). doi:10.1037/11974-000
• Kim B. Serota, Timothy R. Levine and Franklin J. Boster, "The Prevalence of Lying in America: Three Studies of Self-Reported Lies," Human Communication Research 36 (2010). doi:10.1111/j.1468-2958.2009.01366.x
• Victoria Talwar and Kang Lee, "Social and Cognitive Correlates of Children’s Lying Behavior," Child Development 79 (2008). doi:10.1111/j.1467-8624.2008.01164.x PubMed 18717895 open access
• Robert Trivers and Santiago Foz (argentino), The folly of fools (2011). Open Library

See also

• Arguments, the master index
• Ris3n Arguments, the hub of ris3n's original arguments
• The Record-Keeping Argument, physical records and the knowable past; this page concerns personal testimony
• NT Authorship and Eyewitness Apologetics and Richard Bauckham, the gospels as testimony
• Argument from Miracles and David Hume, the debate over testimony to miracles
• Faith is Belief Without Evidence Objection Defeater, why trusting reliable testimony is not blind faith
• Belief Vs Knowledge and Justified True Belief, what testimony gives us
• Witness of the Holy Spirit, the testimony of God within
• Spirit of Deception, Scripture's account of the lie
• Immanuel Kant, whose test shows the lie undermines itself
• Imago Dei, speakers made in the image of a speaking God
• Cumulative Case for Christian Theism, where this argument contributes