Second Response by Peter S. Williams

Do angels such as Gabriel, Michael and Satan exist?

I am enjoying this unique debate. As Proverbs 27:17 says: 'As iron sharpens iron, so one man sharpens another.'

Steven dislikes Dallas Willard's description of angels as being composed of 'non-physical energy', because 'energy is physical.' Steven says that: 'any energy that anybody has ever seen is physical, as Einstein showed.' Einstein showed that energy and mass are convertible (hence atom bombs); thus, Steven implies, 'non-physical energy' is a contradiction in terms. Angels have no mass (being incorporeal) and thus no physical energy. However, I see nothing wrong with the analogical use of 'energy' in describing angels as non-physical 'energy'. Like energy, angels can do work; unlike energy, angels are immaterial. Hence Willard's description is not self-contradictory, simply analogical. Besides, Willard goes beyond his analogy with energy to introduce the notion of spirit: 'spirit is unbodily personal power. It is primarily a substance. . . spirit is something that exists in its own right. . . Thoughts, feelings, willings and their developments are so many dimensions of this spiritual substance, which exercises a power that is outside the physical.'

An angel is a spirit, a creature to whom Boethius' classical definition of a person applies: 'A person is an individual substance of a rational nature.' A person can be more briefly defined as: 'a conscious purposive agent.' As such, 'spirit' means more than 'an immaterial (noncorporeal) substance', for as Antony Flew warns: 'to characterize something as incorporeal is to make an assertion that is at one and the same time both extremely comprehensive and wholly negative.' Hence J.P.Moreland and Scott B. Rae note that: 'A positive characterization of 'spirit' is needed to give content to notions [about] God, angels and disembodied persons.' Moreland and Rae suggest this can be done: 'by saying that spirits are the kinds of substances that possess the ultimate capacities for thought, feeling, consciousness and active volitional power.' An immaterial substance with 'the ultimate capacities for thought, feeling, consciousness and active volitional power' is appropriately called a spirit. The Old Testament term ruach is frequently translated as 'spirit'. If ruach has a core meaning, it is 'a unified centre of unconscious (moving air) or conscious (God, angels, humans, animals) power.' Ruach also refers to the seat of various states of consciousness, including volition (Deut 2:30; Ps 51:10-12; Jer 51:11), cognition (Is 29:24), emotion (Judg 8:3; 1 Kings 21:4) and moral or spiritual disposition (Prov 18:14; Eccles 7:8).' As an unembodied ruach, an angel is just as much a person as I am: 'persons are multiply realizable, that is, persons can be humans, angels, gods, Martians, and so forth.'

In the process of arguing for the existence of the soul, Swinburne writes: 'A person has a body if there is one particular chink of matter through which he has to operate on and learn about the world. But suppose that a person who has been a man now finds himself no longer able to operate on the world, nor to acquire true beliefs about it; yet still to have a full mental life, some of it subject to his voluntary control. He would be disembodied. Or suppose, alternatively, that he finds himself able to operate on and learn about the world within some small finite region, without having to use one particular chunk of matter for this purpose. He might find himself with knowledge of the position of objects in a room (perhaps by having visual sensations, perhaps not), and able to move such objects just like that, in the ways in which we know about the positions of our limbs and can move them. But the room would not be, as it were, the person's body; for we may suppose that simply by choosing to do so he can gradually shift the focus of his knowledge and control, e.g. to the next room. The person would be in no way limited to operating and learning through one particular chunk of matter. Hence he would have no body. The supposition that a person who is currently a man might become disembodied in one or other of these ways seems coherent.'

If this scenario is coherent when applied to an initially embodied human spirit, it is certainly coherent when applied to the conception of an angel. Like every other creature, the existence of an angel: 'depends upon their being created by God and sustained in existence by God.' There is nothing implausible in this; as atheist J.J.C.Smart admits: 'Surely an omnipotent being could have created. . . spirits directly. . .'

Steven asks why we should think of angels as wholly good agents subsequent to their creation and choice for God. Well, Angels are consistently described as 'Holy' (e.g. Mark 8:38, Luke 9:26, Revelation 14:10, Acts 10:22). Angels are instruments of God's wrath and final judgement; and only demons receive judgement. While it may be logically possible that all creatures be partially evil, at least at some stage of their existence (e.g. from their origin), this in no way contradicts the proposition that some creatures are actually wholly good (especially subsequent to their origin). It would be impossible even for God to know which creatures would freely chose to reject or accept him before they existed to make the choice, in which case it was not possible for Him to create only those who would choose to serve Him. Only those spirits who freely chose to serve God were transformed by grace into the wholly good beings angels are today. Those who choose to reject God thereby became demons:

'What glory would God receive if angels and all humans were pre-programmed to worship him? What genuine love relationship could God share with his creatures if they were forced to love him? Both angels and humans can choose what is evil instead of what is good, yet having freedom of choice itself is a good thing. There could be no love for God and no godly character development without the real potential to choose against God. . .'

Richard Swinburne affirms and extends the traditional theory:

'Embodied humans are not the only kind of rational agent with free will there could be. God might well create non-embodied free agents; and some of them might indeed choose the bad. We may perhaps regard the initial choice of character by the angels as one extended in time, and, given that, if it is good that God should give us the ultimate choice over the period of our lives on Earth of being allowed to fix our characters beyond further change, it would seem to be similarly good that God should give to angels also the ultimate choice of being allowed to fix their characters. And he might well also, in giving them that initial choice, have promised them temporary and limited power over the world when they fixed their characters. That would have given a deeper significance to their choice than it would otherwise have. . . If freedom and responsibility are good things, it is good that there be angels who have it, as well as humans. . . angels could only choose the bad if they were tempted so to do, being already subject to bad desires, the bad must have pre-existed any bad choice by angels.'

Steven asserts: 'Peter's sole evidence for such angels [as Gabriel, Michael and Satan] is that somebody has said so.' (my emphasis.) Well, how about when the 'somebody' in question is the omniscient and wholly trustworthy creator of the cosmos, and what if there were good reasons to accept this testimony? Christians believe that the Bible is God's literary revelation to humanity. There are many reasons for believing that the Bible is God's Word, but foremost is the authority of Christ: 'If Jesus is God, then His teachings must be true [and] whatever Jesus taught about Scripture must be true.' The fact is that: 'Jesus not only confirmed the divine authority of the Old Testament but he also guaranteed the inspiration of the New Testament.' As Norman L. Geisler writes: 'a demonstration of the general historical trustworthiness of the [New Testament], particularly of the gospels and Acts, impels one to respond to the claims of Christ which confront us on almost every page. And if one accepts the [New Testament's] portrait of who Jesus is, then one ought to believe his teachings, not least with respect to his high view of Scriptures.' Those scriptures teach the existence of angels and demons (including Michael, Gabriel and Satan), and it is reasonable to accept this teaching: 'Those who take the Bible seriously are obliged to believe in Satan's existence, since the Bible unmistakably refers to the demonic. . . Once the authenticity and divine origin of the Bible are established. . . the existence of Satan follows.'

Peter Kreeft lists five different reasons people don't believe in Satan: 'Believing in the Devil means believing in (1) supernatural, (2) moral, (3) spiritual, (4) personal, (5) evil, and at least one of these five is usually denied.' None of the five positions entailed by the rejection of these five elements is particularly appealing:

'(1) Some are naturalists who do not believe in moral evil (sin), only physical evil (pain). (2) Some are amoralists who have a phobia against judgementalism (i.e., moral law). (3) Most are Gnostics, who identify spirit with goodness (and thus, implicitly, matter with evil, though they no longer say that, as the original Gnostics did), so to them spiritual evil is an oxymoron. (4) Others are Marxists, Socialists, or other leftists who do not believe in personal evil but only evil social and economic systems and institutions (I wonder where they think these came from?). (5) Finally, and most desperately, cockeyed optimists do not believe in evil at all.'

Kreeft goes on to argue: 'The most influential man who ever lived did not fit into any of these five categories, so he was allowed to believe in the Devil. In fact, he not only believed in the Devil, He knew him quite personally from direct experience. Even more, He created him and watched him fall.' Jesus clearly thought that angels and demons were real, as Marcus Borg concludes: 'on historical grounds it is virtually indisputable that Jesus was [an] exorcist.', but 'To be a disciple of Jesus means, among other things, to trust what he said.'

Biblical belief in demons can't be chalked up to a lack of belief in naturalistic causes of mental illness, for as Borg notes: 'The gospels consistently distinguish between exorcisms and healings; not all healings were exorcisms, and not all maladies were caused by evil spirits.' Indeed:

'Belief in angels was not universal in Jesus' day. The Sadducees, for example, disbelieved in angels as well as the resurrection. Jesus went out of his way to side against them on the reality of angels as well (see Mk 12:25), His teaching about angels was unprecedented in the ancient world; he said 'these little ones' - that is, children, and perhaps the uneducated - have angels who 'continually see the face of my father in heaven' (Mt 18:10). No Jew had ever taught that angels behold the face of God. . . If angels do not exist, then Jesus was wrong when he taught these things. And if he was wrong, then he was not a fully trustworthy teacher. Is any Christian ready to believe that?'

There are thus two valid chains of reasoning leading from Jesus to belief in angels, but Steven will reject them because he will not accept their premises. Unfortunately, as I warned in my last response, to give a proper defence these premises goes beyond the remit of this particular debate.


Wesley C. Salmon writes: 'It would be a. . . mistake to suppose that every appeal to authority is illegitimate, for a proper use of authority plays an indispensable role in the accumulation and application of knowledge. . . The appeal to reliable authority is legitimate, for the testimony of a reliable authority is evidence for the conclusion.' Authority is right, not might; it means 'having the right to say'. If anyone has the right to say whether or not immaterial beings exist, it is philosophers (not scientists). The majority of the great philosophers have believed in the existence of immaterial agents. Then again, 'As a part of formulating a biblical and systematic theology. . . the main contours of church history should be consulted, and a burden of proof should be placed on any view that is at odds with what the majority of great thinkers have held throughout church history.' This principle does not suggest 'that the voice of church history is univocal or infallible', but simply that 'the teachings of the great intellectual leaders of the past provide insights that should be taken seriously.' As Stephen T. Davies writes: 'Respect for Christian tradition must (or so I would argue) grant great weight to views held by virtually all the fathers of the church unless there is serious reason to depart from what they say.' The existence of angels is clearly taught by Jesus, the Bible, the Church fathers and Christian thinkers throughout church history. Therefore Christians ought to believe in angels unless they have good reason not to do so. However, there is no good reason to disbelieve in angels (certainly not if one accepts, as Christians should accept, the existence of God and the human soul). Therefore the authority of Christian tradition is good reason for the Christian to believe in angels.

Common Consent

The fact that a lot of people believe something is evidence for its truth, because all these people have made judgements about their beliefs, and have lived by them - testing them in the arena of life, including intellectual life. Such evidence may be very weak evidence, but it is evidence. As philosophers Joshua Hoffman and Gary S. Rosenkrantz affirm: 'if entities of a certain kind belong to folk ontology [the ontological presumptions of our commonsense conceptual scheme], then there is prima facie presumption in favour of their reality. . . Those who deny their existence assume the burden of proof.' This certainly goes for belief in angels: 'Belief in demons is not connected with any particular view of the cosmos. Demons have a very wide geographical and lengthy historical role as spiritual beings influencing man in his relationship to the sacred or holy. . .'


The argument for angels by extrapolation suggests that since nature seems to contain life at every possible level, and since there is an obvious gap between the infinite, immaterial God and finite, embodied human beings, one might reasonably expect the existence of finite but immaterial beings. The argument from extrapolation can be seen as an extension into metaphysics of scientific experience: 'It has generally been the experience of scientists that there are few rules or processes consistent with the laws of nature that fail to be instantiated somewhere in nature. . . if there is a place for the description of a certain sort of particle in [the mathematics of particle physics], then the actual physical particle is found to exist in suitable circumstances.' There is no reason to think that God's taste changes at the limits of the natural world. As Etienne Gilson wrote: 'the general plan of creation would display a manifest gap, if there were no angels.'

The Occult & Occam's Razor

Steven hasn't responded as yet to the suggestion that the demon hypothesis provides a single explanation for a whole range of otherwise separate occult phenomena (such as possession, ouiji boards, apparent past-life memories of some who believe in reincarnation, etc), given the reality of such phenomena.


'Whereas twenty-five years ago the suggestion of demonic activity would have been immediately dismissed, many psychologists are beginning to recognize that maybe there are more things in heaven and earth than our philosophies can account for.' - Dr. Gary R. Collins.

Steven dismissed Dr. Brewer's experience (related in my opening statement) as 'a committed Christian interpreting something in his world view.' I replied by asking whether we should dismiss naturalist's purported experience of natural objects (e.g. coffee cups) simply because 'this is a committed naturalist interpreting something in his world view'?! Steven responded to this reductio by arguing that 'no, materialists should not be trusted'! He points out that: 'in science, the idea that materialists should be trusted when talking about material objects would be laughed at. If the scientist announces that he has discovered a new planet, ten more scientists will double-check his observations, eager to prove him wrong.' But of course, if we should distrust or discount the observations of a materialist about material things simply because he is a materialist, then we should distrust the observations of the ten materialists who check his observations! One shouldn't doubt someone's experience simply because they believe in the possibility or even the reality of what they experience! Besides, Brewer was initially sceptical, and there are (as we will see) other cases of possession observed by more than one person, including sceptics.

Responding to my question as to what evidence he would accept as evincing a genuine case of possession, Steven protests that he 'would be happy to see and accept such evidence. I am not a doctor, but I am sure few doctors would look at the phenomena exhibited in a film like 'The Exorcist' (assuming such cases of demonic possession really happened), and declare it to be a temper-tantrum or growing pains.' That Steven is open to accepting evidence of the type displayed in The Exorcist as demonstrating a genuine instance of possession demonstrates his agreement with me that the existence of angels is possible.

In his book, People of the Lie, Dr. M. Scott Peck relates his involvement with two cases of (apparently Satanic) possession. Having come to a point of belief in God and in the reality of human evil, Peck says he 'was left facing an obvious intellectual question: Is there such a thing as evil spirit?' He says: 'I thought not. . . Still priding myself on being an open-minded scientist, I felt I had to examine the evidence that might challenge my inclination in the matter. . . So I decided to go out and look for a case.' Peck made it known that he was interested in observing cases of purported possession for evaluation:

'The first two cases turned out to be suffering from standard psychiatric disorders, as I had suspected, and I began making marks on my scientific pistol. The third case turned out to be the real thing. Since then I have also been deeply involved with another case of genuine possession. In both cases I was privileged to be present at their successful exorcisms. . . As a hard-headed scientist - which I assume myself to be - I can explain 95 percent of what went on in these two cases by traditional psychiatric dynamics. . . But I am left with a critical five percent I cannot explain in such ways. I am left with the supernatural - or better yet, subnatural.'

Peck argues that while people like to ask whether a patient is possessed or mentally ill, this is 'not a valid question, for 'there has to be a significant emotional problem for the possession to occur in the first place. Then the possession itself will both enhance that problem and create new ones. The proper question is: 'Is the patient just mentally ill or is he or she mentally ill and possessed?'' Peck confirms the central importance of free will in both possession and exorcism:

'From both these cases I would conclude that possession is no accident. . . Possession appears to be a gradual process in which the possessed person repeatedly sells out for one reason or another. . . Free will is basic. It takes precedence over healing. Even God cannot heal a person who does not want to be healed. At the moment of expulsion both these patients voluntarily took the crucifix, held it to their chests and prayed for deliverance. Both chose that moment to cast their lots with God. Ultimately it is the patient himself or herself who is the exorcist.'

Peck concludes: 'Given the severity of their psychopathology before their exorcisms, the rapidity of their progress to health is not explainable in terms of what we know about the ordinary psychotherapeutic process.'

Consider the following two cases of possession recorded by the Anglican Christian Deliverance Study Group. This group helps train those who advise Anglican Bishops in this area and is well aware of those psychological conditions (such as Schizophrenia and depressive psychosis) that might be mistaken for possession:

'A twenty-four-year-old girl was admitted to a psychiatric clinic, claiming that she was possessed. She was showing some of the traditional signs of possession, for example the ability to speak in a foreign language of which she had no previous knowledge, and an unusual knowledge of events. The psychiatrists were divided as to whether she was suffering from a neurotic or psychotic condition, but agreed that a priest should be involved in the case. The priest saw the patient interviewed by two psychiatrists and then he interviewed the patient himself. When the priest entered the room, the patient (without knowing of his visit) knew his name and where he was from and that he was an exorcist. It was agreed that the rite of exorcism should be carried out. During this, the patient convulsed and spoke in the voices of three different men, claiming to be Lust, Greed and Death [notice the judicious 'claiming to be']. These spirits were exorcised one at a time, after which the girl collapsed and lost consciousness for a short while. On gaining consciousness, she asked for something to eat, and appeared to be quite normal.'

'The exorcist was called in by the relatives of the possessed woman, who had been for many years a member of a witch coven centred in the country village in which she had lived all her life. The possessed was a middle-aged woman of limited intelligence and working-class status. She had never left the isolated village in her life except for shopping in the neighbouring market town. Immediately the exorcist entered the room, she started calling out details of his past life which he thought he had forgotten and which were relevant to a wild youth. The second priest in the team had served for many years in the Middle East and was an Arabic scholar. He questioned the woman in various Arabic dialects, and she replied in those dialects. Neither the past of the exorcist nor the dialects could have been known to this woman by any rational process. After exorcism she lost these extrasensory powers and renounced her witchcraft.'

Weighing the Evidence

In recent years, philosophers of science 'have proposed three criteria that must be satisfied for [a hypothesis] to constitute the best explanation of [the evidence].' First, the hypothesis h must be consonant with the evidence e. Instead of injecting discord into our understanding of the evidence, the hypothesis must harmonize with it and with the network of beliefs of which e is a part: 'To say that [h] is consonant with [e] implies that [e] confirms [h]. . .' Consonance is more than a coherentist requirement: 'Consonance involves both goodness of fit and aesthetic or theoretical judgement.' Second, h must contribute to e: 'Thus [h] must perform some useful work in helping to explain [e].' This is achieved by solving problems or answering questions pertinent to e 'which could not be handled otherwise.' This requirement 'is a corollary of Occam's razor, ensuring that adding [h] to our stock of beliefs will not be superfluous.' Third, as the best explanation, h must have a comparative advantage over its rivals: '[h] must simply do a better job of explaining [e] than any of its current competitors.'

The hypothesis of angels is clearly consonant with the evidence. Instead of injecting discord into our understanding of the evidence, the angelic hypothesis harmonizes with it (and with the network of background knowledge Christians bring to the table, e.g. God exists, the human soul is real). Indeed, the angelic hypothesis has a particularly good theoretical and aesthetic 'fit' with the Christian worldview.

Second, the angelic hypothesis certainly contributes to e, performing useful work in explaining the evidence. If angels exist, then many otherwise unexpected things (such as the supernatural behavior of the possessed) would be a matter of course. The existence of angels would explain a wide range of phenomena. In particular, only the actual existence of angels can explain why the Word of God (both in Scripture and in Jesus) teaches their existence.

Thirdly, the angelic hypothesis has a comparative advantage over its rivals. For example, consider the evidence of demonic possession. The principle rival to a in this instance is the admittedly simpler naturalistic explanation in terms of some sort of mental disorder. However, the psychological explanation looses out to a on two counts. First, the psychological explanation is inadequate. While a psychological explanation of purported possession may sometimes (or even mainly) be the best explanation, there do seem to be cases where it is inadequate to the task, as Brewer, Peck and others testify. As James LeBarr says: 'there comes a point, when somebody is climbing up the wall or floating on the ceiling or talking a language they've never studied, when it's harder to put it in the 'psychological-problem' bin.' (This is exactly the 'Exorcist' type of evidence Steven says he would accept for possession.) Second, the psychological explanation cannot account for as wide a range of data as a. Here we touch on the final criteria that '[h] must simply do a better job of explaining [e] than any of its current competitors.' On all criteria, a is the best explanation available. Conclusion

Steven and I agree that:

1) 'science cannot disprove supernatural agents',

2) Angels are logically possible (Steven doesn't discount evidence a priori).

3) Well documented experience offers the strongest independent evidence for the reality of angelic beings.

4) Angels cannot have been wholly good when first created.

In my last response I suggested that Steven might also feel able to agree with me that no resolution on the question of angels can be reached independently of world-view presuppositions, that the evidence for angels is insufficient on its own to overturn a non-theistic worldview (as a non-theist he must agree with this), and that the evidence makes belief in angels rational for anyone who is an agnostic or theist. Indeed, I propose that the evidence makes belief in angels (even in the specific angels mentioned by Steven) not only rationally permissible, but obligatory, for anyone who accepts that God exists and has communicated with us through the Bible and/or Jesus. What Steven thinks to these latter propositions remains to be seen, but I agree with J.J.Haldane: 'Given the assumption of a creator God I find it entirely plausible that we are not alone as rational beings; and scripture and the writings of the saints offer evidence for the existence of angels - pure spiritual beings, not the androgynous chorus line of popular culture.'

Peter William's Opening Statement

Steven Carr's First Response

Peter Williams's First Response

Steven Carr's First Response

Peter William's Final Response

Debate Page

Comments to Steven Carr

General messages (not for publication) can be sent to me using Not for Publication

OR Use the Comments page ,if you do not want to use email