Mediumship, like humor, comes in different styles. One way to distinguish the varieties of mediumship is with respect to the medium’s degree of altered state or “trance.” Some mediums relay messages from (or describe) communicators without significantly modifying their normal waking state. They claim that their mediumistic activities are as routine, clear, and natural as reporting statements of (or describing) people standing next to them. Other mediums experience a light trance, in which they are slightly “spacy” or distracted but are nevertheless able to go about their usual business (e.g., washing dishes in the kitchen). Still other mediums experience a much deeper alteration of consciousness, similar (at least on the surface) to what occurs in cases of multiple personality (or dissociative identity) disorder (MPD/DID). Like multiples, these mediums “lose time,” and their normal waking state is replaced by another state of consciousness (that of the ostensible communicator). Usually, the medium has no knowledge of what transpired during the communication and apparent possession of her body.

In addition to these variations in the mediumistic state, we can also discern variations in the mediumistic process. Some mediums seem almost to be taking dictation, as if they were listening to someone and then simply repeating or interpreting what they were told. For example, the medium might say, “Your Uncle Harry is speaking to me now, and he wants you to know that…” Other mediums, rather than “hearing” communicators, experience mental images which they then describe or interpret. For example, the medium might say, “I see an older, bearded man, fairly short, and wearing a dark suit of the kind worn in the late 1800s. The suit is torn, and the man is pointing to a large bruise on his forehead. I feel as if he has had an accident involving heavy machinery.”

But the most dramatic cases are those in which mediums seem to be physically controlled by a communicator. Some relinquish command only of parts of their bodies, as in automatic writing or drawing. Others apparently have their entire body possessed or controlled by a so-called trance-personality (or persona), who speaks in a different voice from that of the medium and whose behavior (in the best cases) resembles that of a formerly living person. In the most evidential cases, these similarities extend to subtle verbal mannerisms, voice quality, and characteristic facial expressions. Although trance personalities often supplant a medium’s waking consciousness, others leave the medium moderately functional. For example, some mediums can do housework and even write letters while communicators speak or write through them. But not all heavy-trance mediums deliver their messages through trance-personalities. For example, Mrs. Piper would sometimes swoon and drop her head onto pillows arranged in front of her on the table, after which she would produce automatic writing. The sitters then had to move the paper to prevent her words from running off or piling up.

Mediumship can also be a group experience—for example, when sitters gather around a ouija or planchette board, or when they assemble around a seance table for table-tilting. In the case of ouija board communications, the pointer spells out messages one letter at a time (sometimes at a very rapid pace) as it moves around the board. When the planchette is used, sitters lightly touch a pointer attached to a pencil, which writes messages on an underlying piece of paper. And in the case of table-tilting, sitters rest their hands on top of a table. Messages can then be spelled out according to a simple code, or the table can indicate “yes” or “no” to questions posed by the sitters. Generally speaking, participants in these forms of group mediumship feel that some power other than their own moves the objects beneath their fingers. And as with other forms of mediumship, the most interesting cases are when the received messages convey information known normally to no one present at the séance.

Of course, in some instances it seems clear that the material originated from the deep strata of a sitter’s mind. The following example from Gauld is too good not to be repeated.

I was once a sitter in a circle which received pungent communications from Goering and Goebbels and other deceased Nazi leaders. They favoured us with such interesting pieces of information as that Hitler was alive and well and operating a petrol pump in the town of Clifton, Arkansas, and that Martin Bormann was in Gothenburg disguised as a priest by the name of Father Odo. They favoured us also with various apologias for Nazism. After several sessions it became apparent that this little band of unrepentant sinners only communicated when the finger of one particular person was on the glass. Very reluctantly he admitted that many years before he had gone through a phase of admiration for certain features of Hitler’s Germany, and he had joined an extreme right-wing political organization. Now he repudiated, even abhorred, his former paltering with Nazism. None the less these views were clearly still alive in him somewhere, and slipped out when his conscious censorship was circumvented by the ouija board. I am absolutely certain that he was not deliberately manipulating the glass—his embarrassment was too great, and he refused to participate further. (Gauld, 1982, pp. 26–27)

Granted, the material here was initially implausible and relatively easy to trace to a sitter’s mind. And Gauld assures me that nobody at the time took these communications to be evidence of survival. Still, the case illustrates nicely how unconscious needs and interests of the living can manifest in mediumistic contexts. Clearly, what we need to consider is whether similar unconscious processes are at work when the ostensible communications are less obviously bogus.

Mediums often work through so-called controls or control-personalities. These are recurrent and self-consistent characters who act as interpreters or intermediaries (or masters-of-ceremony) between sitters and communicators. Typically, they also look after the medium’s interests, but some controls belligerently assert their own self-interest (e.g., by throwing tantrums and refusing to cooperate with sitters unless appeased). Most controls are flagrantly artificial personalities, often claiming to be from locales that would be exotic to the medium, and often claiming to be and acting like children. Eileen Garrett’s control, Uvani, claimed to be the spirit of a deceased Arab. Mrs. Garrett would speak in a low, masculine tone, and Uvani would usually introduce himself by saying “It is I, Uvani. I bring you greetings, friends.” Mrs. Leonard’s principal control, Feda, was childlike and unpredictable, but often charming. During Feda’s periods of control, Mrs. Leonard was very animated, gesturing frequently and broadly, and she spoke in a childlike, girlish voice, in a foreign accent and with unusual pronunciation of many words. But Uvani and Feda were merely colorful. Other controls are more comical and bizarre. Often, controls of English mediums have claimed to be Native Americans, Black Africans, Arabs, or Chinese, and their opinions, behavior, and diction had the “stilted and stylized” appearance of a caricature or cinematic stereotype (Gauld, 1982, p. 115). Moreover, some controls exhibit extreme ethnological confusions. One, claiming to be a Black African child, asked C. D. Broad for the key to his wigwam. Another, who claimed to be a Native American chief, requested the sitters to encourage him by singing “Swanee River” (Broad, 1962, p. 254).

Control personalities provide ammunition for those opposed to survivalist interpretations of mediumship. Given their obvious artificiality, there can be little doubt that the medium constructed them subconsciously. In fact, as Gauld observed in connection with Mrs. Piper,

Even the most life-like and realistic controls, such as GP, show signs of being impersonations (not deliberate ones). They break down at just the point where Mrs. Piper’s own stock of knowledge runs out, viz. when they are required to talk coherently of science, philosophy and literature (which the living GP could readily have done). (Gauld, 1982, p. 114)

Some of Mrs. Piper’s controls offered a familiar but unsatisfactory explanation of these deficiencies. They claimed to be confused and disoriented by coming into the medium’s “light.” But, as Gauld recognized,

The confusion which obliterates the controls’ grasp of science and philosophy does not prevent them from spouting reams of pompous nonsense upon religious and philosophical topics and presenting it as the profoundest truth…; so that we have to attribute to them not just confusion but downright tale-spinning, which was certainly not a habit of the purported communicators in life, nor yet of the normal Mrs. Piper. (Gauld, 1982, p. 115)

Gauld continues,

Similar tale-spinning tendencies are manifested in the way in which controls cover up their mistakes. Controls will, generally speaking, not admit their blunders. They will rationalize, explain away, concoct any excuse, however tenuous and childish. All other considerations seem subordinated to an overwhelming urge to keep the drama flowing without pause or hiccup. (Gauld, 1982, p. 115)

Granted, some controls are compellingly lifelike. But as Gauld also notes, it doesn’t help that the most convincing communicators adamantly vouch for the authenticity of the least plausible controls. That makes it seem as if “the authenticity of the former is inextricably and disadvantageously tied up with the authenticity of the latter” (p. 115). So it doesn’t require much of a leap to suspect that realistic communicators are likewise creative constructs, possibly based on information the medium acquired psychically.

I should add that the communicators themselves are also sometimes clearly fictional, although seldom as extravagantly contrived as the controls who apparently transmit their messages. For example, in 1909 Stanley Hall deceived Mrs. Piper’s Hodgson control by asking for, and then receiving, messages from a niece, Bessie Beals, who never existed. When confronted by Hall, the Hodgson-control could only offer lame explanations. The interesting exchange between investigator and spirit-control was as follows (Tanner, 1910, p. 254; Sidgwick, 1915, pp. 177–178) (The expression “Hodgsonp” refers, not to Hodgson himself, who was dead, but to the mediumistic Hodgson-persona).

Dr. Hall: Well, what do you say to this, Hodgson. I asked you to call Bessie Beals, and there is no such person. How do you explain that?

Hodgsonp: Bessie Beals is here, and not the—(Note by Miss Tanner) [At this point we laughed and I made some remark to the effect that that was just what we had said Hodgson would do, and the hand continued thus,]

Hodgsonp: I know a Bessie Beals. Her mother asked about her before. Mother asked about her before.

Dr. Hall: I don’t know about that, Hodgson. Bessie Beals is a pure fiction.

Hodgsonp: I refer to a lady who asked me the same thing and the same name.

Dr. Hall: Guess you are wrong about that, Hodgson.

Hodgsonp: Yes, I am mistaken in her. I am mistaken. Her name was not Bessie, but Jessie Beals

Mrs. Sidgwick’s comment on this exchange was understated, to say the least. She wrote, “We can only say about this explanation that it is not plausible” (Sidgwick, 1915, p. 178). Nevertheless, it’s difficult to know what, exactly, was going on. For example, we may need to account for the role played by the investigators’ skeptically contemptuous and deceptive attitude. It’s possible, if not likely, that the results Tanner and Hall obtained with Mrs. Piper represent a kind of experimenter effect found throughout the history of parapsychology and elsewhere in the behavioral sciences. In fact, these investigators seemed clearly to have a chilling effect on Mrs. Piper’s mediumistic abilities. In the next chapter I’ll return to that episode and discuss other sorts of outside influences on mediumistic performance.

Of course, not all mediumistic communicators are dramatically ludicrous or otherwise easily dismissed. One reason mediumship is interesting is that some cases provide compelling evidence suggesting survival, and very clear evidence that something paranormal was going on. Obviously, then, the crucial issue is whether this residue of mediumistic phenomena favors a survivalist explanation even more than a super-psi explanation.

2. Outline of the Issues

In the previous chapter I noted that subjects in good mediumistic and reincarnation cases demonstrate anomalous sorts of knowledge—either in the form of information (knowledge-that) or as embodied in skills and capacities (knowledge-how). In many of the mediumistic cases, this knowledge is supplied by what Ian Stevenson has termed “drop-in” communicators. As the name suggests, drop-in communicators arrive uninvited, and often neither medium nor sitters know who they are. Of course, in the most intriguing cases, drop-ins make statements about themselves which are later verified and which nobody present at the seance knew to be true. And occasionally, the drop-in’s behavior (as expressed through the medium) resembles that of the communicator when alive.

Although drop-in communications are fairly common in mediumistic settings, only a small number seem to have been verified (e.g., Gauld, 1971, 1993; Gibbes, 1937; Haraldsson & Stevenson, 1975a; Myers, 1903, vol. 2, pp. 471–477; Stevenson, 1970, 1973; Tyrrell, 1939; Zorab, 1940). Presumably, that’s because verifying a drop-in case is a complex and laborious business. Typically, it demands careful note taking at the outset, and then the painstaking processes of conducting detailed interviews, and locating and inspecting public records. Probably, most sitters at casual seances would rather avoid such a time-consuming investigation. Moreover, sitters may be more interested in apparent personal communications from deceased friends and relatives, and of course they may be convinced already about the reality of survival. So they would probably find it either irrelevant or unnecessary to gather careful evidence of survival.

As Gauld noted, good drop-in cases discourage super-psi explanations for two main reasons (Gauld, 1982, pp. 58ff). The first concerns the identity and apparent purpose of the communicator. We need to explain why the medium (or someone else present at the sitting) would use ESP to obtain information about an individual who was unknown to those present at the séance. Generally speaking, those attending séances are interested primarily in “contacting” individuals they knew. Why, then, would a medium apparently waste time providing information about a total stranger, one whose story can’t be verified without further (and probably tedious) investigation? If the medium is using psi or normal sleuthing to obtain information about purported communicators, why not just gather information about those likely to be targeted by the sitters? We also need to explain why the communicator supplies information of no apparent interest to the sitters but of understandably serious concern to the communicator. (A good example is the case of Runki’s leg, which we examine below.) By contrast, the survival hypothesis seems appealingly straightforward. To put it simply, the seance provides a forum for communication which the drop-in exploits for urgent personal reasons (e.g., to console a grieving relative or to take care of important unfinished business).

The second obstacle facing super-psi explanations concerns the obscurity and diffuseness of the information provided by the drop-in. According to the super-psi version of events, sitters at the seance use ESP to obtain this information themselves. However, in the best cases that information is hidden and also seemingly irrelevant to any living person’s concerns. So in addition to considering why a living person would have been motivated to dig up that obscure information, advocates of super psi must explain how the medium or sitters were able to locate it in the first place. Since the information was apparently personally meaningless, what pointed them in the right direction? Moreover, although in most cases the information would need to come from only a single source (e.g., a written record, or one living person’s memory), in others it would have to be assembled from separate and equally obscure sources (e.g., different written records and memories). To many, that attributes an implausible degree of accurate psychic functioning to those at the seance. By contrast, on the survival hypothesis the necessary information may all be reasonably and conveniently attributed to a single, conspicuous source: the drop-in communicator.

Certainly, we can agree with Gauld that the first of these problems is substantial. At least on the surface, the psychodynamics of drop-in cases are most prominent and straightforward from the communicator’s point of view. The second problem, however, may be overrated. One reason (noted in chapter 1) is that we presently have no grounds for imposing any limits on the scope or refinement of psychic functioning. But another problem concerns the notion of obscure information. Ordinarily, we understand (roughly at least) what it means to say that some information is obscure. But that conception of obscurity applies only to normal methods of acquiring information. For example, we consider information to be obscure when it’s outside our perceptual field or otherwise difficult to access physically (e.g., if it’s behind layers of security or other barriers, or if it’s remote geographically and not accessible electronically). By contrast, we don’t understand how any physically or perceptually remote information might be acquired by ESP, whether it’s the carefully sealed picture on the table before us or an object thousands of miles away. But then we’re in no position to insist that normally obscure information is also psychically obscure. And in fact, most targets identified by ESP satisfy our ordinary criteria of obscurity. So right from the start, it’s implausible to insist that normal forms of obscurity are barriers to the operation of ESP. Similarly, we’re in no position to insist that the diffuseness of information is a barrier to successful ESP. As far as we know, psychically accessing multiple sources of normally obscure information is no more imposing than accessing one.

Furthermore, as we evaluate the possibility of super psi, we should be careful not to assume that super psi is merely a collection of really good psi, of the kind we apparently see in limited forms in some lab experiments. When we do that, it’s all too easy to think that psychic functioning involves an effort of some kind, so that if one psi performance is difficult, several ought to be out of the question. Instead (as I noted in the last chapter), we need to consider the super-psi hypothesis in all its intimidating richness. In fact, we need to look beyond multiple-process forms of super psi and consider seriously a kind of magic wand hypothesis. According to that hypothesis, (a) psi agency requires nothing more than an efficacious need or wish (under favorable conditions), and (b) given such an efficacious need or wish, virtually anything at all can happen. But in that case, we needn’t suppose that fine-grained ESP requires complex search procedures—for example, of the sort used in looking up references in a library, acquiring information over the internet, or foraging for clues in a police investigation. And we needn’t suppose that refined PK demands constant ESP monitoring of the results of one’s activities—for example, as steering a car and brain surgery require sensory feedback. It may be enough merely to wish for something to happen, and then it does. Task complexity may simply not be an issue. 1

I also noted in the last chapter that from this perspective, successful psychic functioning would resemble placebo effects or success in biofeedback studies, where subjects produce remarkable physiological changes without conscious effort and without knowing how they accomplished the task. In fact, several converging lines of parapsychological research suggest an intriguing parallel with biofeedback studies. In both cases it appears that subjects do best under conditions of passive volition —that is, when (instead of actively trying to succeed) they simply wish, want, or expect, a result to occur (e.g., Palmer, 1978, pp. 90–92; Schmeidler, 1994, pp. 175–176).2 Furthermore, it’s not just in connection with the super-psi hypothesis that the apparent irrelevance of task complexity is an issue. Conventional experiments with random number generators also suggest that task complexity has little if any effect on the success of psi tasks (e.g., Braude, 2002; Schmidt, 1975, 1976.) So it’s hardly unprecedented to suggest, in this context, that successful psychic functioning may be insensitive to task complexity, or at least more so than is commonly supposed.

The first problem, that of explaining the identity and apparent purpose of the communicator, raises more vexing (if less bizarre) issues. In fact, the survival hypothesis has obvious advantages when it comes to explaining why the medium “connects with” one unknown deceased person rather than another. Whereas advocates of super psi need to explain why a living person selects an unknown deceased person out of an unlimited pool, survivalists tell an apparently simple story. The communicator self-selects (Gauld, 1982, p. 61). Indeed, as Stevenson once remarked, “Some ‘drop-in’ communicators have explained their presence very well” (Stevenson, 1970, p. 63). But according to Gauld, on the super-psi hypothesis “we seem reduced…to supposing that selection of communicator depends upon the random operation of wholly unknown factors” (p. 59). Stevenson concurs, and his way of stating the point brings its weakness squarely into the open. He writes,

Since the [super-psi] theory assumes that discarnate personalities do not exist, it has to attribute motive for a particular mediumistic communication or apparitional experience to the subject. But evidence of such a motive is not always available, and we should not assume that one exists in the absence of such evidence. (Stevenson, 1984, p. 159)

The proper reply to this has two parts: First, we shouldn’t assume that the evidence of motive is absent unless we look for it; and second, hardly anyone looks for it, except in the casual or relatively superficial way mentioned in the last chapter. If the motives in question exist, they’re unlikely to reveal themselves to the sorts of investigations Stevenson and others conduct. Without an extensive and penetrating examination into the lives of clearly relevant (and perhaps even seemingly peripheral) personnel, we have no basis for rejecting explanations in terms of motivated super psi.

Some might feel that this criticism is unfair. First, sometimes a cigar is just a cigar. That is, sometimes sitters’ motives seem quite clear, and additional competent questioning turns up nothing worth pursuing (the sitter group described by Gauld, 1971, 1993, may well belong in that category). Moreover, our goals and interests are often unconscious and difficult to discern, and in actual case investigations we may have no real prospect of ferreting out subjects’ deepest needs and concerns. That’s especially true for older cases, where we can no longer interrogate medium and sitters.

Now I grant that some cases seem psychologically straightforward. And I grant that in many (if not most) cases we may never get a handle on the potentially relevant underlying psychodynamics, no matter how hard we try. Still, that’s no reason for not trying, even in cases that seem clear. And often we don’t have to probe very far to glimpse some of the significant psychological activity simmering beneath the surface. Indeed, a few case studies reveal clearly how much we stand to learn from psychological detective work, and they show why even sensitive questioning of the sitters may not reveal relevant dispositions and behavioral patterns.

3. The Cagliostro Case

Possibly the best example is Eisenbud’s intriguing and brilliant analysis of Mrs. Chenoweth’s Cagliostro persona (Eisenbud, 1992, chapter 14). The case is complex and deserves to be read in its entirety. As in all the best cases, it’s the details that count, and no summary can do justice to the subtlety and shrewdness of Eisenbud’s review. Unfortunately, however, we have no choice but to summarize.

James Hyslop was Professor of logic and ethics at Columbia University from 1889 to 1902 and also one of the founders of the American Society for Psychical Research. A staunch advocate of the survival hypothesis, Hyslop wrote extensively on psychical research and conducted many detailed investigations of mediums. In 1914, he held a series of sittings with one of his favorite mediums, Mrs. Minnie M. Soule, whose pseudonym was “Mrs. Chenoweth.” Another sitter was Doris Fischer, the subject of W. F. Prince’s monumental study of multiple personality (Prince, 1915/16). Hyslop, too, had a professional interest in Doris, because he suspected that multiple personality was a disguised form of mediumship (Hyslop, 1917).

Over the series of sittings, several trance personalities manifested through Mrs. Chenoweth. One of the most interesting was a drop-in who claimed to be Count Alessandro Cagliostro, the notorious eighteenth-century mystic, healer, and (as some have alleged) con artist. The behavior of this trance personality was vivid and flamboyantly salacious, but nevertheless rather one-dimensional. Cagliostro came through as a vigorous defender of sexual freedom, including that of women, and as a severe critic of Christianity. In fact, the Cagliostro persona behaved like “a reckless blasphemer who wouldn’t have lasted forty-eight hours in the Church-dominated Europe of the time” (Eisenbud, p. 230). More important, however, the behavior of the Cagliostro persona was probably quite different from that of the actual Cagliostro.

The real Cagliostro was arrested in Rome in 1789 and brought to trial by the Holy Inquisition. Charged with freemasonry, heresy, and promulgating magic and superstition, he was condemned to death, although that sentence was later reduced to life imprisonment. But Mrs. Chenoweth’s Cagliostro-persona gave no indication of that history. Nor did it match the personality of Cagliostro emerging from available biographies. Instead, it corresponded more closely to a very different portrayal of the Count, one that apparently originated with the Vatican. Eisenbud claimed that the Vatican’s description of Cagliostro remained buried in its archives and that the Cagliostro-persona’s behavior matched none of the accounts of Cagliostro’s life published up to the time of the sittings. According to Eisenbud, it wasn’t until 1972 that an account of Cagliostro appeared which presented the Vatican’s version of the trial, and that account wasn’t translated into English until 1974.

So this case presents a surprising puzzle. The Cagliostro trance persona failed to match the picture of Cagliostro painted by all the reliable sources available, not only at the time of the sittings, but until the present time as well. Outside of the Vatican’s version of the facts, not even critical accounts of Cagliostro accused him of being lascivious or religiously cynical. Indeed, there is reason to think that Cagliostro’s trial was rigged and that it was simply expedient for the Vatican to charge him with blasphemy and rampaging licentiousness.

It seems clear, then, that Mrs. Chenoweth’s trance impersonation presented no evidence for the survival of Cagliostro. But in that case, what was the function, psychodynamically speaking, of the Cagliostro-persona? Why should a colorful but historically inaccurate trance personality emerge who was so flagrantly sexual and religiously cynical? Whose needs might that have satisfied? Eisenbud recognized that these were the appropriate questions to ask, and he offered intriguing reasons for thinking that the Cagliostro-persona had a great deal to do with, among other things, the sitters’ sexual repressions and religious upbringing. Here, of course, is where details count most; but we’ll have to settle for a tasteful selection.

For example, Hyslop, who “apparently devoted much of his life to spiritual and moral development” (Eisenbud, pp. 233–234), predictably found the Count to be a deplorable figure. So it seems particularly interesting and significant that Hyslop—by his own admission—repeatedly encountered nonspiritual, immoral, and “sensuous” characters in sittings he conducted. As Eisenbud nicely put it, “Hyslop seemed to find himself locked into some sort of epic Manichean struggle against the forces of evil” (p. 234). And as Eisenbud also recognized, in those days few took seriously as a “psychological possibility” that there might be “an internal dynamic connection between the spiritually inclined professor and the low ‘sensuous’ characters who kept turning up at the séances he attended” (p. 234). In fact, in those days few appreciated the possibility of experimenter effects in case investigations. But as many now recognize, investigators can’t be viewed simply as passive outside observers and note takers, playing no role in the production of the phenomena. Of course, this problem is especially serious in the behavioral sciences, where there are many opportunities for subtle and uncontrollable forms of experimenter influence and subject-experimenter interactions. But in that case, experimenter effects, both normal and paranormal, are presumably as likely in the study of mediums as in more conventional laboratory tests.

But Hyslop wasn’t the only seance participant who may have had a deep investment in the Cagliostro-persona. Doris (like Hyslop) was a model of moral propriety. In fact, she seemed almost to be a caricature of naive virtue. According to W. F. Prince, “A purer, more guileless soul it was never the writer’s good fortune to know.” Prince also notes that Doris had a “notable lack of sex-instinct.” Now it apparently never occurred to Prince (or, for that matter, Hyslop) that Doris’s lack of sexuality may have indicated an inhibition of powerful sexual desires. And as I noted above, in 1914 few considered seriously the possibility of experimenter-influence, and in particular the possibility that persons other than the medium might play an active role in shaping the material presented by a medium.

Furthermore, Mrs. Chenoweth betrayed a surprisingly intense attachment to the Count when it looked as if other communicators might banish him from the scene. (To appreciate the powerful emotions behind her words, I suggest reading her comments aloud.) Crying to the other ostensible communicators who tried to exorcise him, Mrs. Chenoweth said, “You give him back [Pause] You give him back…Give the Count back to me.” Hyslop asked who wanted the Count, and Mrs. Chenoweth replied, “We all do. We are lost. We are lost, we are lost, we are lost [Pause] Oh, Devils, to take him away from us. [Distress and crying]…I won’t stand it [Pause] I don’t want your old God. I want the Count.”

For these and other reasons, Eisenbud proposes that the Cagliostro-persona might have been a composite “dream figure omnibus for the repressed unconscious hankerings of all the principals at the sittings” (pp. 237–238). And considering some of the startlingly close correspondences between the trance personality and the behavior attributed to Cagliostro by the Church, it appears as if one or more of Mrs. Chenoweth, Hyslop, and Doris Fischer psychically raided an extremely obscure portrayal of Count Cagliostro to provide some material for the sitting.

Alan Gauld, whose mastery of both contemporary and historical sources is nothing short of breathtaking, has expressed some doubt about Eisenbud’s conjectures. He informs me that the Vatican view of Cagliostro was initially published, at least in part, in a book printed in Rome in 1790. That book was then translated into French the following year. According to Gauld, the book “appears to contain quite enough about [Cagliostro’s] licentiousness and heresy to have supplied the source for much subsequent scuttlebut about him” (personal communication, May 15, 1999). Gauld adds that “in addition to historical works about Cagliostro…he appeared in quite a lot of novels and plays, some by well-known authors (including one who had known him, Catherine II of Russia). (He was also a communicator through at least one other medium!) So the scope for medium or sitters to have heard all kinds of things about him, good and bad, was very large even at that date.”

Now Gauld may be right. There may have been several sources, besides the apparently spurious and obscure Vatican version, from which members of Mrs. Chenoweth’s circle could have constructed the Cagliostro persona. And it may be that no psi was required to provide the relevant information. But for present purposes, it doesn’t matter particularly whether Eisenbud’s psychoanalytic conjectures are correct. And it doesn’t even matter whether the correspondences between the Cagliostro-persona and the Vatican account are accidental, attributable to a familiar pool of information, or arranged through psychic sleuthing. What matters is the level at which Eisenbud attempted to evaluate the data.

In psi research, it’s always something of a mystery how best to trace the putative psychic causal lines. The value of Eisenbud’s speculations is the match he seems to have uncovered between (on the one hand) the sitters’ “unconscious hankerings” and life issues and (on the other) the character of the vivid but one-dimensional trance persona. The Cagliostro-persona offers nothing convincing in the way of evidence of survival, and it may well be historically inaccurate even if Gauld’s concerns are justified. But it was at least as dramatically satisfying as most mediumistic trance-personalities, and for that reason it’s precisely the kind of mediumistic performance that makes a strong prima facie case for survival. So despite Gauld’s reservations, Eisenbud’s essay illustrates the level of psychological probing required for illuminating survival research. By understanding how compelling, but non-evidential, mediumistic evidence is produced, we can be more alert for the operation of similar processes in more evidential cases. Still, Eisenbud’s proposals do make good sense of the evidence, including peculiar and otherwise unexplained bits of behavior on the part of the sitters. So even if we reject Eisenbud’s view concerning the origin of the trance-persona’s characteristics, it’s still plausible that the Cagliostro-persona makes most sense as a dramatic creation by one or more of the sitters.3

I suppose some might think that psychological archeology is especially critical in cases such as this, which are emotionally and dramatically potent but also non-evidential. However, I’d argue that they matter most in cases that are apparently evidential, where it’s all too easy to overlook underlying and relatively pedestrian human concerns. After all, both evidential and non-evidential cases may be emotionally charged, and participants may care deeply about the form and content of the phenomena. The Cagliostro case seems to demonstrate both a high level of dramatic creativity and possibly also a kind of psychic collaboration among some of the sitters. But with that sort of evidence staring us in the face, it’s both presumptuous and naive to rule out super-psi conjectures in cases where no comparable depth-psychological study has been conducted. Regrettably, however, by comparison to Eisenbud’s standard of analysis, most other case investigations are unacceptably superficial.

4. Runki’s Leg: The Case

As I noted earlier, there are several related respects in which drop-in communicators are particularly intriguing, and in virtue of which super-psi explanations may seem less plausible than survivalist alternatives. First, the communicators are unknown to the sitters at the time of the sitting. Therefore (and second), it’s hard to see why that particular communicator came to be dramatized or represented during the séance. What pointed the medium’s (or sitters’) psi in that direction? And third, in the best cases the communicator’s motive for communicating seems both greater and clearer than any living person’s motive to receive those communications. That is, the communicator in these cases has a much clearer agenda than any we might reasonably attribute to medium, sitter, or anyone else. So a viable super-psi explanation of a good drop-in case faces several challenges. Naturally, it must account for the verifiable information provided in the sittings and (if necessary) the accurate dramatic representation of the previous personality’s behavior. But more important, it must also explain away the drop-in’s apparent motivations. And for the super-psi explanation to be preferable to the straightforward survivalist explanation, it must specify which living person(s) had needs, even stronger than those of the drop-in, for the séance to unfold as it did. That’s a tall order.

Probably the best drop-in case of all time comes from Iceland, a nation with a rich and distinguished tradition of mysticism, spiritism, and mediumship. The medium in this case, Hafsteinn Bjornsson (1914-1977), is arguably Iceland’s most famous medium (his main competitor would be the physical medium Indridi Indridsson). Hafsteinn’s psychic abilities first surfaced in childhood, and apparently they remained strong thereafter. He began holding regular seances in 1937, and although he didn’t earn a living from these activities, he did accept fees for his services. Hafsteinn was a trance medium, and communicators as well as regular controls spoke through him. (For additional information on Hafsteinn, see Haraldsson, Pratt, & Kristjansson, 1978; Haraldsson & Stevenson, 1974; 1975a; 1975b.)

The case we’re now considering began in the autumn of 1937, during a series of seances held at the home of E. H. Kvaran in Reykjavik. A drop-in appeared at one of the séances, and when asked to give his name, he responded by identifying himself with a stereotypically Icelandic male name intended clearly to be fictitious. He then added, “What the hell does it matter to you what my name is?” One of the sitters asked what he wanted, and the drop-in replied, “I am looking for my leg. I want to have my leg.” His leg, he then said, was in the sea.

For the next year this communicator continued to appear at the seances in Kvaran’s home, continuing to ask for his leg and still withholding his identity. In the autumn of 1938 the séances moved to the home of Lilja Kristjansdottir (with a few changes in personnel), and again the communicator manifested, still demanding his leg and still refusing to give his name. On January 1, 1939, a new sitter joined the circle. He was Ludvik Gudmundsson, a fish merchant and owner of a fish processing factory in Sandgerdi, a village about thirty-six miles from Reykjavik. Although Ludvik and his wife owned a house in Sandgerdi, they lived in Reykjavik. Apparently, Ludvik was introduced to the sittings through a relative and one of the recent additions to the circle, Niels Carlsson. Ludvik had never met Hafsteinn, and the medium apparently knew nothing about Ludvik or his family.

When Ludvik joined the circle, the drop-in said he was glad to meet him. Ludvik didn’t know what to make of this, and he asked the communicator to reveal his identity. Although the communicator continued to refuse, he mentioned that Ludvik knew about his missing leg, which he said was in Ludvik’s house in Sandgerdi.

The drop-in’s behavior during this period differed considerably from that of Hafsteinn. Unlike the medium, he was brusque and rude, and in addition to demanding coffee and alcohol, he often asked for snuff (which Hafsteinn never used). Frequently, he would go through the motions of lifting his hand to his (i.e., the medium’s) nose and sniffing. Moreover, whereas Hafsteinn drank only one or two glasses of wine a year, the communicator’s demand for alcohol corresponds to his later intimation (and some independent evidence) that, in life, he had been a heavy drinker.

After additional sittings in which the drop-in continued to conceal his identity, Ludvik and Niels presented an ultimatum. They said they would do nothing to help him so long as he refused to say who he was. Apparently, that annoyed the drop-in, who then made no appearance for a while. Finally, he returned, probably during the late winter or early spring of 1939 (for some reason the date of that event was not recorded), and he did so by abruptly and aggressively ousting another communicator from the scene. The drop-in then told the following story.

He said that his name was Runolfur Runolfsson (nickname, Runki) and that he was fifty-two years old when he died. Runki lived with his wife at Kolga or Klappakot, near Sandgerdi, and he had been walking home, drunk, from Keflavik (about six miles from Sandgerdi) in the latter part of the day. When he reached Sandgerdi he stopped at a friend’s house and had some more to drink. When he was ready to continue his journey home, his friends protested. Because Runki was inebriated and because the weather was so bad, they said Runki shouldn’t leave unless someone went along with him. But that offer of a designated walker angered Runki, who said he wouldn’t go at all if he couldn’t go alone. So, since Runki’s house was only about fifteen minutes’ walk away, he left by himself. (Evidently, Runki’s friends were ready for him to leave.) At one point, wet and tired, Runki sat on a rock near the sea for a rest, and for another drink from the flask he carried with him. He then fell asleep, was carried away by the tide, and drowned. Runki said this happened in October 1879. The following January, his body washed ashore, and dogs and ravens then tore it to pieces. The remnants of Runki’s body were recovered and buried in the graveyard in Utskalar, about four miles from Sandgerdi and six miles from Keflavik. But a thigh bone was missing from Runki’s remains. It was carried out again to sea, and later washed onshore at Sandgerdi. Then, after being passed around for a while, it ended up in Ludvik’s house. Runki also mentioned that he had been very tall, but it’s not clear from the records whether he mentioned that detail at this sitting or at an earlier time.

Runki claimed that his story could be confirmed by checking the church book at Utskalar Church. So the sitters located the church book and found the record of someone named Runolfur Runolfsson, whose date of death and age at the time of death matched the story told by the drop-in. Runki’s claim about his height was confirmed by Runki’s grandson, who said his grandfather had been more than six feet tall. In the meantime, Ludvik asked elderly residents of Sandgerdi if they knew anything about an unclaimed leg bone in the vicinity. Some recalled vaguely that during the early 1920s a thigh bone (femur) had been “going around” and that it had been washed up by the sea. But they didn’t know whose bone it was or what had become of it. However, one person said he seemed to remember that a bone, not associated with any particular person, had been placed in the wall of Ludvik’s house by a carpenter who had built one of the inner walls downstairs. After an unsuccessful search in one of Ludvik’s walls, an employee of the fish factory helped identify the correct wall. At one point he had lived in a room in Ludvik’s house, and he said he knew of the carpenter placing a femur between two walls. Ludvik tore down the wall he indicated and found what was clearly the femur of a tall person. So, a bone that seemed to be Runki’s was found more than forty years after Runki’s death and approximately three years after Runki’s first appearance.

I imagine that many readers will be puzzled by the manner in which the residents of Sandgerdi handled the unclaimed femur. But according to Haraldsson (a native of Iceland), in that culture and community “it would be considered disrespectful, if not sacrilegious…simply to throw a bone away. At the same time, it would be infeasible to bury a bone in the consecrated ground of a cemetery without knowing the identity of its owner” (Haraldsson & Stevenson, 1975b, p. 40, n.13).

Examination of records from Utskalar parish and elsewhere confirmed various details of Runki’s story (see Haraldsson & Stevenson, 1975b for specifics). One of the most interesting documents is the following, from the Utskalar clergyman’s record book.

On October 16, 1879, Runolfur Runolfsson, living in Klappakot, was missing on account of some accidental or unnatural occurrence on his way home from Keflavik during a storm with rain near his farm, in the middle of the night. He is believed to have been carried along by the storm down to the beach south of the farm boundary at Flankastadir from where the sea carried him away, because his bones were found dismembered much later and his clothes were also washed up separated [i.e., apart from his bones]. (Haraldsson & Stevenson, 1975b, p. 42)

The clergyman also noted that Runki’s remains were buried on January 8, 1880, and that Runki was fifty-two years old when he died.

A second record of Runki’s death, also written by the clergyman at Utskalar, appeared later in a book, Annals of Sudurnes. This book was unpublished and virtually unknown at the time of the sittings. The manuscript was held in Reykjavik’s National Library and was finally published in 1953. Both accounts claim that Runki’s body was dismembered, and neither states that a leg bone was missing from the remains recovered near the shore and buried the following January. But the account in Annals differs in some respects from that found in the church record book. For example, it notes that Runki had been drinking alcohol around the time of his death. Moreover, the Annals account fails to mention Runki’s last name, or the fact that his remains were buried at Utskalar. So if the mediumistic communication was derived psychically (or normally) from existing accounts, it couldn’t have come from just one of those written by the clergyman. Runki’s grandson also couldn’t have been the sole source of the confirmed information. Although he knew that his grandfather had been tall, he had never known his grandfather and apparently was ignorant about the bone and other relevant facts of the case.

Unfortunately, the femur found in Ludvik’s house was never conclusively linked with Runki. However, several considerations lend credibility to that connection. We know that Runki’s body was described as “dismembered,” and although no one claimed that bones were missing from the remains found onshore, the femur was clearly the bone of a tall person. Moreover, it’s rare for bones to be washed ashore in that part of Iceland (or anywhere in Iceland, as far as I know). So it’s plausible to associate recovered bones with the few people known to have died along the coast. Haraldsson and Stevenson were prepared to have Runki’s body disinterred, and they even obtained the consent of his grandchildren. But the graves in Utskalar are unmarked, crowded together, and perhaps also layered atop each other at the same plot of ground. So there seemed no way of determining where to look.

After recovering the bone from his wall, Ludvik had a coffin built for it. He kept the bone for a year and then had it buried during a ceremony conducted at Utskalar. Those present at the ceremony believed they were burying Runki’s final remains. The clergyman eulogized Runki, the choir sang, and afterwards the clergyman held a reception with refreshments at his home. That reception was attended by several of Hafsteinn’s regular sitters. At the next seance held by Hafsteinn, Runki came and expressed gratitude for the proper disposal of his leg. He claimed he had been present at the ceremony and reception, and he described those events in detail. Although Runki didn’t disappear after his business was settled, as many drop-ins do, he did mellow and continued to serve as Hafsteinn’s principal control.

5. Runki’s Leg: Theoretical Considerations and Nagging Concerns

Haraldsson and Stevenson investigated this case carefully and considered whether Hafsteinn might have obtained the relevant information by normal means, either by visiting the National Archives in Reykjavik where the Utskalar parish records were kept, or the National Library where the Annals were located. It turns out that Hafsteinn had visited the National Archives during the sittings, but about six months after Runki had provided the account of his demise. Originally, Hafsteinn claimed not to have visited the Archives at all, but after Haraldsson told him that his signature had been found in the guest book for November 24, 1939, he recalled that he had gone there to examine the records which sitters told him they had verified. Haraldsson and Stevenson also determined that the guest book at the National Archives is not an entirely reliable record of visitors, and that some visitors’ signatures are never recorded. Still, for reasons Haraldsson and Stevenson discuss in detail, I’m inclined to agree either (a) that we should interpret Hafsteinn’s initial testimony as an honest memory lapse for an event that took place thirty-two years before he was interrogated, or (b) that Hafsteinn suppressed the information of his visit out of fear that his trip to the Archives would look suspicious. Besides, the documents in the National Archives and Library don’t deal with the matter of the leg found in Ludvik’s home. So at best they cover only part of what makes the case so interesting.

Of course, no case is ideal. There are always vast numbers of details to examine, and omissions may loom larger in the clarity of hindsight. So presumably there will inevitably be annoying respects in which even the best cases could be stronger. Not surprisingly, then, some puzzling features of this case merit attention. Some cast doubt on the survivalist interpretation of the evidence, and others simply deepen the mystery of the case.

First, when Runki gave details about his life, he said he was fifty-two years old when he died. But if the church records are correct, Runki was in fact about two months shy of fifty-one at the time. Curiously, though, the Utskalar parish notebook entry also says that Runki was fifty-two when he died. Haraldsson and Stevenson mention this discrepancy in a footnote, and they offer a reasonable (but not compelling) explanation. They suggest that the clergyman who made the note might have meant to say that, at the time Runki’s remains were recovered and Runki’s death was confirmed (in January 1880), Runki was in his fifty-second year. However, this detail may be more revealing than Haraldsson and Stevenson realized. Consider: If Runki wasn’t actually communicating, and if someone at the seance had, either normally or paranormally, scoured existing records for information, this is the sort of error we could expect to see. To figure out Runki’s actual age at the time of death, one would have to locate the appropriate records and do some calculating. But to identify Runki as fifty-two years old at his death, one would only have to read it off the false or misleading record in the parish notebook.

Now let’s play devil’s advocate and be as sympathetic as we can to the survivalist. From that point of view, we have to concede that Runki might have been confused (both before and after bodily death), either from the ravages of alcoholism or from the possible strain of communicating mediumistically. Even if it opens the door for reckless survivalist speculation, we must admit that, if postmortem communication is possible, we have no idea how difficult or easy it might be. For example, we don’t know what sort of toll it might take on mental acuity, whether we might remain stuck with the cognitive impairments we had at death, or how “noisy” the mediumistic channel might be. And certainly there’s no reason to assume that survival increases (or even preserves) the clarity and accuracy of our memories. After all, the banality and fallibility of most ostensible communications is notorious. So for all we know, the error communicated about Runki’s age when he died may be a typical (if not predictable) lapse and therefore no grounds for suspicion, especially if there are no other lingering doubts about the evidence.

But there are some additional nagging concerns. Originally, Runki said his leg was in the sea. Only after Ludvik joined the circle did Runki say it was in Ludvik’s house and not the sea. How do we make sense of that shift in position? Survivalists have several explanatory options, none of which strike me as compelling or attractive, but none of which we can conclusively rule out. First, survivalists might argue that, before Ludvik joined the circle, nobody at the seance would have been in a position to help Runki find his leg. So perhaps Runki was merely seizing the opportunity to vent his frustration over his missing leg. Then, after Ludvik joined the group, that might have helped Runki focus on the location of his leg, or perhaps it finally gave him reason to direct the sitters to Ludvik’s house. However, I don’t see why Ludvik’s absence from the sitter group would have prevented or discouraged Runki from directing the sitters to the appropriate house in Sandgerdi. Survivalists might also argue (again) that Runki was confused, or that in the struggle to communicate he might not have conveyed clearly that his leg had been in the sea. As we’ve noted, we have no reason to insist that mediumistic communications are easy or noise free.

On the other hand, suppose that living persons were (normally or paranormally) assembling the Runki story as the case progressed. In that case, Ludvik’s arrival and the existence of somebody’s (not necessarily Runki’s) femur in his house, made it viable at that time to construct a more compelling case. It’s possible, but highly unlikely, that once Ludvik had been invited to join the circle, the medium did some quick research and incorporated into the seance the information about the hidden femur. But considering the obscurity of that information and the lengths to which Ludvik had to go in order to learn about and locate the leg, I think we can rule this out. But, on the super-psi hypothesis, the information could have been acquired psychically, once Ludvik joined the circle.

Moreover, although this is not the only case in which a crucial sitter joins the circle after the drop-in’s first appearance,4 it certainly seems to be a striking bit of serendipity that Ludvik arrived on the scene. How is that to be explained on either the survivalist or super-psi reading of the case? As just a piece of good luck, or as a sequence of events orchestrated somehow in order to make the case more convincing? And if the latter, who would have been able to pull it off? Let’s grant, plausibly (but at least for the sake of argument), that the medium is innocent of any normal shenanigans in determining the sequence of events. And in fact, there’s no evidence of any conspiracy involving medium or sitters, and no basis for sustaining a general suspicion about Hafsteinn’s integrity. So could we plausibly regard Runki as the director behind the scenes? I don’t pretend to know how to answer these questions. I consider them simply to be lingering puzzles about the case.

We might also wonder why Runki disappeared for several months after Ludvik and Niels presented their ultimatum. Haraldsson and Stevenson claimed that Runki seemed annoyed, but since they mention no other behavioral signs of annoyance, that conjecture seems charitable at best. We need to be both fair-minded and circumspect with matters of this sort, and it’s not outrageous to interpret Runki’s disappearance with more suspicion. Since the details of Runki’s story were provided only after Runki returned from his hiatus, we might wonder, reasonably, whether that period was needed for some normal or paranormal information gathering. But before we get carried away with skeptical musings, we should also remember that Hafsteinn’s confirmed visit to the National Archives was six months after Runki told his story, and (as I noted) that account tells only part of the story. So although I see no solid reasons for worrying about Runki’s absence during this period, the case would be even more convincing if that absence had not occurred.

Finally, why didn’t the communicator help identify Runki’s unmarked grave? It would have been a powerful addition to the evidence to have found those remains missing a femur of the appropriate size. Haraldsson and Stevenson’s observations about the configuration of unmarked graves, although legitimate, do little to allay concerns. After all, if Runki could identify where his missing leg was located, why couldn’t he also direct investigators to the rest of his bones? Or (to put a super-psi spin on this), if Hafsteinn could paranormally locate the leg, why not also the site of Runki’s other remains? There may well be reasonable answers to these questions, although I don’t know what they are. And although I don’t consider our inability to answer them reason to dismiss an otherwise very provocative case, they remain sources of concern, and they illustrate again how far from ideal this case is.

But despite its weaknesses, the case of Runki’s leg illustrates clearly how drop-in communications lend support to the survival hypothesis. Haraldsson and Stevenson nicely summarize the issues concerning the correct information provided during the séances.

It does not seem feasible to attribute all of this information to any single person or any single written source. And this would be true, we believe, whether the medium acquired the information normally or by extrasensory perception. We think, therefore, that some process of integration of details derived from different persons or other sources must be supposed in the interpretation of the case. It may be simplest to explain this integration as due to Runki’s survival after his physical death with retention of many memories and their subsequent communication through the mediumship of Hafsteinn. On the other hand, sensitives have been known to accomplish remarkable feats of deriving and integrating information without the participation of any purported discarnate personality. (Haraldsson & Stevenson, 1975b, p. 57)

So, as far as the case’s behavioral details and underlying psychology are concerned, there’s both good news and bad news for the survivalist. The bad news is that we have no idea what Runki’s character was like, except for the evidence that he drank heavily. Therefore, we don’t know what to make of Hafsteinn’s Runki trance-persona. Besides, the Cagliostro case reminds us that vivid behavior different from that of the medium needn’t be evidential. The good news is that the drop-in’s motivations to communicate are much clearer and more straightforward than those we would need to ascribe to the sitters, and even to Hafsteinn, who at this early stage in his mediumistic career already had a solid reputation as a psychic. Even if he might have benefitted somewhat from additional good publicity, he didn’t need this case either to establish or cement his reputation. Furthermore, the drop-in’s behavior, after the burial of the femur, adds credibility to the survivalist interpretation. Runki seemed satisfied that his bones were now all properly disposed of, and although it would have been appropriate for him (and typical behavior for a drop-in) to depart once his affairs were settled, his mellower and helpful participation at subsequent seances was no less appropriate.5

6. Concluding Remarks

Although the best cases are by no means coercive, the evidence for drop-ins, overall, seems to strengthen the case for survival. Granted, we can’t conclusively rule out explanations in terms of motivated psi among the living. But as the challenges facing super-psi explanations mount, their antecedent plausibility decreases. Even if we grant that task complexity may be overrated as an obstacle to psi success, and even if we grant that what really motivates people may not be the concerns lying closest to the surface, drop-in cases make particularly good sense in terms of the ostensible communicator’s expressed motives for communicating. As a result, survivalist interpretations of those cases seem more parsimonious than their super-psi alternatives. As we observed earlier, anti-suryivalists need to explain why a séance participant used ESP to gather information about a person known to nobody present. They also need to explain why the communicator’s needs or interests are so much more clear-cut than those we could reasonably attribute to medium or sitter, even after reasonable probing. And of course, whereas communicators supply information they would be likely to know, living persons would have to derive that information from different and often (normally) obscure sources.

Moreover, the very fact that there are drop-in cases seems to strengthen the case for survival. As Gauld correctly observes,

if there were no verified cases of “drop in” communicators the survivalist case would be considerably weakened. For if people do survive death with some at least of their former interests and affections, and if communication is a possibility, we should expect that not a few deceased persons would try to contact living persons for exactly the sorts of plausible-sounding reasons that “drop in” communicators quite often give. (Gauld, 1971, pp. 276–277)

NOTES

1 I’m aware of the intense and deep resistance this conjecture arouses in most people, and I realize that it usually survives even a painstaking appraisal of the issues and examination of the relevant evidence (e.g., of the kind provided in Braude, 1997). But although that resistance is understandable emotionally, it may be indefensible intellectually. Like so much else in this area, this issue is much more interesting and complex than it might seem initially. Still, I don’t expect the reader to take this on faith, and we can’t review all the relevant material here. So I simply urge readers to keep an open mind on the subject and tentatively accept the possibility of unlimited psychic functioning, if only as a thought experiment, just to see where it leads in the context of this discussion.

2 For information on placebos and biofeedback, see, e.g., Basmajian, 1963, 1972; Frank & Frank, 1991; Rossi & Cheek, 1988; White, Tursky, & Schwartz, 1985).

3 In fact, a contemporary case lends indirect support to Eisenbud’s conjecture. In the early 1970s, members of the Toronto Society for Psychical Research received ostensible spirit messages (primarily through table rapping) from a communicator they invented (“Philip”), and in whose fictitious history and personality they immersed themselves in preparation for the séances. Sitter-group members realized that the raps in the table seemed to be, as one member put it, “psychokinesis by committee” (Owen & Sparrow, 1976). Although the invention of Philip was an overt project, the sittings nevertheless support the view that something similar could occur surreptitiously.

4 In the drop-in case of Edward Druce (Gauld, 1971, pp. 301–302), sitter R. W. (who knew Druce) joined the circle after Druce’s initial appearance.

5 Colleagues and students have made very helpful comments and suggestions to me about this case. In particular, I’d like to thank Christian Perring and my student, Pratima Thotakura.

In successful proxy sittings, such as those described in the last chapter, there is still, it is often, however implausibly, argued, some kind of link between the medium and some absent person or persons possessing the relevant information. If necessary an upholder of the super-ESP hypothesis will propose a series of such links, ending up, naturally, with someone who knows whatever it was that the medium came out with. In the class of cases which I shall now discuss even those tenuous and exceedingly ill-defined links are absent. The class concerned is that labelled by Professor Ian Stevenson (153d) ‘drop-in’ communicators.

‘Drop-in’ communicators are, as the name implies, communicators who arrive uninvited, and are ostensibly unknown to medium and sitters. We have on record a number of cases in which ‘drop-in’ communicators have made statements about themselves and their careers which it has subsequently proved possible to verify. Such cases are of obvious theoretical interest and, before proceeding to some actual examples, I shall briefly explore their potential theoretical implications. These implications are, in general, hostile to the super-ESP hypothesis, and favourable to some form of survival theory. The difficulties for the super-ESP hypothesis may be spelled out under two headings, viz. selection of communicator and locating of materials.

The first of these sets of problems becomes apparent if one asks why, in any verified drop-in case, the medium’s supposed ESP should have lit upon facts about that particular deceased person. The facts about the great majority of ‘drop-in’ communicators are not in any way especially eye-catching. They would not, for instance, be enshrined in unusually striking obituary notices or letters of condolence such as might be supposed to attract the medium’s clairvoyance more than would other such notices or letters. Nor, so far as one can tell, would they stand out with such prominence in the minds of grief-stricken relatives as particularly to arrest her telepathic attention. Nor have medium and sitters any special motive for desiring information about |59| that particular deceased person. Some very broad constraints do seem to be imposed on the selection of ‘drop-in’ communicators, in that most of them (that is most of the ones whose statements about themselves have been verified) come from the medium’s own country and speak her own language. But these constraints aside, we seem reduced, on the super-ESP hypothesis, to supposing that selection of communicator depends upon the random operation of wholly unknown factors.

The second set of difficulties which verified ‘drop-in’ cases may raise for the super-ESP hypothesis, that to do with the location of materials, is much more complex and difficult. In most ‘drop-in’ cases there is, no doubt, some single possible source, such as a printed record, or the organized memory system of a living person, from which the medium could through her supposed extrasensory powers have obtained the whole of her information. But what if (and some cases may at least approximate to this type) the requisite information could have been assembled only through the tapping of a number of discrete sources, e.g. the memory systems of several different living persons or a variety of different printed records? How is the medium, having selected the deceased person she will present to her sitters, to discriminate from amongst all the innumerable items of information telepathically and clairvoyantly available to her, those and only those which are relevant to that person? I do not think that it is possible to give an account of this matter that is even remotely plausible.

Consider first the case where the different items are locked in the memories of a number of different people. The most obvious hypothesis is probably that the various relevant memory-sets in different peoples’ minds are all flagged or marked out by similar quasi-perceptible features, for instance recognisably similar images of the deceased communicator. But even if we set aside for the moment the logical difficulties involved in the suggestion that one person can inspect another person’s images by a kind of quasi-perception, the image-theory remains grossly implausible. It would force one to predict that mediums should be prone to confuse with each other persons who simply happened in life to look alike, and even perhaps confuse real people with fictional characters. It would force one further to maintain that each person’s memory-images are, when not in use, stored away in some internal filing cabinet accessible to the medium’s telepathic rummagings. This is a remarkably implausible idea. Yet if one replaces it with the notion that memories are stored up in the form of subtle structural or functional changes in the brain, one must |60| attribute to the medium the ability to read the neural code in which the memories are represented, a skill which no neurophysiologist is currently anywhere near attaining (or, as I shall later argue, ever could attain). Finally, it is in any case quite clear that it is not any quasi-perceptible features of an image which make that image an image of some particular person, but the reference or intentionality with which the imager invests it. An image of a round and jolly face—the same face—can serve as an image of one’s late Uncle Nat, as an image of John Bull, as an image of Mr Pickwick, as an image of a brand of breakfast cereal, as an image of Jupiter, the bringer of jollity, as an image of jollity in general, and so forth. Psychological processes outside the image determine what the image is an image of.

It might seem as though the case where the different items of information exist in the form of written or other records presents less difficulty than the case just discussed. For obituary notices, letters of condolence, and so forth, commonly carry distinctive headings or addresses which, clairvoyantly perceived, would at once indicate that the same person was involved. Perhaps this does simplify the problem; but it is far from making it simple. Consider the case where a number of newspaper notices have to be clairvoyantly collated. Let us assume that (as is commonly the case) the newspapers concerned are old ones, and not current issues lying on breakfast tables throughout the country. Then it has to be supposed that the medium, in the course of her incessant clairvoyant but presumably unconscious browsings among the files of old newspapers, picks out from the enormous number of obituary notices thus accessible to her those and only those relating to a certain person, and then juxtaposes and synthesizes them. In other words she must discriminate these obituaries from all obituaries of persons of the same or similar name, from all obituaries of persons who had similar careers, from all obituaries of persons who had the same dates, and so on and so forth. Anyone who has had (as I have had) occasion to study newspaper obituary notices extensively will realize that this is an exceedingly tall order, and a few examples of obvious confusion between newspaper obituaries would greatly strengthen the clairvoyant explanation—especially in view of the fact that so far as I am aware we do not have, from outside the mediumistic situation, a single properly authenticated example of a clairvoyant managing to read a concealed passage of prose in anything like the necessary detail.

It is thus possible to construct an idealized ‘drop-in’ case which |61| pushes the super-ESP hypothesis to the verge of unintelligibility; indeed beyond that verge. Such a case would have the following features:

(a) The ‘drop-in’ communicator in question would have a strong and comprehensible reason for wishing to communicate; a reason clearly stronger than any which the medium might have for wishing to contact him.

(b) The information which he communicates would be such that the medium could not have obtained it all by extrasensory contact with a single living person, obituary notice, etc.

(c) We can be tolerably certain that the medium could not have obtained the information by ordinary means (this is a point to which I shall shortly return).

It is hardly necessary to spell out how great are the advantages of the survivalist theory in respect of cases where the super-ESP hypothesis would have to suppose that the medium had used her extraordinary powers of ESP to locate several disparate sources of information about the communicator and had then put together the information thus gleaned. It also has obvious advantages when it comes to explaining why the medium selects one unknown deceased person rather than another unknown deceased person as the subject for her extrasensory researches. The deceased person selects himself. As Stevenson remarks (153d, p. 63), ‘Some “drop-in” communicators have explained their presence very well and their motivation to communicate is an important part of the whole case which has to be explained as well as the provenance of any information communicated.’ ‘Drop-in’ communicators may represent themselves as wishing to assuage the grief of living friends, as brought along by persons in the next world who have previously communicated through the same medium, as lost in a kind of limbo where the medium is their only means of contact with others, as linked through common interest to persons present, as altruistically trying to help, as simply ‘dropping in’ for a chat. It is difficult indeed to decide how seriously communicators’ own explanations of their presences ought to be taken; but sometimes at least the professed explanations are ‘in character’.

So much for the theoretical implications of drop-in’ communicators and for the ideal (and hence imaginary) case. We must now get to grips with some actual cases and see to what extent (if at all) they measure up to the ideal.

|62| Cases of verified ‘drop-in’ communicators are fairly scarce in the ‘reputable’ literature of psychical research (for examples see 48; 64b, pp. 97–102; 110a, II, pp. 471–477; 153d; 153e; 162a; 174). How far this reflects an overall scarcity it is hard to say. ‘Drop-in’ communicators seem to be much more characteristic of the ‘home circle’, the ouija and planchette boards, and the automatist experimenting out of curiosity and interest, than they are of the professional medium. There could be various reasons for this, one of them being, of course, that there are often pressures on mediums who regularly give sittings to individual clients to exclude communicators other than those with whom the sitter wishes to speak. Since a high proportion of the investigations of mediumship published by the SPR and the ASPR have concerned mediums of this latter sort, cases of ‘drop-in’ communicators have not often figured (for some cases of this kind with Mrs Piper see 66a, pp. 37–42). Also, of course, the verification of ‘drop-in’ cases requires a good deal of time, and also, very often, a working knowledge of the country’s public records system together with access to a large library. ‘Drop-in’ communicators of the utmost veridicality could march into and out of the average home circle without its occurring to anyone that it would be feasible to check up on them. And where such checks have been undertaken, they have often fallen far short of the required standard of thoroughness.

I shall illustrate this last point with an example taken from Sir Lawrence Jones’s Presidential Address to the SPR (76). In the year 1900 Sir Lawrence had for some time at his house in the South of France a home circle centring around the well-known amateur medium, Miss Kate Wingfield (she is referred to as ‘Miss A.’ in F. W. H. Myers’s Human Personality). Communications were received both by raps and by automatic writing. On 8 September 1900 a certain Sarah Willett, of 7 Sydney Street, London, wrote through Miss Wingfield that she had been shot and killed by one Jack Parr, of Green Street, a polisher. On 30 September, Miss Wingfield had a vision of this girl, and on the evening of 2 October saw an ominous ‘figure with a black thing like a sack tied over his head and shoulders’. Raps spelled out ‘John Parr hanged today’. Miss Willett, whose dallyings with another man had precipitated the murder, expressed terror at the revenge which Jack Parr might exact upon his premature transition to the other side. Shortly afterwards John Parr began to communicate regularly, at first breathing vengeance against the doubly persecuted Sarah. Later on, |63| however, he calmed down and dictated a recipe for furniture polish which proved highly serviceable.

John Parr’s execution on 2 October was not mentioned in British newspapers until the following day. However its date could probably have been predicted by any knowledgeable person who had read accounts of the murder and inquest (Morning Post 29 August and 1 September) and of John Parr’s trial (The Times, 14 September). The details given in the ‘communications’ corresponded closely to the newspaper reports. Sir Lawrence Jones, by all accounts a man of the greatest charm, was happy to accept the assurance of the medium, and of her mother (who was also staying with him), that they had not read these newspaper accounts, and normally did not look at The Times and The Morning Post. But a cynic would certainly say that Miss Wingfield had fraudulently ‘got up’ the newspaper accounts and regurgitated them in her automatic writing. A less serious supposition would be that of cryptomnesia; the supposition, that is, that she looked at the reports, forgot them, but retained a latent or hidden memory of them which subsequently found expression in her automatic writing. One or other of these hypotheses would certainly have been suggested if, for instance, the newspaper accounts had proved to contain errors which were reproduced in the communications. Sir Lawrence did not even attempt to discover whether the recipe for furniture polish had been lifted from Aunt Kate’s Home Treasury or some similar compilation. Nor on the other hand did he enquire whether the two addresses given (Sidney Street and Green Street), which do not appear in the Times report of the trial which he quotes, were correct. If they had proved correct, the case for paranormality would have been correspondingly strengthened.

To rule out the fraud and cryptomnesia explanations one would need, not charitably to accept the medium’s say-so that she had never come across the relevant information, but to give reasons for supposing that she could not have come across it. And this involves proving a negative, a notoriously difficult undertaking. The negative cannot, I think, be proved in the case just discussed; certainly it was not proved. But there are other cases in which it may be not proved exactly, for ‘proof’ is a word somewhat strong for any non-mathematical demonstration, but at any rate powerfully supported. I shall now proceed to outline and comment upon three such cases.

The first of these cases was received through a well-known Icelandic |64| trance medium, Hafsteinn Bjornsson. Hafsteinn was not a professional medium in the sense of earning his living through his mediumship, but he did accept fees from sitters. He had a regular control named ‘Finna’, who would relay messages from other communicators; but sometimes the latter would themselves control. The original communications were obtained in 1941, and were investigated shortly afterwards (88). The case was further studied in 1971–2 by Haraldsson and Stevenson, who published their report in 1975 (59b).

On 25 January 1941, Hjalmar Gudjonsson, a visitor from eastern Iceland, had a sitting with Hafsteinn Bjornsson in Reykjavik, which is in the extreme south-west. (It is perhaps worth noting that at that time communications between Reykjavik and eastern Iceland were poor and mainly by boat.) The sitting was held at the home of Gudrun Jonsdottir, an experienced sitter, who was also present, along with another lady, Hansina Hansdottir. Hjalmar Gudjonsson was anxious to contact various persons he had known, but to his annoyance an intruding communicator, who gave the name Gudni Magnusson, monopolized the sitting. Gudni, who was not known to medium or sitters, stated that he had ties with Eskifjordur, in Hjalmar’s part of the world, and addressed himself to Hjalmar for that reason. He said that he had died following internal injuries received while attempting to repair his truck; and he gave various other details about himself which we will come to. Most unfortunately no contemporary notes were made of what was said.

Two days later, Hjalmar’s hostess at the sitting, Gudrun Jonsdottir, told a friend, Asmundur Gestsson, about this intrusive communicator. Asmundur had a cousin, Gudrun Gudmundsdottir, who was the wife of a physician practising in Eskifjordur, the place with which Gudni Magnusson had claimed to be linked. He accordingly wrote to this cousin, asking if she knew of anyone corresponding to the supposed communicator. His letter, dated 26 February 1941, survives—it was unearthed by Erlundur Haraldsson—and is the earliest document which gives details of the communications. It antedates, and in fact led to, verification of the communicator’s statements.

Asmundur Gestsson’s cousin replied on 14 March 1941 confirming that a Gudni Magnusson answering the description given had lived in Eskifjordur and had died in circumstances resembling those given. This letter, which is quite detailed, will be referred to below as the ‘Gudmundsdottir letter’.

At this point Asmundur Gestsson realized that he had an interesting |65| case on his hands and got Hjalmar Gudjonsson and Gudrun Jonsdottir to write out independently their recollections of the sitting and sign them. Hjalmar’s account is dated 30 March 1941, and Gudrun’s, which is fairly full, 6 June 1941. The third sitter, Hansina Hansdottir, signed Gudrun’s statement. There do not seem to be any serious discrepancies between these statements, or between them and our earliest document, Asmundur Gestsson’s letter of 26 February 1941. I think that, despite the absence of contemporary notes, we may safely accept the statements as accurately reflecting what passed at the sitting, especially since they are confirmed by the Asmundur Gestsson letter written before the verifications were received.

In his investigations of 1971–2, Erlundur Haraldsson found further sources of verification for some of the statements made. He interviewed Hjalmar Gudjonsson, and Gudni Magnusson’s brother and sister, Otto Magnusson and Rosa Magnusdottir; he obtained a copy of Gudni’s death certificate (such certificates are not obtainable by the general public in Iceland); he found an obituary notice of Gudni in the issue of Morgunbladid for 7 November 1940; and he interviewed the author of this obituary notice. Putting together all the information thus obtained we can, following Haraldsson and Stevenson, tabulate the communicator’s statements and the verifications as follows:

We now come to the question of whether this material could have been known to the medium or sitters through ordinary channels. In their article on the case Haraldsson and Stevenson give much attention to this issue. They summarize their conclusions about it as follows (59b, pp. 260–261):

The communicator came from a part of Iceland which the medium had never even visited. The sitters, even including the one person present (Hjalmar Gudjonsson) who was from eastern Iceland, had no connection whatever with Gudni or his family. The newspaper obituary could not have furnished the medium with all the correctly communicated details, nor could the writer of the obituary, who then lived in eastern Iceland, which the medium had never visited. The communicator had an uncle in Reykjavik, but as far as we can learn, he had no connection with the medium. Thus despite extensive enquiries we have not been able to find any channel for |67| normal communication to the medium of the correct information he had about Gudni Magnusson and expressed at the seance under consideration.

Haraldsson and Stevenson are here considering, and rejecting, primarily the cryptomnesia hypothesis, that is, that Hafsteinn Bjornsson might have somewhere come across the relevant information, and have retained a latent memory of it which came to the fore only in his trance state. They do not take seriously the hypothesis of outright fraud by Hafsteinn, and there do indeed seem to be quite strong reasons for dismissing it. Hafsteinn’s reputation throughout some forty years of mediumship was generally good; he had no known connection with Eskifjordur, yet to have obtained all the information about Gudni he would have needed not just an agent in Eskifjordur (a remote and sparsely populated place), but an agent who knew Gudni personally; and it was certainly not Hafsteinn who pushed or promoted the investigation of the Gudni communicator—it was in fact a person (Asmundur Gestsson) who was not even present at the sitting. I agree, therefore, with the cautious assessment of Haraldsson and Stevenson: ‘We conclude … that despite its obvious weaknesses [the absence of seance notes], the case justifies an interpretation that includes some paranormal process.’

But what kind of paranormal process? We can rule out clairvoyance at once, because the only relevant record of events which might be supposed to have been clairvoyantly accessible (the obituary notice) contained by no means all of the items given, and would in any case have had to be cognized precognitively. (One cannot, I think, take seriously the idea that by some sort of unconscious clairvoyance Hafsteinn monitored the events at the time when they occurred, and stored up a record of them for future regurgitation; how many other such sets of events must he have been simultaneously monitoring?) The hypothesis of telepathy from the living is more plausible, provided, at least, that one is prepared to believe (despite lack of substantial evidence) that telepathy of such a range and extent can occur; there must at the time of the sitting have been several persons alive who possessed all the requisite information. Thus the case of Gudni Magnusson is not one which strains the super-ESP hypothesis in the way that I indicated when constructing my ‘ideal’ drop-in case earlier in the chapter. It does not require one to suppose that the medium extrasensorially located and then collated relevant information from several different sources. Gudni does, however, offer an intelligible motive for communicating—the desire to talk to someone |68| from his own part of the world—whereas neither medium nor sitters had, so far as can be ascertained, any reason at all for picking out that particular deceased person as a target for super-ESP.

I shall next give a case from a series which I investigated myself (44c). They occurred in the context of a ouija board circle operated by a small group of people in Cambridge during and after the Second World War. Altogether more than two hundred deceased persons (and one living one) communicated through this circle. Most were friends and relations of the sitters. There were, however, eleven instances of verified ‘drop-in’ communicators, plus a rather larger number of unverified ones. Most of the verified cases were first verified by me, from thirteen to twenty-eight years after the original communications. This constitutes, I think, a strong argument against the likelihood of deliberate fraud. No one, however devious, would be likely to cast so much bread upon the waters, without eventually dropping some hints which might facilitate a return. The sitters, it should be noticed, made no attempts to promote the cases, or to obtain publicity, and had themselves not much idea of how one might set about checking them. I shall briefly summarize one of the more interesting cases.

At a number of sittings between 1950 and 1952 a communicator calling himself ‘Harry Stockbridge’ (not the real name) spelled out the following items of information about himself:

Second Loot attached Northumberland Fusiliers. Died Fourteen July sixteen.

Tyneside Scottish.

Tall, dark, thin. Special features large brown eyes.

I hung out in Leicester … Leicester hold[s] a record.

[Asked what were his likes and dislikes] Problems any. Pepys reading. Water colouring.

[Asked if he knew a ‘Powis Street’ about which two sitters had dreamed] I know it well. My association took my memory there.

[Asked if his mother was with him] Yes.

The sitters made one—unsuccessful—attempt to check up on these statements. The matter then rested until 1965, when I began to investigate the Stockbridge case. In an HMSO publication entitled Officers died in the Great War of 1914–19 I found it stated that a Second Lieutenant H. Stockbridge of the Northumberland Fusiliers was killed on 19 (not 14) July 1916. I then sent for Stockbridge’s death certificate. This gives his date of death as 14 July 1916 (as in the scripts) and not 19 |69| July (as in the official list). To resolve the issue, I wrote to the Army Records Centre, and received official confirmation that the death was 14 July.

Stockbridge’s death certificate shows that he was born in Leicester in 1896. This information is also contained in Joseph Keating’s Tyneside Irish Brigade (London, 1896), the only military history I have found which mentions Stockbridge (it does not, however, give the date of his death). This book states that Stockbridge was in one of the Tyneside Irish battalions of the Northumberland Fusiliers. However a record card in the War Office Library, kindly consulted for me by a friend, states that before his death he had been transferred to a Tyneside Scottish battalion.

That Stockbridge was tall, dark and thin, and had large brown eyes, was confirmed by his surviving brothers, and also by a photograph of him preserved in the archives of his old school. Whether he read Pepys or enjoyed water-colouring no-one could say. We may presume that he enjoyed ‘problems any’, since school records show that he won form prizes in mathematics and physics. He later enrolled for a university science course. His mother had died before the period of the communications.

The sitters thought that ‘Leicester hold a record’ meant that Stockbridge’s name might be on a War Memorial in Leicester. They asked a friend who was passing through Leicester to investigate, but she found nothing. His name is in fact on a War Memorial in his old school in Leicester. There was a ‘Powis Street’ near the house in which Stockbridge was born, although the family left the district within a few years.

So much, then, for the verifications of what the Stockbridge communicator said about himself. We have now to ask whether all these correct statements could have originated from a latent memory in the mind of one of the operators of the ouija board. During all the relevant sittings the ouija board was worked by a married couple, whom I shall call Mr and Mrs L. G. (it was quite clear that Mrs G. was the medium). Other persons were present, but did not operate the board. Neither Mr nor Mrs L. G. had any contacts in Leicester or had ever visited it, and I could trace no likely line of contact between either of them and any member of the Stockbridge family. Mr L. G. served in the First World War, but not in Stockbridge’s regiment. Furthermore he did not join up until after Stockbridge had been killed (I have seen his pay book).

Could the relevant information have been hoarded up |70| subconsciously (‘cryptomnesia’), following a glance at some obituary notice of Stockbridge? I was unable to trace a contemporary death notice of Stockbridge in any national newspaper, nor did ‘In Memoriam’ notices appear in later years. Two Leicester newspapers printed an obituary notice of him on 19 July 1916. This obituary notice (which it is highly unlikely that any of the sitters would have seen) gives the correct date of death, but the wrong rank (Lieutenant instead of Second Lieutenant), and makes no reference to the ‘Tyneside Scottish’ battalion. It give none of the details about Stockbridge’s appearance and interests (there is no photograph), and of course says nothing about Powis Street. Keating’s Tyneside Irish Brigade contains several of the items of information which the Stockbridge communicator produced. It is, however, a very out-of-the-way book, and it gives no death date, makes no reference to ‘Tyneside Scottish’, and contains no photograph or description of Stockbridge. Stockbridge’s appearance, and his Tyneside Scottish connections are, in fact, not mentioned, so far as I can discover, in any publicly available source. It does not seem to me that the hypothesis of cryptomnesia can possibly suffice to explain away the correct statements made by the Stockbridge communicator.

If, as I have argued, we can rule out the fraud and the cryptomnesia explanations in this case, we seem left to weigh up the respective merits of some form of survival hypothesis and of some version of the super-ESP hypothesis. The Stockbridge case does appear to stretch the super-ESP hypothesis in some of the ways which I indicated earlier in the chapter when discussing an ‘ideal’ ‘drop-in’ case. Stockbridge advances as his reason for coming that he is to help one of the other sitters (another ex-serviceman). This may not be an especially powerful reason, but it at any rate gives him a stronger reason for communicating than any of the sitters had for singling out facts about him as targets for super-ESP. If we suppose that the medium obtained information about him by clairvoyant apprehension of existing records, we must face the fact that she must have located, and synthesized the contents of, at least four separate sources, including the archives of his old school and the War Office Library. Could there at the time of the communications have been a living person or persons whose minds, telepathically tapped by the medium, might have provided all the requisite items? It is extremely difficult to say. His parents were by that time dead. It was nearly twenty years later that I made contact with two living brothers, and through them with a third brother and a sister. They had only the vaguest recollections of the brother who had died over fifty years before, and it |71| was quite apparent that as a result of following up the seance data I knew more details about his life than they did. My own guess is that the situation would not have been substantially different at the time of the sittings. But in this slippery field a guess is not good enough.

It seems therefore that even the very curious Stockbridge case does not fully measure up to the ideal ‘drop-in’ case for which I suggested criteria earlier in the chapter. Had it been investigated in 1952 it might have done so. There is in the literature, however, at least one carefully investigated case in which a ‘drop-in’ communicator made a series of correct statements, the totality of which could not have been obtained either clairvoyantly from a single document, obituary, etc., or telepathically, from the mind of a single living person. I refer to the case of Runolfur Runollsson (‘Runki’), for which the medium was once again Hafsteinn Bjornsson. and the investigators were once again Haraldsson and Stevenson (59a). The case is a complex and singular one, but it is unfortunately too long to be fully presented here. In outline the story is this. During the years 1937–8, Hafsteinn was acting as medium for what seems to have been a home circle in Reykjavik. In this period a highly eccentric communicator began to manifest through the entranced medium. He showed a yearning for snuff, coffee and alcohol, refused to give his name, and kept reiterating that he was looking for his leg. Asked where his leg was, he replied ‘in the sea’. In short he must have appeared at this time to be one of those comic-relief characters who so frequently brighten up the otherwise sober proceedings at home circles.

In January 1939 the circle was joined by Ludvik Gudmundsson, the owner of a fish factory in the village of Sandgerdi, about 36 miles from Reykjavik. The unknown communicator showed great interest in this new sitter, and eventually stated that his missing leg was in the latter’s house at Sandgerdi. After a good deal of further pressure from the sitters, he made the following statement (59a, p. 39):

My name is Runolfur Runolfsson, and I was 52 years old when I died. I lived with my wife at Kolga or Klappakot, near Sandgerdi. I was on a journey from Keflavik [about six miles from Sandgerdi] in the latter part of the day and I was drunk. I stopped at the house of Sveinbjorn Thordarson in Sandgerdi and accepted some refreshments there. When I went to go, the weather was so bad that they did not wish me to leave unless accompanied by someone else. I became angry and said I would not go at all if I could not go alone. My house was only about 15 minutes’ walk away. So I left by myself, but I was wet and tired. I walked over the kambuin [pebbles] and reached the |72| rock known as Flankastadaklettur which has almost disappeared now. There I sat down, took my bottle, and drank some more. Then I fell asleep. The tide came in and carried me away. This happened in October, 1879. I was not found until January, 1880. I was carried in by the tide, but then dogs and ravens came and tore me to pieces. The remnants [of my body] were found and buried in Utskalar graveyard [about four miles from Sandgerdi]. But then the thigh bone was missing. It was carried out again to sea, but was later washed up again at Sandgerdi. There it was passed around and now it is in Ludvik’s house.

On another occasion the communicator stated that he had been a very tall man. To cut a long story short, Runki’s extraordinary tale was subsequently verified in considerable detail, although it did not appear that he had in fact stopped at the house of Sveinbjorn Thordarson. Ludvik Gudmundsson knew nothing about any thigh bone in his house, but after enquiries among older local inhabitants, he found that sometime in the 1920s such a bone, believed to have been washed up by the sea, had been placed in an interior wall. It was recovered, and turned out to be the femur of a very tall man. No one knew whose bone it was, and there was no record which indicated whether or not the thigh bone was missing from Runki’s remains. One wonders, indeed, why, even if the deceased Runki were the source of the communications, and even if the thigh bone were actually his, he should have had any special knowledge of the matter.

The remaining statements were nearly all verifiable from entries distributed between two manuscript sources, the Church books of Utskalar (in the National Archives at Reykjavik), and the Rev Sigidur Severtsen’s Annals of Sudurnes, which at the time of the sitting rested unpublished and little known in the National Library at Reykjavik. That Runki had been tall was confirmed by his grandson, who, however, had not known him, and was not aware of the bone and of other relevant facts. He could therefore not have been, either through telepathy or through normal channels, a source for all the information communicated. It is possible that the Rev Jon Thorarensen, who in 1953 edited Annals of Sudurnes for publication, was even in 1939 aware of the major details of the story, but he did not know about the bone. Nor did he meet Hafsteinn before 1940.

Haraldsson and Stevenson consider in great detail the possibility that Hafsteinn could have obtained by normal means information from these and other less important sources—it seems extremely unlikely that he would have heard of the Annals of Sudurnes— and sum up the possibilities as follows (59a, p. 57):|73|

… for the medium to have acquired all the correctly communicated information, it does not seem feasible to attribute all of this information to any single person or any single written source. And this would be true, we believe, whether the medium acquired the information normally or by extrasensory perception. We think, therefore, that some process of integration of details derived from different persons or other sources must be supposed in the interpretation of the case. It may be simplest to explain this integration as due to Runki’s survival after his physical death with the retention of many memories and their subsequent communication through the mediumship of Hafsteinn. On the other hand, sensitives have been known to achieve remarkable feats of deriving and integrating information without the participation of any purported discarnate personality.

The last remark brings us to the crux of the matter. If sensitives operating in a non-mediumistic context can perform feats of location and integration of detailed information from discrete sources which, duplicated in the mediumistic sphere, would permit the construction of such communicators as Runki, Harry Stockbridge, Mr Aitken’s son, or Lodge’s Uncle Jerry, then the super-ESP hypothesis, fantastic though this is, will be rendered more plausible. To this issue I shall return in later chapters. With regard to the survivalist hypothesis, the following observation may be made. If communication between the living and the dead is possible, and can be carried on through the agency of mediums, we should expect to meet with ‘drop-in’ communicators, for there must be many recently deceased persons who earnestly desire to send messages of comfort, reassurance and advice to their bereaved relations. Had there been no records at all of verified ‘drop-in’ communicators, the survivalist position would necessarily have been seriously weakened. As it is, the onus is still on the survivalist either to explain away, or else to present reasons for denying, the supposed fact that such cases are relatively rare. I briefly discussed this matter earlier in the chapter.|74| 

In 1971, psi researcher Alan Gauld published a detailed paper on drop-in communicators who appeared in mediumistic sittings held by an English group in Cambridgeshire.4  The sittings were organized by a member of the Society for Psychical Research whom Gauld refers to as ‘LG’, and his wife, ‘WG’.  From 1937 to 1943 they used a Ouija board, later using a darkened room in order to obtain physical phenomena.. The group was active, on and off, until 1964.  Records with dates and attendee lists as well as the communications were kept, though some were lost.5  Gauld viewed the surviving records and interviewed the principal sitters several times each, noting their recollections of the alleged communicators.

In 470 sittings of which records survive, some 240 alleged communicators appeared, at least 37 of whom were apparent drop-ins.  Of these, thirteen did not give sufficient details about themselves for their identities to be verified. However, fifteen did give sufficient details for Gauld to match to a deceased individual; a further ten gave statements about themselves that he was able at least partly to verify.

Some cases, of varying degrees of evidentiality, are summarized here (Gauld used pseudonyms for the sake of the living relatives).

Duncan Stevens

‘Duncan Stevens’ communicated at about 40 sittings between 1942 and 1950. He first identified himself on July 14, 1942, at which time he also brought news of the husband of a sitter, who had been killed in 1941 on a training flight. Over time, he further revealed

• he had lived on Hinckley Road in Nottingham

• he had been a curate at Frinton Parish Church, then became a RAF pilot

• his full name was Reverend Duncan Stevens

• he had died in a plane crash into water in a Blenheim aircraft at age 28, about 10 months prior to the sitting

• his favourite composer was Brahms

• he had interest in ‘many religious orders’

The first three statements were verified by the sitters from clerical records.  Gauld visited Stevens’s sister, who verified the other details except the favourite composer.  No normal means for the sitters to have received all this information was found.

Edward Druce

This communicator first appeared on September 4, 1942, He revealed only that his surname was Druce, that he had died some time before, and that he lived on Hartington Drive. He also referred to ‘Grantchester Rive Xmas,’ and ‘university laboratory or library’.  LG found a Mrs Druce living on Hartington Drive in Cambridge, and learned that her husband Edward had been a laboratory worker who had drowned himself in the River Granta at Christmas some years previously.  His death had been described in newspaper accounts at the time, but Gauld noted that none of the sitters were regular readers, and that these events occurred before the group first met to hold sittings.

Robert Fletcher

‘Robert Fletcher’ first appeared on September 28, 1942.  He disclosed that

• his name was Robert Fletcher

• he had died two years before on a ship that had been torpedoed, on which he had been a crew member

• his parents and brother John were still alive

• he had lived in Tenterden

• his birthday was July 8 and his age (given at a sitting in 1943) was 21

Between a newspaper account and a journal account, both circulated in Tenterden, all these details were confirmed, except the birthday which was actually July 3 and so could be classed as a near miss.  Tenterden is in Kent, south-east of London and far from Cambridge, but Gauld noted that WG had grown up nearby and had a relative there. On the other hand, she had left fifteen years prior and the relative did not send her newspaper articles.

Gustav Adolf Biedermann

This communicator was belligerent at his first appearance on January 4, 1943, railing against religion. However, the group was patient with him, and on two subsequent appearances he was friendly and forthcoming.  Facts about him were confirmed by Gauld, who recalled having read writings by a psychologist of the same name.  Biedermann’s correct statements were:

• he had lived in London

• his house had been Charnwood Lodge

• he was of German birth, having moved to England in 1887

• Gustav Adolf Biedermann was his full name

• he was a rationalist

• he was past 70 when he died, about a year prior to the sitting

• he had his own business

• he was associated with the London University (having worked in the psychology department of University College)

It was also noted by Sir Cyril Burt, who had been friends with Biedermann, that he had a ‘blunt, arrogant, obstinate and aggressive manner’, but was a ‘pleasant companion’ once one got past the façade.  He also liked to denounce religion.

Walter Leggatt

This communicator first appeared on May 10, 1943, and while at first confused somewhat, said his memory had improved during a later sitting.  Over three sittings, he revealed the following facts, which were verified by Gauld:

• his full name was Walter Leggatt

• he had been a sergeant in the RAF

• he had worked as a rates clerk for the town of Acton (name of town changed)

With so few details, Gauld did not consider this case to be a strong one.

Josephine Street

This communicator appeared on May 17, 1943 was clearly motivated to contact her husband, in order to reassure him that she was still with him.  She identified him as Archie Street, and said they had lived on Lauriston Road in Cambridge.  LG was able to contact Archie Street, an administrator at a Cambridge college, and invited him to the next sitting, at which Josephine sent him a loving message for him and their daughter. Street’s skepticism was fuelled when the daughter’s name was given incorrectly. However, he was impressed when Josephine’s second name, Eugenia, was given.  The records for further sittings were lost. Gauld calls the case ‘rather unsatisfactory’, as a death notice giving all these details had been published a few days previously.

Max Cheyne

On June 28, 1943, a ‘control’ communicator acting as intermediary mentioned a ‘Max’ who had been in the RAF, lived in ‘Ditton Park’ and whose surname ‘has some connection with cables or chains’. He wished to send his love to his ‘wife and babe’.   In a second sitting the intermediary said the name was ‘Cheynes’ and the plural might be wrong.  The sitters were unable to verify his existence. However, Gauld discovered that Max Langdon Cheyne, who had lived in Ditton Fields, Cambridge and had been in the RAF, had been killed in October 1942, leaving behind a wife and young daughter.

Kate Clarke

On September 20, 1943, a communicator named Kathleen offered her services as a ‘helper’.  She said that her full name was Kathleen Clarke and everyone called her Kate, that she had been the eldest of eleven children, was British and had lived in Poplar, and that she had died during the war when George V was king (World War I), in childbirth at the age of seventeen.

Gauld found records of several Kathleen and Kate Clarkes who died in their teens during World War I. Only one had died in childbirth, but not in Poplar, and she was nineteen not seventeen.  Gauld also traced the birth records of seven other children in the family.  He concluded that the correspondences were too many to attribute to chance, and if a link to Poplar could be found the case would become a strong one; however, he had not succeeded at the time of writing.

The sitters made a single attempt to verify his identity, which failed.  Gauld did better, however, finding a Second Lieutenant H Stockbridge of the Scottish Tyneside battalion of the Northumberland Fusiliers who had been killed in action on July 14, 1916. His birth certificate showed he had been born in Leicester.  A relative confirmed that Stockbridge had been tall, dark and thin, and Gauld confirmed from a photograph that he had large dark eyes.  He also learned that there was a Powis Street close to where Stockbridge was born.  His death had been mentioned in some publications and on a memorial, but Gauld determined that it was highly unlikely any of the sitters would have seen these.

Gauld concludes that the material contained much inaccurate information, but also many correct details which he doubts could have been acquired by normal means. He notes that the information as a whole revealed no pattern, such as coming mostly from a single source; that in some cases it was known only to relatives that Gauld contacted later; and that in two cases the communicators did not reproduce errors found in written records, but instead corrected them.   

This case emerged via the reputable Icelandic medium Hafsteinn Björnsson. The  unknown communicator did not ‘drop out’ after one or a few sittings, as is usual in such cases, but instead developed a long-standing relationship with Hafsteinn. The case developed over numerous sittings during the years 1937-38 and was described in a 1946 book by Icelandic author Elinborg Larusdottir.7 Haraldsson and Stevenson published a paper on it in 1975.8

The communicator intruded into the sitting, refusing to give his name. He asked brusquely for snuff, coffee and rum, and persistently demanded ‘I am looking for my leg, where is my leg?’. The sitters began to lose patience, causing him eventually to yield to their request for information about himself:

his full name was Runolfur Runolfsson

he had lived with his wife at Kolga/Klappakot, near Sandgerdi

he had been very tall

he was 52 when he died, in October 1879

he died after attempting to walk home from a visit in Keflavit during very bad weather while severely inebriated; after lying down by the shore to drink more, he fell asleep and was carried away by the tide and drowned

his body was not found until 1880, by which time it had been torn to pieces by dogs and ravens

the remains were buried in Utskalar graveyard, except for a missing thigh bone

the bone washed up again at Sandgerdi, where it was passed around, and was now somewhere in the house of a man attending the sitting, Ludvik Gudmundsson

the sitters would be able to verify the accuracy of his words by checking the church book of Utskalar Church

Doing so, they found a record with the correct name, date of death and age of death. Ludvik Gudmundsson consulted with elderly men of Sangerdi, and learned that a thigh bone had been placed between the inner and outer walls of his house when it had been renovated by a previous owner. After some searching, the bone was found, and discovered to be very long, matching the communicator’s claimed stature.  It was buried in traditional Icelandic fashion, and at the next sitting, Runolfur said he had been present at the rite and reception afterwards and gave some veridical details, including the types of cakes served.  Further investigations by Larusdottir using the Utskalar parish records and a cleric’s diary revealed that Runolfur had lived in Kolga/Klappakot, and had died and been dismembered as the communicator had described.

These verifications were double-checked by Erlendur Haraldsson and Ian Stevenson working together on the case in the early 1970s. They inverviewed 23 witnesses and examined records, from which they were able to establish that neither Hafsteinn nor other sitters had gained access to these records or known anyone in the area prior to the sittings.  They noted also that the communicator’s behaviour and manner during the sittings closely matched the known personality traits of the living Runolfur.

The spirit continued to communicate through Hafsteinn, eventually becoming the medium’s main control and acting as go-between for other discarnates.  The case is described in a short documentary by Keith Parsons which may be seen here.

In June, two other sitters confirmed that these recollections were correct, and added more:

• Gudni had living parents

• he had managed to get home before being taken by boat to the doctor

• he had died four or five months prior to the sitting

These details were eventually found to closely match the life and death of Gudni Magnusson, a truck driver living in Eskifjordur, who had died the previous fall. His truck, which had not been running well, ran out of gasoline on a mountain pass between Eskifjordur and Reydarfjordur forcing Gudni to walk eight miles to fetch a refill. He returned home exhausted, then during the night suffered extreme stomach pain, which doctors later diagnosed as caused by an internal rupture or obstruction. He was rushed by motorboat to hospital but died on the way.

Haraldsson and Stevenson, investigating in the early 1970s, received further confirmation of the communicator’s description of himself from his brother and sister. The death certificate gave cause of death as intestinal perforation and peritonitis, possibly aggravated by an intestinal weakness resulting from a childhood operation. The birth and death certificates revealed his age at death as 24. It was not possible to confirm the communicator’s statement that he had been trying to repair his vehicle when the rupture occurred, but the fact that it had been running badly at the time made that a plausible conjecture. The investigators also ascertained that neither medium nor sitters had connections with Gudni, his home locale being in a remote part of the country. 

A weakness of this case is that the statements about the sitting were based not on notes recorded at the time but on recollections some weeks and months later. In the investigators’ view, this is somewhat mitigated by the fact that the February 26 letter recorded some of the key facts before any attempt was made to verify them. A newspaper had published an obituary of Gudni the previous November, but it did not contain some of the key details revealed at the sitting.