More Options

Fiery Tales That Spontaneously Destruct

Investigative Files

Joe Nickell

Skeptical Inquirer Volume 22.2, March / April 1998

Spontaneous human combustion (SHC) cases continue to spark controversy (so to speak), largely due to the efforts of nonscientist authors and journalists. These include self-styled British paranormal researchers Jenny Randles and Peter Hough (Spontaneous Human Combustion, 1992), Pennsylvania school bus driver Larry Arnold (Ablaze!, 1995), English coal-miner-turned-constable John E. Heymer (The Entrancing Flame, 1996), and — more recently — the producers of A&E network’s TV series The Unexplained.

The continued lack of scientific evidence for SHC [see Mark Benecke’s article “Spontaneous Human Combustion: Thoughts of a Forensic Biologist,” in this issue, pp. 47-51] keeps proponents desperately looking for cases they can attribute to the alleged phenomenon — cases that are often quite disparate. They assign instances of unusual burning deaths to SHC rather like one might blame freak auto accidents on the Highway Gremlin. (For a discussion, see Nickell and Fischer 1987 and Nickell 1996.) Even cases with an obvious solution — like the one illustrated on page 17 — are sometimes hyped. That case, involving the 1980 death of an elderly Englishwoman who fell headfirst into a burning fireplace, was duly reported by Randles and Hough (1992, 84-85, 91) and by Arnold (1995, 130).

Another case, that of Jeannie Saffin, included in Heymer (1996, 179-188) and Arnold (1995, 208-209), is quite instructive. Because the source of the body’s ignition is not obvious in Saffin’s death, paranormalists are especially quick to propose SHC. In doing so, of course, they engage in the logical fallacy called argumentum ad ignorantiam ("arguing from ignorance”), since one cannot prove a cause from a lack of facts. The case also illustrates how crucial details may be omitted and how accounts become exaggerated over time, and therefore it demonstrates the consequent need to return to original sources.

Jean Lucille “Jeannie” Saffin was a sixty-one-year-old woman with the mental age of a child, due to brain damage from a forceps delivery at birth. Her mother having died the previous year, she lived with her eighty-two-year-old father and a brother at the family home in Edmonton, in northern London. On Wednesday, September 15, 1982, a hot, humid day, Jeannie was sitting with her father in the kitchen. The windows were open. Suddenly, about 4:15 p.m., Jack Saffin’s attention was directed to his daughter who was ablaze. He shouted to his son-in-law Don Carroll, and the two men put out the fire with water. Carroll phoned for an ambulance, which arrived quickly, and Jeannie was transported to North Middlesex Hospital. She was later transferred to the burn unit at Mount Vernon Hospital. She died there, nearly eight days later, at 8:10 a.m. on September 23. The cause of death was listed rather perfunctorily as “broncho-pneumonia due to burns.”

To proponents of SHC, the case is a spine-tingling mystery. In a chapter devoted to Jeannie Saffin’s death, Heymer expresses “puzzlement about the source of her burns” (1996, 186), and he includes a statement by Don Carroll (1994) who says “there was nothing alight in the kitchen except the pilot light on the grill.” Even so, Carroll insists that he saw “flames coming out of her mouth and her midriff.” Indeed, he says, “the flames were coming from her mouth like a dragon and they were making a roaring noise.” Yet, he insists, “Her clothes did not burn much at all” (Carroll 1994). Heymer emphasizes the latter point, insisting it is “a mystery how she came to be burned inside unburned clothes” (original emphasis) (Heymer 1996, 187).

Arnold essentially repeats the claims, obtaining his information largely from Heymer and apparently doing little actual investigation of his own. He writes: “As the men battled to save Lucille [sic], the son-in-law swore that ‘she had flames roaring from her mouth like a dragon.'” He adds: “Ambulance men . . . noticed there was no smoke damage in the kitchen; that her clothing was not burned. Only a portion of her red nylon cardigan only [sic] had melted” (Arnold 1995, 208-209).

Certainly the case sounds impressive — at least until investigation takes us back to original sources. First, regarding the allegedly unburned clothing, there is the signed statement Carroll gave to authorities soon after his sister-in-law’s death. In that statement from twelve years earlier, he noted that “[h]er clothes were in ribbons and were charred black. She was black as well. She started to try to pick her clothes off herself but I told her to stop” (Carroll 1982). In addition, a typed account by Constable Leigh Marsden stated: “The clothes were still burning when I [Marsden] got there. I pulled off the rest of her clothes. She and her clothes were burning. I put it out with a towel” (Marsden 1982). The ambulance paramedics supposedly reported that Saffin’s clothing had not burned, but what they actually wrote was that she was “wearing nylon clothes, not on fire” (Heymer 1996, 186) — obviously meaning “no longer on fire,” not “unburned.” It is disingenuous to state, as Heymer does, that the nylon cardigan was “melted not burned” (Heymer 1996, 186). In addition to the cardigan, her clothing consisted of a cotton apron and dress (Marsden 1982; Heymer 1996, 196).

As to Carroll’s statements about the fire, the flames probably did appear to come from Saffin’s midriff. That may have been where the nylon cardigan began burning. Also, flaming blobs of melting nylon may have caused the burns on the victim’s “front of left thigh” and, after she stood up, on her “left buttock” and “patches on the right knee” — as related in the autopsy report. Since damage is typically greatest above a flame rather than below it, it is not surprising that there were also “[f]ull to partial thickness burns on the face, neck, both shoulders, front of upper chest” and “patchily distributed on the abdomen” as well as “affecting both hands” (Post-mortem 1982). ("Full thickness burns” means the skin is destroyed down to the underlying fat.)

As to the flames issuing from her mouth “like a dragon,” that claim is not supported by the medical evidence. A report from Mount Vernon Hospital to the coroner’s office stated that when the victim arrived at the burn unit, “There was soot in her nose, but the back of the mouth appeared undamaged” (Whitlock 1982). This was confirmed by the autopsy. Except for the bronchopneumonia (with the inflammation of the trachea and bronchi), there was “no evidence of natural disease"; nor was there any indication of internal combustion. To the contrary, the autopsy report confirmed: “Total body surface burns being about 30-40%” (emphasis added) (Post-mortem 1982).

But what of Carroll’s description of the flames being expelled from Saffin’s mouth and “making a roaring noise"? That may have been the effect on Carroll — especially after twelve years’ reflection. Those details are absent from his original statement to the police. However, Carroll does say in his later statement that, at the hospital, despite her head being swathed in bandages, “I could see into Jeannie’s mouth and the inside of her mouth was burnt” (Carroll 1994). It is possible that Saffin was breathing excitedly so that the flames attacking her face were partially drawn into, then expelled from, her mouth. Heymer agrees with this possibility (Heymer 1996, 195). As to the alleged “roaring,” although a doctor reportedly told Carroll he must be mistaken (Arnold 1995, 208), and even though he is admittedly technically deaf, “Even so, I heard the sound of the flames coming from Jeannie,” he says (Carroll 1994). Possibly due to expectation and the interrelationship of the senses, he simply thought the flames roared.

The medical evidence makes clear that the fire was not internal but, instead, that Saffin suffered external burning as a result of her clothing catching fire. As usual, SHC proponents are unable to theorize how that could have occurred. But a clue comes from Carroll’s original report in which he states, “I made a point of checking on the gas cooker and saw that it was not on and saw that my father-in-law had his pipe in his hand and I checked it and saw that it was fresh tobacco which had not been lit” (Carroll 1982). This seems to rule out the pipe, and, indeed, there is no further mention of it — by Carroll (1994), Heymer (1996), or Arnold (1995). Yet Her Majesty’s coroner for Greater London (Western District), John Burton, told Arnold, “we usually find some smoking material, particularly in [the case of] the immobile or elderly victim” (Burton 1996).

The pipe represents just the type of “smoking material” one looks for, and Carroll’s insistence that it was freshly filled and unlit overlooks an obvious possibility. Did the elderly Mr. Saffin previously knock the hot ashes from his pipe, and did an ember land in Jeannie Saffin’s lap? To this very plausible scenario we must add the fact that the kitchen window and door were open, as was the back door, so that there was the potential for a draft. This could easily have caused the smoldering clothing to flare up.

If we accept this possibility, only minor mysteries remain, and we may clear them up as well. Jeannie Saffin was sitting on some newspapers in a wooden Windsor chair and SHC proponents wonder why the paper suffered no fire damage (Heymer 1996, 185). The simple answer is that Saffin’s body actually protected the papers; obviously the flames did not extend to her buttock until she stood up and moved away from the chair. Proponents also wonder why there was no smoke damage to the room (Heymer 1996, 186-187). The obvious answer would be that smoke was minimized because the fire was confined to the victim and the open windows and draft effectively helped dispel what smoke there was. Finally, paranormalists wonder whether it was “normal” that the victim was not in pain — either at the time of the accident or subsequently (Heymer 1996, 187, 194). In fact, at the time of the fire, while she failed to cry out, she had “whimpered,"according to her father (Saffin 1982). Her mental condition, her body’s production of endorphins (pain-reducing chemicals), the subsequent shock, and her eventual semi-conscious state may all have played a part in minimizing her response to pain.

A&E’s The Unexplained series included the Saffin case in its hour-long discussion of SHC (September 18, 1997). My brief comments on the cases were included, and, overall, the presentation might have been considered balanced. Unfortunately, at the closing of the program the narrator spoke of “an acupuncturist with remarkable talents. Through meditation and practice, he had learned to harness the electrical currents in his body. . . .” Suiting action to words, the supposed marvel crumpled a newspaper which — with a wave of his hand (and a bit of editing to cover the delay) — burst into flames. Alas, it appears the producers were snookered by a well-known magic trick billed variously as the “Yogi’s Gaze” (Miller 1978) and “fire by mental power” (Premanand 1994). Ironically, the feat depends on the secret combining of two chemicals that do, actually, spontaneously combust.


I am indebted to Tim Binga, Director of the Center for Inquiry Libraries; CSICOP Fellow Ray Hyman, for supplying me with the “Yogi’s Gaze” material; and Ranjit Sandhu, CSICOP Research Associate.


Joe Nickell

Joe Nickell's photo

Joe Nickell, Ph.D., is Senior Research Fellow of the Committee for Skeptical Inquiry (CSI) and "Investigative Files" Columnist for Skeptical Inquirer. A former stage magician, private investigator, and teacher, he is author of numerous books, including Inquest on the Shroud of Turin (1998), Pen, Ink and Evidence (2003), Unsolved History (2005) and Adventures in Paranormal Investigation (2007). He has appeared in many television documentaries and has been profiled in The New Yorker and on NBC's Today Show. His personal website is at