# How would characters communicate under surveillance

I know the question is rather broad but I hope to find a somewhat general solution. Here is a sample scenario:

Two characters are in a room. One of them starts discussing something private, the other says "Shh, don't you know the walls have ears?" signifying that they could be under surveillance.

How can they communicate in relative safety? Here are my assumptions so far:

• the setting is of oppressive almost constant surveillance, akin to George Orwell 1984
• the surveillance could be mystical or high tech or simply just guys listening and watching. The exact nature doesn't matter as much, the characters possess similar means. But it's not a matter of hopping to the local supermarket and getting an anti-surveillance kit.
• the characters also do not possess any abnormal means of defeating the spying, for example they won't have telepathic communication super secret wireless communication devices embedded in their skulls if these don't exist in the world.
• the observation is not absolute. However, the characters do not know what is monitored and how. It can be assumed that simply passing notes is not automatically safe (they could be watching, not just listening) but it could be that it is safe in, some environments. Going by 1984 again, Winston's methods may have succeeded, not just ruled out to be completely subverted.
• the characters need not necessarily know each other, so they don't have an pre-established code or cipher or something. They can certainly do this as things progress but it shouldn't be a given.

So, I am looking for how would the characters go about establishing any number of these things

• where is safe or not - if they could at least be fairly sure about either, they might find places to communicate
• what the nature of the surveillance is - audio, visual, both, maybe others kinds, like wiretaps, or invisible spirits watching a room.
• how to subvert it - would probably rely on the above first as the characters would need to work out a scheme of communication without being spied on. Unless there is a safe way to communicate without establishing the method of communication first.

The characters could use ciphers or encryption but note that these could be cracked or compromised, if they establish what the scheme is in a non-safe place. Ideally, I would want the characters to be able to communicate face to face (or at least being in the same place) in some way.

I am interested in how people would go about understanding the limits of prevailing surveillance without too much risk to themselves.

• I'm thinking of game theory... Yes he's the one who did it not me. – user6760 Apr 5 '15 at 8:48
• Do the characters have to conceal what they are communicating about or do they have to conceal that they are communicating at all? Take a pen and paper, a blanket to hide and a candle to read under the blanket and to burn the message afterwards ... – o.m. Apr 5 '15 at 8:57
• @o.m. They'd want to hide it in as much as to not arouse suspicion. hiding under a blanket within view of a camera would hide the communication but would not avoid the suspicion. I'll edit the question to clarify some points. – VLAZ Apr 5 '15 at 17:04
• Please be careful to not turn this into a question about individual-character actions or character building (both of which are specifically off topic here). It seems to me to be fine as it is now, just something to watch out for. – a CVn Apr 5 '15 at 18:32
• @MichaelKjörling Thanks, I do not intend it to be for a specific character but for any of the inhabitants of the setting - I am not sure how would they operate below the radar which would be an important starting point for overthrowing it. I hope this is OK. It should fall under the third of the welcome topics "How to achieve a specified effect in a defined world" only I've not yet defined the world, hence I want the general strategy rather than specifically how to deal with magic or technology or aliens or something else. – VLAZ Apr 5 '15 at 18:55

If you go to the Information Security Exchange you will see over and over again:

There is no such thing as perfect security.


An adversary with sufficient desire and means will compromise your communication pathway. That is what makes 1984 (and real life big brother states like China) so frightening.

These big brother states are limited by their ability to filter out the normal from the interesting. During the Vietnam War, American POWs were isolated from each other in order to wear them down. The POWs began communicating using everyday sounds. Since everyone in the Hanoi Hilton was sick, the POWs would cough and sneeze in code to pass messages. Chores became means of communicating - the delay between sweeps on a broom could be varied to produce Morse code.

These methods are not fool proof, but they were sufficient for the POWs. A more aggressive "big brother" would have certainly understood every message passed, but the North Vietnamese were distracted by the whole fighting a war thing: they could only devoted so much manpower / time / energy to the POW camp.

So the key to your story will be three-fold: 1) What is normal behavior?

2) How can I modify that behavior to pass information?

3) (most importantly) What limits Big Brother? - with unlimited time and money nothing is secret.

• I guess this makes sense. To answer point number 3) I am unsure yet what would limit Big Brother, but so would the characters in-universe. I was more questioning how would one try to test that specific thing without relying on blind luck. Thank you for the answer, I think yours did point me at the right direction of how to handle it - while I like the rest, I think I'd go with yours. I'd wait for the weekend for a "last chance" to answer. – VLAZ Apr 10 '15 at 16:19
• Money and personnel are the limiting factor for authoritarian states. Governments exist to provide services - security from external threats, roads, market regulation (FDA, EPA, etc). Every person working surveillance is not working on those tasks, and the services suffer. If government fails to provide those basics, people will revolt and the people outnumber the government, so a large revolution will always be successful. What level of resources can Big Brother expend without roads crumbling and poisonous food routinely making it to market? People will revolt over security, not privacy. – codeMonkey Apr 13 '15 at 13:53

There are two ways to beat this sort of system.

Firstly from without:

Essentially you need to be able to hide the message in among seemingly innocent interactions.

For example I read a sci-fi novel where an undercover agent always went and had sex with the same person whenever he got back into port. What wasn't known to the local authorities is that they were both agents and they had a language that could be "spoken" between them by applying various levels of pressure with various parts of their hands while they seemed to be paying attention to other matters.

The example in another answer of inserting key words into phrases is also a good example, as is using pauses between words.

The key thing is to bury your communications inside something unrelated so that people do not see anything untoward either at the time or later on replaying the surveillance.

You should also make a habit of going and having similar conversations without any hidden meaning to all sorts of other people. This means that if one person is uncovered then it becomes much harder to work out who out of all their contacts is actually a contact.

Secondly from within:

Get people inside the system and they can work out where is monitored and where is not, knowledge of those black spots can then be distributed to others and information slowly spreads as more people are recruited.

The weakest point in most systems are the humans running it, not the systems themselves. Subvert the people and you subvert the system.

Combined approach

Perhaps the best solution would be the combined approach. Insiders provide some privacy in one area which is then used to develop codes and other methods that allow you to communicate even while under surveillance.

• ISTR that the book you mentioned is The Algebraist by Iain M. Banks. – o.m. Apr 5 '15 at 8:54
• @o.m. That sounds right, I can't remember exactly what book/author it was but it was definitely that book or one similar. – Tim B Apr 5 '15 at 13:04

Sign language? Can't be overheard, and you can use your body to block your communications so they can't be seen. (You can whisper to someone in sign by making your signs really small so that only they can see).

Alternatively, morse code using touch. Morse code already exists, you wouldn't have to spend too much time developing it. You might not even have to ever broach the idea of using it out loud - everyone knows SOS. if someone starting tapping that on my knee in an orwellian world i would go out and learn the rest of morse code to find out what else they had to say. Light taps can be easily obscured by clothes/blankets/screens.

I think there would be an element of risk no matter what system you choose. The consciousness of this risk can only improve your story by adding tension.

In that environment I would just assume that I am always under visual/audio surveillance 100% of the time. Anything I read or write will be read by the bad guys.

The best form of communication would be something that you can do in plain sight in front of your observers.

One TV show had a bit where two stage magicians communicated using Morse code with the length of the spaces between spoken words. It's clever, but if the bad guys play back a video recording there's a chance they could pick up on this.

If your two people have a safe way to make arrangements prior to being under surveillance, they could set up a set of One-Time Pads.

Rather than going through the math described in the Wikipedia article, I'd just memorize a simple lookup table such as

• fish: get me out of this place!
• wrench: I need more time
• potato: I need more stuff/tools
• ring: I need more people
• sock: I quit, this is too complicated.

The interesting bit is that you don't get to reuse the table. You must have a new one with completely different (& totally random) key words for every communication.

Anyone listening to your conversation might not even notice that you're sending secret messages. If they do notice, how are they going to figure out what "fish" means, when in the next message you might use the word "carbon" to say the same thing and you never re-use the word "fish"?

• Yes, this is more or less what I've been thinking of, but it requires the characters to find a safe place and/or mode of communication first before they establish any sort of code. I'm not sure how would that be done. – VLAZ Apr 4 '15 at 16:58
• That's not really a one-time pad, it's a code book, and they work surprisingly well when the number of messages that need conferring is sufficiently small. – a CVn Jun 22 '16 at 12:38