In my post on why mass surveillance is not normal, I referenced how the Wikipedia page for the Nothing to hide argument labels the argument as a “logical fallacy.” On October 19th, user Gratecznik edited the Wikipedia page to remove the “logical fallacy” text. I am here to prove that the “Nothing to hide” argument is indeed a logical fallacy and go through some arguments against it.
The “Nothing to hide” argument is an intuitive but misleading argument, stating that if a person has done nothing unethical, unlawful, immoral, etc., then there is no reason to hide any of their actions or information. However, this argument has been well covered already and debunked many times (here is one example).
Besides the cost of what it takes for someone to never hide anything, there are many reasons why a person may not want to share information about themselves, even if no misconduct has taken place. The “Nothing to hide” argument intuitively (but not explicitly) assumes that those whom you share your information with will handle it with care and not falsely use it against you. Unfortunately, that is not how it currently works in the real world.
You don’t get to make the rules on what is and is not deemed unlawful. Something you do may be ethical or moral, but unlawful and could cost you if you aren’t able to hide those actions. For example, whistleblowers try to expose government misconduct. That is an ethical and moral goal, but it does not align with government interests. Therefor, if the whistleblower is not able to hide their actions, they will have reason to fear the government or other parties. The whistleblower has something to hide, even though it is not unethical or immoral.
You are likely not a whistleblower, so you have nothing to hide, right? As stated before, you don’t get to make the rules on what is and is not deemed unlawful. Anything you say or do could be used against you. Having a certain religion or viewpoint may be legal now, but if one day those become outlawed, you will have wished you hid it.
Just because you have nothing to hide doesn’t mean it is justified to share everything. Privacy is a basic human right (at least until someone edits Wikipedia to say otherwise), so you shouldn’t be forced to trust whoever just because you have nothing to hide.
For completeness, here is a proof that the “Nothing to hide” argument is a logical fallacy by using propositional calculus:
Let p be the proposition “I have nothing to hide”
Let q be the proposition “I should not be concerned about surveillance”
You can represent the “Nothing to hide” argument as follows:
p → q
I will be providing a proof by counterexample. Suppose p is true, but q is false (i.e. “I have nothing to hide” and “I am concerned about surveillance”):
p ∧ ¬q
Someone may have nothing to hide, but still be concerned about the state of surveillance. Since that is a viable scenario, we can conclude that the “Nothing to hide” argument is invalid (a logical fallacy).
I know someone is going to try to rip that proof apart. If anyone is an editor on Wikipedia, please revert the edit that removed the “logical fallacy” text, as it provides a very easy and direct way for people to cite that the “Nothing to hide” argument is false.
Thanks for reading!
- The 8232 Project
I do agree with you point and opinion, but that “logical proof” is one of the worst I’ve read.
The “Nothing to Hide” argument could be restated that way:
Axioms:
A1: Surveillance reveals hidden thingsA2: If I have something to hide, I would be concerned if it’s revealedPropositions
p: I have something to hideq: I should be concerned about surveillanceWe deduce from the axioms that
p => q: “if I have something to hide I must be concerned about surveillance”.The logical fallacy of the nothing to hide is to deduce
!p => !q: “If I have nothing to hide I should not fear surveillance”. Which is a case of Denying the antecedent fallacy.Another fallacy of the argument is that they suppose
!pis true, which is a debunked fact.What was wrong with your proof was that you used another human to disprove a fact about the first one. The I may not be switchable because the other human may not have the same axioms. Moreover, you statement was about “should” but if someone doesn’t do something they only should do, it’s not a contradiction
but that “logical proof” is one of the worst I’ve read.
😅 I’m not very experienced in proofs like these yet, as you can tell. Thank you for submitting your own proof, I greatly appreciate it!
You’re welcome :) to be honest it’s my first for this as well 😂, but I do have experience with math.
The one thing that ticked me with your proof, was about your phrasing. You were trying to prove
!(p=>q)i.e.p^!qby a counter example, but your wrote “suppose we havep^!q”, which is already the thesis of the proof. So what you wrote is essentially “We will proof A is false. Suppose !A, then !A.” which is not proving !A. What you should have done is to remove the “suppose” part and say ifp=>qthen if I nothing to hide I should not be concerned, but I can have nothing to hide and be concerned, which is a contradiction. Then your proof would be somewhat correct but my last two arguments still hold. The issue could be solved woth some modals or quantifiers to express the different people.
No need to make things complicated. I’m from Brazil and I live everyday the consequences of having our data exposed out there by mismanagement and abuse:
There are criminals out there using our data in several ways, like extorting money, taking banking loans in your name, using your credit card online, etc etc etc. We can’t rest in peace
Nothing to hide…my answer to that is: “Then why do you close your curtains at night if you have nothing to hide…?”
I’ve asked those people “why do you lock the toilet door?”
“Ok, so would you send me nudes? I’m not looking at them, I would just like to have them.”
Here’s a little experiment: next time you hear someone defends surveillance because they have nothing to hide ask them if you can have a look through their chat and browser history. Most likely they’ll reply “that’s private” and maybe after some time they will understand.
That’s of course by far not the only argument for privacy but it has a certain effect with most people
The issue that arises from this approach, as I’ve found, is that people have something to hide from you, but not the government/large corporations. When they feel as if they are in a pool, they feel less important compared to being singled out by you.
You could instead do something similar: “Why does the FBI need to know what color of underwear you wear?” etc. to help them realize that surveillance goes much deeper than they realize, and not everything is relevant information.
yep. I use that argument sometimes , but it really depends on the person whether the “then give me your email/chat history/etc.” argumeng will work.
and just like you said, people don’t you reading it. They wouldn’t want to see you looking through their phone. But in the context of technology, it’s very abstract. Like, when Instagram’s chats weren’t encrypted, telling someone “Instagram can read them” may sound vague. They don’t imagine how like, Meta employees maybe could have access to it, or the softwares behind it that could analyze their chats. that could all be happening, but “out of sight, out of mind!” really helps people tolerate those possibilities. Thhe frontend, the chat interface, looks OK, so yeah.
I don’t think I could explain it very well, so my bad (not a native speaker), but yeah, I feel like because we are not encoruaged to think about how the software works behind the curtains, it’s easy to assume that “well, I’m not a target, so why should I worry about doing this [privacy thing]?”
A logical fallacy is a specific technical term pointing to an issue with the logic itself. What you are accurately pointing out is the missing information and hidden premises. In other words, you are finding issue with the facts, not the logic.
Some may have nothing to hide, but still be concerned about the state of surveillance
This is where your proof falls apart. It follows from nothing you’ve established and relies on context outside of our proof, which does not work with propositional logic. Another commenter goes into a bit more detail with some pre-defined axioms; with the right axioms you can wave away anything. However you have to agree on your axioms to begin with (this is the foundation of things like non-Euclidean geometry; choose to accept normally unacceptable axioms).
A rigorous proof using propositional calculus would have to start with the definitions of what things are, what hiding means, what surveillance is, how it relates to hiding, and slowly work your way to showing, based on the definitions and lemmas you’ve built along the way, how this actually works. Understanding how to build arithmetic from the Peano Axioms is a good foundation.
However, by attempting to represent this conversation in formal logic, we fall prey to Gödel’s Incompleteness Theorems, which means something beyond the axioms in our system has to be based on faith. This arguably leads us back to the beginning, where “nothing to hide” and “state surveillance” fall under personal preference.
Please note that I think “nothing to hide” is bullshit always and do not support heavy surveillance. I like the discussion you’ve started.
Sure you can always infinitely define what is behind but I don’t think it is relevant here or you couldn’t do any moral logic.
The two axioms I assumed are A1 a proven fact and A2 the very defintion of having something to hide. It is enough for this specific problem.
I don’t see how Gödel’s theorems are useful since they say that a given system of actions is either incomplete or inconsistent. With these two axioms it’s hardly inconsistent and we don’t care about it being incomplete since we only have one theorem to prove
This is some BS. What the OP haplessly tries to say is simply modus ponens. What Gödel are you talking about.
I’m not sure how you prove by negation in this case just via modus ponens. Care to enlighten me? I opened with something that doesn’t follow so that would be a great place to start.
Give me a consistent formal system with a list of theorems to prove OP’s conjecture and I’ll show you how we have gaps in the system. My analytic philosophy is pretty rusty; I think there are a few 20th century folks you can start from for this.
In modus ponens you have four cases:
A B A -> B a 0 0 TRUE b 0 1 TRUE c 1 0 FALSE d 1 1 TRUE Here, A is “Having sth to hide”, and B “Caring about encryption”. Obviously case b says that although people having something to hide seek out encrypted methods of communication, it is logically accepted that there might be other reasons, even unknown. A more silly example is this: the grass is wet does not necessarily means it has rained. There might be other reasons. But this does not mean that rain does not make the grass wet.
To sum up, the OP could have just said that. It does not change anything anyway. You can’t beat a propaganda apparatus with this “fallacy talk”.
You made the same leap that OP did.
[I]t is logically accepted that there might be other reasons, even unknown.
No, it’s not. That’s what I’m calling out. This doesn’t follow from A or B and requires further definition. While you’re using to explain case b, OP tried to use it to explain case c. In both cases, you are assuming some sort of framework that allows you to build these truth tables from real life. That’s where my ask for a consistent formal system comes from.
In your case b, we have not(I have something to hide) and (I am not concerned about surveillance). Since OP is not saying that the two are necessary and sufficient, we don’t really care. However, in your case c, where we have I have nothing to hide and not(I am not concerned about surveillance), both of you say we are logically allowed to force that to make sense. It’s now an axiom that A and not B cannot be; it has not come from within our proof or our formal system. We waved our hands and said there’s no way for that to happen. Remember, we started with the assumption we could prove A -> B by negation, not that A -> B was guaranteed.
If you’ll notice my last paragraph in my first post basically says the same thing your last paragraph says.
It’s now an axiom that A and not B cannot be
How so?
Remember, we started with the assumption we could prove A -> B by negation, not that A -> B was guaranteed.
It is rather that the fact that people who do have something to hide will probably use encryption cannot be refuted by an instance of someone using encryption without having something to hide.
We waved our hands and said there’s no way for that to happen.
This is textbook modus ponens, sorry if you find that disturbing.
you are assuming some sort of framework that allows you to build these truth tables from real life
This is unproductive and eventually relativistic. I can’t fathom how you dare bring advanced topics of math/logic fundamentals in a discussion like this. We are talking the kind of stuff that takes 200 pages to prove 1 + 1 = 2, and why it is not correct, or absolute. What is the purpose of that level of meta in a discussion about flipping privacy?
How so?
OP said that, given A and B, they would prove A -> B via negation, meaning the truth table you built does not yet exist and must be proved.
It is rather…
OP is not trying to use language, OP is trying to use propositional calculus. Using language unattached to propositional calculus is meaningless in this context.
This is textbook modus ponens
No, it’s not. Textbook modus ponens is when you are given A -> B. We are given A and B and are trying to prove A -> B. Never in any of my reading have I ever seen someone say “We want to prove A -> B ergo given A and B, A -> B.” I mean, had I graded symbolic logic papers, I probably would have because it’s a textbook mistake to write a proof that just has the conclusion with none of the work. As the in group, we may assume A -> B in this situation; OP was taking some new tools they’ve picked up and applying them to something OP appears passionate about to prove our assumptions.
how dare you
I was responding to OP. Why are you getting mad at me instead of getting mad at OP? OP brought propositional logic to a relativistic conversation. My goal was show why that’s a bad idea. You have proven my point incredibly well.
We want to prove A -> B ergo given A and B, A -> B.
Still failing to see that we aren’t proving A -> B, but getting its truth value within a proof.
OP brought propositional logic to a relativistic conversation. My goal was show why that’s a bad idea.
I think your goal was the equivalent of what any postmodernist does in deconstructing any given field:
- “Nothing is real”
- “you can’t prove the first axioms within the system”
- “it is all in the historical context”
- “No truth statements are possible”
By the same coin, all the other logical fallacies go out of the window, together with boolean logic and what have you. Even the valid ones.
people are missing the mark; the “not hiding” is an imposed narrative, a straw man argument and a false dichotomy, all in one.
the issue is not whether you have or don’t something to hide, as this “hide” part implies something inherently sinister. the issue is you being forced to share stuff you haven’t decided to do so.
when I’m not sharing the quality of my morning’s stool across all my social media outlets, it’s not something I’m hiding, it’s something I haven’t decided on sharing with the public. consequently, I don’t allow my software, hardware, service provider, government, or whoever-the-fuck to do it for me.
so what this false equivalency is doing is moving the onus from the evildoer to me, forcing me to explain why I don’t like what this fucker is doing. fuck him and the horse he rode in on.
Pregnancy, abortion seeking, sexual orientation of clergy, being trans, all have become matter of life and death level reasons for caring about privacy.
Ah, another one: insurance company might profile you as XYZ subcategory and discriminate against you.
“Yes but you know what data brokers are hiding from you?” I haven’t tried this one, but I will.
I usually respond to the phrase “I’ve nothing to hide” with the request to then provide me with the checking account number, the medical history, to let me read the email and chats, to tell me when you are going to be on vacation and the home. empty… these are exactly the things that these pages collect to sell. We don’t allow it in real life either, let alone online.
Surveillance advertising is a crime, period.
Nothing to hide? Okay, send pic of your bussy.
The “Nothing to hide” argument isn’t really an argument, it’s more of a conclusion. That conclusion is then taken to support mass surveillance. It’s also not a logical fallacy (even if it’s wrong). It may be “proven” using logical fallacies, but that doesn’t make it a logical fallacy on its own. So I think it’s correct to remove the logical fallacy text.
I think the more effective defense against this one is to provide counterexamples for why you might care about mass surveillance:
- People do have something to hide. E.g. browser history, religious/political beliefs, etc…
- You may not have something to hide now, but in the future you may wish it was still hidden. You can’t unpublish information these days.
- People you care about may have something to hide, and not caring about mass surveillance puts them at risk.
- Relatively harmless individual datapoints can be combined to create harmful datasets that allow for mass exploitation.
- Governments may abuse mass surveillance, whereby you may experience negative effects from journalists/political dissidents being silenced
- Etc…
Nobody’s business that I’ve got nothing to hide.
Something like a joke about this:
If you’re working on something genius great, no matter who’s watching you, they won’t understand anything, but if you’re doing ordinary things, it’s really not that important that someone is watching.
But of course, annoying comprehensive surveillance, especially from commercial companies, certainly should not be justified.
Nothing to hide if no one is willing to exploit
I would add to the conversation with the questions;
Should all information be known? Just because something doesn’t need to be hidden doesn’t imply that it should be known broadly. It’s not okay for somebody to know what color underwear I’m wearing right now.
Is all information equal in value? Presuming one kind of data point is okay to be public does not mean that all data points are okay to be public. My address is public record (unfortunately) but that doesn’t mean my social security number, ID number, and passport number should be public as well.
I would add to the conversation with the questions;
I love fostering discussions, and I’m glad to see you do too!
Should all information be known?
Obviously nothing good can come if we all turn into the Borg. However, the question becomes more interesting when you consider different people. Should all of your information be known? No, obviously not. Should all of the CIA’s information (besides the personal information of others) be known?
Is all information equal in value?
My previous counter question assumes that no, not all information is of equal value. However, even if the value of information differs, your ability to control if that information is shared should not be diminished.
I had a boss that would remember everything you told to them, and would make incomplete assumptions about it. If you told them you liked dogs, they would assume dogs are your favorite animal. If you told them you got scratched by a cat, they would assume you absolutely hate cats. In this context, “mundane” information such as my personal preferences (favorite animal, etc.) is not something I would want to share with my boss, since nothing good comes from it. Even though the information has “less value,” the value of it was raised depending on who I told.
Your social security number is high value to say, your neighbor, but not necessarily the DMV. CIA documents may be high value against other countries, but it might be worth making it available to national citizens. So, information itself does not have a set value above any other piece of information, but it does have differing value depending on who you share it with.
I know nothing about Wikipedia drama, but the term “logical fallacy” usually indicates a reasoning pattern and does not apply to specific arguments, regardless of how fallacious they may happen be. See https://en.wikipedia.org/wiki/List_of_fallacies
