1.9: Testimony

In the testimony case a person comes to know something when he is told about it by an eyewitness or when he reads it in the newspaper. . . . No obvious deductive inference leads to a probabilistic conclusion in this case; the acceptance of testimony can be based on two consecutive inferences to the best explanation. . . . First, we would infer that the speaker so testifies because he believes what he says (and not because he has something to gain by so testifying, or because he has gotten confused and has said the opposite of what he means, etc.). Second we would infer that he believes as he does because in fact he witnessed what he described (and not because he has suffered an hallucination, or because his memory deceived him, etc.).

—GILBERT HARMAN1

A Letter of Recommendation

I have a good friend in the psychology program. He has asked me to write a letter of recommendation in his search for a new job. I know him pretty well—we have collaborated on a short article and have team-taught on two occasions. I tell his prospective employers that he is a fine teacher and a great colleague and that he will go on to be a major figure in academic psychology someday. Suppose you read my letter and wonder what kind of evidence it provides about the job candidate.

Gilbert Harman, in the previous quote, provides a succinct characterization of how inference to the best explanation can be used to unpack the reasoning involved in accepting the word of others. In most cases where we assess testimony, we have more data to explain than simply what has been said. Minimally, we will know something about the speaker and something about the context in which the statement was made. The abstract model looks something like the following. First of all, we have the information contained in the language:

e₁. Linguistic statement—“He is a fine teacher . . . major figure.”

Almost as important in this case is the context in which the statement is offered:

e₂. Context—letter of recommendation.

Finally, we know something about the letter writer himself:

e₃. Relevant biography—philosophy professor at a small state university.

The explanatory or interpretive question is, Why did this speaker (biography) in this circumstance (context) say this (statement)? In ordinary communicative contexts, we generally give people the benefit of the doubt. Until given reason not to, we are presupposed to believe that our interlocutors are being straight with us. The conventions of normal linguistic communication ask you to first consider a theory that explains all these data in terms of sincerity:

t₀. The letter writer said it because he believed it to be true—he believed that his friend was a good teacher and a great colleague and had the potential to make significant contributions to his field.

Unfortunately, years of reading these sorts of letters have made some of us a little cynical. We can immediately conceive of two alternative explanations of the letter writer’s linguistic behavior.

t₁. The letter writer said it because he wants to get his friend a job.

t₂. The speaker said it to get rid of an undesirable colleague.

Inference to the best explanation asks us at some point to commit ourselves to a judgment of explanatory plausibility. What is the best explanation of what the letter writer said? Basically, our answers fall into two categories. We will either judge that the best explanation of the statement is the original one that normal communication recommends—he said it because he believes it; he is sincere. Or we will prefer one of the rival explanatory accounts that offer some other reason for his having made the statement. In this latter case, his testimony is of no use to us, indeed we should discount it. Even if it turns out that his friend is a great candidate for the job, if we judge that the letter writer is insincere or dishonest, his testimony is unreliable evidence about this.

If we do give him the benefit of the doubt on the question of sincerity, we must go through a whole other level of assessment before we can put complete confidence in the truth of his statement. The first level of evidence evaluation yields some new data that must also be explained.

e₄. The letter writer is saying these great things about his colleague because he sincerely believes them to be true.

Why does this person (biography) believe these things (the content of the statement)? Once again, the presuppositions of normal communication ask us to endorse a standard explanation for most sincere communicative attempts.

t*₀. The speaker believes this because he knows what he is talking about—he believes it because it is, in fact, true.

Thus when we accept information through the testimony or authority of others, we tacitly engage in a dual explanatory inference. We explain the linguistic act as a sincere attempt to communicate the speaker’s belief and then explain the speaker’s having the belief in terms of the speaker knowing what they’re talking about.

Larry Wright has helpfully distinguished two quite different things that can go wrong when someone communicates a sincerely held belief. Sometimes people have unreliable access to the information they are trying to communicate. Thus a rival explanation of my belief that my friend is a good teacher might be that I have only observed him in specialized upper-division courses that would be of interest to philosophy and psychology majors—I have never observed him, for example, in large introductory courses.

t*₁. The letter writer believes that his friend is a good teacher because he has never observed his lousy teaching in large introductory courses.

Even when authorities possess excellent access to information, we still worry sometimes about their ability to reliably interpret this information. In this context, the cautious letter reader might have at least two potential worries about my testimony. The first has to do with specialized training. Obviously, my claims presuppose some fairly technical knowledge about pedagogy, academia, and research standards in contemporary psychology. One would like to think that expectations for teaching and collegiality would not vary across the humanities and natural and behavioral sciences. I, hopefully, have the necessary background to provide relevant information about these aspects of my friend’s career. But what about the prediction for professional distinction with respect to his research? I am trained as a philosopher, not a psychologist. Perhaps his psychological “insights” I observed in the course of our collaborative teaching and writing are common knowledge in the field. Or worse, maybe they are discredited or eccentric. Am I really qualified to say? A rival explanation once again suggests itself.

t*₂. The letter writer believes his colleague will make a name for himself because of his lack of knowledge about contemporary academic psychology.

A very different worry about the reliability of my belief focuses on my ability to “objectively” process the information to which I do have reliable access. Basically, the worry here is one of perceptual or interpretive bias. Perhaps I so admire his pedagogic technique because it is so similar to my own ineffective classroom style. Or maybe I am so impressed with his psychological hypotheses because they nicely coincide with my own half-baked notions. He is, after all, my good friend—might I not be guilty of “seeing more with my heart than with my eyes?” So we have yet another category of rival explanation:

t*₃. The letter writer believes these grossly inflated things about his friend because of some sort of perceptual bias.

None of the above should be taken to suggest that testimony is inherently unreliable. What could be more obvious than the fact that almost everything we claim to know comes to us secondhand through the word of others? What I am suggesting is that our assessment of testimony can be structured and critically evaluated as a kind of evidence—evidence that perfectly fits the inference-to-the-best-explanation recipe.

Testimony regarding Miracles

In David Hume’s monumental book, An Enquiry Concerning Human Understanding, there is a very short little argument with astounding religious implications. Hume argues that we are never justified in accepting the testimony of others that a truly miraculous event has transpired. But since the three great theistic religions—Judaism, Christianity, and Islam—all depend, at some foundational level, on reports of miracles, Hume’s argument seems to threaten their intellectual legitimacy. A full analysis of Hume’s argument, let alone a full philosophical investigation of miracles, would be the subject of a whole book, maybe a whole career. Nevertheless, we now possess the tools to at least lay out the structure of Hume’s argument and perhaps to begin the process of evaluating his evidence. So what exactly is the argument?

A miracle is a violation of the laws of nature; and as a firm and unalterable experience has established these laws, the proof against a miracle, from the very nature of the fact, is as entire as any argument from experience can possibly be imagined. Why is it more than probable, that all men must die; that lead cannot, of itself, remain suspended in the air; that fire consumes wood, and is extinguished by water; unless it be, that these events are found agreeable to the laws of nature, and there is required a violation of these laws, or in other words, a miracle to prevent them? Nothing is esteemed a miracle, if it ever happened in the common course of nature. It is no miracle that a man, seemingly in good health, should die on a sudden: because such a kind of death, though more unusual than any other, has yet been frequently observed to happen. But it is a miracle, that a dead man should come to life; because that has never been observed in any age or country. There must, therefore, be a uniform experience against every miraculous event, otherwise the event would not merit that appellation. . . .

The plain consequence is (and it is a general maxim worthy of our attention), “That no testimony is sufficient to establish a miracle, unless the testimony be of such a kind, that its falsehood would be more miraculous, than the fact, which it endeavours to establish. . . .” When anyone tells me, that he saw a dead man restored to life, I immediately consider with myself, whether it be more probable, that this person should either deceive or be deceived, or that the fact, which he relates, should really have happened. I weigh the one miracle against the other; and according to the superiority, which I discover, I pronounce my decision, and always reject the greater miracle. If the falsehood of his testimony would be more miraculous, than the event which he relates; then, and not till then, can he pretend to command my belief or opinion.

In the foregoing reasoning we have supposed, that the testimony, upon which a miracle is founded, may possibly amount to an entire proof, and that the falsehood of that testimony would be a real prodigy: But it is easy to shew, that we have been a great deal too liberal in our concession, and that there never was a miraculous event established on so full an evidence.2

Let’s begin with the middle paragraph. Someone reports seeing a dead man restored to life. If we treat this report as potential evidence a genuine miracle has occurred, we would schematize this testimony as follows.

e₁. Linguistic statement—“I saw a dead man restored to life.”

e₂. Context—where, when, and how we were told

e₃. Relevant biography—whatever we know about the person who tells us this

---

t₀. He genuinely believes that he saw a dead man restored to life.

---

t*₀ A dead man was restored to life.

Hume now considers two rival explanations, one for each of the inferences: “This person . . . deceive[s].”

t₁. He does not really believe that he saw a dead man restored to life.

Or “this person . . . [was] deceived.”

t*₁. He was mistaken in thinking he saw a dead man restored to life.

Hume proceeds to combine t₁ and t*₁ into what logicians call a disjunction—“t₁ or t*₁”—“he deceives, or he was deceived.” Although he was not explicitly using inference to the best explanation, we can take Hume as proposing a rival explanation.

t*₂. He does not really believe that he saw a dead man restored to life, or he was mistaken in thinking he saw a dead man restored to life.

Hume then implicitly moves to step 3 in the recipe and offers a rank ordering of the original explanation of the testimony regarding a miracle and the disjunctive rival explanation. Theory t*₂ is a better explanation of what was said than t*₀.

Why is he so confident of this ranking? The answer is what Hume, and almost every philosopher and theologian since, means by something being a miracle. Miracles are violations of laws of nature. Given the laws of physics, biochemistry, and biology, the natural world dictates that death is permanent. The very evidence that establishes these laws of nature automatically counts against the reported miracle. For Hume, it’s obvious that the various law of nature hypotheses are so much better explanations than rivals that allow for exceptions to these laws that miracles are doomed to be exceedingly implausible.

I agree with Hume about this so far. If a casual stranger tells me that she has witnessed a miracle, I would almost certainly judge that she is either lying or honestly mistaken and not that there has been an interruption in the operations of the natural world. But my judgment is based on a subjective assessment of the plausibility of differing explanatory accounts—classic application of the inference-to-the-best-explanation recipe—and not the meaning of the term miracle. Theists are not claiming, in my judgment, that laws of nature don’t hold or that the evidence for them is in anyway flawed. They believe, rather, that an omnipotent God created the entire physical world, including those laws of nature, and can, if He chooses, supersede those laws by the exercise of His omnipotence. Miracles, therefore, presuppose the existence of God. Testimony about the occurrence a miracle might count as good evidence for the existence of God, but as Hume saw, it will always face serious difficulty. For me to take seriously this kind of testimony, it would need to come from very special sources, so e₃ would be very important. Further, I suspect that regardless of the qualities of the source, I would need the corroboration of lots of equally good sources. Finally, I think I would need some independent evidence that God might exist. To see whether such independent evidence available is one of the oldest questions in Western philosophy. If you are curious about using evidence, indeed using inference to the best explanation (IBE), to present arguments for and against the existence of God, I invite you to read my next book, Religious Evidence, or perhaps some of my articles that attempt to apply inference to the best explanation as a means of assessing the quality of evidence we have for the existence of God.3

EXERCISES

1. 1. Why do the conventions of normal communication recommend that we assume honesty when someone tells us something?
2. 2. Assuming that I was being totally honest in my letter of recommendation, what were the two of worries that one might have that I knew what I was talking about?
3. 3. You read the following on a Facebook post: “I just observed a miracle—I saw a dead man restored to life!” Use the tools we have developed in this chapter to assess the quality of evidence you have for the claim that a dead man has been restored to life.

QUIZ NINE

On November 6, 2012, Donald Trump tweeted the following: “The concept of global warming was created by and for the Chinese in order to make U.S. manufacturing non-competitive” (twitter.com/realdonaldtrump/status/265895292191248385?lang=en).4

Use the tools we have developed in this chapter to assess the quality of evidence you have for the claim that global warming is a hoax perpetrated by the Chinese to damage US manufacturing.

Notes

1. Gilbert Harman, “Knowledge, Inference, and Explanation,” American Philosophical Quarterly 5 (1968): 167.

2. David Hume, An Inquiry Concerning Human Understanding, ed. L. A. Selby Bigge (Oxford: Calendon, 1902), 114–16.

3. See Jeff Johnson, “Inference to the Best Explanation and the Problem of Evil,” Journal of Religion 64, no. 1 (1984): 54–72; Jeffery L. Johnson, “Inference to the Best Explanation and the New Teleological Argument,” Southern Journal of Philosophy 31 (1993): 193–203; Jeffery L. Johnson, “Explanation, Evidence, and Mystical Experience,” Minerva—an Internet Journal of Philosophy 5 (2001): 69–93; Jeffery L. Johnson, “From Friendly Atheism to Friendly Natural Theology,” Minerva—an Internet Journal of Philosophy 7 (2003): 125–42; and Jeffery L. Johnson and Joyclynn Potter, “The Argument from Language and the Existence of God,” Journal of Religion 85, no. 1 (January 2005): 83–93.

4. Donald J. Trump (@realDonaldTrump), “The concept of global warming was created by and for the Chinese in order to make U.S. manufacturing non-competitive,” Twitter, November 6, 2012, 11:15 a.m., twitter.com/realdonaldtrump/status/265895292191248385?lang=en.