In looking up something I'd written awhile back, I ran across this comment, which I posted in another thread two years ago, in response to a query whether I categorically disbelieved all testimony. I've done some light editing, mostly to change from first-person to third-person.
* * *
In a nutshell, my view of testimony is basically that of my profession, viz., Anglo-American jurisprudence.
We should be initially skeptical of all testimony, and should indeed initially doubt it on principle. We shouldn't take any consequential action in reliance on particular testimony until we're shown sufficient other evidence to persuade us that, for the matter in question, the potential consequences of error make it reasonably safe to take a chance on the testimony.
Circumstances alter the case, of course. For a decision with comparatively little downside, it should take less to persuade us of the testimony's reliability.
At a minimum, though, we shouldn't credit someone's testimony, allowing it to influence our decisions — and even that's not the same as giving the testimony conclusive weight, that is, letting it make up our minds for us about what to do — unless at least the following conditions are met:
(a) Does the testimony have sufficient foundation? Before we rely on testimony in a consequential matter, we should insist on being presented with sufficient evidence to persuade us that the witness has reason to know what s/he's talking about. This is called "laying a foundation"; a common event in trial work is "objection, your honor, lack of foundation."
We impose the foundation requirement because we want to know how the witness came to know what s/he claims to know. From millennia of experience, we know that people can readily say things that they think they "know," whereas on closer scrutiny it turns out that they really didn't know it at all; instead, they were imagining things, or they were repeating what they'd heard from someone else, or they were simply remembering incorrectly.
Here's a canonical example: If the witness wants to testify that the light was green, the other lawyer will object, and the judge won't allow the jury to hear the testimony, unless it's first established — e.g., by the witness's own sworn testimony — that the witness was in fact in a position to see the light at the time in question. If the witness is wearing glasses while testifying, we're going to ask whether he was wearing glasses at the time in question (cf. the cross-examination scene from the movie My Cousin Vinnie). And so on.
Now of course, in everyday non-courtroom life, we routinely apply a looser variation of the foundation requirement — assuming, that is, that the evidence otherwise seems sufficiently trustworthy for its intended purpose.
Here's an example. Suppose my wife says "take your umbrella, it's going to rain." Normally I'm not going to cross-examine her about how she came to that conclusion. I'm going to assume that she looked at the newspaper, or heard the weather report on the radio, or looked outside and saw threatening clouds. In that case, I'm willing to get by with less foundational evidence because the downside of an error is essentially zero. (If my wife is on an out-of-town business trip when she says it's going to rain, I'm going to be at least mildly curious how she knows what she's talking about.)
In contrast, if my doctor tells me I've got cancer of the testicles and need to have them immediately cut off, you can bet I'm going to explore the foundation for his assertion before I let the surgeon take a scalpel to me.
The Prologue of the Gospel of John is a classic example of lack of foundation. The author asserts that certain things happened at the beginning of time. But we have no evidence, direct or indirect, to suggest that "John" knew what he was talking about. Scripturalists want us to rely on that passage from John (and many, many other foundationless passages in the Bible), in making perhaps the biggest decision of our lives, namely to accept the assertion that Jesus was God incarnate. IMHO, that decision is far too important to base on foundationless assertions like that.
The factual testimony of the New Testament documents is replete with similar examples of lack of foundation. Some of the testimony I'm willing to accept, despite its lack of foundation, for reasons I'd have to look at on a case-by-case basis. As to other foundationless testimony, however, I'm not willing to accept it.
(b) Is the likelihood of unacceptable distortion sufficiently low? Before we rely on testimony, the surrounding evidence should make us comfortable that the testimony is likely to be "sufficiently" free from distortion to be trustworthy for its intended purpose.
Again from long experience, we know that testimony can often be distorted, in a variety of ways.
(i) Testimony can be distorted as it is passed along from person to person, such as in the "telephone game" — that's why we have a hearsay rule; or
(ii) Testimony can be distorted by the witness having an apparent bias, or an agenda, or an axe to grind, or any of a number of possible memory problems — that's why we have a requirement for cross-examination, so that we can explore these possibilities to get the complete picture. Cf. the rule in Numbers and Deut. prohibiting putting anyone to death except on the testimony of at least two witnesses, as well as the Commandment against bearing false witness against thy neighbor.
(Elsewhere, I've written about indications of possible distortion in the various New Testament documents, including for example indications of possible bias in the Gospel of John and in Paul's letters.)
Again, circumstances alter the case. For a highly consequential decision, though, we should want to be very confident that the testimony hasn't been distorted. (Cf. again the OT rule about not putting someone to death except on the testimony of at least two witnesses.) For a less-consequential decision, we needn't be quite so cautious.
(c) Have alternative possibilities been given sufficient weight? When a doctor diagnoses an illness and prescribes treatment, she normally tries to rule out problematic alternatives. When a patient comes in and "testifies" that his throat is sore, it could be just a symptom of a common cold that can be treated by gargling with salt water. But it could also be strep throat, so the doctor will often order a strep culture, and sometimes will preemptively put the patient on an antibiotic just in case.
Similarly, when a New Testament witness offers the conclusory testimony that (for example) Jesus was raised from the dead, we need to identify exactly what the witnesses perceived, and to explore possible alternative explanations for the evidence, before we make major, life-changing decisions on that basis.