The Christian Doctrine of the Resurrection of Jesus

Let’s represent the Christian doctrine of the resurrection of Jesus as follows:

(C) The Christian doctrine of the resurrection of Jesus is true.

Sometimes the Christian doctrine of the resurrection of Jesus is summed up this way:

(R) God raised Jesus from the dead.

The Christian doctrine of the resurrection asserts that the resurrection of Jesus was a miracle, and that God caused it to happen:

12 Now if Christ is proclaimed as raised from the dead, how can some of you say there is no resurrection of the dead?
13 If there is no resurrection of the dead, then Christ has not been raised;
14 and if Christ has not been raised, then our proclamation has been in vain and your faith has been in vain.
15 We are even found to be misrepresenting God, because we testified of God that he raised Christ—whom he did not raise if it is true that the dead are not raised.

(1 Corinthians 15:12-15, NRSV, emphasis added)

The belief that God raised Jesus from the dead is even declared to be a requirement for salvation:

9 because if you confess with your lips that Jesus is Lord and believe in your heart that God raised him from the dead, you will be saved.
10 For one believes with the heart and so is justified, and one confesses with the mouth and so is saved.
11 The scripture says, “No one who believes in him will be put to shame.”
12 For there is no distinction between Jew and Greek; the same Lord is Lord of all and is generous to all who call on him.
13 For, “Everyone who calls on the name of the Lord shall be saved.”

(Romans 10:9-13, NRSV, emphasis added)

Clearly, (R) must be true in order for (C) to be true. (R) is a necessary condition for (C):

C -> R

However, (R) is NOT equivalent to (C). Taken literally, (R) only captures a part of the Christian doctrine of the resurrection of Jesus. For example, if Jesus had died on a cross in Jerusalem in 30 CE, and remained dead for 1,983 years, and then God brought Jesus back to life yesterday, on November 4th, 2013, then (R) would be true, but the Christian doctrine of the resurrection of Jesus would be false, because the Christian doctrine implies that Jesus came back to life early on a Sunday morning less than 48 hours after Jesus was crucified. Or to be a little less precise, the Christian doctrine of the resurrection implies that Jesus was dead for less than one week, and then came back to life.

Similarly, if Jesus had been stoned to death by Jews in Nazareth in 30 CE, and then brought back to life by God a few days later, (R) would be true but the Christian doctrine of the resurrection would be false, because the Christian doctrine implies that Jesus died on a cross in Jerusalem, not by being stoned to death in Nazareth.

Alternatively, suppose that Jesus was stabbed to death in Rome in 30 CE, and then God brought him back to life a couple of days later. Again, (R) would be true, but (C) would be false.

Suppose that Jesus was crucified in Jerusalem by the Romans in 70 CE and he died and God brought him back to life a couple of days later. In this case (R) would be true, but (C) would be false. Note that Pilate would not have been running the show in Jerusalem in 70 CE, and Paul would have already written his letters (about the death and alleged resurrection of Jesus!) and been executed before 70 CE.

So, we see that although the truth of (R) is a necessary condition for the truth of the Christian doctrine of the resurrection of Jesus, it is not a sufficient condition. (R) encompasses other possibilities in which (C) would be false. Therefore, the probability of (R) is greater than the probability of the Christian doctrine of the resurrection:

P(R) > P(C)

The claim that ‘God raised Jesus from the dead’ assumes or presupposes a couple of basic Christian beliefs:

(G) God exists.

(J) Jesus existed (as a flesh-and-blood human being).

Each of these basic Christian beliefs is a necessary condition for the truth of (R):

R -> G

R -> J

If (R) is true, then both (G) and (J) must also be true:

R -> (G & J)

Let’s summarize the two basic Christian assumptions as a single claim:

(B) God exists AND Jesus existed (as a flesh-and-blood human being).

This conjuctive claim is a necessary condition for the truth of (R):

R -> B

Either (B) is true or it is not (assuming that ‘God exists’ makes a coherent claim). If (B) is not true, then (R) is also not true:

~B -> ~R

We can divide the probability of (R) into two possible cases:

P(R) = [P(R|B) x P(B)] + [P(R|~B) x P(~B)]

In English:

The probability of (R) is equal to the sum of the following two probabilities:
1. The product of the probability of (R) given that (B) is the case and the probability that (B) is the case.
2. The product of the probability of (R) given that (B) is NOT the case and the probability that (B) is not the case.

We already know the probability of (R) given that (B) is NOT the case is ZERO, because (B) is a necessary condition for the truth of (R). So, if (B) is NOT the case, then it follows that (R) is false. This means that we can simply ignore the second case, since (R) has no chance of being true unless (B) is the case:

P(R) = P(R|B) x P(B)

In my opinion the probability of (R) given that (B) is very low, and the probability that (B) is the case is also very low. Based on these assumptions, the probabiity of (R) is very very low, and given that (R) is a necessary condition for (C) and that there are various possibilities in which (R) could be true while (C) is false, the probability of (C) would be something less than the very very low probability of (R).

I’m not going to try to prove that my estimate of the probability of (R) is true, at least not in this post. Richard Swinburne wrote a fairly long and very dense book developing a philosophical argument for the claim that the probabiltiy of the existence of God is (at least) a bit higher than .5, and so I won’t attempt to build a case for a lower probability in just one short blog post. But I do want to illustrate the implications of the above simple probability equation.

In my view, the question of the existence of God is not one about which one can arrive at a conclusion with certainty. So, if knowledge requires certainty, then I would be correctly categorized an ‘agnostic’. However, I don’t believe that knowledge requires certainty, especially when the question at issue is ‘Does God exist?’ or ‘Did Jesus really exist?’ So, I don’t think of myself as an agnostic on either question. I prefer to think of both of these questions in terms of evidence and probability. It is possible that by examining the available relevant evidence one could arrive at a justified true belief about the existence of God, or about the existence of Jesus. But the justification would not be one that makes the belief certain.

Probability is generally measured on a scale from 0 to 1. If the probability of (G) was 0, that means that it is certain that God does NOT exist. If the probability of (G) was 1, that means that it is certain that God does exist. Given my misgivings about certainty on these questions, I like to focus in on nine other possible positions:

P(G) = .9
P(G) = .8
P(G) = .7
P(G) = .6
P(G) = .5
P(G) = .4
P(G) = .3
P(G) = .2
P(G) = .1

If the probability of (G) is estimated as .9, this means that (G) is very probable.
If the probability of (G) is estimaged as .1, this means that (G) is very improbable.

There are various other positions between these extremes. If the probability of (G) is estimated as .5, this means that (G) being true is about as probable as (G) being false.

Currently, I favor the view that Jesus existed as a flesh-and-blood human being, but there are grounds for doubt about this, so I would estimate the probability of (J) to be about .8. I’m much more skeptical about the existence of God, so I would estimate a probability of no more than .1 for (G). Since (B) is simply the conjunction of (G) and (J), the probabity of (B) equals the probability of the conjunction of (G) and (J):

P(B) = P(G & J)

The probability of a conjunction is calculated this way:

P(G & J) = P(G|J) x P(J)

What is the probability that God exists given that Jesus existed (as a flesh-and-blood human being)? I believe that this probability is equal to, or is very close to being equal to, the probability that God exists, period. In other words, the existence (or non-existence) of a flesh-and-blood Jesus is irrelevant to the question of whether God exists. If it could be proven that Jesus performed miracles, or that Jesus was omnipotent or omniscient, then those facts might well be relevant to the issue of the existence of God, but we are not talking about such claims here. What is in view here is the bare-bones claim that Jesus existed as a flesh-and-blood human being, and this claim tells us nothing about whether Jesus performed miracles or demonstrated amazing powers. The mere existence of an historical Jesus does not help decide the question ‘Does God exist?’ Thus, we can simplify the above equation:

P(G & J) = P(G) x P(J)

This equation is not entailed by the more complex equation, but based on our understanding of the relationship between (G) and (J), we are able to substitute ‘P(G)’ for ‘P(G|J)’.

I would estimate the probability of (G) to be very low, and would represent this as follows:

P(G) = .1

I would estimate the probability of (J) to be high, but not very high, so I would represent this as follows:

P(J) = .8

P(G & J) = P(G) x P(J) = .1 x .8 = .08

Since P(B) = P(G & J), we can infer that:

P(B) = .08

In order to calculate the probability of (R), I need to come up with an estimated probability for it being the case that God raised Jesus from the dead given that God exists AND Jesus existed (as a flesh-and-blood human being). I believe that I can assign this scenario a very low probability on the following grounds:

1. God (if God exists) would not raise a false prophet from the dead.
2. Jesus (if Jesus existed) was a false prophet.
Therefore:
3. God (if God exists) would not raise Jesus (if Jesus existed) from the dead.

Based on this argument, and my great confidence in the correctness of the premises, I would estimate the probability that God raised Jesus from the dead given that God exists and that Jesus existed to be very low:

P(R|B) = .1

So, given my judgments, my estimated probabilities, the overall equation goes like this:

P(R) = P(R|B) x P(B) = .1 x .08 = .008

So, the probability that God raised Jesus from the dead would be about .01 (rounding the calculated answer), or one chance in 100, and the probability that the Christian doctrine of the resurrection was true would be something less than that, because P(R) > P(C):

P(C) < .01

OK. Enough about me and my opinions. Let’s consider some other possible viewpoints, and see how the probability equation works in other cases.

Some people might be more skeptical than me concerning the existence of Jesus. Suppose some skeptical person agrees with me that the existence of God is very improbable, but is convinced that it is also very improbable that Jesus existed as a flesh-and-blood human being. The following would be reasonable probability estimates for such a person:

P(G) = .1

P(J) = .1

Given these estimates we could calculate the probability of (B):

P(B) = P(G & J) = P(G) x P(J) = .1 x .1 = .01

Suppose this skeptic agreed with me that Jesus (if Jesus existed) was a false prophet, and that God (if God existed) would never raise a false prophet from the dead. In that case this skeptical person might well agree that it was very improbable that God raised Jesus from the dead given that God exists and the Jesus existed:

P(R|B) = .1

Now we can plug this skeptic’s estimated probabilities into the equation:

P(R) = P(R|B) x P(B) = .1 x .01 = .001

Given that the probability of the truth of the Christian doctrine of the resurrection is lower than the probability of it being the case that God raised Jesus from the dead, this skeptic should conclude that the probability of the Christian doctrine of the resurrection being true is less than one chance in a thousand:

P(C) < .001

Now let’s consider a person who was not as skeptical as I am, and how the probability equation would work for such a person.
Let’s suppose that this person read Swinburne’s case for God in the book The Existence of God and agreed with the conclusion that the probability of the existence of God was greater than .5. Suppose this person agreed with my view that Jesus probably existed but that his existence was less than very probable. In that case, the probability estimates for this person might well be as follows:

P(G) = .6

P(J) = .8

In this case the probability of (B) could be calculated this way:

P(B) = P(G) x P(J) = .6 x .8 = .48

Suppose this person was unconvinced by my argument concerning Jesus being a false prophet, and was inclined to say that given the existence of God and of an historical Jesus, it would be somewhat probable that God raised Jesus from the dead, but not very probable. In that case this person might well agree with this probability estimate:

P(R|B) = .7

Now we can use the equation to calculate a conclusion:

P(R) = P(R|B) x P(B) = .7 x .48 = .336

If we round the conclusion off, the probability of (R) would be .3 or three chances in ten. Given that the truth of the Christian doctrine of the resurrection of Jesus is less probable than (R), this person, would properly draw this conclusion:

P(C) < .3

So, even this person who was much less skeptical than I am, ought not to accept the Christian doctrine of the resurrection.

Let's consider a person who was even more inclined towards Christian faith, and see how the probability equation works for this person. Suppose this person believes that it is very probable that God exists, and believes that it is probable that Jesus existed, but not very probable. In that case this person might well accept the following probability estimates:

P(G) = .9

P(J) = .8

We can now calculate the probability of (B):

P(B) = P(G) x P(J) = .9 x .8 = .72

Suppose this person rejected my argument about Jesus being a false prophet, and agreed with the above view that it was somewhat probable that God raised Jesus from the dead given that God exists and that Jesus existed. This person might well agree with the following probability estimate:

P(R|B) = .7

Now we have the input required to calculate a conclusion:

P(R) = P(R|B) x P(B) = .7 x .72 = .504

So, this person should conclude that the probability of (R) is about .5, and since the probability of the truth of the Christian doctrine of the resurrection is less than the probability of (R), this person, who is much more inclined towards Christian faith than I am, ought to draw the conclusion that the probability of the truth of the Christian doctrine of the resurrection is less than five chances in ten:

P(C) < .5

Clearly, in order to rationally arrive at a positive conclusion about the Christian doctrine of the resurrection of Jesus, one must believe that each of the three key items are very probable:

P(G) = .9

P(J) = .9

P(B) = P(G) x P(J) = .9 x .9 = .81

P(R|B) = .9

Here is how the calculation would work for such a person with such a strong inclination towards the Christian faith:

P(R) = P(R|B) x P(B) = .9 x .81 = .729

If we round off the calculated probability, we see that even starting with the assumption that each key item was very probable (i.e. probability of .9), the conclusion would be that there were about seven chances in ten that God raised Jesus from the dead, and the Christian doctrine of the resurrection of Jesus would be a bit less probable than that:

P(C) < .7

Based on these examples and calculations, it seems to me that nobody has a right to be dogmatic about the truth of the Christian doctrine of the resurrection of Jesus, nor even about the weaker claim that ‘God raised Jesus from the dead’. The best case scenario for Christianity is that a reasonable person could justifiably believe that it was somewhat probable that the Christian doctrine of the resurrection of Jesus was true. Such a conclusion would be based on assumptions that each of three key probability estimates concerning controversial claims were in the ‘very probable’ range.

  • Greg G.

    The probability that I agree with your calculations could be represented by:

    P (A) > .9

  • Ron

    Your measure of P(R) seems to be the prior probability of R. Couldn’t someone assign very low probabilities G, J, and B, but still think R has a high posterior probability in light of certain historical evidence (E)? In other words Pr(R) isn’t the end of the story, we still need to consider P(R/E). This means that P(C) can be higher than P(R). No?

    • Ron

      To clarify, let’s say G=.1, J=.5, so B=0.05. And let’s say P(B/R)=.1, so P(R)0.005. Your argument would imply P(C)> Pr(E/~R). Let’s say Pr(E/R)=1 and Pr(E/~R)= 0.00001. This would make P(R/E)=.9. In that case P(C) could be WAY above P(R).

      • Bradley Bowen

        Ron,
        Thanks for your thoughtful comments. I will have to think about this and get back to you later.
        One quick response for now: C entails R, so the probability of CANNOT be higher than the probability of R. Or to use conditional probability: the probability of C|E cannot be higher than the probability of R|E.

        • Ron

          Absolutely. I think it’s helpful to mention the implicit background knowledge (B). P(C/B&E) can be greater than P(R/B), but P(C/B&E) can never be greater than P(R/B&E).

          My only disagreement is “Such a conclusion would be based on assumptions that each of three key probability estimates concerning controversial claims were in the ‘very probable’ range.” Strong evidence can make R probable even if those 3 estimates are small (but I agree with you that the evidence is not nearly that strong).

          • Bradley Bowen

            I think we still disagree, but I appreciate your clarification.
            I think it would help (me) to incorporate a reference to E (for relevant historical data) into my probability equation, and to walk though it again, considering a scenario where we imagine finding new very powerful historical evidence (that does not suffer from the problems of the Gospels and NT evidence), and ask the question: How would this powerful historical evidence impact the probability estimates in my original equation and also how would it impact probability estimates in a modified version of it that explicitly references E?
            I think you raise an excellent question: How does historical evidence for the resurrection of Jesus get taken into account in the simple probability equation that I put forward?

          • Bradley Bowen

            You asked “Couldn’t someone assign very low probabilities G, J, and B, but still think R has a high posterior probability in light of certain historical evidence (E)?”
            If I am in possession of certain historical evidence (E), then when I estimate the probability of (B) I will (or ought to) take that evidence (E) into account in estimating the probability of (B), assuming (E) is relevant to (B).
            (R) entails (B). So, whenever (R) is true, (B) must also be true. So, if (E) entails (R), then (E) also entails (B). If (E) proves that (R) is true, then (E) proves that (B) is true too.
            I think that a similar relation holds if (E) provides powerful evidence for (R) that makes (R) very probable. I’m thinking that…
            IF (R) entails (B), but (B) does NOT entail (R), and there are real possibilities (not just logical possibilities) in which (B) is true and (R) is false, then P(B) > P(R), and also:
            P(B|E) > P(R|E). I have not tried to prove these claims, they just seem true to me, but I might well be mistaken.

          • Bradley Bowen

            Some additional background…
            I have in mind Classical Apologetics, which was used by, and perhaps originated with, Thomas Aquinas. The basic idea is that the case for Christianity is made in two steps:
            1. Prove that God exists.
            2. Show how miracles point to Christianity as the true releigion.
            As an atheist, this strategy fails in the first step. There is no God, in my view, so there cannot be any miracles, and I immediately reject the Big 3 theistic religions (Judaism, Christianity, and Islam).
            But I do want to engage Christians on the issue of miracles, and particularly on the issue of the resurrection. Trying to prove that God does not exist is one way of objecting to the claim that ‘God raised Jesus from the dead’ but Christian believers and non-Christians who are inclined to believe in God will be hard to persuade by this strategy, so it would be helpful to raise other sorts of objections to the resurrection claim.
            Prooving the existence of God like one proves a principle of geometry is an old idea that has been rejected by many philosophers. Instead most philosophers think in terms of probability and inductive reasoning and cumulative cases for (or against) the existence of God. Swinburne, for example, puts several inductive arguments together in order to justify the claim that the probability of the existence of God is something greater than .5.
            So, instead of making part 1 of the Classical Apologetics a black-or-white, yes-or-no, all-or-nothing matter, it seems reasonable to approach that first step in terms of probability. One examines the various arguments for and against the existence of God, and then synthesizes this information and arrives at a probability estimate:
            P(G) = ?
            Then one can set aside that first step, and move on to the second step, and a skeptic like me can avoid begging the question in dealing with the second step, because I simply grant the assumption that God exists, when dealing with the second step about miracles, and about the alleged resurrection of Jesus. My skepticism about God is taken into account by my low estimated probability for the existence of God (i.e. P(G) = .1). I then ask the question: Assuming that God exists, what is the probability that God raised Jesus from the dead? This question can be represented in terms of conditional probability:
            P(R|G) = ?
            When I ask myself ‘What is the probability that God exists? I’m NOT interested in the prior probability of (G). Rather, I’m interested in the probability that God exists, given the available relevant evidence, at least, the available relevant evidence of which I am currently aware.
            Similarly, when I ask myself, ‘What is the probability that God raised Jesus from the dead?” I’m NOT interested in the prior probability of (R). Rather, I’m interested in the probability of (R) given the available relevant evidence, at least the available relevant evidence of which I am aware.
            Same goes for the question ‘What is the probability that Jesus existed (as a flesh-and-blood human)?’

          • Bradley Bowen

            Either God exists or God does not exist. So, the probability of R can be divided into two cases:

            P(R) = [P(R|G) x P(G)] + [P(R|~G) x P(~G)]

            But since R entails G, ~G entails ~R, so we know that the probability of R given ~G is zero:

            P(R|~G) = 0

            So we can ignore the second case (where we consider the possibility that God does not exist), Thus, we can simplify the above equation:

            P(R) = P(R|G) x P(G)

            What I did in my post was to add in a second key issue, the question of the existence of an historical Jesus. J represents the claim that Jesus existed as a flesh-and-blood human being. B is the conjunction of G and J. So, we can think of the step 1 of Classical Apologetics as focused on B, and assign an estimated probability to B (instead of G by itself):
            P(B) = ?
            Then the previous probability formula can be revised accordingly:
            P(R) = P(R|B) x P(B)

  • Bradley Bowen

    The way I have been thinking about the following probability equation assumes that historical evidence is taken into account in assessing both factors on the right-hand side of the equation:

    P(R) = P(R|B) x P(B)

    But let’s make this assumption explicit by using E to represent the available relevant historical evidence. I would divide the possibilities into two again, based on the fact that either B is true or B is not true:

    P(R|E) = [P(R|(B&E)) x P(B|E)] + [P(R|(~B & E)) x P(~B|E)]

    NOTE: I have not derived this formula mathematically, it is just based on my intuitions about the meaning of P(R|E), so I (or somebody else) needs to see if this equation can be derived from basic probability formulas.

    Once again, when ~B is the case, then the probability of R given the relevant historical evidence is zero. This is because R entails B, so ~B entails ~ R:

    P(R|(~B & E)) = 0

    So, as with the simpler probability equation in my post, we can eliminate the second half of the right-hand side of the equation, because the only chance of R being true is in the case where B is also true:

    P(R|E) = P(R|(B&E)) x P(B|E)

    Now lets imagine that new historical evidence is discovered that provides powerful proof that Jesus died on the cross on the day he was crucified, and powerful proof that Jesus was alive and walking around Jerusalem less than 48 hours after being crucified:

    (D) Jesus was crucified in Jerusalem in about 30 CE and died on the cross the same day he was crucified.

    (A) Jesus was alive and walking around in Jerusalem less than 48 hours after he was crucified.

    To be continued…

  • Bradley Bowen

    In my post, I was assuming that in assigning estimated probabilities to the factors on the right side of the equation, one would take into account the relevant available historical evidence:
    P(R) = P(R|B) x P(B)
    If so, then it appears that it would NOT be the case that P(R) could be greater than P(C).
    But we can make my assumption explicity by reference to relevant historical data as E:
    P(R|E) = P(R|(B&E)) x P(B|E)
    If the relevant historical evidence strongly supports the claim that Jesus died on the cross in Jerusalem about 30CE on the same day he was crucified, and the claim that Jesus was alive and walking around in Jerusalem less than 48 hours after he was crucified, then this historical evidence would significantly increase the probability of both factors on the right side of the equation, thus bumping up the value of P(R|E).
    Furthermore, the value of P(R|E) would clearly be greater than the value of P(C|E).
    So, on the revised version of the equation, where the use of historical evidence is made explicit, I don’t see how P(C|E) could be greater than P(R|E). And if this revised equation simply makes explicit what was already implicit in the equation from my post, then I don’t see how P(C) could be greater than P(R), even if the relevant available historical evidence strongly supported the death of Jesus on the cross and the claim that Jesus was alive and walking around less than 48 hours after being crucified.
    However, in making the relationships to the historical evidence explicit, there is an objection that occurs to me that could be raised to a part of my post. Here is the part of the post that I have in mind:
    =========================
    The probability of a conjunction is calculated this way:

    P(G & J) = P(G|J) x P(J)

    What is the probability that God exists given that Jesus existed (as a flesh-and-blood human being)? I believe that this probability is equal to, or is very close to being equal to, the probability that God exists, period. In other words, the existence (or non-existence) of a flesh-and-blood Jesus is irrelevant to the question of whether God exists. If it could be proven that Jesus performed miracles, or that Jesus was omnipotent or omniscient, then those facts might well be relevant to the issue of the existence of God, but we are not talking about such claims here. What is in view here is the bare-bones claim that Jesus existed as a flesh-and-blood human being, and this claim tells us nothing about whether Jesus performed miracles or demonstrated amazing powers. The mere existence of an historical Jesus does not help decide the question ‘Does God exist?’ Thus, we can simplify the above equation:

    P(G & J) = P(G) x P(J)

    ==============================
    If P(G|J) is determined in light of the relevant available historical evidence, then this would be equivalent to the following:
    P(G|(J&E))
    But adding in reference to the historical data E, casts doubt on my simplification of P(G|J) to the following expression:
    P(G)
    A Christian apologist might insist that the available historical evidence shows that Jesus performed miracles, and that this raises the probability of the existence of God significantly, so elimination of reference to the existence of Jesus appears to beg the question about whether Jesus performed miracles.
    So, this apears to be a good reason to make the use of historical evidence explicit by means of the letter E.
    One possible response to this objection would be that supernatural events may rule out naturalism, but they do not prove the existence of God. One might further argue that supernatural events do not significantly increase the probability that God exists, since there are many other possible explanations, and since we also need to take into consideration the meaning or significance of the particular alleged supernatural event, to determine whether it is really the sort of thing that a perfectly good person would want to bring about. For example, the resurrection of Jesus is NOT something that a perfectly good person would bring about, because Jesus was a false prophet, and it would be a great deception for God to raise a false prophet from the dead.


CLOSE | X

HIDE | X