Many times have the annals of the paranormal been graced with descriptions of Frank's ghost box, an electronic device that is claimed to allow ghosts to communicate with us through randomly tuned radio broadcasts. Imagine turning the tuning knob on a radio (first, imagine a radio that still works that way), and recording the sound output. You'd expect a bunch of random noise from across all those stations you just swept past, and that's exactly what Frank's Box produces. But inventor Frank Sumption, and many imitators who have built similar devices, believe they receive intelligent communication in that noise. It's basically an iteration of what ghost hunters call Electronic Voice Phenomena, the idea that ghosts communicate with us through electronics.
I should note that the impetus for this episode was an email I received from Frank Sumption's niece. I won't give her name or what she said in the email, but her point is that Frank is a good, honest, average guy who is notable only in that he believes he's stumbled onto something. He doesn't scam anyone; he's neither selling his box with unproven claims, nor is he charging people money to talk to their dead relatives. Frank makes the plans to build his device available for free as a downloadable PDF file, and he has a page on SoundCloud where he posts his most interesting recordings, for everyone to hear. He's a guy who has some bad information, and has failed to apply critical thinking to it within the context of scientific literacy. In other words, he's exactly like 99% of the people out there. Nevertheless, most of the time when Frank Sumption has been mentioned anywhere, online, in print, or on a radio program or podcast, he's dismissed as a crazy nutcase — and that's precisely what prompted his niece to write. So today we're going to look at the phenomenon of Frank's ghost box and talk about how and why it occupies the place it does in pop paranormal culture, and hopefully find something more intelligent to discuss than hurling personal insults.
So let's get to what you really want to hear, some of Frank's received messages. Let's get started with one just to get the idea. On his SoundCloud page, Frank posts clips that he has already edited down to just the moment that he considers significant, and then repeats each one several times as he adds equalization and/or changes the pitch to make the speech more distinct. Here's one, and I'll just tell you in advance that Frank has identified the words "Frank we'll save you" in here:
Frank's Box, or the "ghost box" as Frank prefers it to be called, is often described as a random radio tuner, but there's a bit more to it than that. He first developed the idea in 2002, and there are three basic parts to his device. First is a component that generates either white noise, like this:
Or, a sweeping tone like this:
Then the voltage of that signal is moderated to the correct voltage that can be used to control the tuner of a radio taken from a car stereo, which is the second component. Car stereos are used whose tuners are voltage controlled, which in their original factory condition, would have come from the tuning knob; but in Frank's case, it's either a random signal that constantly tunes the radio all over the dial, or a sweeping signal that tunes the radio all the way from one end to the other. Frank has said he believes that the sweeping method gives better results than the random, or white noise controlled, method.
The third component is what he calls the echo box. It's a box about the size of a shoe box with speakers and a microphone inside. The radio's audio output is played over the speakers inside the chamber, and picked up by the microphone. The signal coming from this microphone is what produces the final product. So what you hear from the box is not the direct output of what the radio tuner picks up; you're actually hearing it one generation away. It's played over speakers into a small box and then that live audio is recorded and played for you.
Frank describes the purpose of the echo box:
The randomly tuned radio modules provide a source of random audio that is sent to an enclosure I call an echo box, for lack of a better name. The idea of the echo box was received by what I can only call telepathy, which indicated to me that there is at least some "outside" guidance working in tandem with the electrical components of this system.
For me, this analysis is the most illogical part of the system. If the purpose of the echo box is to give the ghosts a chance to interact with and alter the audio to say what they want it to say, then what's the point of the radio tuner part to begin with? Most EVP enthusiasts believe that the ghosts are able to interact with the electronics of the recording system itself, thus producing sound that was inaudible in the room where it was recorded. If this is the case with the ghost box, then either the radio tuner or the echo box is redundant.
Logically, all of this is predicated on two massive assumptions. Assumption #1 is that ghosts exist, or at least some sort of spiritual entities that are actively trying to communicate with us. Assumption #2 is that these ghosts (or other beings) are able to communicate with us via electronic equipment. Both of these assumptions, in order to be true, would require radical revisions to much of what we know about the universe. This is where Occam's Razor comes into play. Occam's Razor tells us that of two possible explanations for an observation, the one requiring the fewest new assumptions is more likely to be the true one.
Frank's explanation for his observation is that ghosts are talking through his box. Another explanation, founded in scientific skepticism, suggests that Frank is simply honestly mistaken. Frank's explanation requires radical new assumptions about our universe; the skeptical explanation requires no new assumptions at all. People are mistaken about their observations all day, every day. I am, you are, and Frank is. Everyone is.
Proponents of the ghost box have argued that the evidence of the voices being real is so overwhelming that the skeptical explanation becomes the one requiring the most new assumptions. And, granted, if the ghost boxes were indeed producing unambiguous communications, they might have an argument there. So let's go back to SoundCloud and check out another example — keeping in mind that the ones Frank has chosen to post have already been carefully edited and selected for being the best quality. This time, I'm not going to tell you in advance what Frank thinks this one says. Listen:
Besides being actually quite nice musically, this one, according to Frank, says "We love Obama." Can you hear it now?
But when I listened to it, I thought it said "We kill Obama." A slightly different concept. Which do you think is closer?
This illustrates the largest problem with not only the ghost box voices, but with all paranormal electronic voice phenomena: they are completely subjective. It's not printed on a page with an unequivocal meaning; it's a collection of ambiguous, vague sounds that could mean anything to anyone, depending on where they're from, what accent they're used to, what's their native language, and so on. Consequently, random sounds are always going to be interpreted differently by different people who pick up different patterns. We call it audio pareidolia, or apophenia. It's what happens when your brain correctly tries to discern patterns in randomness, but comes up with an incorrect identification.
Thus, anyone who has called Frank Sumption crazy is wrong — at least, based only on his work with the ghost box. Frank's brain is doing exactly what it's supposed to do, as was mine when I listened to his recording and heard "We kill Obama." In reality, there was no one on the other end of the line saying anything specific — so far as we know, and so far as rational investigation of how the box works will lead us.
To his own misfortune, Frank has placed an amount of faith in his analysis that most of us would characterize as too much. It's normal and healthy to misinterpret minor observations when you can integrate those interpretations into your life; but when they start to take over your life or change its course, then you've got a real problem. One of the most popular themes to the messages he believes he has received are warnings of tremendous natural disasters. Try this:
Frank's analysis is "In space, it brings the big rock, it kills." Listen again:
How ghosts would have astronomical knowledge and be able to express it in English, yet not know the word for asteroid, is not convincingly argued. Here's another. Frank believes the ghosts address him personally as "Purple", and to save time, I'll prime you first with Frank's interpretation. It is "Purple, look in space, a rock is coming, big rock," and then he plays the same clip backwards and gets "and my Jesus, he wields it, rock will hit here":
Well, I didn't hear it either. But Frank has an explanation for that. He wrote in an email to the Committee for Skeptical Inquiry's Karen Stollznow:
It seems not everyone can hear it, especially just starting out. It takes time to tune in the ear, and maybe develop some intuition. The box is extremely complicated, and talking about it for a mere two hours just is not enough time, two years would be better. It takes time to develop a feel for the communications. I get very frustrated when I get voices that seem perfectly clear, and no one else can hear it.
In logic, we call this a "special pleading". It's an appeal to a higher power or greater ability than you have or can comprehend, thus your experiences are inadequate and not useful as evidence. This is how Frank has convinced himself why other people not hearing his messages is not a problem for his device. He also has a way to twist the ambiguity of each recording's meaning into evidence in favor of the process. He wrote this on a Yahoo Groups message board:
It's not the device, it's the user. The messages received are for the user, and rarely does anyone hear the same thing as the original listener... No two people hear the same, so [the idea of] an "objective listener" is an insult to most people doing this work.
Again, a special pleading. If you didn't hear the same thing as I did, it's because the message wasn't meant for you. If this were true, it would make every analysis of every recording by every EVP practitioner immune to question. Fortunately, the scientific method does not allow us to accept concepts that are immune to testing. If an idea is not verifiable, then it's not scientific. Frank's ghost box is not scientific. It is a fine toy and an amusing gadget to fool around with, but with all of its fundamental premises grounded firmly in the anti-logical, it is not something that is yet worthy of any scientific attention.
Cite this article:
Dunning, B. "The Many Voices of Frank's Box." Skeptoid Podcast. Skeptoid Media,
26 Aug 2014. Web.
20 Oct 2017. <http://skeptoid.com/episodes/4429>
References & Further Reading
Edward, M. "Box of Fiends." SkepticBlog. Skeptologist Partners, 25 Apr. 2009. Web. 28 Jun. 2014. <http://www.skepticblog.org/2009/04/25/box-of-fiends/>
Shermer, M. "Telephone to the Dead." Scientific American. Skeptic Society, 1 Jan. 2009. Web. 28 Jun. 2014. <http://www.michaelshermer.com/2009/01/telephone-to-the-dead/>
Stollznow, K. "Frank's Box: The Broken Radio." The Good Word. Committee for Skeptical Inquiry, 28 Jan. 2010. Web. 28 Jun. 2014. <http://www.csicop.org/specialarticles/show/franks_box_the_broken_radio/>
Sumption, F. "OldManAlien." SoundCloud. SoundCloud Limited, 11 May 2011. Web. 28 Jun. 2014. <https://soundcloud.com/oldmanalien>
Sumption, F. "Frank's Boxes." Tripod. Lycos Inc., 11 May 2011. Web. 28 Jun. 2014. <http://purplealiengirl.tripod.com>
Sumption, F. "The Ghost Box: A Notebook." Ghost-Tech. Craig Telesha, 1 Jan. 2007. Web. 28 Jun. 2014. <http://www.ghost-tech.com/adobe/Franks_box_6-19.pdf>