This is a draft
This post starts the same way as a previous post about communication problems vs information problems. Here is the beginning again:
Misinformation is consensually defined as untruthful, misleading messages. We would rather have information than misinformation.
Meanwhile, the task of defining information remains classified as a challenge or a matter of context. 21 years ago, Wikipedia started. The same year, an outline for the encyclopedic Wikipedia article on ‘Information’ was drafted. 21 years of discussion later, Wikipedians are still trying to agree on what information is.
“Conceptually, information is the message (utterance or expression) being conveyed. Therefore, in a general sense, information is “Knowledge communicated or received concerning a particular fact or circumstance”.”
“Viewing information as the conveyor of a message contradicts my experience on how this word is used. But your experience may be different from mine.
“Information, in its most restricted technical sense, is a sequence of symbols that can be interpreted as a message. Information can be recorded as signs, or transmitted as signals.”
In this post, I will not assume nor propose a definition of information. I don’t want to add yet another necessarily opinionated theory of what information is. I mean to illustrate in a later post how analytical theories and fully formalized definitions aren’t always appropriate at a given time in the history of human understanding. Sometimes we are not ready for them yet. Our understanding first needs to mature through a combination of intuitive and half-baked analytical steps. This, I think, is where our understanding of information presently is at. We’re not ready for a consensual formal analytical definition of what information is. So let’s humbly respect the present, ambient state of confusion and indecision around information and cautiously explore our logical options.
Let’s start simple:
Is information the container or the content, or a bit of both?
The question is inspired by Wikipedians’ traditional worry (cf quotes above): is information the conveyor of meaning or is it itself the meaning conveyed.
The container ≈ the type of things like symbols, language, encoding that contain/convey/communicate meaning
The content ≈ the message / the interpretation or meaning of symbols / the sense we make of them
What about misinformation?
In the spirit of starting simple:
ASSUMPTION NO 1: Misinformation is different from information.
And:
ASSUMPTION NO 2: We prefer information to misinformation.
ASSUMPTION NO 2 (reformulated): We would prefer misinformation be replaced with information.
Based on Assumption No 2, we have:
THEOREM NO 1: Misinformation is of the same type as information.
where :
DEFINITION NO 1: X and Y are of the same type if and only if X can replace Y and Y can replace X.
| Information |
Misinformation |
|
 |
 |
Same type: okay! |
|
 |
Not the same type: not okay! Violates Theorem 1 |
 |
|
|
|
|
Same type: okay! |
DEFINITION NO 2: We say that X is of type 'container' if X contains/conveys/expresses meaning.
DEFINITION NO 3: We say X is of type 'content' if X is meaning.
See above, first figure of this post.
Of course all of these considerations are only worth having under the following assumption:
ASSUMPTION NO 3: Meaning is different from the expression of meaning.
Ideally, one is distinguishable from the other in practice. Content is distinguishable from container of content.
Now contrary to the message itself (cf fig 1), symbols and series of symbols conveying the message can be jumbled and unreadable, but are neither true nor false in themselves.
ASSUMPTION NO 4: If X can be qualified of being either true or false, then X is not of type container.
Assumption NO 4 reinforces the parallel with glasses of wine: it’s not the glass itself that may be good or bad, it’s the wine in it. Similarly it’s not a series of symbols that are true or false, it’s our interpretation of them, the meaning we give to them.
That said, you are extremely welcome to pinpoint fallacies in any of this post’s assumptions and reasoning!
Based on Theorem No 1, we have 3 possible ways of answering the two questions raised above about (mis)information and their definitions in terms of container/content:
1. Either information and misinformation are both of type container
2. Either they are both of type content
3. Or they are both a bit of both.
| Information |
Misinformation |
|
|
|
Both container |
|
|
Both content |
 |
 |
Both both |
Information is of type container.
Information is a certain way in which we arrange symbols to express a message. It isn’t the meaning of the message (that would be the content conveyed by the symbols).
Misinformation is comparable to information yet different from it (Assumption 1 + Theorem 1). Misinformation is also of type container. What we want to do with it is eradicate it or replace it with information (Assumption 2).
We are beating around the bush…
LEMMA NO 1: If information is of type container then, misinformation can't be false.
Proof: If information is of type container then, misinformation is of type container by Theorem 1. And by Assumption 4, misinformation can’t be qualified of false, nor of true for that matter.
Deep down if we are honest with ourselves, our problem with misinformation has nothing to do with untruthfulness. What we really are out to eradicate is a language, not the French language nor Javascript, but still a means of expressing operational content. Too bad we’re still unclear as to how misinformation as a container, a language can be recognised.
The things we say about misinformation being wrong are just to give ourselves good conscience. The dirty truth is that it’s what we are doing that is wrong.
No wonder there isn’t a UN meeting about misinformation that doesn’t cycle back to the topic of censorship and freedom of speech! Those are all indeed the same topic under the previous set of assumptions.
Both information and misinformation are of type content. But I’m assuming that by now, everyone equipped with a relatively functional mind, has noticed that the notion of truth becomes very silly, very slippery and vacuous and dysfunctional when it is generalised and stripped from the context in which it makes sense.
Of course, the container conveys part of the context.
So assuming information and misinformation are of the content type i.e. pure meaning makes us say really stupid things — drawing on an extremist view of Absolute Truthhood / Falsehood.
ASSUMPTION NO 5: There is no sense in discussing things that are absolutely false (respectively absolutely true) independently of how they are expressed.
Forget about identifying them, forget about arguing that they exist or that they don’t. Things that are true/false independently of how you express them, by definition are not things that can even be discussed. Something that is true/false independently of its experssion, is experienced independently of any form of language. It is essentially ineffable.
LEMMA NO 2: If information is of type content and not ineffable, then misinformation can't be false in itself.
Proof: By Theorem 1, misinformation is like information: of type content and not ineffable. Lemma No 2 follows from Assumption No 5 and the fact that it is presently making sense to us to discuss misinformation.
Information is neither purely container nor purely content. It’s a bit of both. As is misinformation (by Theorem No 1).
I hate drinking good wine out of a water glass. I don’t know if it changes the taste enough for me to notice or if I can’t help but imagine I’m drinking cheap table wine, the kind my grandpa used to drink out of a water glass at every meal.
Perhaps the problem with misinformation is similar. It’s not so much a problem of container or of content. It rather is a misfit between container and content.
Maybe my palate is sensitive enough to detect a difference in the degree of ventilation of the wine depending on which glass it is served in. Maybe my mind is suggestible enough to change the way the wine tastes.
If there is something similar happening with information, then once again, things are not as simple as a true/false dichotomy. If misinformation denotes a misfit between container and content, then qualifying misinformation of misleading is more appropriate than qualifying it of untrue. Perhaps the activists of our time fighting against misinformation should make sure they use the right language and clear out the ambiguity. “Misleading information” has a very different vibe than “false information”.
NB: No relativism here. Some wines definitely are better than others. Further, some palates certainly are more trained than mine. And some minds more suggestible than mine. The most important remains that not all wines are created equal. Similarly, not all information is created equal. Shitty wine exists. Wine can be spoiled but it also can be shitty from the start. Of course taste is very subjective. But there is no logical implication between subjectivity and the absence of consensual criteria of evaluation. If anything, my subjective preference for dry wine is a motivation to formulate a dryness criteria useful to all wine drinkers who want to communicate their wine preferences with others.
I’ve been writing posts about misinformation because I am sick of prehistoric dichotomies underlying common discourse on misinformation. The great true/false dichotomy is turning us into intellectual savages, incisively oscillating between two uncivilised extremes: (1) rudimentary relativism which grants expertise the same amount of speaking time it grants each kind of thoughtlessness wallowing anywhere on the informational landscape; (2) inquisitional thought police ready to trade free speech for more of their own ideas of what is right.
LEMMA NO 3: If information is a combination of both container and content, then misinformation can be misleading but not false.
Proof: If information is a combination of both content and container, the same goes for misinformation (Theorem 1). A combination in itself can’t be false. The combination can be relevant, efficient… or misleading. Not false.
If misinformation is a misfit between container and content, what does it mean to replace misinformation with information? It means refitting the content to the container or vice versa. It doesn’t mean that we necessarily want to change the meaning of the content. We want to replace the way it fits with its container. The meaning can stay the same. The refitting can preserve the falsehood or truth of the content.
Perhaps what we don’t want is something true to seem false and something false to seem true?
(Re)fitting is different from sorting content (Option 2). Fitting requires subjectivity to be addressed. The fit is to be appreciated by a recipient. The recipient is the one who decodes the container to access the content. They must have the decoding manual corresponding to the particular fit that they receive.
If misinformation is a misfit between container and content, we should be going about it very differently than we are. Fact-checking might be a solution to some problem. But it’s not a solution to the misinformation problem under this assumption. Fact-checking is like hunting down all glasses of shitty wine to toss their content, and possibly fill the glasses back up with good wine instead, regardless of parameters such as people’s palates and imagination.
We don’t necessarily need to settle on a definition of information right now. But we do need to be mindful that as long as we are not ready for formal functioning definitions, we are lying in a grey zone, with consequences. Our half-baked understanding of information can make or break our solutions to misinformation. Depending on how we look at information, it can look obvious that our solutions to misinformation are silly, evil or besides the point.
Having said all that perhaps this whole discussion is founded on invalid assumptions. Perhaps the silliness we want to eradicate is captured in one or several of the assumptions above. You tell me…
