# Is there an algorithm to hide text in another text, while preserving the meaning of the latter?

Is there a way to hide a text message $A$ inside another text $B$ using a cryptographic cipher/algorithm, while preserving the sense/meaning of $B$ and without anybody being able to notice there is a message $A$ hidden in text $B$?

I am looking for something that can be processed either manually or by a computer.

For clarification: I don't mean to hide a message like “Attack X” saying “I went to X on holiday this summer” but an actual cipher/algorithm.

I know for sure there must be something, but being new to cryptographic and not knowing much about it, I don’t know where to look or what to look for. Is there a way to achieve this? If there is, could you (besides a reference) also provide a short example?

• What you are looking for is called steganography. Questions about steganography can be on topic here, if it is the right kind of question. – mikeazo Nov 4 '15 at 19:24
• No, I know stenography... I don't want to add hidden messages in (image/audio/any) files but in text that can be printed or written on paper. I'm not looking to use that – Zorgatone Nov 4 '15 at 19:39
• Steganography can be in text, text that can be printed or written on paper. For example, hide your message in the first letter of every paragraph. Here is another example – mikeazo Nov 4 '15 at 19:49
• I never said anything about PGP/GPG. Where did that come from? The link I sent you to mentions PGP/GPG as not a solution to the problem they are trying to solve. – mikeazo Nov 4 '15 at 20:41
• How about this one. Try "Linguistic Steganography" in google scholar for more, that aren't too obvious. – mikeazo Nov 4 '15 at 20:44

I think you can embed message $$A$$ into message $$B$$ with simple $$LSB$$, but you will have message $$B'$$ and maybe alter the meaning. If you want the message $$B'$$ to retain the original meaning, you should design $$B$$ to do that.
• Just to be sure, do you mean Least Significant Bit by LSB? Also, it seems the last $B$ is supposed to be $B'$ – Artjom B. Jan 10 '16 at 16:42