By Russell White
P.I.T.T. / Paranormal Investigation Team of Tulsa

This article originally appeared in the August, 2004 online issue of Ghost Magazine in the Spirited Science: Mainstream Science Meets Paranormal Studies column. Posted with permission of Russell White.

The day was hot and still as we approached the abandoned mental hospital. The old door screeched as we pushed it open and walked into the cool, musty darkness. Our footsteps seemed to echo forever. I was interviewing one of those crazy “ghost hunters”, and working on my Experimental Psychology class project, hoping to prove that these people suffered from delusions, and that the “paranormal” had no place in the world of legitimate science.

A local “paranormal researcher”, as she insisted on being called, had agreed to assist me. She claimed she had nothing to hide and welcomed my interest. This particular day she was going to show me her procedures for data collection and basically how she went about an investigation.

I had my handheld tape recorder that I used for taking notes in class with me as we went out to the hospital. I was recording everything for later review. When we reached the interior of the building I had already noted that no one else was able to gain access to the place and that the closest people where approximately 75 to 100 yards away and consisted of three members of the grounds crew. She and I were the only ones present in the structure.

She went about setting up her equipment as I stood watching. At one point, she was basically talking to herself trying to decide if she should set up her tripod.

“I wonder if I should leave it on, or not ?” her voice echoed through the room.

We walked around the place for several hours and left with me thinking that nothing remarkable had happened. Until, that is, I was reviewing the tape later that night. At the point where she had been talking to herself about her tripod I clearly and distinctly heard the voice of a third person who was not present in the room at the time, answer her.

Here it is (in it’s raw form):
http://ghouli.com/evp/nothern_ok_mental/youllwantto.wav

When I heard this I stared unbelieving at my recorder and rewound it and listened to it over and over again. The next day I drove to her house and had her listen to it.

“Oh, yeah”, she says nonchalantly, “That’s an EVP.”

From that moment on I was hooked!

Electronic Voice Phenomenon (EVP) has been around as long as man has had the capability of recording sound. Thomas Edison, at the time of his death was working on what he called a device to record “spirit voices”. Other notable researchers in the 20th century included Friedrich Juergenson, with his work in the 1950’s and George W. Meek, noted for building the Spiricom device. One would think that with this many distinguished scientist interested in this phenomenon that it would have garnered more attention. But alas, like much of the paranormal world EVP has been brushed aside as well, a hoax, or intercepted radio transmissions, natural noises etc.

EVP in my mind, however, remains the most highly probable among paranormal phenomenon to be replicated in a laboratory setting. If you believe the work of George Meek and or, Otto Koenig, it in fact has been replicated in this type of setting.

Surfing the Internet today one can find a limitless amount of EVP scattered among the infinite universe of paranormal websites. One can find anything from the sound of radio static that is alleged to be reciting “The Pledge of Allegiance”, to the recorded sound of running water that is said to hold secret transmissions from the planet QRTEG-7.

So with all this evidence, why isn't EVP considered to be a proven scientific fact at this point??? As with many other aspects of the paranormal, to date it has failed to meet the rigorous standards of the scientific method. What can be done about it? Well, it can be approached with a logical, scientific viewpoint. Instead of wildly shouting that this, faint muttering that had to be amplified 100 times is an EVP, let’s throw that out and adopt some type of standard procedures for data collection.

EVP Collection

Ok! Here it is Saturday night and once again instead of being out clubbing with your friends, you've spent it wandering around an empty, dirty building sweating like a long-tailed cat in a room full of rocking chairs. You skinned your knee when you fell through that step coming up from the basement. Your hair is full of spider webs and you seem to have a very definite paranormal smell emanating from your body. You have 4- 90 minutes tapes full of recordings, so how do you go about approaching this in a scientific manner?

First, here are a few things to consider during the actual recording phase:

1. Use only one side of a tape — The goal is to eliminate any possibility of a false positive. By using only one side of the tape we automatically eliminate the possibility of bleed through from the other side of the tape. Personally, I have never witnessed this but I have heard of it happening and by using only one side of the tape this possibility is taken away.

2. Empty your pockets — Change, keys etc. can make some awfully strange sounds upon playback. Also wear soft-soled shoes.

3. Utilize headphones — Just listen in real time while you are recording. I adopted this method about a year ago. Now I will admit it is somewhat startling to hear an EVP right up in your face. In fact normally the next sounds coming through the headset are sudden sounds of scurrying feet heading off in any direction followed by a string of expletives that would make a Norwegian merchant seaman blush. As spooky as EVP is on playback, it cannot touch the eeriness of hearing one in real time. Although potentially it might be a bit frightening, it is also a very good idea to familiarize yourself with what certain noises will sound like on playback and be aware of your surroundings. But just being aware of the environment I can rule out decrease in potential anomalous sounds ahead of time. Believe me things sound very different when you are not hearing spatially like we do under normal conditions. Put some headphones on and you will see what I am talking about.

4. Limit the number of people — The smaller number of people around equals a smaller number of sounds and voices to sort through. For instance this past weekend I had another trial for my cognition study. I didn't even attempt to record for EVP as there were 17 participants, so I waited and recorded at times when I was alone or with one or two other people As an aside never ever whisper and try to refrain from spontaneous eruptions of noise and speak clearly when you must talk; mumbling is something that will sound anomalous on tape.

5. Always carry your recorder with you — I catch some flak for this and some people tell me that this will cause false positives. I agree to an extent, but any noise that could be created by the movement of a hand on the microphone, the microphone brushing my clothes, air from movement hitting the mic, none of this will sound like a vocalization. It could sound like a whisper or a mutter, but I'm not interested in that anyway and will discard it during analysis. There are a couple of reasons I think the recorder should be carried along. It has been my experience that certain people seem to always get EVP, while others never get any. In studying this on our team I notice that the people who are more garrulous and open are the ones collecting EVP. While people who are harder to get to know, or more closed off don’t seem to be as successful. To wit I feel that EVP could possibly come easier to certain personality types. I have also advanced many times my thoughts that paranormal phenomenon occurs on a scalar level, or at a certain place in space-time. Logically if the recording device is carried around it will have a higher probability of collecting EVP. Also I have never, ever to date gotten a valid EVP from leaving a recording device sitting stationary somewhere.

EVP Analysis

The next stage centers on the analysis of the recordings. Out of all these tapes and all these sounds how does one go about separating the regular noises from potential EVP? It simply isn't enough to say that something is a voice from beyond because it sounds creepy, or because no one remembers making that sound. Remember science demands more.

To begin with the tapes should be reviewed as soon as possible. This way the events of an evening will still be fresh and not clouded by time. Upon pinpointing a potential EVP, the first step should be to allow the rest of those that were present to listen to it to ascertain if they remember anything that might have caused it. In addition, all other recording devices including video cameras should be checked to see if the sound shows up on them as well. Some people believe that EVP collected on multiple devices strengthens the probability that it is indeed paranormal, but I feel that it probably is not as likely to be paranormal in this case. Think about this if you get a man’s voice yelling something on just one recorder when there were in fact 3 recorders and a video camera in a room, I think it is highly possible that this voice is of a paranormal origin.

Once a potential EVP is isolated you should classify or categorize it in some manner. Below is the classification scale my team uses:

Class A EVP
100% audible, clear voice & intelligent or reactive responses, not random noise or words. Example: having direct questions responded to or a reaction statement. Audible to the human ear no enhancements needed, no question to what the voice is saying. Low Hz levels do not always apply with true Class A EVP. Some have registered on a higher frequency as the team present, other data to support the potential EVP claim, such as other tape recorders, and or video gear.

Class B EVP
Practically audible, some doubt to the interpretation, syllables & consonant pronunciation is questionable. Appropriate tone & timing to question or statement made by team member. Frequency levels are also undetermined, but usually found on the low end of the spectrum. Non English EVP are placed in this category.

Class C EVP
Frequency levels support this unexplainable voice, not always audible, but some constants are present. Timing & tone apply to this category. Direct response is not required. Random voices or tones apply in this category. Out of text words or phrases that are not explained by correlating data. Audible, but faint, not a whisper or murmur those will fall into the Class G category.

Class D EVP
Non published or announced EVP. This may be an EVP, but it is too questionable to declare as paranormal. Whispers, faint murmurs, breathe like sounds and airy sounding voices. This classification is not the trash barrel for EVP; it is simply a holding area until the EVP has been validated by all data. Usually Class D EVP are not published or presented as paranormal phenomena.

Class G EVP
If you have to work that hard to clarify or understand the potential EVP, then it’s probably not an EVP. G stands for GARBAGE. Trash it & keep trying.

By establishing and adhering to a scale such as this a team can streamline the EVP analysis procedure.

Even after isolating a potential EVP and eliminating all other possible false positives there still remains a problem. How does one eliminate the voices of all living humans at this location? This seems to be a problem, but using logic and deduction, this problem can be circumvented.

Most research into EVP seems to agree that many of these “voices from beyond” share one common bond; they oscillate at a much lower frequency than the normal human speaking voice. So, in order to eliminate the human voice anything within that range must be discarded as a false positive.

Researchers agree that the average speaking range of the human voice is anywhere from 280 hertz (Hz) to 3400 Hz. So in essence if any voice within these limits is discarded then only truly anomalous recordings should remain. I would also recommend establishing Hz levels on all team members, to use during EVP analysis; it doesn't hurt to know each individuals range.

Well Russ, that's a good idea but how exactly do we do that? Well ideally all teams should have the FBI's voice print analysis software, but as most of us can’t afford that there are some alternatives.

There are freeware versions of several sound editing software programs available on the internet. Audacity, Cool Edit and Spectrogram all have the capability of measuring Hz levels and there may even be more that I am not aware of. I would advise however downloading one and trying it before you go buy the full version.

Finally, and above all else, EVP should not have to be enhanced or manipulated in order to be heard. If some sort of protocol and classification scale has been established then this should never be a problem. I see teams all the time that use programs to “enhance” EVP, this is just plain unscientific. If any data at any time is manipulated in any fashion then that in itself invalidates the data. If it is that hard to hear, then it probably should be discarded as trash.