Redundantly encoded information is a concept that involves the encoding of information in a way that it can be observed several times, even when some parts are removed or altered. This concept is present in various fields like linguistics, genetics, and computer science.
Imagine that you are exploring a jungle. You want to learn more about its composition and the relationships between its animals. Specifically, you want to learn about the food chain in it. You spend your days in the jungle, examining the consumption behaviour of several different animals.
Sooner or later you will find the same instances of predator-prey instances all over the jungle – whenever some animal A is eating plant B or eating animal C, it’s likely that you will find the same behaviour in some other part of the jungle.
Thus, predator-prey relationships are redundantly encoded in the jungle. You can infer from one instance to many others. You throw away one observation (animal A eats plant B) and still remain with the same amount of information about predator-prey relationships.
In linguistics, redundancy is the presence of extra information that can be removed without affecting the meaning of a statement. For example, consider the following sentences:
- Will John leave the room?
- Should John leave the room?
- I dare John to leave the room!
- When will John leave the room?
In all these statements, the core piece of information is that John leaves the room. The other parts of the sentence provide context or extra information, but are not strictly necessary to convey the main idea. This redundancy helps to ensure effective communication, as it allows for the message to be understood even if some parts of the sentence are lost or misheard. Thus, the proposition of these sentences is redundantly encoded: John leaves the room.
John Wentworth explored further examples in his post about redundantly encoded information and abstractions.
Source: Wentworth, J. (2022) ‘Abstractions as Redundant Information’. Available at: https://www.alignmentforum.org/posts/vvEebH5jEvxnJEvBC/abstractions-as-redundant-information (Accessed: 20 March 2023).
Chan, L., Lang, L. and Jenner, E. (no date) ‘Natural Abstractions: Key claims, Theorems, and Critiques’. Available at: https://www.alignmentforum.org/posts/gvzW46Z3BsaZsLc25/natural-abstractions-key-claims-theorems-and-critiques-1 (Accessed: 19 March 2023).