In this piece I argue that the Hallucinations, Biases, and Socialisation Model (henceforth HBS model, which I outline here) provides a superior explanation for the resurrection appearances of Jesus of Nazareth than does the competing Christian explanation (i.e. that Jesus was raised by God). In making this argument, I first present an account of what I mean by an ‘explanation’, and how one explanation can be judged superior to another. I argue that an explanation has greater explanatory power to the degree to which it can explain diverse phenomena (‘explanatory scope’), and to the degree to which it does not need to introduce antecedently unknown entities (‘plausibility’).
I then argue that the HBS model is both more plausible and has wider explanatory scope than the Christian explanation. I argue that it is more plausible since it depends only on the existence of psychological and sociological processes which are known to exist, whereas the Christian explanation must make contentious and uncertain assumptions about the existence and motivations of God. I argue that is has wider scope because it is capable (with minor adjustments) of explaining a wide range of miracle claims across different religions, whereas the Christian account is specific to the Resurrection appearances only. I thus conclude by arguing that, since the HBS model provides a superior explanation for the resurrection appearances without needing to posit the divinity of Jesus, the alleged superior explanatory power of the Christian explanation (as argued by apologists like William Lane Craig or Mike Licona) cannot in fact be appealed to as a significant argument to support the probable truth of Christianity.
What is an Explanation?
I will begin by assuming that our objective is to provide an explanatory account of the resurrection appearances, including other associated details like the empty tomb and the conversion of Paul. The first step, I think, is define what we mean by an ‘explanation’, since different people use this word in different ways. In my view, an ‘explanation’ is more than just a satisfying story, or an account that seems to ‘make sense’. An explanation needs to get at the ‘underlying truth’ of the situation; what we might call the ‘causal structure’ of what is occurring. I know words like ‘truth’ and ‘causal’ are themselves problematic, but I’m trying to gesture at a very tricky concept here by using terms that I hope people have some existing familiarity with.
In light of these considerations, let me provide what I think is a suitable first-order approximate definition which will be sufficient for our purposes here: “an explanation of some phenomena X consists of a set of events, entities, and processes, which taken together provide/entail the causes which gave rise to X”. Put simply, an explanation of X is an answer to the question “what made X be the case?”, or “why X and not something else?”
Quality of Explanations
Explanations are not all or nothing; they come in varying degrees of higher and lower quality. In assessing the relative quality of different explanations, I believe that essentially what we are doing is maximising some abstract quantity, which for the sake of argument I will call the ‘power’ of the explanation. That is, better explanations have greater ‘explanatory power’. Explanatory power is a difficult and abstract concept which eludes simple definitions. Here I propose (again for the sake of conceptual clarity and without pretence of comprehensiveness) to think of explanatory power as being the combination (in a vaguely mathematical manner, analogous to multiplication) of two additional concepts: ‘scope’ and ‘plausibility’. Let me explain each of these in turn.
Explanatory scope refers to the size and extent of the phenomena that a given explanation can explain. Thus, given a particular explanation, the more different things that are in X (the set of things which are explained), the greater is the scope of that explanation. Special Relativity has greater explanatory scope than classical Newtonian Mechanics, as the latter is only applicable when velocities are considerably lower than the speed of light, while the former is applicable with any velocities. Greater explanatory scope is to be preferred, as it means that the explanation yields a greater insight into the underlying causal processes at work; it ‘tells us more’ about what is going on. However, greater explanatory scope does not by itself mean that an explanation is a good one – for instance, conspiracy theories tend to have very large explanatory scope, as they provide causal explanations for (often) a very diverse range of social, political, and economic phenomena. Such explanations, however, generally score poorly on the criteria of plausibility, to which I will now turn.
The plausibility of an explanation refers to its ‘simplicity’ or (more loosely) its ‘elegance’. This is closely related to the idea of Occam’s razor, which some people state as being the principle that ‘simple explanations are to be preferred’ or ‘the simplest explanation is most likely to be correct’, however I think this is a misleading characterisation. As I believe the idea is generally understood and applied in science and elsewhere, the notion of ‘simplicity’ has little or nothing to do with how easy an explanation is to understand, or how long it takes to explain, or even how many entities or processes it needs to appeal to. Rather, the version of the razor which I prefer, and which I think is most accurately descriptive of good inferential practise, is ‘entities must not be multiplied beyond necessity’. That is, given a particular phenomena to be explained, the explanation which requires the fewest assumptions or premises that are new (that is, not known antecedently) is (all else equal) to be preferred as an explanation. Understood in this way, the value of simplicity in an explanation is that with every new assumption we introduce about something existing or some process working in a particular way, we also introduce another place where we might make a mistake or go wrong. The more of these there are in our explanation, the more likely it is that at least one of them is incorrect, and hence the less likely the explanation is to actually be true.
Now that I have outlined the notions of ‘scope’ and ‘simplicity’, I will return to articulating the concept of ‘explanatory power’. As I stated earlier, I believe that explanatory power can be profitably understood as combination (loosely speaking, like the mathematical product) of scope and plausibility. That is, an explanation is said to have greater explanatory power to the degree to which it has greater scope, and the degree to which it has greater plausibility. Explanations with greater explanatory scope are to be preferred because they tell us more about the underlying causal processes at work, and more plausible explanations are to be preferred because they are ceteris paribus less likely to introduce a false assumption or premise which would invalidate the explanation.
Many explanations in science, and I also think some in history and even philosophy, have both a wider scope and high plausibility, and so consequently have high explanatory power. Some explanations, like conspiracy theories, have wide scope but immensely low plausibility (as they must posit a very large number of people working behind the scenes, competence to avoid detection, presence of immense resources, motivations to act, and many other such things that we do not antecedently know to exist, and indeed I think often have good reason to believe do not and even cannot exist). Other explanations may lack explanatory power for the opposite reason: although they have high plausibility in the sense of not needing to posit many new entities or processes, they may be so circumscribed and restricted in the class of phenomena which they can explain, that their explanatory scope is very narrow (arguably many historical explanations are of this sort). The sort of explanations which have the least explanatory power of all are those with both narrow scope and low plausibility (I think many paranormal explanations fit into this category, as they often only apply to specific events or a small class of events, and also make reference to ghosts and other such entities which are not antecedently known to exist).
Degrees of Plausibility
Before moving on, there are two final points to make. First, when I talk about ‘positing new entities and processes that are not antecedently known to exist’, this should be interpreted properly be interpreted as also being a matter of degrees. Entities or processes are seldom known for certain to exist, but are antecedently established with varying degrees of probability. Likewise, one entity or process cannot necessarily be assumed to be equal in plausibility to another merely because they are both referred to by a single word. Positing a new type of fundamental particle, or a new Neolithic culture in some part of the world, will in general be much less ‘extravagant’, and hence much more plausible, than positing the existence of ghosts or big foot, even if the latter are capable of providing a causal account of (i.e. an explanation for) the same set of phenomena. Of course, making this determination about the relative degrees of plausibility of different entities or processes is often quite difficult, but in principle I believe this is what we ought to attempt when constructing a plausible explanation.
Second, many people in discussing explanations make reference to the consistency of an explanation; both the consistency of the explanation with the specific events or processes to be explained, and also more generally its consistency with our existing background knowledge about the world. Personally, however, I do not think it is necessary to introduce ‘general consistency with background beliefs’ as a separate criterion in judging explanatory power (or the quality of explanations generally), as I believe the idea of an explanation being consistent with our ‘background knowledge’ about the world is already incorporated into the notion of simplicity, in the form of the number of ‘new entities’ that a proposed explanation must posit. As to the question of consistency of the explanation with the specific phenomena to be explained, I think that if the explanation is inconsistent with the phenomena to be explained, then it is simply not an explanation of those phenomena (though it may be a partial explanation of sum subset of those phenomena). This sort of specific consistency, however, is relatively easy to obtain, simply by introducing additional ad hoc auxiliary hypothesis into an explanation (e.g. in an extreme example, one could simply say the explanation works one way on Mondays and another way on Tuesdays. Obviously this has very low plausibility, but it is nonetheless consistent with the specific phenomena to be explained).
The Resurrection Appearances
The HBS Model
We are now in a position to analyse competing explanatory accounts of the resurrection appearances of Jesus. Given the above considerations, we have established that our goal is to discover or develop an explanation with the maximum amount of explanatory power. Such an explanation allows us to understand the most about why things happened as they did, at the lowest ‘cost’ in terms of introducing new, antecedently unknown entities or processes (and thus multiplying the chances for error to creep in).
I believe that my HBS model (probably with some tweaks and additions, as its only a first draft, and I’ve had much less time to work at it and expertise spent on it than have the apologists on their arguments) possess greater explanatory power as an explanation for the resurrection appearances (and related events) on both accounts: I believe it has wider scope, and also greater plausibility. I will now defend each of these claims in turn.
Scope of the HBS Model
I believe the HBS model has reasonably wide scope because, with relatively small adjustments of details, it can serve as an account for the development and propagation of many different miracle claims and other paranormal beliefs throughout history. The psychological and sociological processes that it refers to are, given their widespread documentation and repeated validation, largely universal (in broad terms, obviously specifics vary), and so can be appealed to in many different cultural and historical circumstances to explain how people’s memories are reshaped over time, and how large groups of people can come to believe very unusual things even in spite of overwhelming evidence to the contrary. As such, since it is able to provide an account of a wide range of phenomena, the HBS model has reasonably wide explanatory scope.
Plausibility of the HBS Model
I also believe the HBS model has reasonably high plausibility, as it does not require the introduction of many new entities or processes. The model is based upon known psychological and sociological phenomena which have been generally quite well documented (though more work remains to be done on many details of course), and thus are antecedently known to exist. The main posit necessary in the model is in extrapolating these processes beyond the specific environments in which they have been originally studied, and applying them in collectively to explain a particular complex event in history (i.e. the resurrection appearances). In extrapolating and applying such phenomena, there is of course a degree of uncertainty. The HBS model assumes that the processes operate in broadly the way they have been observed to in various other contexts, and also assumes that they can interact and play off each other in the way I outlined in the model. I believe that these are reasonable assumptions to make, as the processes I document are sufficiently robust, and have been observed in sufficiently many contexts, that extrapolating them in the manner in which I have done in the HBS model is reasonably plausible, and consistent with other such ‘extrapolation’ practices in science and history.
Explanatory Power of the HBS Model
Thus, taken together, I believe that the HBS model of the resurrection appearances exhibits a fairly high degree of explanatory power. Its antecedently unknown assumptions are relatively few, mostly restricted to extrapolating and applying processes which I believe are already quite well documented. As such, it has fairly high plausibility. Likewise, its explanatory scope is reasonably high, as (with some appropriate modifications of specifics) the broad account can be applied to explain many other miracles and supernatural claims throughout history.
Plausibility of the Christian Explanation
I will now contrast the HBS model, with the Christian explanation of the resurrection appearances – namely that God resurrected Jesus, who then went on to appear to his various followers. First, I believe this account has relatively low plausibility. As far as I can tell, it requires three assumptions or premises which are not antecedently established: 1) that there is a God, 2) that this God desires to intervene in human affairs, and 3) that Jesus was the/a means by which this God desired to intervene in human affairs. I have chosen this tripartite division because I think it facilitates greater conceptual clarity: God could exist but not care to intervene in the world, or he could exist and be interventionist, but not be interested in resurrecting Jesus because in fact he is the Islamic God or the Hindu god (or whatever else). Of course, one could subsume all three assumptions into a single premise, for example simply “Jesus was God”, but I think this is essentially just stating the same three things in a different way. The key point is not how many sentences we write, but how many distinct conditions there are, each are separately controversial: some people believe 1) only, some believe 1) and 2), some all three, and others none.
So how plausible are propositions 1-3? I don’t know. I have argued elsewhere that our best guess for the probability of 1) is something like 10%, however I think even values north of 50% are also defensible (though not, say, 90%). The other two are considerably harder to put numbers on. Regardless, the real point is simply that I believe a Christian should agree that, antecedently to considering the resurrection, all of these three propositions are at best uncertain. They are a long way from firmly established. By contrast, I think most of the psychological and sociological processes utilized by the HBS model are quite firmly established, and the extrapolations made in applying them to the particular case of the resurrection are relatively small. This is, of course, a question of weighing up relative plausibilities, which is not easy to do. But I do think a strong case can be made that the processes and entities which the HBS model must posit in order to explain the resurrection appearances are antecedently known to exist with considerably higher confidence than the entities and processes required by the Christian account. As such, it is my view that the HBS model has greater plausibility than the Christian explanation.
Explanatory Scope of the Christian Explanation
I also think that the HBS model has greater explanatory scope than the Christian explanation. As noted before, the HBS model (with minor adjustments) can explain a diverse range of supernatural and miracle claims from all over the world, as it relies on psychological and sociological processes which (in general terms) are known or reasonably thought to operate in sufficiently similar ways across different times and cultures (there is, of course, a degree of extrapolation here as noted above, but I believe it is reasonably small). In contrast, the Christian explanation is so specific that it can only account for the Resurrection appearances, and perhaps also (with minor adjustments to extend the account to Jesus also appearing at other times and places in history) at least some subset of other Christian miracle claims throughout history. It cannot, however, provide any explanation for the many other miracles reported in Buddhist, Hindu, Mormon, Islamic, Pagan, and many other religious and spiritual traditions. As such the Christian explanation has narrower explanatory scope than the HBS model.
I am not saying here that a Christian worldview cannot provide an explanation for non-Christian miracle claims or paranormal occurrences. Rather, what I am saying is that the Christian account of the resurrection appearances, or any simple extrapolation thereof, does not itself provide such an explanation. Perhaps by introducing further assumptions about God appearing in other ways throughout history, or demons acting to deceive mankind, or even by appealing to some of the very same psychological and sociological mechanisms which the HBS account is based on, a Christian would be able to provide an explanation for these other miracle claims that is consistent with their worldview. But my point is precisely that this would require positing additional entities or processes (demons who can appear to people, or God choosing to reveal himself in additional ways to other peoples, etc) which are not entailed by the original explanation of the resurrection appearances itself.
Summing up, I have argued that the HBS model of the resurrection appearances possess greater explanatory power than does the Christian explanation. As such, I believe that we ought to prefer the HBS model over the Christian explanation, and judge that the former is more likely than the latter to be a correct, ‘true’ account of the causal processes which accounted for these sequences of events. If this is correct, it follows that the inference from the resurrection appearances to the probable divinity of Jesus (and hence the truth of Christianity) is an unsound one. Such an inference cannot validly be drawn, because in fact a more satisfactory causal account of these events can be given which does not entail the divinity of Jesus or the truth of Christianity.
It is very important to emphasise that here I am not in any way making an argument for the falsity of Christianity. Indeed, I believe a perfectly orthodox Christian could agree with my entire argument here. I am saying only that the Christian explanation for certain historical facts concerning the resurrection appearances (and related matters like the empty tomb and conversion of Paul) does not constitute by itself a strong reason to believe in the truth of Christianity, as there exists a superior explanation which does not entail this conclusion (namely, the HBS model). In spite of this, Christianity could nonetheless be true, since the HBS model does not rule out the divinity of Jesus or the reality of the resurrection accounts; it simply renders them unnecessary to explain said phenomena. Indeed, I believe (though I don’t have any firm data on this) that the majority of Christians both in the present and throughout history have not believed on the basis of this sort of historical argument. As such, I certainly don’t think that refuting this argument is a refutation of Christianity. It is merely a refutation of this particular argument in favour of Christianity.
A final point that I wish to make is that this isn’t merely some sort of intellectual game. It’s about finding the truth. If we wish to honestly seek the truth, we cannot decide on our conclusion beforehand and work out what evidence or arguments will get us there. We must examine the evidence and arguments as objectively as we can (with perfect objectively always remaining elusive), and attempt to arrive at the conclusion which is best supported by said facts and arguments. I believe that the conclusion which is best supported by the facts and arguments available, in the light of the analysis I have given, is that the resurrection appearances can be better explained naturalistically rather than supernaturally, and that as such the Christian explanation of the resurrection appearances does not constitute any substantial reason for belief in the truth of Christianity. I might be wrong about this conclusion, and so I invite everyone reading this to honestly and politely critique my arguments to expose errors or gaps in my reasoning. May we all be enriched in this joint search for the truth concerning this most important question.