Logical Fallacies
logicalfallacies.org
logicalfallacies.org
For example, using "appeal to authority", it is logically true that just because a group of Fields Medalists think your proof is wrong, does not necessarily mean that your proof is wrong. But in practice, there is a very high chance you are in the wrong.
Logical fallacies are binary, but the world is more stochastic, so I may be wrong, but I think these logical fallacies do not actually apply to reasoning about the real world. They only work when applied to thought experiments.
But I am no philosopher so I do not know for sure.
Whether, when you're making a business decision, you go with the lone whacko or the Fields medalists is a different question.
They're meant to do exactly what you are asking for here: eliminate a binary rule from deduction.
Regarding the appeal to authority fallacy you mentioned. I personally do not believe this is a fallacy. I do agree that the appeal to false authority is a fallacy. We all have to rely on the expertise of others for much of what we believe to be true. Of course said beliefs might be wrong but relying on expert knowledge is not a fallacy in and of itself in my opinion.
That doesn't sound right. An argument is sound iff it is logically valid and its premises are true, right? From Wikipedia:
> In deductive reasoning, a sound argument is an argument that is both valid, and all of whose premises are true (and as a consequence its conclusion is true as well)
An argument can correctly use the rules of logic and have an incorrect conclusion. This can happen if some of the premises are false. For example:
All men are immortal. Socrates is a man. Therefore Socrates is immortal.
The rules of logic have been correctly followed but the conclusion is false.
I think you're just mixing up the terminology a bit.
They aren't the same. These are standard terms in logic, as explained in the Wikipedia article.
Besides logically sound and sound argument aren't the same thing in my mind. One refers to soundness of logic and the other to soundness of the argument. And that was the essence of my point. There's a difference in talking about an argument and talking about the logic used. Logical fallacies are only useful in determining if the reasoning was flawed.
I didn't disagree with anything you wrote and even clarified my meaning so as to be clear. There is also the fact that, linguistically speaking, every person has a unique language. The meaning of words can vary slightly between people who nominally speak the same language. That's why I mentioned to me in my previous post.
Clearly I was not writing in a mathematically precise way. I took mathematical logic in graduate school many years ago but am not qualified to use all of the terminology in a mathematically precise way. But this is an internet forum and not a course in mathematical logic. The standards of rigor are not so strict. Hence my phrase to me ought to have sufficed as far as clarification.
Private languages are useless for communication, which depends on shared semantics. This thread is a clear example of that, as your first post here is very misleading except to anyone who a) knows that you are using your own private language, and b) actually knows what you mean by your words. In that post, using your own private definition of the various terms under discussion is worse than useless, as anyone who does not know what the accepted meanings are will end up worse off for reading it.
“When I use a word,’ Humpty Dumpty said in rather a scornful tone, ‘it means just what I choose it to mean — neither more nor less.’
’The question is,’ said Alice, ‘whether you can make words mean so many different things.’
’The question is,’ said Humpty Dumpty, ‘which is to be master — that’s all.”
― Lewis Carroll, Through the Looking Glass
What is interesting is that the responses to me have been about the meaning of sound argument and I never used this phrase. I used logically sound. Do you personally think logically sound and sound argument mean the same thing? I'm asking out of curiosity.
EDIT: Private languages and the acknowledgement of them is not useless. It's a basis for being able to clarify things without a discussion devolving arguments over definitions of words. We all have slightly different ideas of what certain words mean.
Well, the real point is that it is not clear what you mean here, and consequently what you are writing is likely to cause confusion and misunderstanding. In particular, as I pointed out in an earlier post, you were inconsistent in your own usage, sometimes using 'sound' correctly, and sometimes incorrectly.
You also seem to think that because, in a series of follow-up posts, you can explain what you meant, then it follows that there was nothing wrong with your original post - but if no-one had replied to it, you would not have corrected it on your own, and it would have remained there, sowing confusion among those who did not know it was wrong (and, of course, being useless to those who do know it is wrong.)
The size of the responses to your original post is a clear testament to the way private usage inhibits understanding and spreads confusion (though there would be considerably less if you had simply acknowledged your mistake immediately, and henceforward used 'sound' in its conventional sense.)
> Private languages and the acknowledgement of them is not useless.
I said that they are useless for communication (of course, they can be useful to the extent that they are not entirely private, but that caveat actually bolsters my point.)
>It's a basis for being able to clarify things...
Quite the opposite, as shown here: it's most often a basis for confusion and misunderstanding - and how could it not be? You have to achieve a shared meaning before there can be clarity and mutual understanding.
One nice thing about doing so is that an understanding of validity vs soundness, necessity vs sufficiency, and the truth tables themselves is all you need to understand all the formal fallacies.
All you need to understand about the informal fallacies is, if someone appears to need to discuss informal fallacies, what they really need is a hug. Or a kick in the pants.
Validity means the form of an argument is such that it's impossible for its premises to be true while the conclusion is false i.e. if the premises are true then the conclulsion must be true. Validity is a formal property of arguments and the relationships between premises. An argument can be valid even if its premises are false.
Soundness means that an argument is valid and that the premises are true (and therefore the conclusion must be true).
These are standard terms that have been used for over a century. You're going to generate a lot of confusion if you choose to use them differently to everyone who has ever studied elementary logic.
sound argument
logically sound
I'm not aware of these phrases being the same. I took mathematical logic in graduate school many years ago and I've used logically sound to be something different than sound argument.
I'm not disagreeing that valid would have been a better word choice. I did clarify my meaning in a subsequent post. I rewrote my point without using either sound or valid in an effort to be more precise.
Your first post, however, uses 'sound' in a manner that is inconsistent with both this example and correct usage. In particular, where you wrote "An argument can be logically sound but with a false conclusion" you presumably meant 'valid' instead of 'logically sound', as, by definition, sound arguments have true conclusions.
Do you see that you have taken my phrase logically sound and then switched to sound arguments? Do you think these mean the same thing?
I have not "taken [your] phrase logically sound and then switched to sound arguments". What I did, in the clause "sound arguments have true conclusions", was to make a statement of fact about the meaning of the word 'sound' in the domain of logic. On account of that being so, it would have been unequivocally wrong for you to have written "An argument can be sound but with a false conclusion."
Ah, but you say (this is where it gets really pedantic) you did not write precisely that; you wrote "An argument can be logically sound but with a false conclusion" - but how do you think the modifier logically will be taken here, given that 'sound' has a specific meaning in the domain of logic? Don't tell us that (obviously!) 'logically sound' actually means 'valid'!
So for example I defer to Roger Penrose on anything to do with mathematics, and will absolutely he has earned a voice on astrophysics#, but quantum consciousness microtubules? Nobody is an authority on that, so who knows?
# And to be clear part of his credibility there is in making it clear what he believes with a high degree of certainty and what is more speculative.
The more nuanced versions in the examples make an uncritical assertion, based on an appeal to an authority in a closely related domain, to overturn an established set of “validated facts” (for the domain, say what you want about a religious creed).
Such conversations are almost always about changing someone's view of the world - sometimes through debate, where conflicting positions are taken and argued upon, and sometimes simply because someone (e.g. you) asks for a clarification in order to know which position to take.
So I think that the idea is that people can at least take well-reasoned positions, not necessarily equal positions, in a world that is not black and white.
The list of fallacies can then be a useful guide to judge the persuasiveness of the arguments of the different sides, including of course your own arguments, and to help you see whether a position might need to change because it is not well-reasoned.
In your example, I can say "I trust Fields Medallists" and therefore judge the proof to be likely wrong. That is a valid position to take. You may say that "the author of the proof is a very talented mathematician" so the proof is likely right, which would be equally valid. What we cannot do, without further information, is to claim that the proof must with absolute certainty be right or wrong - that I believe would be the fallacy here.
Generally speaking, I like to use fallacies for myself, just to be aware about certain pitfalls. Not so much about discrediting others, but rather not fall into one myself.
If I realize other people are using fallacies to support their argument, it helps me to build an argument against it. Just don't use it naively: "you're wrong because of fallacy number 42". Using it like this is a bad habit, because of the wast number of fallacies, you can probably kill any discussion.
What interesting is that in the example of Bitcoin, since currency is little more than a shared belief in something, this would not really constitute a logical fallacy.
Regarding your second: you’re making essentially the same mistake. There’s no objective argument here, no deliberations on abstract truth that could be easily countered by pointing out a logical fallacy. Just because it’s a shared belief doesn’t mean it’s not beholden to valid yet purely subjective arguments such as “I don’t have any BTC” or “I don’t have the technical ability to use BTC.” In reality, nobody is going to use your product or service because you can make a fallacy-free argument for it. On the other hand, popularity is a very real and useful indicator of utility. You may hate jQuery, for instance, but it’s going to have a plethora of documentation, blog posts, and StackOverflow answers.
Basically - the real world is tolerant and even encouraging of logical fallacies.
I think it also depends on precisely what statement you’re trying to prove. In the jQuery example you have to get really specific, because just talking about the best can be pretty vague. There are many examples of best where merely pointing to the popularity of jQuery is a logical fallacy, and perhaps some where it isn’t (e.g. the breadth of documentation making development easier). This is really a different class of problem to something more abstract such as the belief in alien civilisations, where the fallacy is more clear cut.
The currency example is a slightly different case though, because it’s the belief in the currency itself that creates the currency. So the statement everyone should replace their USD with Bitcoin is undermined by the overwhelming depth and breadth of belief in USD (misplaced or not). So I don’t think it counts as a logical fallacy, because it goes beyond merely trying to leverage the majority opinion on the subject; the beliefs themselves are absolutely instrumental in driving the relative utility of USD and Bitcoin.
"orange_tee made an appeal to authority, therefore every conclusion he is attempting to draw is false and can be rejected out of hand."
It's a very jejune debate tactic, by people who have just dipped their toes into logic for the first time, and have discovered a brand new way to shut down arguments.
"The ad hominem fallacy fallacy"
EDIT: debunking -> clarification https://laurencetennant.com/bonds/adhominem.html
"You are wrong because you are an idiot" is ad hominem. "You are wrong, and also an idiot" is not.
It’s the abuse of crying “ad hominem” that’s the problem.
Not every conversation is a scientific debate. If you want to learn, you should still participate in them.
Logical fallacies are easy to misunderstand, which makes them very dangerous. That's why this website has examples, I guess. This is a case of knowledge that might require a teacher who can give feedback and guide understanding.
The probability that an assertion is true drastically increases if there is a clear consensus among the subject matter experts in that field.
This is important when certain assertions cannot be personally verified.
For example, I cannot personally verify that Pluto has a moon. ...but I can show people pictures of an object that the experts claim is Pluto's moon, and point to various statements that Pluto has a moon.
The probability that Pluto does not have a moon would require a massive conspiracy, or some unimaginable misunderstanding, that the probability is extremely low.
Is it still a "fallacy" to point to the statements of those scientists?
Now people refer to him expert in a facebook thread.
That's the "appeal to authority" fallacy, which you could call out.
Extensive use of fallacies is in the active repertoire of demagogues. That's why it helps to study them, a bit, so you recognize a skunk if you meet it in the wild.
It certainly doesn't mean "listen to the experts". There is no such intellectual guidance there. The perspective of "logical fallacy" only cares about whether something must be true, and it has no opinion on how to judge whether something is probably true, or how to develop trust.
Formal methods became fashionable only late in mathematical history, yet mathematicians somehow achieved productivity up to that point.
The proof is wrong not because the status of those Field Medalists but instead because of the content of their argument. That's all that the appeal to authority is saying.
I'd agree that it's a mistake to relate the existence of a logical fallacies to the probability of a statement being correct. Instead, it only gives information about whether the reasoning of the statement is on solid ground.
Do mathematicians even use logic as a reasoning tool, as opposed to a proving tool?
I don't think you can just throw away the word "logic" like that. We try to make arguments using deduction and inference, so there is a logical structure behind them. It's just that it isn't as well-defined and doesn't have the mathematical rigor of formal logic.
Mathematicians can scarcely describe their own thinking process, and proving is often the last step to mathematical reasoning.
Talking about logical fallacy in ordinary conversation is backward.
ad hominem - someone you know to receive large sums of money from cigarette manufacturers gives you several plausible sounding reasons why cigarettes don't cause cancer. It's an adhominem to doubt his arguments based on your knowledge of how he's getting paid, but it's also sensible.
argumentum ad populum - an argument isn't necessarily true just because lots of people believe it, but in the absence of you being able to work out the absolute truth from first principles, it's completely sensible to take into account what other people believe when you evaluate your options, or what people in your situation have typically believed (appeal to tradition)
denying the antecedent - red sky in the morning implies shepherds should take warning. It's not true to say that just because there isn't red sky in the morning, then shepherds should assume the weather will be fine, but excluding one of the conditions in which shepherds need to be careful is weak evidence that shepherds will be fine.
The old 'absence of evidence is not evidence of absence' one, although I didn't see it on this page. Actually it is. It is weak and incomplete evidence of absence, but in the right circumstances it can be perfectly reasonable to change your opinion on whether something is present or not based on a lack of evidence for it being present.
In the real world, weak evidence is often the best we're going to get and we're often going to be in situations where thought shortcuts are necessary and valuable.
Pointing out fallacies is usually poor argumentation. My pet peeve on this one is the 'burden of proof'. Normal human debate isn't a game with strict rules and a judge that awards points, and I'm not sure it'd be any better if it were. You engage in a debate with someone when you both want to change the other persons mind. If you didn't want to change their mind you shouldn't be debating with them. You're very unlikely to change their minds by telling them that the burden of proof is on them not you (much more likely to end up arguing about who really has the burden of proof, a fruitless and uninteresting topic).
It looks like that it intermixed a few things that are not logical fallacies in order to appear to have more content to sell books that are more than a few pages thick.
Example:. https://www.logicalfallacies.org/shotgun-argumentation.html
<<A student argues that he didn't do his homework because, he had lost his backpack, and when he found it the notebook was not in there, and it turned out that the dog had eaten his notebook.>>
It is clearly not a logical fallacies case:
either it is a lie, or if it is the truth, it is a logical explanation of the chain of unfortunate events that resulted in the student not to able to deliver.
I’d go further and say that the linked page doesn’t accurately describe the fallacy. The first part of the fallacy is offering so many arguments (or even just arguments swamped in endless text) your opponent can’t or won’t respond to everything. (There’s also generally an implication that these arguments are known by the “spewer” to be tenuous or outright fallacious.) The second part is claiming that an opponent’s lack of response to any or all of the arguments means the opponent is wrong.
To me, the first part is just bad faith, but it’s not necessarily illogical. It’s the second part that makes it a fallacy.
Are you sure we're reading the same page? Most things described as "logical fallacies" are informal anyway, and have nothing to do with "logic": they're not expressed in the form of a formal Quinean logical syllogism that maps out truth tables with elemental logical operators like AND, XAND, IFF etc.
The only work the word "logic" is doing here is not as a formal term but as a vague complement that approximately means something like "rational." And invoking that term, or witholding it is just a personal expression of whether you like something enough to award it a merit badge that has the word "logical" on it. It's just playing calvinball.
>The second part is claiming that an opponent’s lack of response to any or all of the arguments means the opponent is wrong.
This is just incoherent to me. Second part? The page isn't grouped into parts. I'm assuming you mean the second half of the sentence explaining the shotgun fallacy. But the second half of the sentence isn't expressing a second, isolated idea. The whole thing together describes a rhetorical strategy for expressing an argument that attempts to escape refutation. It's as real as any other informal fallacy. If it falls short of qualifying for your personal complement that it's "logical", well, either you're referring to formal logic or you're not, and if you're not, then this whole exercise in deeming it "logical" is just a game of calvinball that doesn't have anything to do with anything.
I don't know how responding to your own claims could possibly put me on the wrong side of a distinction between the fallacy itself and the fallacy the page is describing unless you were also on the wrong side of it, too.
I suppose you're saying my response addresses itself to the idea of the fallacy and not to your criticisms of the way it's illustrated on the page - but I find that to be quite obviously wrong, because I am talking about what's on the page. Perhaps there's an argument than I'm not seeing, but you're being too cryptic and difficult to interpret for me to figure out what that argument is supposed to be.
Aka the Gish gallop:
https://en.wikipedia.org/wiki/Gish_gallop
(And, yes, agreed, it is not a logical fallacy, but a (duplicitous) debate tactic.)
You can always fall back to Wikipedia's extensive list of fallacies: https://en.wikipedia.org/wiki/List_of_fallacies
As with anything, the first step is awareness of the flaws, followed by working to correct those to whatever extent one can.
SCNR
[1] https://en.wikipedia.org/wiki/Faulty_generalization#Hasty_ge...
An argument can be profoundly fallacious or not. A solid argument, pretty much by definition, is solid because there aren't obvious fallacies in it.
> dismissing every fallacious argument will allow you to "win" every argument simply by pointing out all the ways in which your opponent is wrong
I should also add that one of the most successful modern fallacies is the straw man, and it crushes meaningful political discourse. How can pointing out a straw man "permit everyone to avoid thinking"? Fallacy awareness actually helps people avoid getting caught in traps on both sides of an argument. I would say this knowledge adds to discourse when used appropriately.
We could say this about very nearly everything that exists.
Because pointing out fallacies is not argument, but feels a lot like it is. Pointing out fallacies allows you to defeat any other argument, but it does not allow you to take a position for anything. It is the perfect tool for allowing you to entrench yourself in your original view, by handily defeating those who would assail you in a fully general way that applies however correct you are. Merely saying "but to do so is to commit the fallacy fallacy" doesn't change the fact that it happens extremely commonly.
It's the "when used appropriately" that's the issue. A list of fallacies is probably a good thing to have access to after you've somehow had epistemic humility installed in you. But epistemic humility does not come naturally to most people, and must be taught.
Once an analogy has been made, the conversation always turns to contorting it to matching the viewpoints by both sides. It's always distracting.
If you're reaching for an analogy, spend a bit more time to try to make your point without it.
Except in the case where your analogy is effectively an argument to generalisation, attacked via a specialised example (but arguing only over the generalisable properties).
Any other type of analogy immediately falls flat as soon as someone calls out specialised property x that does not hold for oroginal argument y.
If humans were infinitely smart, analogies wouldn't be needed. But since we aren't, we often understand things better by seeing a comparison to something that is more visceral, or that we already know.
If I want to discuss the merits of different woods in cavaquinho construction with someone that's spent all of their life in an office, then it is almost always easier to work in an analogistic manner that spend hours upon hundreds of hours building up their knowledge.
If it's a _debate_ with two parties trying to convince the other of a point, then analogies begin to fall short.
You did say _conversation_ though. Not all conversations are debates.
Arming yourself with rhetorical weapons like this, especially if you rather view them as tools for thinking, will get you some way but I think rarely aid in reaching consensus or understanding where disagreement comes from.
It goes through a list of "dishonest" debate techniques (not necessarily relying on fallacies) and possible techniques to disarm them.
The definition offered is effectively a rewording of the formal fallacy.
In fact the very thing that makes an informal fallacy INformal is the fact that from a formal point of view it is not fallacious (i.e. the argument is technically valid)
Informality comes from the fact that while technically valid, an informally fallacious argument does not further the conversation, either because there is no universe where the premises could ever be true (i.e. a valid but unsound argument), or because the validity is the result of a "vacuous truth", or because it is structured in such a way that the conclusion, while not invalid, is designed to derail rather than further the conversation in a useful way (i.e. trolling)
I would go so far as to say it's not a legitimate fallacy. People can both overestimate or underestimate the probability that A will lead to B, but we call the former fallacious and not the latter, which makes no sense. If we think that the former should be validly called the slippery slope fallacy then we need a fallacy for the latter too (inertia fallacy? fallacy of lack of imagination?)
Kidding aside, the point of the Slippery Slope argument isn’t that a chain of events can’t happen or won’t happen - it’s that we can’t assume they will happen.
Also I believe my other point stands with respect to how this fallacy is abused in common discourse to score debate points, at least according to my experience. In practice, not many commit this fallacy, since everyone except the most dishonest will admit there's a tiny probability that their causal chain won't play out.
This is generally the domain of logic as a formal tool for argumentative analysis. In that context, one is trying to nail down whether an argument is valid or not. It’s important for us to make a distinction between what is logically sound, and what is a hypothetical that can’t be proven.
It comes down to feels not reals. Logic is reals. It is a formal system. It helps us ground our thinking. A slope may be indeed slippery, if it can be proven that A leads to Z every time. But usually a Slippery Slope is utilized as a way of scaring people into the assumption that A will lead to Z, even when there’s plenty of off ramps along the way.
It’s about certainty vs speculation.
To say that, “hey if we start doing A we might end up at Z” is not wrong because we say “might”, and it’s worthy of debating the risks and possibilities. But often people present it as a Truth. And it’s fair to point out that it’s neither a certainty, nor logically sound.
A statement based on a logical fallacy can turn out to be true. A prediction based on nothing at all can turn out to be true as well. All a logical fallacy means is that the statement will not NECESSARILY turn out to be true.
- They help people to understand why a superficially appealing argument might not be a good argument. It's a useful innoculation against bullshit. Someone who has internalized informal fallacies can use them to spot bad arguments in the media or from politicians, for example.
- They help people to avoid using falacious arguments in their own conversations. If a person wants to make a strong case for a position, at work for example, they're more likely to be taken seriously if they avoid bad reasoning.
The harmful motivation for someone to be interested in a list like this is that they're effective rhetorical strategies when used with the right audience. If you're interested in persuading people and you aren't very moral, fallacies can be your friend.
That's very much not true. For one thing, academic debate in high school has pretty much the same criteria as it does in a high school. But more importantly:
> where you need to actually engage with the substance of their argument
Fallacies are, without exceptions, weaknesses in the substance of an argument, or places where it fails to engage with the substance of the argument it is offered against. Identifying them is very useful if you are in a context (like, say, high school or collegiate debate, or, say, a court room) where that arguments engage with each other's substance is critical. Now, in most cases you will leverage the weakness revealed by having identified the fallacy rather than just pointing it out by name, but especially when the audience is familiar with them and the argument is in writing, naming them makes a convenient counterargument macro, as each fallacy is a compact name for a defect in the argument in serving it's purpose.
An example might be: "Listening to the scientists is just an appeal to authority"
What it refers to is the fact that an argument being fallacious is no guarantee that the conclusion doesn't hold in reality.
(What you're describing is probably closer to motte and bailey, straw man, and appeal to definition.)
This is given as the second example for circular reasoning, but I don't agree. I think it's a simple logical statement that doesn't prove its prerequisite.
I agree with many here that just pointing out fallacies does not advance the debate, however I believe this only applies to informal fallacies (which are often much less clear cut). To me informal fallacies are often just poor debating style.
However formal fallacies should absolutely always be challenged, because the make the argument formally invalid (non sequitur).
That being said, where's the favorite of the moment, "guilt by association?"
X: We shouldn't listen to health experts! They got it wrong at the beginning of the pandemic when they told us not to buy or wear masks, and therefore they are wrong now too!
Y: Wait, that doesn't make sense.
X: Strawman!
Y: Oh sorry sir/madam; you are correct. Please carry on.
Elsewhere, this line of reasoning is called the "fallacy fallacy". But the linked site describes the fallacy fallacy as "an argument that is based on false claims, but is logically coherent", which sounds more like a false premise to me.
Joking aside, whataboutism is a "logical fallacy" that I see commonly invoked here on HN, most often when the topic is political in nature.
It seems like many people want to assert their opponents questions are whataboutism so that they do not have to consider the broader context of the issue, which almost always would invalidate the narrow, one sided, siloed analysis which they are desperately clinging to.