There are reports of blind people being able to develop an echolocation ability, by listening to the echo of sounds emitted by them.

Since this ability is pretty handy for spies and soldiers, the army of a middle-age kingdom has trained some non-blind specialists to use this ability while moving in dark environments without the need of using a light source.

Of couse moving in the dark in a castle or in a town while shouting is not that smart if you want to stay hidden.

What is a sound which can be used, and how could the specialist produce it?

  • $\begingroup$ I think it uses vibrations inside the skull as well so you're stuck with whatever the operative itself can produce. $\endgroup$ – Mormacil Mar 30 '17 at 14:50
  • 23
    $\begingroup$ I don't think there's any scenario where you're going to be able to make useful sounds that the spy can hear and use to echo-locate that won't also just announce his presence to anyone within a few dozen yards. The clicks blind people use are very loud, and one of the answers suggested bird song, but birds don't tend to sing at night in the complete darkness, and anyone who's used to living in the quiet of a pre-industrial village is going to find periodic bird songs in the night quite strange. Any useful sound will let other people locate you far more easily than you echo-locate them. $\endgroup$ – Shufflepants Mar 30 '17 at 21:30
  • $\begingroup$ I absolutely love this & all the comments - perhaps the range of answers and solutions are represented in the story when told as perspectives from various groups in the story. $\endgroup$ – Mirv - Matt Mar 31 '17 at 18:36
  • $\begingroup$ Also....what if the spy just used noises of anything happening in the area, of course you need some method to judge how far the originating noise...or a wall, "...a truly talented student of the blindschool could listen to the sound travel past herself & then use the noise going past a second time to determine proportionately how far things are from them based on their knowledge of the wall they just walked past..." $\endgroup$ – Mirv - Matt Mar 31 '17 at 20:00

Given my own experiences with night sounds, I'd go with cricket chirps.

People who live among those creatures barely give the noise a second thought. Attempting to track the source down is rarely worth it, because the cricket will stop for a while (long enough to frustrate a searcher, far to short to alleviate the aggravation) once something big like a human starts moving about. Due to this, I once spent an entire night frustrated by a cricket in my room, only to discover the next morning it was a malfunctioning digital watch.

I'm noticing from some of the other answers that cricket chirps are quite compatible with them. They are fairly high-frequency, very similar to the noises echo-locating humans have been observed to use, and a natural sound that is not likely to be investigated.

  • $\begingroup$ I've never seen someone do a convincing cricket chirp. This seems like a great way to attract attention to yourself. $\endgroup$ – sphennings Mar 30 '17 at 20:05
  • $\begingroup$ @sphennings - There are some people on reddit who claim it can be done (some with A/V examples). I suppose a device could be used too. Sadly, I no longer have that watch. :-) $\endgroup$ – T.E.D. Mar 30 '17 at 20:06
  • 4
    $\begingroup$ This would be especially effective if the spy was eavesdropping on an awkward conversation. $\endgroup$ – SPavel Mar 31 '17 at 2:39
  • 1
    $\begingroup$ @sphennings I can do a dang good cricket chirp. It's good enough to fool people who 1 - don't know I can make the noise and 2 - don't see my mouth when I'm chirping, even indoors or in places where there aren't crickets. A good cricket chirp can be approximated by purring against the hard palate while whistling a very high pitch (have to overhang your upper lip and whistle in to hit it) $\endgroup$ – Jeutnarg Mar 31 '17 at 18:16
  • 1
    $\begingroup$ @Jeutnarg Must be different species of local crickets then. :) $\endgroup$ – sphennings Mar 31 '17 at 18:18

According to this paper, clicking the tongue and hissing are the most common sound used for human echolocation. There are also reports of people tapping hard surfaces with their canes and using that as the sound source. Assuming your spies were trained by blind people, who developed this skill out of necessity, they would probably use similar sounds.

There is a small scale study where 10 sighted people were taught basic navigation skills within a few days. This study also tested the efficacy of various sounds and concluded that palatal clicks are the most effective.

Just remember that the reason for moving in the dark without a light-source is to not be noticed. Making loud clicking noises and listening for the echo will announce your presence to nearby guards.

  • 8
    $\begingroup$ @sphennings It looks good now. Basically answers need to stand by themselves even if all the linked-to websites were to disappear which this now does. $\endgroup$ – Tim B Mar 30 '17 at 15:22

Let's go by a basic sound principle: higher-frequency sounds are absorbed quicker than low frequency sounds. For stealthy purposes, this means that someone in the next room is less likely to hear a squeak as opposed to a burp, volumes being equal. This is also beneficial for our spy as it means that the echoes from their high-pitched emission are more localised to close-by objects - they don't get as much cross-talk from further objects. Bats use ultrasound for echo location; I believe recorded cases of human echo location also uses high-pitched clicks.

Now, is a high-pitched click appropriate for a spy? Probably not. But what were common in the middle ages? Rats. Lots of rats. And rats use high-pitched sounds themselves.

A spy sneaking around and making squeaking rat sounds would probably be safely ignored - unless they had broken into the kitchen!

  • 5
    $\begingroup$ A sound with a definite frequency has a longer duration and will have a longer duration echo. A click has many (mostly high) frequencies, and a very short duration which makes the echo more distinct. A more distinct echo will give the echolocator a better sense of the room. $\endgroup$ – user121330 Mar 30 '17 at 22:14
  • $\begingroup$ "A sound with a definite frequency has a longer duration" - that doesn't really make sense; frequency, amplitude and length of phrase (envelope) aren't inherently related, they are all independently controllable. A "Ssssh" sound is a very wide-band sound (i.e. not a definite frequency) but I can make one for a long time. The vocal mechanism we have available to generate a "click" is limited to short durations, is this what you meant? $\endgroup$ – Hugh Nolan Mar 31 '17 at 16:44
  • $\begingroup$ Frequency and amplitude are independent. The envelope matters a lot if it's on a similar scale to the wavelength though. A Fourier transform of a sine wave over a short duration yields a wide range of frequencies. The frequency's uncertainty decreases predictably as the envelope gets wider. $\endgroup$ – user121330 Mar 31 '17 at 17:53

Your own question, a prior answer, and a comment collective already give us several parameters:

  1. The human body must be able to produce the sound.
  2. The sound must not attract the notice of others.
  3. The sound must produce an echo that can be used to interpret the surroundings.

I believe the solution is bird songs.

It takes practice, but many people have been able to produce very realistic mimicry of a wide variety of common backyard birds. That last part is essential, as when someone hears such a sound, he should assume the sound is coming from something that commonly visits his yard, not from an enemy performing surveillance on him. Finally, again with a lot of practice, the high-pitched warbling tones should reliably rebound to the ear, allowing you to determine approximate size and shape of nearby obstacles.

  • 6
    $\begingroup$ At some point someone is going to come along and make a joke about this answer being given by a duck, and refer to that old rumor that a duck's quack doesn't echo. So I may as well do it myself. $\endgroup$ – cobaltduck Mar 30 '17 at 15:43
  • $\begingroup$ Except the physics of this means that you would have absolutely zero useful information from a quack. $\endgroup$ – Aron Mar 31 '17 at 1:13
  • $\begingroup$ Quacks do too echo, I can testify. If you have a big river running through your city there will be ducks. They do not normally quack in the night, though. Unless an animal comes and disturbs them, then they can quack loud enough to wake the dead. $\endgroup$ – RedSonja Mar 31 '17 at 9:03

Many people commented that the spies would give themselves away by producing the sounds.

A way out of that is externally produced sound. Someone else 'on your side' engulfs the area in sound patterns; retrain your spies to listen for the reflections and obstructions ('shadows') of those.

This may also allow you to vary the sound frequency, thereby providing more information to the spies, because the refraction/reflection patterns change with frequency.

  • $\begingroup$ While there are real world examples of people using externally produced sound, either through a clicker or by tapping a surface with a cane, neither of these is a stealthy option. $\endgroup$ – sphennings Mar 31 '17 at 13:07
  • $\begingroup$ Someone else (on your side) is generating the sounds, not the spies. The only information that is given away is that there are possible spies in the entire area. No that is not stealth, but the spy himself is not giving away his location (which seems to be your objection). I did not mean 'externally' as 50-100 cm away. $\endgroup$ – Jan Doggen Mar 31 '17 at 13:16
  • $\begingroup$ There's a maximum range of about 100 feet. That doesn't provide much leeway for an extra person making sounds to stay hidden. $\endgroup$ – sphennings Mar 31 '17 at 13:21
  • $\begingroup$ How about - the spies are trained to use noises around them to judge distances... $\endgroup$ – Mirv - Matt Mar 31 '17 at 19:57
  • $\begingroup$ @Mirv Could be, but than you are almost back to 'normal' situations. Plus, judging interference with just any noise is more difficult then judging interference with a specific type of noise that you can train with. $\endgroup$ – Jan Doggen Apr 1 '17 at 9:46

There's two ways to hide, be unobserved or be unnoticed ... perhaps a range of sounds would be best - rather than just one, to blend in ... like do that popping noise which would sound like dripping water or as another said - rats, crickets etc ...

If the person has hands, why not use a tool to create common sounds too? Like crickets could be made with pieces of wood where the edges rub together


Wavelength determines fidelity. For high, surface detail you need high frequency, while for lower, but deeper detail you need low frequencies.

Infrasound frequencies travel further, and so can be used as a long distance communication channel without any need for electromagnetic radiation. Low frequencies produce fewer reflections. They are great for seeing through walls, not so good for seeing in the dark. Being infrasound they cannot be heard without assistance. Like elephants. Elephants have a better ability to hear infrasound than humans.

Ultrasound frequencies travel less far, but have very high fidelity, and produce more reflections. They are great for seeing in the dark and around walls, but not so good for 'seeing' through walls. Being ultrasound they cannot be heard without assistance. Like dogs. Dogs have a better ability to hear high frequencies than humans.

Both are great. Because echolocation is main,y concerned with not bumping into things, high frequencies - and ultrasound - are more commonly found in nature. But elephants (and possibly whales) have the ability to use infrasound for long distance 'phone calls'.

Without mechanical assistance it may be really hard to achieve either ultrasound or infrasound frequencies - hence why palatial clicks (high frequency) or cane taps are most used. A portable infrasound generator would be the equivalent of a subwoofer. But an electronic cane that you could press hard against the floor could provide great infrasound frequencies.

So a double action comms. cane with an ultrasound chirp and an infrasound tip could really work. You would need earbuds to decode the information, but it would be effectively silent.

Being able to generate two frequencies from each hand would provide a far easier access to 3D resolution. Even though your ears will provide excellent 3D awareness, the brain can use whatever it's given, and two precisely known sources provides a clearer image.

So, you want the body to produce it.. Maybe there's a way of training yourself to generate and sense infrasound, or even ultrasound. You may need a prosthesis such as a special tooth to pick up ultrasound. Infrasound can sometimes be sensed, even though it cannot be heard.


As with any imaging system in the world, our ability to resolve detail is dependant on the bandwidth of the signal (and of course the speed the wave). Given that sound has a much lower velocity than light, a much higher bandwidth is required to produce any meaningful images.

So any answer describing any melodic sounds would fail to resolve anything smaller than a castle.

Conversely anything with enough bandwidth to resolve, say a person, would be instantly recognizable.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.