Thursday, June 16, 2011

The Hallucination Hypothesis

For several years now, I've been wanting to write a series of blogs where the first half of the series argued that Jesus considered himself to be the Jewish messiah and the second half of the series argued that Jesus was raised from the dead. I read so much on the subject and had so many thoughts on it that my stack of notes became formidable, which made it difficult for me to muster the enthusiasm to finally write it all down. So I've been putting off blogging about little pieces of the puzzle individually because I thought I'd eventually write this whole series. Well, since it's been so many years and I have nothing to show for it, I thought I'd go ahead and start blogging on some of the pieces. For all I know, I may never write the whole series, and I hate for that bee in my bonnet to keep buzzing around.

I'm pretty convinced that the apostles saw what they took to be a living breathing Jesus some time after he had died. I think that is a really good explanation for why they continued to believe that Jesus was the promised messiah even after he had been killed by the Romans. But the question is this: What did they actually see? The other question is this: Why did seeing it cause them to believe Jesus had risen from the dead?

One of the common responses apologists give to the hallucination hypothesis is that hallucinations are, by their nature, subjective experiences. You can no more see somebody else's hallucination than you can see somebody else's dream, and the reason is because hallucinations and dreams only happen in the mind. There's no objective reality outside the mind that more than one person could see. And since the appearances of Jesus happened in groups, it could not have been a hallucination.

There are two responses to that argument I have heard. I came up with one of them myself a few years ago. In my theory, Peter was actually the only one who saw Jesus. The others believed his story and soon began to claim that they had seen Jesus as well even though they didn't. But they each believed each other's testimony, and that strengthened the faith of each of them that Jesus really had risen from the dead.

This theory is basically a denial that the appearances happened in groups. Since there's no group appearances, the hallucination hypothesis remains viable. However; it appears from the gospels and 1 Corinthians that the appearances happened in groups. How do we account for that? Well, it isn't hard to imagine how that might've happened given my theory. Suppose one of the apostles is preaching the gospel in a new town, and he tells them that Jesus was raised from the dead. Somebody says, "Did you see him yourself?" The apostle replies, "Oh yeah. We ALL saw him." That would be consistent with my theory because, in fact, all of them claimed to see Jesus. But the listeners mistakenly infer that the apostles saw him at the same time. So when some of the listeners begin to tell others about the gospel, they portray the appearances as if they were group appearances. It's an easy mistake, and it could've happened early on. You might think that if this misunderstanding was widespread during the lifetime of the apostles they would've corrected it, but I don't know if it would've been worth correcting from their point of view. From their point of view, would it really matter whether they all saw Jesus individually or at the same time? I can't think of a motive for the apostles to go to the trouble of straightening out this detail.

I think the appearances probably did happen in groups, though, for a couple of reasons. In Paul's appearance traditions he quotes in 1 Corinthians 15, he says that Jesus appeared to Peter, then to the twelve. He also says Jesus appeared to James, then to all the apostles. These two traditions seem to be reports of the same appearances but from different sources. And I've read this in some of the scholarly literature. They think the first tradition came from Peter or some of his followers, and the other one came from James or some of his followers. That's why you have Peter at the head of one tradition and James at the head of the other even though "the apostles" and "the twelve" refer to the same group of people. So these appearance traditions come right from the source. They aren't the end product of a long telephone game. Pinchas Lapide went so far as to say that they can be taken as eye-witness testimony from the apostles themselves. These two appearance traditions don't necessarily entail group appearances, but they do strongly imply it. That's one reason I think the appearances happened in groups.

The other reason is because Paul mentions these 500 witnesses. I don't think Paul is quoting an oral tradition at this point because he makes a parenthetical comment about them, that most of them were still living, though some had fallen asleep. That couldn't have been an oral tradition because if it was, it would have to be continually updated as more and more of them died. Apologists often take this as Paul challenging the Corinthians to check out the claim. They could ask these witnesses themselves. The weakness, of course, is that Paul doesn't identify any of these witnesses. Still, the parenthetical comment makes me think that a large number of people must've claimed to have seen Jesus. Paul would have to have been intentionally making things up to claim that some of them had died. It's possible Paul may have been exaggerating the number or just guestimating, but his parenthetical comment makes me believe they really existed. I don't think it's likely that such a large number of people would claim they saw Jesus individually. Given such a large number of people, I think it's likely that at least some of them claimed to see Jesus in groups. After all, how could Paul even come up with an estimate if there were just random individuals here and there claiming to have seen Jesus? I think it's far less likely that a group of people would lie about an appearance than that an individual would lie about an appearance.

N.T. Wright makes another good point. If people were individually claiming to see the risen Jesus, it's inexplicable that these appearances seem to have suddenly stopped. From all our sources, it appears that Paul's appearance is odd in the fact that it happened much later than everybody else's, and nobody after Paul ever claimed to have seen the risen Jesus. If people were going about claiming to have seen Jesus just to be cool, we shouldn't expect those claims to all of a sudden stop.

So I think there probably were group appearances. But I don't think that necessarily undermines the hallucination hypothesis, which brings me to my second response to the claim that hallucinations don't happen in groups. The most common counter-example is Marian apparitions, which involve groups of people claiming to see Mary. Now let's face it. People often do claim to see things in groups even though there's nothing really there to see. You see this kind of stuff all the time. If I saw a face in a cloud that nobody else saw, but then I pointed it out to other people, pretty soon, they'd see the face, too. They'd see it because they were looking for it. They expected to see it. That's the only reason anybody ever saw Mother Theresa in a cinnamon roll. Maybe the apostles saw something or somebody who resembled Jesus, and they believed it was him.

One problem with this line of reasoning is that the apostles likely were not expecting to see Jesus. One reason, as Bill Craig often points out, is that Jews who believed in resurrection all seemed to think of resurrection as an eschatological event. It was something that happened on the last day, not in the middle of history. And the resurrection was supposed to be general, not individual. I went into detail about this point in November and December 2005 in my series on the resurrection in parts 1-5, so I won't repeat the arguments here. The other reason the apostles probably weren't expecting the resurrection is because resurrection wasn't an obvious part of what the messiah was supposed to do. The messiah was supposed to sit on the throne of David, reestablish national sovereignty, and usher in an era of peace and prosperity. Instead of running the Romans out, Jesus was crucified by them and subjected to publish humiliation. To any Jew, that would've signified that he was a failure. And, in fact, no messianic movement other than Jesus' survived the death of its leader, and there were quite a few of them in the first and second centuries. Given common Jewish expectations about the messiah, we should expect the apostles to have given up hope after Jesus was arrested and crucified. And, as it turns out, that's exactly what the gospels report.

On the other hand, the gospels also report that Jesus predicted his resurrection. It says the apostles couldn't really wrap their heads around what he was saying, but if Jesus really did predict his resurrection, then maybe they were holding out some remote hope that he might rise from the dead. If so, then maybe they were looking for him. If so, then maybe they saw him just like people see faces in clouds or Mother Theresa in a cinnamon roll.

C.S. Lewis pointed something out in his book on Miracles that seems to count against the idea that the apostles were expecting to see Jesus. He pointed out that on a few occasions where Jesus appeared to the apostles, they did not immediate recognize him. Assuming that really did happen, I agree with him that it counts against the notion that the apostles saw Jesus because they were expecting to see him or because they were looking for him. It would seem odd not to recognize the object of your own mental projection. The only difficulty is trying to establish that particular detail of the narrative. It's one thing to say Jesus appeared to the apostles; it's another thing to say he appeared in a particular way at a particular place and had a particular conversation, etc. I haven't thought about it that much, but it seems to me you'd have to come up with some possible motives for why the authors would include that detail if it didn't happen. Does it serve some apologetic or rhetorical purpose?

The major problem I have with the hallucination hypothesis is that it doesn't really explain why the apostles came to believe Jesus had risen from the dead. Think about this for a minute. Think of somebody you know to have died, like a relative or something. Maybe your parents. What would you honestly think if you saw that person standing right in front of you right now? It seems like you'd have a few options:

1. You're dreaming.
2. You're hallucinating.
3. You're seeing a ghost.
4. The person never died to begin with.
5. The person has risen from the dead.

I put this question to my daughter a while back when I first thought of it. She has a cousin who died three years ago, so I asked Grace what she would think if she saw Madeline standing right in front of her, and I gave her some of these options. I left out the dreaming one because I didn't think of it at the time, but Grace said she'd think she was dreaming. That wasn't what I expected her to say. I expected her to say it was a ghost because that's probably what I would think. Plus, Grace's mother used to be a ghost hunter, and Grace was kind of interested in the subject. I don't think I'd assume I was dreaming. I might entertain the thought at first, but it wouldn't take much to convince me that I was awake. I'd probably rub my eyes a lot and look really carefully. My mind would be reeling, trying to make sense out of it. I've seen weird stuff before, so I know this is probably how I'd react. The very last thing I would think was that she had risen from the dead. The second to last thing I'd think was that she had never died. The first thing I'd probably think once I came to my senses was that I was hallucinating. But after rubbing my eyes for a while, walking around and looking at different angles, and maybe even talking to her, I'd probably come to the conclusion that I was seeing a ghost. I suspect that's what most people would think in a similar situation.

And, it turns out that's exactly what the gospels report that the apostles thought (Luke 24:37). They thought they were seeing a spirit. Luke goes on to say that Jesus corrected their misunderstanding by pointing to the scars on his hands and feet and then eating in front of them. John's gospels reports that Thomas wanted to actually touch Jesus before he'd believe. 1 John begins with "what we have heard, what we have seen with our eyes, what we have looked at and touched with our hands..." Why all this emphasis on the physicality of Jesus? Why the emphasis on touching and eating to prove that Jesus really had risen from the dead? Well, I think it's only natural because nobody would conclude that a dead person was physically alive, walking and talking, just because they saw a vision or hallucination. It would take a lot more than that. Seeing Elvis made people think he never died, not that he had risen from the dead. Seeing Mary was only consistent with what Catholics already believed--that Mary never died, not that she had risen from the dead. If I saw my niece alive who I know to be dead, it would take a lot to convince me she was really alive. I'd have to touch her with my own hands and see that she was tangible, and it would help to see her eat something. And that's probably what I'd say to people if I were telling them about it. I'd say, "Holy cow, I didn't believe it myself until I actually touched her. I mean I felt her with my own hands, and she was as tangible as anything! She sat right here and ate a veggie burger." Madeline was a vegetarian, by the way.

I want to point out a weakness in relying on the appearance traditions that Paul quoted in 1 Corinthians 15. I wanted to stick this somewhere in the body of my blog here, but I couldn't find a place for it that didn't destroy the flow. So I'm just sticking it right here at the end. The Mormons have, at the beginning of The Book of Mormon, a testimony signed by several people saying that they saw the golden plates from which Joseph Smith translated the Book of Mormon. Mormons believe these testimonies provide evidence that the golden plates really existed even though we no longer have them because the angel, Moroni, took them away once Joseph Smith finished the translation. If all we had was this written testimony, it might seem persuasive. After all, these people signed it. They endorsed it. But that isn't all the information we have. It turns out that a few of them wrote about their experiences later on. When you read their individual accounts, it becomes ambiguous whether anybody actually saw the plates at all. The three witnesses didn't see the plates. They went out in the woods and prayed to be able to see them and ended up only seeing a vision of the plates, not the plates themselves. A few of the other witnesses said they only got to feel the plates which were hidden under some cloth or something. They didn't actually see them. If all we had was the appearance traditions quoted by Paul, then even if these appearances came from Peter and James themselves, we'd be justified in questioning what it is they actually saw. Maybe they DID just see a hallucination or something like Mother Theresa in a cinnamon roll. Of course I already explained why I don't buy that.

I haven't said anything about the appearance to Paul. I'm really hungry, though, so I'm not going to talk about the appearance to Paul. I don't have much to say about it anyway. There is the argument that since the appearance to Paul was visionary, and since Paul thinks his appearance was just like the appearance to the other apostles, that Paul must've understood their appearances to be visionary as well. That's an argument worth responding to. Maybe I'll do that some other time.


Boz said...

I like that you show uncertainty, and give some credence to opposing arguemnts. It is refreshing, after reading most apologists who give the impression of 100% certainty.

I agree with you overall, that a ressurection is the best explanation of the artifacts that we have. Yet I am not persuaded that a ressurection occured. let me explain why.

The artifacts that we have are the List_of_New_Testament_papyri (wikipedia). For simplicity, lets assume that we have the original manuscripts, in full, of the NT and the apocrypha. Primary among these are the dozens of gospels, and pauls genuine letters.

Now, what are some possible explanations of the fact that these documents exist?

We can think of all kinds of scenarios ranging from these documents being completely fictional novels, to being 100% correct in every detail. There are an uncountable number of scenarios.

Each scenario would include either: no ressurection; a bodily/corporeal ressurection; a spiritual/ethereal/ghostlike ressurection; or a brain-only image of a ressurection.

Now, I think a scenario that involves a bodily ressurection is the best explanation of the fact that these pieces of papyrus exist.

Given the amount and scale of uncertanties involved, I would say that all of the scenarios that involve a bodily ressurection, when combined, add to a 80% probability. or 90%. or 75 or 95. This means that all of the scenarios that do not involve a bodily ressurection, when combined, add to a 20% probability, or 10 or 25 or 5. The precise number is not important.

So I would say that a ressurection is the best explanation of the artifacts that we have.

Now, consider the prior probability of a ressurection. If some anonymous commenter on the internet claimed that their friend bodily ressurected, what probability would you assign to that claim? 1 in a million? 1 in a billion? 1 in a trillion? 1 in a quadrillion? Lets say 1 in a trillion.

Now, using bayes theorem, the final probability of a bodily ressurection is 0.000000000004. the final probability of no bodily ressurection is 0.999999999996.

This is why I am not persuaded, because the evidence cannot overcome the miniscule prior probability.

Sam Harper said...


I had to do a Bayesian assessment for the probability of the resurrection of Jesus for my application with Reasonable Faith to start a Reasonable Faith chapter. I came up with 50% probability using very round numbers. Here's what I wrote in my application:

9. Construct an argument for a Christian hypothesis like “God raised Jesus from the dead” using Bayes’ Theorem. (41-44)

Pr(H/E) = [Pr(H) x Pr(E/H)]/[Pr(H) X Pr(E/H) + Pr(¬H) x Pr(E/¬H)

H = God raised Jesus from the dead

Pr(H/E) = [0.1 x 0.9]/[0.1 x 0.9 + 0.9 x 0.1] = 0.5

Explanation: Since in our general experience, God doesn’t usually raise people from the dead (or it’s extremely rare if he does), I gave Pr(H) a 0.1. Since the tomb would be empty, the disciples would be converted, etc almost definitely if Jesus had been raised from the dead and appeared to them, I gave Pr(E/H) a 0.9. Since Pr(H) was 0.1, it seemed like the Pr(¬H) should be 0.9. Since it seems highly unlikely the tomb would be empty and almost impossible that the disciples would’ve been converted if Jesus had not risen from the dead, I gave Pr(E/¬H) a 0.1. So, the result is that there’s a 50% chance that Jesus was raised from the dead.

I don't really have a good handle on Baye's theorem. You seem to be looking at the background information (or prior probability) a little differently than I did. I was just looking at it statistically. Assuming there have been 12 billion people in the last 20 thousand years or so, and maybe one of them has risen from the dead, there's statistically a 1 in 12 billion chance that somebody will rise from the dead.

But instead of looking at it purely statisticalyl like that, you referred to somebody claiming that somebody rose from the dead. I think that would change the probability considerably, because now we're asking a different question: What is the probability that somebody would claim their friend rose from the dead if their friend really didn't? The vast majority of people never make claims like that. So I think if somebody did make a claim like that, the probability that it actually happened would be a lot higher than what you gave it.

Boz said...

We agree on Pr(E|H), being about 0.9, which is good.

And you are right, that our disagreement is about prior probabilities.

I have some statistics background, so let me explain my understanding of bayesian prior probabilities.

1) It is to a ressurection-proponent's advantage to use a different hypothesis to what you used in your previous comment. hypothesis should be = "Jesus was bodily ressurected by any method at all", or just "jesus was bodily ressurected". Because this hypothesis is more likely than the previous hypothesis of "God raised Jesus from the dead". This is because it is possible that jesus was ressurected, but not due to Yahweh's actions. For example, he could of done it himself. Or aliens, or Baal, or another human might have done it. This is explained here: (you did not commit this fallacy)

so we should use the hypothesis: "jesus was bodily ressurected"

2) now, when doing a bayesian analysis, evidence is split in to two groups - background information, and evidence under active consideration (The Evidence). In our example, The Evidence is the dozens of manuscripts. Background information is everything else in the world ever.

3) a prior probability has to be the same for every person who makes the same claim. (i.e. "Boz was ressurected", 'Sam was ressurected", and "Jesus was ressurected" must all have the same prior probability). This is because everyone has the same background information. also, when we are calculating a prior probability we are looking at the background information only, and ignoring The Evidence. If we allow The Evidence to affect our prior probability, then we are double-counting The Evidence.

For example, I can't say: "the prior probability for "Boz was ressurected" is higher than for everyone else, because Boz was signed up for Cryogenics". This would be double counting The Evidence about Boz and cryogenics. Similarly, I can't say that the prior probability for "Jesus was ressurected" is higher than for everyone else, because Jesus was a religious leader. We do not know that Jesus was a religious leader (until we consider The Evidence).

This means that, if you are puting a prior probability on "X person was ressurected" of 0.1, then if I came to you with 1,000,000 ressurection stories, and no evidence, you would be forced to say that: "roughly 100,000 ressurections have occured."

4) The best way to calculate a prior for "X person has done Y action" is: (Total number of people that have done Y)/(total number of people ever). This can be shown by example.

X person ran 100m in less than 10 seconds. the prior is 100/12b
X person was a citizen of USA. the prior is 600m/12b
X person was a female. the prior is 6b/12b
X person was over 2m tall. the prior is 20m/12b

5)In response to your point about this question: "What is the probability that somebody would claim their friend rose from the dead if their friend really didn't?", I can make this probability equal to 1/12b by making 12b claims, that every person ever, was ressurected. So, we shouldn't talk about claims made, because it is a bit pointless.


so, in summary, I disagree with you prior of 0.1 because:
a)the prior must be the same for all people. (no one is special)
b) a prior that is more generous than 1 in a million, means that there are thousands of ressurected people walking around.
c) the best way to calculate a bayesian prior in this instance is via historical frequency.

DagoodS said...


I have always been impressed with Sam’s careful handling regarding both sides of an issue.


Although I could spend hours and hours on the various topics here, one question did stand out. Why do you think “the apostles” and “the Twelve” are the same group of people in 1 Cor. 15? Thanks.

Sam Harper said...


Thank you for giving such a lucid explanation of Baye's theorem. I had some questions about your third and fourth points, though.

I don't see how it's possible to come up with a prior probability for a resurrection that would be the same for everybody because I don't see how it's possible to come up with a number that wouldn't be influenced by a person's worldview. If a person didn't believe in anything supernatural and that the physical cosmos is all there is or ever was, then that person would probably think resurrections are almost impossible. The only possibility is that something about the indeterminism in quantum physics resulted in a bunch of subatomic particles all moving in just the right direction spontaneously to create a resurrection. That chance would be next to zero. But if a person believes in God, then that person would probably assign a much higher probability to a resurrection because resurrections wouldn't be impossible.

There's another difficulty, too. We might both agree to put 12 billion people in our denominator, but how do we determine what to put in the numerator? We've just got to guess how many resurrections have happened. I arbitrarily chose one just for the sake of having a non-zero number. A physicalist might put a zero in there.

I also don't understand why the denominator must be all the people who have ever existed. Why couldn't it be all the people who have ever been in a certain situation? Using your running example, why not use all the people who have run in a race or who have timed themselves, or something like that? Earlier, you said, "If some anonymous commenter on the internet claimed that their friend bodily resurrected, what probability would you assign to that claim?" It seems to me the probability would be the number of people who have ever been resurrected compared to the number of people who have ever had somebody claim that they were raised from the dead. I said earlier that you'd get a much higher number that way, but you said you could easily get it back to 1 in 12 billion just by making the claim yourself that all 12 billion people have risen from the dead. But that doesn't really answer my question because then I'd just tweak the situation and say "the number of people who have ever been in a situation where somebody they knew claimed they had risen from the dead, and weren't just being hypothetical about it." Since you haven't known 12 billion people, you couldn't just get around that situation by claiming yourself that all 12 billion people had been resurrected.

I talked to Tim McGrew about some of these issues one time, and he seemed to have a slightly different take on it than you. He said that prior probabilities depend on background information. Your background information may contain evidence you've already collected. You use Baye's theorem to evaluate new evidence. The formula compares the new evidence to what you already know.

I think that makes some sense because even in the case of using all the people who have ever existed, you've first got to have some evidence for how many people have ever existed. So you'd be smuggling some evidence into your denominator, and that seems unavoidable. And if different people gave different estimates for how many people have ever existed, then the prior probability couldn't be the same for everybody. I can see how it would be ideal for the prior probability to be the same for everybody, but practically, that would be hard to come by.

Since we already know that people who knew Jesus claimed that he had risen from the dead, why can't we include that in our background information?

Sam Harper said...


I have always been impressed with Sam’s careful handling regarding both sides of an issue.

Thank you.

Why do you think “the apostles” and “the Twelve” are the same group of people in 1 Cor. 15?

Because of all the references to the "twelve apostles" in the new testament. If you search BibleGateway, that exact phrase shows up in Matthew 10:2, Luke 16:13, and Revelation 21:14. Of course there were other apostles outside of the twelve, but I think if Jesus appeared to "all the apostles," that would've at least included the twelve since they were among the apostles.

DagoodS said...

Thanks, Sam.

After following through on some research I had already done, and reviewing those verses, I am unconvinced the 1 Cor. 15 creed was implying the Twelve and “all the Apostles” were the same. Neither the same people, nor the same occurrence. I agree the twelve Disciples eventually became known as apostles; I agree there were additional apostles beyond the twelve (Paul being the most famous.)

How that came into being, though, is a bit of a mystery.

Psiomniac said...

I have always been impressed with Sam’s careful handling regarding both sides of an issue.
Me too.

Boz said...

WOLDVIEWS - Yes, I agree. To clarify, I was saying that the prior probability for a ressurection must be the same for all subjects under consideration. This is because the background information is the same for all subjects under consideration. I might calculate a probability of X%, and you might calculate a probability of Y% - but X% and Y% apply equally to Barack Obama, my neighbour, your neighbour, L.Ron Hubbard, Jesus, Cher, etc.


NUMERATOR - yes, I agree also. We have to use our best judgement to estimate the numerator.

Note that, due to the way the formula works, a prior probability of 0% or 100% cannot be changed by any evidence. So, these prior probabilities are inappropriate.

Bayes' Theorem is an iterative process, so we have to start somewhere.


DENOMINATOR - The intention behind the numerator and denominator of the prior probability is to calculate the background probability of the event occuring. The most accurate way to do this is to conduct a census. A representative sample of a sub-set of the full population will give a very close approximation.

In the running example, the claim is that "person X has run 100m in under 10 seconds" the problem with using a denominator/sub-set of all the people who have run in a race, is that Person X may have never run a race!

Using only our background information, we have no knowledge about Person X. We have no idea if they ahve ever run a race. so, "all the people who have run in a race" is an unrepresentative sample.


A FRIEND'S CLAIM - our hypothesis is "Person X (jesus) ressurected". All information must be split in to "background information" and "New Evidence". I suggest that our "New Evidence" be the manuscripts, and our "Background Information" be everything else (because this is the easiest partition).

Using only our background information, we have no knowledge about jesus. We don't know if jesus is a boy or a girl. We don't know if jesus is alive or dead. We don't know when jesus was born. We don't know how or if jesus died. We don't know which country or town that jesus was born in. We don't know if jesus is tall or short, brown or blue eyes. We don't know if jesus had any friends. We don't know if jesus is rich or poor. We know nothing about jesus. This is because all the information about jesus is in the "New Evidence".

Now, we are perfectly free to move some manuscripts from the category 'New Evidence' to 'Background Information'. Such that "a friend(s) claimed that jesus ressurected" is in our 'Background Information'. This would increase the prior probability, because we would know that a friend claimed that he ressurected. However, this would reduce the power of the New Evidence, because some manuscripts have been removed. The outcome would be the same in both circumstances, because the increase in the prior probability would be matched by the reduction in power of the New Evidence.


TIM MCGREW - yes I agree with your paraphrase of his position. The "Background Information" contains a lot of information/evidence. In fact, it contains 99.9999% of of all the evidence in the world. Because the "New Evidence" contains only ~50(?) manuscripts

(I can't work out how to do a quote)

Gary said...

Were Jesus' disciples capable of differentiating between a vivid dream and reality?

In the Gospel of Matthew, an angel appears to Joseph twice, once to tell him that he should go ahead and marry Mary, even though she is pregnant (not by him), and then again a couple of years later to warn him of Herod's plan to kill Jesus and that he should take the family to Egypt. The author of Matthew tells us that both of these "appearances" occurred in dreams.

The question is: Did Joseph believe that God had sent a real angel to him to give him real messages?

If first century Jews were truly able to distinguish dreams/visions from reality, why would Joseph marry a woman who had been impregnated by someone else just because an angel "appeared" to him in a dream? If first century Jews knew that dreams are not reality, Joseph would have ignored the imaginary angel and his imaginary message. For Joseph to go through with his marriage to a pregnant Mary was a very rare exception to the behavior of people in an Honor-Shame society. His act of obeying an angel in a dream is solid proof that he believed that the angel was real and the message was real.

And if Joseph understood that dreams are not reality, why would he move his family to a foreign country based only on a dream?

And how about Paul's dream/vision? Paul saw and heard a talking bright light in a dream. Paul saw the men accompanying him to Damascus collapse to the ground with a dream. Paul reported that these men also saw the light but didn't hear the voice...or heard some kind of noise but didn't see the a dream....depending which passage of Acts you read.

So it is obvious that first century Jews were just as likely to believe that a dream is reality as some people do today! People have been seeing angels, bright lights and dead people for thousands of their dreams...and have believed that these events are reality.

So the fact that four, anonymous, first century books contain stories of people "seeing" dead people and even "seeing" large groups of people "seeing" dead people, should come as no surprise.

They were vivid dreams. Visions. Nothing more.