Finnish version not currently available.
Why is it that different people can look at the same evidence and see it as proving entirely different things? This post introduces what I see as being one reason.
Here’s an example:
But to analyse it explicitly…
The above encapsulates the idea very well, but proofs of God viewed really objectively is something I’ve been meaning to write about separately, so I won’t go further into that now. Let’s take another related example: I have heard of someone being convinced of the truth of Christianity because they knew that the first Christians died for their religion, and they would not have done so if they did not believe in it. That last part, given as the reason, is not a bad assumption.
The inference, then, goes something like this:
- The first Christians would not have died for their beliefs if their beliefs were not true.
- The first Christians did die for their beliefs.
- Therefore, their beliefs were true.
But is the first premise really justified? Obviously, it’s as bit more complicated than that. The above should be:
- The first Christians would not have died for their professed beliefs if they did not really believe they were true.
- The first Christians did die for their beliefs.
- Therefore, they really believed.
This first premise isn’t unquestionable, but it’s pretty solid. Barring other information, it seems like the best and simplest explanation for what they did. But now the conclusion isn’t what we wanted. They believed it, but was it true?
Real historians (at least from what I’ve read) can actually come to the aid of the Christian apologist here. Americans seem to believe, where they don’t simply trust everything in the Bible, that there is no real historical evidence of the existence of the person the English-speaking world now calls “Jesus”, probably because everything there is so politicised that their education system is forbidden from giving a decent basic education about religious matters. However, the evidence of such a person does exist. (The actual name he was called by when he lived was something like “Yeshua”, though.) Historians also have opinions such as this: the evidence suggests that Jesus’ disciples became convinced he had been resurrected, as they would not have acted as they did otherwise.
So, it looks like historians might back up this:
- The first Christians would not have acted as they did if Jesus wasn’t really resurrected.
- The first Christians did act as they did.
- Therefore, Jesus was really resurrected.
But there are two things wrong here, compared what we needed to prove the truth of Christianity. First, as above, it should really be like this:
- The first Christians would not have acted as they did if they didn’t believe Jesus was resurrected.
- The first Christians did act as they did.
- Therefore, they believed Jesus was really resurrected.
I’ll get back to the second problem below, but there’s more to be done here first. We can strengthen this further by referring to something some historians have also said: the whole business in belief in the resurrection, and therefore Christianity starting up in general, would not have worked if Jesus’ body had been in the tomb, because, apparently, people could have gone and checked. (Edited to add: Of course, it’s just some historians; I’ve also read historians have considered it unlikely that he would have got his own personal tomb in the first place, instead being placed in a group grave of some sort, which would appear to nullify this argument. But let’s ignore that for the sake of the example.) So we can say this:
- The first Christians would not have acted as they did if Jesus’ body had not disappeared from the tomb.
- The first Christians did act that way.
- Therefore, Jesus’ body disappeared from the tomb.
That sounds like such a solid case that I can imagine some atheist reading this already denying the premises. (At least before I added the edit above.) But we’re not nearly done here.
Now, you could hardly say that the only way the body could have disappeared was via resurrection. But maybe you could say this:
- The most likely explanation for Jesus’ body disappearing would be his resurrection.
- His body did disappear.
- Therefore, the most likely explanation is that Jesus was resurrected.
The first premise is obviously questionable: such a resurrection would presumably be a miracle (consider this, though: is it necessarily so?), and how could a miracle be the most likely explanation? Historians usually automatically dismiss such explanations as being beyond what they can say anything about and try to find mundane ones.
Well, one could say, if it is suggested by the evidence, then who cares if it is a miracle? Take the facts as they are. I’ve heard of an argument, though I have only read a short summary of it so I may not do it full justice, that one could certainly find evidence of such a thing as the Resurrection if one was not begging the question against it in the first place, not just choosing to consider such explanations as inadmissible.
This argument doesn’t work for the following reason: the historian dismissing supernatural explanations need not be begging the question against them. The one accepting them needs to be begging the question for them. Consider the evidence we have, and the previous formally expressed argument above. Suppose a historian reasoned like that. What would their reason for accepting the first premise be? Presumably that all other explanations would be unlikely based on what we know: it was unlikely people would have behaved as they did if there was a body, it was unlikely that someone would have stolen a body… All of these could be unlikely based on what we know about people (though I’m not commenting on whether they really are). Except, you know, people rising from the dead is even more unlikely. Immensely more so. If you eliminate the more unlikely explanations first, you will not be left with resurrection as a candidate, because it was the most unlikely explanation. You have to ignore its greater unlikeliness and act as if only the other things were unlikely. Begging the question. You’d need a huge amount of evidence pointing to its likelihood and to other explanations being very unlikely before you’d have cause to accept such an explanation. There is a point at which you would have enough evidence. But it’s not reached here, and indeed it would be exceedingly hard in such a distant historical case to ever reach it. We’ve simply got different possible explanations that would fit the facts at hand, and some people are choosing to believe the most unlikely one.
Now, someone might react to this with a new version of the same argument: If we treat evidence like this, we’re making it impossible to come to consider things with such evidence proven. But what if they are true? Then we’ll never know. The answer to this is discussed elsewhere, and it is that the less something is supported by evidence, the less likely it is to be true, quickly reaching the point where it can be safely dismissed. Without evidence, we’re talking about random guesses. Why would we need to account for the possibility that extremely unlikely random guesses might be true? The reason one would is just that they already want to believe something, regardless of evidence and therefore regardless of whether there is any real chance it is true. A method that makes us dismiss possibilities that almost certainly aren’t true is a good one, one that makes us treat them as likely true is a bad one. Stop thinking “This is true, how can I use the evidence to support it?” and start thinking “This is the evidence, what is likely to be true based on it?”
So, I said there were two things that were problematic about the previous argument. The first one was claiming that the first premise was justified by an inference to the best explanation. It wasn’t, as just shown. The second is that we were supposed to be proving the truth of Christianity — but all we got as a conclusion was that Jesus was resurrected.
Since the first premise isn’t sound, the conclusion isn’t either. There is no reason in anything given above to think that Jesus was even likely resurrected. But suppose, just as a thought experiment, that we had such reason and that it really happened. What would follow?
The Resurrection is pretty central to Christianity. But there is a lot more to it all than the mere supposed fact of the resurrection of the historical person called “Yeshua”. You cannot say the following:
- Jesus could have been resurrected only if Christianity was exactly true.
You can’t even say this:
- Jesus could have been resurrected only if he was the son of God.
In fact, you can’t even say this:
- Jesus could have been resurrected only through God’s involvement.
- Jesus could have been resurrected only if God exists.
These were all part of the story Christians were telling, sure. (Well… they really weren’t sure about his being the son of God at that time. Maybe not even thinking about it. Even in the Gospels that theme develops sort of gradually.) But, and this is really the point of the whole article, you cannot use the following as an implicit premise, though people do so:
- If a given story explains certain facts, then those facts can only be explained by that story.
This premise is obviously untrue. If you say you walked here and I say, no, wait, you must have grown wings and flown, that would explain why you are here now, then obviously this is not proof about your wings even though it explains your being here. Besides, your walking here (and having no wings) also explains it. But they cannot both be true. So immediately we get an example of two stories that explain some facts about the situation which cannot possibly both be true. So there is at least one story that explains some facts about the situation and which is not true. This besides the fact that the story about wings was obviously absurd in its own right.
Obvious is obvious. Now, about the resurrecting Jesus example: If he actually had been resurrected (and recall that that’s extremely unlikely), it could have been for other reasons. He might have had an extremely rare condition where you die for a moment and then come back to life. (Now, you might say that’s not real resurrection, or real death, but it would explain about all the same facts, so as an explanation it’s about equivalent.) He might have been an alien playing a prank on us who could do such a thing by biology or technology. He could have been an incarnation of Vishnu instead of Yahweh. None of these explanations are less likely than the elaborate story about God just because that story is what people were and are telling. Remember the part said above in bold: it also and specifically applies when we’re talking about your favourite theory. We’re aware of cultural reasons for people telling this particular story, but not of their having any special further knowledge that would justify its details.
Finally, enough of that example. This is why I’m speaking of “Small Evidence, Big Theory”. The general point is that people like to “prove” their favourite story by finding some isolated facts that it fits (perhaps because it was made to explain them in the first place) and then declare the whole story proven. Time to look at the comic again:
Yes, like that. We have this story with details A, B, C, D, E, F, G, H, I, J and Ö, and then we find proof of B, which sort of maybe implies C but none of the other things, and then we declare the whole thing proven. This should be evidently nonsense. So we could restate the point I made above like this:
- If a piece of evidence proves a small part of your theory but not the rest of it, then that evidence does not prove your theory.
(This, of course, also entails that the rest of the theory is not somehow directly implied by that small part, because then the evidence would prove all of it.)
Consider your typical Agatha Christie story. The detective gathers everyone around and starts explaining to them what actually happened, telling a story and revealing the murderer in the process.
And then the murderer smirks and says: “That’s a nice story. But where’s the proof?”
The story already explained everything. It explained all the clues and brought the whole thing together. Even without further proof admissible in court, it might be seen as an inference to the best explanation.
But it’s not the singular best explanation. Inference to the best explanation is good for creating tentative hypotheses but not for proving them. You can make up all sorts of stories, even plausible-sounding ones, to explain a collection of facts — and importantly, do this by postulating a lot of new facts that are not proven. And that’s why the detective will usually say something like, hey, I know you were wiping out your fingerprints, but there’s this specific place where I bet you forgot to wipe them from. Shall we check? And the murderer goes pale. Because this evidence proves something much more specific, like that they were in a particular place at a particular time, which actually is unlikely to be explained any other way than by their having done it. No other plausible explanations account for it. Unlike a random wild story with unproven details, it has specific evidence pointing to it.
So in sum, the point is that a hypothesis explaining the facts is not enough. To reach the truth, to have a real chance of it being true, you need one that the facts constrain you to accept. And you won’t even notice your hypothesis isn’t like that unless you set out to try to disprove it. People can prove pretty much anything to their satisfaction if they only look for confirmation. Don’t be too fond of your first guesses or some received but not evidence-based wisdom, either; they’re likely untrue.