Defending the Hallucination Theory – Part 9: Clarification of the Hallucination Principle


In Part 8 of this series, I focused on Peter Kreeft’s VERY UNCLEAR argument constituting his Objection #1 (“Too Many Witnesses”) against the Hallucination Theory.    I argued that this was a brief and UNCLEAR version of Josh McDowell’s “Very Personal” objection against the Hallucination Theory (found in his book The Resurrection Factor, hereafter: TRF).  On that basis, I was able to make sense out of Kreeft’s VERY UNCLEAR argument.

The core argument constituting Kreeft’s Objection #1 in my clarified version of his argument goes as follows:

B. IF on multiple occasions more than two persons had the same experience of an alleged appearance of the risen Jesus at the same time, THEN it is extremely unlikely that those experiences on ALL of those occasions were hallucinations.

3a. On multiple occasions more than two persons had the same experience of an alleged appearance of the risen Jesus at the same time.


C. It is extremely unlikely that the experiences on ALL of the occasions when more than two persons had the same experience of an alleged appearance of the risen Jesus at the same time were hallucinations.

This is a modus ponens argument; it has the form:

 IF P, THEN Q.   




So the logic of this core argument is fine (assuming that the meanings of the key terms don’t change between the premises or between the premises and the conclusion).  We only need to evaluate the truth or falsity of the two premises in order to determine whether this argument is a strong and solid argument against the Hallucination Theory (although there is one further inference required to arrive at the ultimate conclusion that it is very likely that the Hallucination Theory is FALSE.)


Premise (B) states a principle about hallucinations, particularly about “group” hallucinations.  Initially, this principle seems plausible and reasonable.   So, it is natural to focus on premise (3a), which asserts a historical claim about alleged experiences of alleged appearances of the risen Jesus.  I will argue that Kreeft FAILS to show that premise (3a) is true, and thus that this premise is DUBIOUS.  However, there is still some UNCLARITY in premise (B), and when that premise is further CLARIFIED it will cease to be plausible and reasonable.  So, I expect that in the end, I will argue that both premises of this core argument should be rejected, and thus that Kreeft’s Objection #1 (Too Many Witnesses) FAILS to refute the Hallucination Theory,  just like Josh McDowell’s “Very Personal” objection (upon which Kreeft’s Objection #1 is based) FAILED.


There are at least three UNCLEAR terms in premise (B):

  • the same experience
  • extremely unlikely
  • hallucination

Because premise (B) contains these UNCLEAR terms, it cannot be rationally evaluated as it stands.  These expressions need to be CLARIFIED before one can rationally evaluate the truth or falsehood of premise (B).


Conceptual vs. Empirical Claims about “the same experience”

First of all, “the same experience” cannot be had by even two people, in the sense that any experience, like a hallucination, is a SUBJECTIVE event.  My experiences are MINE, and your experiences are YOURS, and you CANNOT literally have “the same experience” that I just had.  Josh McDowell confuses this conceptual point about experiences and hallucinations with an empirical claim about experiences and hallucinations.  You CAN have experiences that are similar to mine, in that my DESCRIPTION of my experience can closely match your DESCRIPTION of your experience, and my DESCRIPTION of my hallucination can closely match your DESCRIPTION of your hallucination.

For example, a man can experience a dream and describe the contents of the dream this way: “I saw a full-grown orange tabby cat walk slowly across the foot of my bed.”  If that man’s wife also has a dream, and she describes the contents of her dream this way: “I saw a full-grown orange tabby cat walk slowly across the foot of my bed.”, then we can reasonably conclude that the man and his wife both had “the same dream” or “very similar dreams”.

In other words, I can give detailed descriptions of my own experiences, dreams, and hallucinations, and if those detailed descriptions match up with a detailed description that someone else gives of his/her experience, dream, or hallucination, then we have good reason to conclude that my experience, dream, or hallucination is “the same” or “very similar to” the other person’s experience, dream, or hallucination.  This is so, even though my experience CANNOT be someone else’s experience, because MY experiences occur in MY mind and CANNOT occur in anyone else’s mind.

An Experience vs. a Description of an Experience

A second important point of clarification is that experiences, especially visual experiences, cannot be fully captured in words or DESCRIPTIONS.  At any rate, the old saying that “a picture is worth a thousand words” is actually an extreme UNDERSTATEMENT.

A photographic image on my PC monitor has a resolution of 3840 by 2160.  That means that the color image on my monitor consists of 3,840 pixels across and 2,160 pixels vertically, and thus that this image consists of 8,249,400 individual pixels.  Each pixel can be a different color.  Because my monitor is 4K Ultra High Definition, each subpixel is 10 bits, and each pixel is 30 bits, which means there are over 1 billion different possible colors for each pixel. If we think of the 10 bits per subpixel as a “word” describing the type of RED or GREEN or BLUE that is part of the overall color of one pixel, then three “words” are used to define the color of each pixel.  Based on this analogy, there would need to be 3 “words” to define each of the 8,249,400 pixels = 24,748,200 “words” to define one high-definition color image on my PC monitor.  So, we need about twenty-five million words to fully define one high-definition color image.

But when we see an event, the visual data is more like a video or movie.  We see objects moving through space, changes in shadows, colors, shades, brightness, and shapes.  For a movie to look realistic you need between 30 frames per second and 60 frames per second.  So, ten seconds of a  60-frame-per-second movie would require 600 frames or images.  If each frame or image was of the 4K Ultra High Definition kind (like on my PC monitor), then each frame or image would require about 25 million “words” to define, so a ten-second portion of a 60-frame-per-second movie in Ultra High Definition would require 600 frames times 25 million “words” per frame = 15 billion “words”.  So, a picture, especially a moving picture, is worth a hell of a lot more than just 1,000 words.

Normally, when we DESCRIBE what we saw and experienced during an event, we do NOT use millions or billions of words.  So, the information contained in verbal DESCRIPTIONS of an experience, dream, or hallucination normally captures only a tiny fraction of the information contained in the original experience, dream, or hallucination.

Because when we compare experiences, dreams, or hallucinations between different people, we are actually comparing the DESCRIPTIONS of those experiences, dreams, or hallucinations, and because descriptions are almost always given in dozens of words, or hundreds of words, or in some cases thousands of words, and NOT in millions of words, nor in billions of words, we are comparing only a tiny fraction of the information contained in the original experiences, dreams, or hallucinations.  Therefore, it is virtually impossible to prove that an experience had by person A was “exactly the same” as an experience that was had by person B.  It is, of course, theoretically possible that person A had “exactly the same” experience as person B had, but verbal DESCRIPTIONS of these experiences only give us a high-level summary of the experiences, which does not allow us to compare experiences at the lowest level of details.

Point of View Affects Experiences

There is one more important point about the experiences of two people being “the same”.    Things look different from a different point of view.  The following images are of the same object but from different points of view:


The image of the bowl is very different depending on the point of view one has of it.   The same is true of people, plants, animals, and physical objects.  How they look depends on the point of view one has while observing the person or thing in question.

If John is standing behind Jesus, and Peter is standing in front of Jesus, and Thomas is standing to the side of Jesus, on Jesus’ left, then if Jesus is looking straight ahead, John will see the back of Jesus’ head, Peter will see Jesus’ full face, and Thomas will see only the left side of Jesus’ face.  They will all have different visual experiences of Jesus even if they are all looking at a physically present Jesus at the same time.

So, having “the same experience” of Jesus at the same time does NOT mean having the exact same visual experiences of Jesus at the same time.  What it means is that the people in question have visual experiences of Jesus that we would expect them to have IF Jesus was actually and physically present, given their different points of view.   In other words, we understand that in three-dimensional space, different points of view of actual physically present people or objects produce different visual experiences, but the variations between those different visual experiences coordinate with each other in predictable ways.


At the very least “Extremely Unlikely” means SIGNIFICANTLY MORE UNLIKELY than events that are just “Very Unlikely”.  But what does “Very Unlikely” mean? and exactly how much MORE unlikely does something have to be in order to be SIGNIFICANTLY MORE UNLIKELY?  In short, the expression “Very Unlikely” and the expression “Extremely Unlikely” are both VAGUE.  Furthermore, it makes a big difference what the precise meanings of these terms are because Kreeft is NOT merely trying to show that the Hallucination Theory is somewhat improbable; he is trying to DISPROVE the Hallucination Theory; he is trying to PROVE it to be FALSE.  One might reasonably argue that the qualified conclusion of the clarified version of Kreeft’s argument is TOO WEAK, given that his goal was to DISPROVE or REFUTE the Hallucination Theory:

A1.  It is very likely that the Hallucination Theory is FALSE.

If “very likely” means, for example, that there is an 80% chance that the Hallucination Theory is FALSE, then Kreeft’s argument is TOO WEAK to be considered a REFUTATION because it leaves open a 20% chance (or one chance in five) that the Hallucination Theory is TRUE.  I would take that as a victory for skepticism.  If each of the four skeptical theories that Kreeft attacks have a 20% chance of being TRUE, then the disjunction of those theories could potentially have an 80% chance of being TRUE!  If there is an 80% chance that either the Hallucination Theory or the Conspiracy Theory or the Apparent Death Theory or the Myth Theory is TRUE, then skepticism about the resurrection is clearly the most reasonable position.

This is why Kreeft and other Christian apologists NEED to REFUTE or DISPROVE each one of the various skeptical theories about the resurrection.  There are a number of skeptical theories (actually many more than Kreeft realizes) and if each skeptical theory has some significant chance of being TRUE, then the disjunction of those skeptical theories can potentially be probable, or even “very likely”.

However, it is NAIVE and UNREASONABLE to expect that any historical argument about alleged events in the life of Jesus (or alleged events related to Jesus’ death) could be PROVEN or KNOWN to be TRUE.  Given the nature of ancient history in general, and the generally poor quality and the limited quantity of historical evidence available about the life (and death) of Jesus, we can only reasonably expect to arrive at conclusions that are PROBABLE, not conclusions that are CERTAIN.  So, Kreeft has a very narrow range of probabilities that will allow him to be successful in his apologetic quest.  Showing that a skeptical theory only has a 20% chance of being true is NOT GOOD ENOUGH!  But he has no reasonable hope of showing that a skeptical theory only has a 1% chance of being TRUE (or a 99% chance of being FALSE).

How close to showing that there is a 99% chance that the Hallucination Theory is FALSE does Kreeft need to get in order to be successful?  Would showing that there is a 90% chance that the Hallucination Theory is FALSE be good enough?  That would mean that there might well be a 10% chance that the Hallucination Theory is TRUE.  Once again, if each of the four skeptical theories has a 10% chance of being true, then that leaves open the possibility that the disjunction of the four skeptical theories that Kreeft rejects has a 40% chance of being TRUE.  That hardly amounts to PROVING that the Christian Theory is TRUE, and so this would NOT be good enough for Kreeft to obtain his apologetic goal.

Thus, Kreeft needs to show that the chance of the Hallucination Theory being FALSE is at least 95% (greater than 90% but less than 99%).   There is NO WAY that the weak dubious evidence available on this subject (mostly from the biblical Gospels) will support such a high level of probability.  I don’t think that ANYONE can even show that there is a 95% chance that Jesus actually existed, so showing that there is a 95% chance that the Hallucination Theory is FALSE is a wild fantasy.

In any case, neither McDowell nor Kreeft give us any indication of what they mean by “very unlikely” or “very likely” or “extremely unlikely” or “extremely likely”.  So, in order to evaluate claims in their arguments that use these terms, we need to make educated guesses (like I’m doing here) about what these terms NEED to mean in order for their apologetic arguments to be successful.


In his book The Resurrection Factor, Josh McDowell quotes three different definitions of the word “hallucination” and then provides a similar definition of his own:

…a hallucination is an apparent act of vision for which there is no corresponding external object.   (TRF, 1981 edition, p.84)

This is a fairly BROAD definition of “hallucination” and, although McDowell probably did not realize this, it includes DREAMS.  We have visual experiences when we dream, and “there is no corresponding external object” to the visual experiences of people, animals, and objects that we “see” in our dreams.  So, on McDowell’s definition of “hallucination”, every dream anyone experiences (that involves visual experiences) is a hallucination.

But what does “hallucination” mean to Kreeft?  Unfortunately, because Kreeft’s presentation of his objections against the Hallucination Theory is ridiculously brief, Kreeft provides NO DEFINITION of this key term.  However, since Kreeft appears to have borrowed Objection #1 from Josh McDowell, namely from McDowell’s “Very Personal” objection in The Resurrection Factor, it is reasonable to assume that Kreeft accepts McDowell’s broad definition of “hallucination”, and thus that the term “hallucination” correctly applies to DREAM experiences, as well as to other more typical kinds of hallucinatory experiences, like when a person who has taken LSD and “sees” a fire-breathing dragon is sitting in the middle of a freeway.

McDowell’s definition is, however, clearly wrong because many hallucinations do NOT involve vision or visual experiences.  One of the most common sorts of hallucination is audio, hearing sounds or voices that are not actually present.  But we can easily fix this problem with McDowell’s definition so that it includes other senses besides sight:

An apparent sensory experience S that seems to be of a person or object is a hallucination IF AND ONLY IF

there is no corresponding external object or actual person present during apparent sensory experience S.

This definition is still a broad one that includes DREAMS as being a subset of hallucinations.

As I pointed out when critically evaluating McDowell’s objections against the Hallucination Theory, it is important that “hallucinations” include DREAMS, because if they don’t, then McDowell’s argument for the resurrection of Jesus FAILS.  The same is true of Kreeft’s short and unclear version of McDowell’s “Very Personal” objection.

If “hallucinations” does NOT include DREAMS, then there is a significant skeptical theory that both McDowell and Kreeft have FAILED to address and thus FAILED to refute:  the theory that some disciples of Jesus had a DREAM about Jesus and mistakenly believed that the dream was a real experience of an actually present Jesus who had risen from the dead.  Neither McDowell nor Kreeft explicitly considers such a theory.  So, in order for their cases for the resurrection of Jesus to be successful, their objections against the Hallucination Theory must work against this DREAM theory.

Given that McDowell defined the word “hallucination” in a way that includes DREAMS, his argument ought to apply to the skeptical DREAM theory.  Given that Kreeft presumably accepts McDowell’s definition of “hallucination”, particularly for his “Too Many Witnesses” objection (since that objection was borrowed from McDowell), Kreeft’s Objection #1 (Too Many Witnesses) also ought to apply to the skeptical DREAM theory.

If Kreeft wants to define “hallucination” more narrowly than McDowell, so that it excludes DREAMS, then his case for the resurrection of Jesus will immediately FAIL, because Kreeft has provided no objections against, and thus no refutation of, this skeptical DREAM theory.  So, if Kreeft’s case for the resurrection is to be successful as it stands, then Kreeft NEEDS to define “hallucination” in the broad manner that McDowell did so that DREAMS will count as examples of “hallucinations”.