Take My Word for It: On Testimony

Author: Spencer Case
Category: Epistemology
Word Count: 975

Consider two lives. Helen Keller was born blind and deaf, but her access to the written word enabled her to succeed as a writer and activist. Peter the Wild Boy, discovered near Hanover, Germany in 1719, possessed sight and hearing, but was innocent of language. He left the world as innocent as he came. “Without artifice, particularly the shared human artifice of speech, an unmeaning silence traps Peter in an unvarying bestiality” (Newton 2002, 44. Emphasis mine).

The stark comparison illustrates the degree of our dependence on the word of others. Without the ability to give and receive testimony, we cannot transcend the limitations of our individual faculties to borrow from the thoughts, observations, and experiences of others. The kind of knowledge that separates human beings from the rest of the animal world disappears. While a deficiency in one’s sensory faculties can perhaps be compensated for, inability to draw from the knowledge of others forecloses all possibility of higher knowledge.

The contemporary philosophical literature on testimony can be viewed as an elaborate set of footnotes to the debate between the great rivals of the Scottish Enlightenment: David Hume and Thomas Reid. The positions of Hume and Reid are not exhaustive, but the rift between them represents the continental divide in the intellectual landscape. The main point of dispute between them is whether testimony is a derivative source of knowledge (the “reductive theory”), or an irreducible, basic source of knowledge (the so-called “non-reductive theory”).

To the outsider this debate might seem pedantic. Even most academic philosophers spend relatively little time on it. So what is the big deal? At stake is whether we should accept a vision of human knowledge that prioritizes autonomy, avoidance of gullibility, and skepticism of external authority, or one that prioritizes trust and acknowledges our inescapable dependence on authority.

For Hume, we are justified in accepting testimony only to the extent that we have independent reasons for thinking the speaker is telling the truth (Hume 1967, 196). We must have prior knowledge that the speaker is honest, or recognition that what is said coheres with other things we know. To believe testimony that has not been so accredited is to be guilty of the sin of gullibility.

Reid, by contrast, claims there is a standing entitlement to accept testimony, though the entitlement can be nullified by sufficient counter-evidence. This “Principle of Credulity” results in a more egalitarian relationship between testimony and the senses (Reid 1997, 195-200). Being told by a fellow student that Professor Smith has just entered his office and seeing him enter his office are both sufficient reasons for believing that he is there, though they are not necessarily of equal weight.

In claiming that testimony is a basic source of justification for the hearer, Reid isn’t denying that testimony can work only in conjunction with other sources of knowledge. Just as memory extends pre-existent knowledge across time, testimony transfers pre-existent knowledge from one person to another. Testimony, unlike other sources of knowledge, is irreducibly social and morally significant. It cannot operate without trust on the part of the hearer and honesty on the part of the speaker.

In the contest between these two views, the reductive theory (Hume’s view) enjoys the advantage of simplicity. It postulates one fewer basic sources of knowledge. Theoretically, we don’t want to add sources of knowledge willy-nilly—each one must be explanatory or we’ll end up postulating psychic powers and the like. So the friend of the non-reductive theory is under pressure to explain how the additional basic source earns its keep.

C.A.J. Coady, whose work on testimony played a major role in reviving interest in the topic, claims the non-reductive theory of testimony is essential to avoiding skepticism. Few, if any, have witnessed the complete path of one letter, yet everyone knows how the postal service works. “Similarly,” Coady writes, “that babies are born of women in a certain way, is known to all of us and it is a fact of observation, but very many of us have not observed even one birth for ourselves” (Coady 1992, 81).

We flatter ourselves by thinking knowledge comes by our own devices, and not though implicit reliance on the vast, un-vetted pool of collective experience we take on trust. Ronald Reagan said, “Trust, but verify.” The Gipper would probably concede Coady’s point, though, that we place our trust in far more things than we can possibly verify “off of our own bats.” Given our extensive dependence on trust, attempting that would lead us to skepticism about ordinary things. Positing testimony as a basic source of knowledge is justified because it is necessary to help us avoid this unhappy result.

In theory, defenders of the reductive theory could simply “bite the bullet” and embrace skepticism. In practice, that never happens. Philosophers hate to admit that they don’t know things. So instead, they retort that testimony can deliver the goods as a non-basic source of knowledge. Rational acceptance testimony doesn’t require vast amounts of fieldwork, only the recognition that human beings normally tell the truth—something that individual experience can establish.

Armed with that knowledge, we can infer that particular pieces of testimony are likely to be true without positing testimony as a basic source of knowledge. Variations of this line of response have been convincingly developed by Paul Faulkner (Faulkner 2011, 32-39), Elizabeth Fricker (Fricker 1994, pp. 125-161), and David Lyons (Lyons 1997, 171) among others.

Such replies serve to blunt the main offensive of the non-reductive theory, but at a cost. Once friends of the reductive theory endorse a norm of accepting what we are told, they are in less of a position to distance themselves from their rivals’ theory on the grounds that it sanctions gullibility. Neutralizing the observation argument does not win the game, it only resets the board. The reductive theorist plays white; his one-tempo starting advantage is the appeal to simplicity. Given the strength of both sides’ arguments, the game will probably continue for some time.


Coady, C.A.J. 1992. Testimony: a Philosophical Study. New York: Oxford University Press.

Faulkner, Paul. 2011 Knowledge on Trust. New York: Oxford University Press.

Fricker, Elizabeth. “Against Gullibility.” In Knowing from Words. Ed. B.K. Matilal and Arindam Chakrabarti. 1994. Dordrecht: Klewar.

Hume, David. Treatise on Human Nature. 1967. Ed.  L.A. Selby-Bigge. USA: Oxford University Press.

Hume, David. An Enquiry Concerning Human Nature.  1958 ed. L.A. Selby-Bigge. USA: Oxford University Press.

Lyons, Jack. “Testimony, Induction, and Folk Psychology.” Australasian Journal of Philosophy. 75:2, 163-178. 1997.

Newton, Michael. 2003. Savage Girls and Wild Boys. Thomas Dunne Books. 1st U.S. Edition.

Reid, Thomas. 1997. An Inquiry into the Human Mind on the Principles of Common Sense: a Critical edition. Ed. Derek R. Brookes. Pennsylvania State University Press.

Spencer is currently pursuing a doctorate in philosophy at University of Colorado at Boulder. Previously, he served as a public affairs specialist for the U.S. Army in both the Iraq and Afghanistan wars. He also spent nine months in Egypt as a Fulbright Student Grant recipient to study Islamic philosophy. His main interests are ethics, meta-ethics, and comparative philosophy.