Posted: Jul 11, 2010 9:24 pm
by jerome
Defender's #6 post: 2929 words - Mr.Samsa

Thanks to Campermon for his useful and engaging survey of the discussions so far. In today's response I plan to concentrate upon two of the issues we have discussed – namely Dr. Barrie Colvin's analysis of acoustic materials associated with purported poltergeist phenomena, and the psychomanteum data.

At the time I brought up Dr.Colvin's research it was unpublished: now fortuitously it is in print, and we can discuss it in a little more depth. Campermon (I am not sure if he has seen the paper in full yet, but I shall ensure he does) cites the SPR news item on the article

SPR wrote:
“The sample involved 10 separate recordings recorded on different recording apparatus. In each of the recordings, when subjected to acoustic analysis, a particular sound pattern is detected which so far remains unexplained. Attempts to replicate this pattern in ordinary ways have so far been unsuccessful.”

He then raises a sensible objection

Campermon wrote:
So, we have ’10 separate recordings’; not really a good sample size but I let that one roll. However, this sample was ‘recorded on different recording apparatus’. Surely, this could invalidate the experiment? Or has the Dr taken this into his account in his analysis?

Well, any reader with experience of sound engineering or acoustics will be aware that the characteristics of a given recording can be shaped by the recording device used. For example, a certain type of microphone may produce distinctive 'patterns' of distortion, suppression or in other ways 'process' the signal. Dr Colvin was certainly aware of this possibility – if the unusual characteristics detected were all recorded with the same make, model and type of microphone, we would be entirely justified in believing that the recording equipment was the source of the artefact.

Remember, this is not a lab experiment. Even if it was, we would want a plethora of different microphones to rule out the possibility that the equipment, not the sound itself, was causing the unusual acoustic characteristics. Instead we are looking at ten recordings from 'the field', from cases in which Dr Colvin was for the most part not involved – Andover (1981) being the exception. The other cases which provided recordings were

Sauchie, Scotland (1960) – from Broadcasting House, from the BBC recordings taken at the time. ... icle23.htm has more on the case for those not familiar with it - Owen, A.R.G. Can We Explain the Poltergeist? New York: Helix Press / Garrett Publications, 1964 gives a full account by the principle investigator.

Thun, near Bern, Switzerland (1967)
The recordings were taken from a CD. ... iale+Musik

Schleswig , Switzerland, (1967
) taken from a CD. ... iale+Musik

Pursruck, Germany (1971) – from a recording by Prof. Hans Bender (16-bit stereo, 44100Hz)

Ipiranga, Brazil (1973)
– recording by Guy Lyon Playfair taken during the IBPP investigation. More on the case can be found in Playfair's 1975 book The Flying Cow.

La Machine, France (1973) –
recording by Dr Alfred Krantz.

Enfield, England (1977)
– from original reel to reel tapes, which was running "at the rather slow speed of 15/16 of inch per second" (Colvin 2010); recording taken by SPR investigator Maurice Grosse. A recent Channel 4 documentary on the case well worth watching can be seen here -- ... oltergeist - you can see the equipment used and context.

Andover, England (1981) - already discussed in this debate.

Santa Rosa, Brazil (1988)
– taken from a recording made of a television broadcast (by TV Globo) on the case.

Euston Square, England (2000)
This case has recordings by both Maurice Grosse and Mary Rose Barrington available.

Ten cases, none recent, because recordings of acoustic phenomena associated with poltergeists are by their nature difficult to collect: one need a poltergeist after all! The two Swiss cases are from a digital CD recording commercially available of recordings of parapsychological phenomena – it is impossible to say to what extent they have been edited and processed, so I would say they were VERY weak evidence. The Brazilian cases rely on recordings taken by Guy Lyon Playfair at the time, and by him off the TV, and he was present at Enfield – yet fraud seems unthinkable, given the dates, unless he somehow had access to very high end studio equipment and knowledge of acoustics in a pre-digital sound age. Therefore, I think that so long as we trust Dr Colvin's acoustic analysis, the sound signatures he claims to detect in his varied collection of poltergeist sounds are authentic. Colvin's claims are checkable -- at least some of these recordings – the Enfield sounds and the two from the CD, and possibly if you are willing to approach the BBC Sauchie – are available in their original form to interested independent parties wishing to check the results. I suspect someone with appropriate acoustics knowledge could acquire copies of all the recordings by request to the SPR. ( Adobe Audition was used for the analysis, in case anyone fancies trying a replication. I do favour a hands on approach as you all know by now!

So what does Colvin claim to have found? Well let's start with a normal waveform. It follows a characteristic pattern – a wave form showing a sharp rise in amplitude or immediately to maximum amplitude, followed by a gradual decrease to zero. Adobe Audition has a free trial, but there is plenty of freeware on the web you can download which allows you to experiment with banging various substances yourself. I did so, analysing some sounds submitted by Wayne Morris from his paranormal investigation at Landguard Fort, Felixstowe last year, and found that the banging noises there followed the same type of acoustic signature I could get by kicking the wall or banging my desk – the above pattern, suggesting a normal explanation for those (non-poltergeist) sounds. Simple experimentation with a large number of substances demonstrated that the pattern is consistent, and that Colvin's comment on this is correct. I encourage everyone reading to try this for themselves, to familiarise themselves with the standard way the amplitude and frequency can be analysed and the common pattern one sees.

Colvin gives a couple examples of frequency ranges in mundane sounds – a hammer hitting an oak desk gives a frequency band of mainly 50Hz to 300Hz – a teaspoon on a crystal glass 300Hz to 3000Hz, with a decay of amplitude lasting three seconds. What one might expect in short. However, once again I strongly suggest a few minutes experimentation at home, and posting the results???

So how do the acoustic properties of the raps in the ten cases in question vary? They show a consistently odd rise in amplitude, a waveform that slowly rises rather building to a sudden peak and then falling away. One can test this on the knocks from the Channel 4 shows recordings from Enfield I guess, or armed with some money, order the CD Colvin took the Swiss cases from: I have too date done neither. Given the fact the JSPR article is clearly copyright, I shall simply reproduce two of the figures here, showing a knock deliberately made by Grosse at Enfield as he tried to replicate the noises,and one of the anomalous raps...


So why do the waveforms show these unusual characteristics? Colvin thought of a possibility, which takes us right back to Campermon's original problem, and my response with which I opened this post…

Dr. Barrie Colvin, [i]JSPR[/i] 73.2, Number 899, April 2010 wrote:
One of the possible normal explanations put forward to explain the results is that certain types of microphones may give rise to the anomalous results because of their inherent qualities and mode of operation. A microphone is simply a sensor that converts sound in to an electrical signal. The most common types consist of a thin membrane that vibrates in response to sound pressure.

As you will recall this is how I opened my post…

Dr. Barrie Colvin, [i]JSPR [/i]73.2, Number 899, April 2010 wrote:
This movement is subsequently translated in to an electrical signal using one of several techniques. Most examples use electromagnetic induction, capacitance change, piezoelectric generation or light modulation to convert the mechanical vibration of the signal to an electrical signal. The question that arises is relation to a short impulse such as a rap is whether or not there could a be a delay between production of the sound and vibration of the membrane. Could the inertia of the membrane, particularly with microphones dating back to the 1970's, lead to a relatively slow increase to maximum amplitude when subjected to a short burst of acoustic energy?

This is why I suggested in a lab set up we would require several microphones, of different makes, models and manufacture. Colvin experimented making raps with a number of microphones dating from 1959 (including the Phillips EL3549A & the EL3782 with impedance 583 ohms) to present day microphones, looking at the waveforms, to falsify this hypothesis. Again, with old microphones common in attics if my house is anything to go by, I suggest interested parties can do at home…

However there is another reason to believe the results are not an artefact of the microphones. Three of the recordings include the investigators making their own raps. These investigator produced raps possess the normal waveform, not the slow rise in amplitude associated with the "poltergeist knockings". As such, we have an inadvertent control, which demonstrates the microphone was NOT the source of the unusual waveforms.

Colvin has managed to find similar acoustic waveforms to those recorded in these ten cases – in recordings of seismic activity. His paper gives two examples – a recording of an earthquake at Ascension Island in 2007, and a British Geological Survey recording of a seismic event at Folkestone in 2009, described as being "like an explosion". Colvin theorises that the sound signatures associated with the poltergeist events imply they are caused by a sudden release of tension or alteration in the substance of an object, not with as one would assume a rapping of one thing on another. An intriguing suggestion, but clearly one that requires further high quality recordings to test adequately. Perhaps Campermon and Twistor can think of some experiments to create waveforms from stress induced noises? One of my parapsychology interested friends is a DLitt (or some postdoc) in metallurgy with special reference to airframe fatigue, and happens to have a background in sound engineering, so I might ask him to run some tests for next post in generating waveforms.

Campermon's next comment unfortunately could easily consume my entire response!

Campermon wrote:
So, where can the career physicist get hold of this paper?

“Dr Colvin’s research is published in his article ‘The Acoustic Properties of Unexplained Rapping Sounds’ in the Journal of the Society for Psychical Research [2010] Vol 73.2 Number 899 pp 65-93.”

Well, the career physicist could email me, or join the SPR ( ) or subscribe to LEXSCIEN or wander in to 49 Marloes Rd and ask the SPR librarian if they could have a gander. The question really is "why not publish in Nature, or a physics or acoustic journal?" the answer is simple: try it! Nature has a long standing policy of rejecting parapsychology papers on principle – a few are published, but they are extremely rare, and often result in a heated debate. Pretty much no physics journal will touch parapsychology papers – yet the SPR has no shortage of physicists, now or historically (though currently only one Nobel winner in the field I think – Brian Josephson – earlier ones being Marie Curie, Lord Rayleigh, and JJ Thomson – just an amusing historical side, I'm not saying that means anything!). Professor Bernard Carr was of course the previous President. So anyone interested in the physics of poltergeist papers has to publish in a dedicated parapsychology journal, and as the EJP Is more psychology orientated, that means the JSPR for a peer reviewed physics article on something like this. (Well you could publish here like these gentleman - - but Brian Josephson described that paper as "flaky" in New Scientist April 2008, and I think it's a parody ;). I assumed it was an April Fool's joke!

So, in short, if you are writing a parapsychology article, you publish in a peer reviewed parapsychology journal – though if you flip back through this debate and look at the articles I reference you will note that is far from always the case – in fact I have provided a fairly inter-disciplinary set of journal citations, in keeping with the nature of the subject. Someone joked years ago that parapsychology is a game for polymaths, sadly I'm not even bright -- I think I might count as just "not as dumb as my wife seem to think!"

So in response to Campermon's sensible question

Campermon wrote:
Why is Dr Barrie Colvin reluctant to publish his piece of physics research in a physics journal?

I think the answer is, speaking for Dr. Colvin which is always unwise, find one interested and I'm sure he will!

Ok, time to turn to the psychomanteum research. Firstly, here is Radin & Rebman's set up from Lange & Houran's Multidisciplinary Perspectives on Hauntings & Poltergeists (2001) ISBN 0786409843.


If you want to try this at home, here is the equipment list with specifications for these (1996) trials

JSPR Vol 61, No 843 wrote:
List of Equipment Used in Pilot Experiment

FLIR Prism SP high resolution infrared video spectral range 3.6 to 5 microns, O.l" C sensitivity
FLIRIQ-810 high resolution infrared video spectral range 8 to 12 microns, 0.06° C sensitivity
Astroscope night vision light amplifier visual range, with IO" {4} lux sensitivity
Canon LI video camcorder high resolution camera used with Astroscope
Applied Physics 3-axis fluxgate magnetometers 1 μθ3Η55 sensitivity, DC - 300 Hz response
Systems 534 (2)
Hewlett Packard 500 MHz digital oscilloscope used to display magnetometer outputs
Hewlett Packard spectrum analyzer continuously scanned electromagnetic spectrum,
8561E 30 Hz - 6 GHz
JVC GR-AX75 VHS-C camcorder recorded display of EM signal analyzer
Sony WM-D6C audio cassette recorder recorded sounds inside psychomanteum
RNG random number generator truly random, based on electronic noise
EFM battery-powered AC magnetic (0 - 300 Hz), DC magnetic, AC electric
EMDEX-C electromagnetic field monitor (0 - 300 Hz)
Ertco-Hart 850 temperature monitor computer-based, millidegree sensitivity
Aware RM-60 Geiger counter computer-based, sensitive to α, β, γ and x-rays
Panasonic video quad system combined infrared, light-amplified videos, and normal
WJ-450 video signals into one display
Panasonic 20" video monitor used to display quad video signal
RCA VR 323 video cassette recorder usd to record combined video signals
Computers controlled computer-based two 80486DX2 computers, one 80286 computer;
equipment custom-designed software
J&JI-410 computer-based physiological electrodermal (EDA), blood volume pulse (BVP), heart
monitoring system rate (HR), skin temperature (temp),
electroencephalograph power (beta, alpha & theta)

As you can see, it is slightly more sophisticated than our "Blue Peter" try-this-at-home version, though doubtless could be improved a decade on. Radin & Rebman (1996) collected what one might expect during the sessions – subjective data as to the percipients experiences (the percipient activated a microswitch during the apparitional/hallucinatory parts of the trial), physiological monitoring of the subject; but they also decided to do environmental monitoring, to check if the experience correlated with any changes in the physical environment. Their results were to put it mildly surprising --- a weak but significant correlation was found between the physiological and environmental variables. The authors postulate some kind of reciprocal feedback relationship between body, mind and environment.

The most common physiological versus environmental correlation was probably caused by a simple experimental flaw – three people were in the psychomanteum before the beginning of each session, then only one for the session. The temperature inside therefore declined, and it would not be surprising if this was to result in a physiological response as Radin and Rebman note. However to quote Radin

Radin wrote:
After exclusion of all Atemp cross-correlations, 152 correlations remained, of which 7.6 were expected to be significant by chance. Because 24 significant correlations were observed among the remaining physical variables, there is suggestive evidence that some of the physical vs. physiological correlations reflect genuine relationships.

Anyway here are the data tables from Houran and Lange, so you can see for yourself what was recorded, and think of possible explanations. When I get my book back I'll go in to this in more detail, and hopefully briefly talk about Parsons and O Keefe's 2006 replication – I have emailed Ciaran to ask him if he can let me have a summary of his findings for the debate. Here are the figures from Radin and Rebman's 1996 trials…


So does the psychomanteum produce something more than just hallucinations? Is the mind actually effecting the environment, or is it simply a one-way relationship? I'll discuss this in my next post. For now I just wanted to give Campermon something he could get his teeth in to, as people keep accusing me of just "telling ghost stories". I have aimed for a spirited defence of my position, but in my next post I will bring some more scientific studies to bear, from names familiar to many readers of this forum (Richard Wiseman among others) and then start to tie my argument together in to a defence of my apparently unlikely claim, and to make what I hope will be a strong defence of my position.