An Illustrated Book of Bad Arguments (2013)
bookofbadarguments.com
bookofbadarguments.com
While it's important to be able to recognize a fallacious proposition, it's equally important to realize that the presence of an informal fallacy in an argument does not imply that the argument is either invalid or that the conclusion is false. To give two examples: ad hominems are frequently used in moral characterizations, and most moral analogies are functional strawmen by virtue of exposing the circumstances under which a particular claim is weakest.
At best, the presence of informal fallacies indicates that an argument needs more attention from its audience.
If you are able to give something a name and study it, it helps you to recognise it faster and be more confident dealing with it.
But I agree in some sense. Just yelling them out when you see them like it's a game of bingo is rarely productive.
Shorthands also have a couple of obvious pitfalls. One of them being in this case that neither party might recognize particular leakiness in a generally agreed-upon abstraction.
What you (IMHO) should be learning--and the vocabulary sort of comes implied with this, but is not necessary--is how to recognize the shape of such a concept or structure (in this case, fallacies) and how to "disarm" them, deal with them, in the context of an argument. Pointing out the fallacy by name should only be done maybe after disarming the fallacy in context of the argument, as further clarification.
If doing that is too long and convoluted, it really goes to show that learning a common vocabulary is training for the wrong thing. Instead, train to make your ways of dealing with fallacies shorter and to the point.
Say you know and studied all the moves in karate; During a match, your opponent comes at you with a certain attack, and you know the perfect defence / counter attack to it because you studied it all. You call out the name of your opponent's attack and shout the name of your counter attack. Having agreed upon a common vocabulary, your opponent knows they are defeated by your counter. A shouting match ensues.
I'm not saying that having a common vocabulary is not useful, it's very useful. Imagine two karate trainers sharing tips on what techniques best to learn students of a certain level. Imagine how they are even able to talk about "students of a certain level". But that's all meta game. In a real fight the moves have no names, and in a real argument neither do the fallacies.
That's fine but it also leads to a shift of burden of proof on the counter argument rather than the initial claim. In turn, this leads to "proving a negative" and it is generally quite impossible to do so. Therefore, one needs to be careful with the line of arguments.
Personally, I have also found that the best way to counter a fallacy is to ask the claimant to substantiate his/her stance.
Great comic though
However, they might turn useful in understanding why somebody told you not to "put up a straw man", or used other apparently weird expressions. (I think I've never found an occurrence of the "No true scotsman" argument, but it sounds quite funny!) Though I doubt I would hunt for this specific book in that case, just googling around would be probably faster.
Thanks for sharing the comic; this looks like a fantastic weekly read.
I think there is a difference how those lists are presented and particularly which audience they are intended for.
I think the worst possible way is if a list is intended to be read by participants of a debate - and then the author (or one of the participants) magically expects everyone to obey to the list and complains if they don't.
That's pretty much like the bully who as been outwitted and demands the other to "fight like a man". For people not familiar with them, those lists are in the end only another set of arbitrary set of rules thought up by humans. (Yes, this may even apply to the whole principle of logical reasoning). Sticking to those rules may in fact give you anormous insight to learn facts about the world, but this is in no way obvious. In particular, if the one who tells you about those rules appears to be biased against your own views, you have even less incentive to learn about them.
The second-worst way is to memorize the fallacies, recognize them in an argument, call them out and by that expect them to be disputed. This might be useful if you're an undecided bystander and try to keep a clear head. However, if you're yourself part of the debate, this probably isn't going to convince your opponent or any bystanders: You're calling out the fallacies, but you're not explaining why they invalidate the particular argument.
So in the end all that's left in your actual counter-argument is "the list says that kind of argument is invalid" - which, ironically, is itself a pretty bad appeal to authority.
Fallacy Man has demonstrated that nicely.
I think the good kind of lists go a step further: They show you how to detect the fallacies and how to dispute them. This way you could actually use them like the patterns in the chess example and drive the discussion forward, even if neither your opponent nor your bystanders know about the fallacies.
We have a fallacy for it too, "Fallacy Fallacy".
I love this. I did a few extracurricular courses in philosophy in university (a rather broad "General Overview" and another course on Ethics in particular). Super interesting stuff. What I learned in Ethics is useful to me almost every day (more days than not, anyway).
One thing I took from these courses is that philosophy is not quite the scientific study of learning the Great Answers to The Great Questions. It's not even quite about learning the Great Answers proposed by historical Great Thinkers.
No, The Great Questions don't really have any single correct Great Answers to go with them.
Philosophy is about learning the counterarguments to any of the proposed Great Answers by Great Thinkers, recognizing their shapes in other proposed answers, cementing the Great Questions to stay questions (which might seem disingenuous, but is only when applied without rigour).
I like your post because it applies this idea on a meta-level to the field of philosophy concerned with fallacious arguments (Logic and Eristic dialectic, if I'm not mistaken)--you provide a counterargument to the dialectic strategy of pointing out a fallacy! Nice one :)
> ad hominems are frequently
> used in moral
> characterizations
In the above, we haven't delved into why skinning cats is bad (it seems to be bad self evidently and without need for justification, although that's another story), we're just characterized Bob as himself as bad. That's an ad hominem, but it's not fallacious - it only serves to instantiate the badness of an action in a particular person.
Popular/controversial/recognizable example:
Terrorists are no true believers of <faith>.
Those in the 'you are committing a fallacy camp': The terrorists profess to follow the core tenets(perform the rituals, say the prayers, etc...) of <faith> and adhere to <faith>, therefore they are true believers of <faith>, therefore saying that they are not true believers of faith because of their terrorist actions is a fallacy.
Those of the 'not a fallacy camp': The in/out borders of religious groups are defined by consensus, it's normative. A large group of people may feel that <trait> is actually part of true belief in <faith>, then this is a truth for them. If a <trait> were to be abstinence from violence, then obviously terrorist are not true believers of <faith>.
Truly, language is too unspecific to properly pin down the meanings of normative/subjective facts. This fallacy can probably only truly be committed when one has codified a closed loop of rules, and then breaks those rules. But religion is too nebulous a subject to be clear about such things.
If you say "No X would do Y", and then after an X does Y move the goalposts to say "No true X would do Y", that's the fallacy.
If you simply stated in the first place that "No true X would do Y", then you're off the hook as long as your definition of "true X" doesn't apply to the particular X that did Y.
This is where all ability to have a rational discussion breaks down in many cases, especially when people care a lot about their initial opinions.
Take an issue like evolution. You start with a source like https://www.khanacademy.org/science/biology/her/evolution-an..., which outlines a pretty good case for how evolution is supported by today's evidence. It gives plenty of lines to follow if you're curious to dig deeper, and absent other factors most people would just see it as a boring science thing that's probably true and they don't care much about.
But because some people are bothered by the idea, they've spent a lot of time thinking up objections. A whole lot of time. See https://en.wikipedia.org/wiki/Objections_to_evolution for summaries of just some of the complaints that people have come up with. These have all been thoroughly demolished, of course, but to do so often involves a lot of really labor-intensive science.
As far as science goes, that's all good - it's great to be detailed and thoroughly justify every piece of a theory, chasing any loose ends down at any cost. It's part of the job description.
But casual debate about the issue is devastatingly difficult, even if everyone is acting in good faith. All someone that has a problem with evolution has to do is throw out three or four big complaints (the eye! carbon dating! transitional fossils! micro vs macro-evolution!) and all of a sudden a proper response to their 3 minute "gotcha" attempt means a good half hour sourcing high quality rebuttals and summarizing them in the thread. Which will likely each bring up three or four points which the anti-evolution person can quibble with (and they'll be able to do so quickly). Assuming you make it through that gauntlet and the person accepts that you've shot down those points, a list of ten more fresh "problems", each of which requires detailed research and attention to shoot down, is a low-effort Google search away (say https://www.evolutionnews.org/2012/07/what_are_the_to_1/), so you can get "yeah, but!"-ted into oblivion until you're just not up to the task of responding anymore, even though you absolutely could address every complaint without difficulty, if you had the time.
Like you said, moving the goalposts. When you have an almost limitless sea of low-effort objections to draw from via Google (which you do, about almost anything, as long as a lot of people have spent time coming up with them), it's easy to put up a shitscreen so thick that nobody can ever break through, and not even realize that you're acting badly.
I know that rationalists very strongly believe that with enough good-faith discussion everyone's beliefs should ultimately align, and there should never be such a thing as agreeing to disagree, but I fear that in many cases the relaxation time to that state is prohibitively long.
As I understand it, the no true scotsman boils down to "group A all do this. Person B belongs to group A, but does not do that. Only real members of A do this."
So the terrorist claim wouldn't be part of the fallacy, just a difference in interpretation. The fallacy occurs not in the thesis (group A all do this), but in the operation of arguing itself, when the reaction to the counter is to repeat the thesis but now excluding the counter example. You could imagine that every counterexample was so specially treated until the generalization is true, but meaningless.
The overall argument becoming trivially true: forall i in {A - B}, i has property C
Where A is the whole group
And B is a subset of A, having the property not-C
- The religion has an official leader (for example the Pope for Catholics). He (and the hierarchy below him) gets to decide who is a true Catholic and who is not, and what the rules are. The line is clear.
- The religion has no official leader (Islam, protestants...). In this case you can't claim that terrorists are "not true Muslims". Pretty much anyone can self-identify as a Muslim. You can however fight against generalisations claiming for example that Muslims are terrorists, or in favor of terrorist action "by default" unless they shout their opposition loud enough.
If you try to argue that this person does not represent the majority of feminists, you will be accused of No True Scotsman.
"This person is not a feminist" would be NTS for some definitions but not all.
For example, slippery slope is only argumentative failure if the slope is implied without justification. If you can provide good argument that slippery slope exists, it's not a bad argument.
Good idea to name appeal to authority as appeal to irrelevant authority.
Appeal to relevant authority is often important shorthand. Some issues are too complex to or require too much domain knowledge to go trough. For example, I consider IPCC reports authoritative on the subject of climate change.
Agreed. This is very very common unfortunately.
Generally, I classify any argument in a discussion to consist of two parts - about the central topic and off topic (much like ratio decidendi and ober dictum in case judgments). It is important to ignore "ober dictum" and concentrate on "ratio decidendi" much like in law. Fallacies should be pointed out in on-topic part of the discussion while the fallacies in the off-topic part may be used to point out a general pattern or behavior of the person at the other end of the table. It should be made explicitly clear that off-topic sentence fallacies have no bearing on the content of the on-topic sentences. The best strategy however, is to quote the on-topic part and ignore the off-topic parts.
I'd like to see more books about how people actually argue in life and how to change their view.
First, the point of a discussion isn't necessarily to change the mind of opposing parties. There is very often an audience, often including members receptive to a changed opinion. We need to discuss things rationally and reasonably for the sake of the broader community if nothing else. And, barring influencing third parties, there's something to be said for making proponents of bad arguments work for all the rhetorical ground they take.
Second, even in smaller discussions, I feel it's honorable to continually give people the chance to think better and make better decisions. It's respectful to them to do so, and it's deeply disrespectful to write someone off as unpersuadable or hopeless, not matter how accurate that sentiment is.
Third, it's good for us to give ourselves the chance to improve as well. If we fail to engage with different viewpoints, we only entrench in our current thought patterns. We owe it to ourselves to expose those thoughts to the light of day and see if they can be improved or if they need to be abandoned altogether.
It's not, really. A person who understands the notions of necessity and sufficiency, knows what a counterexample is, and knows their truth tables (with or without really knowing the textbook definitions of these things) should be able to function just fine without knowing the names of all the various fallacies.
(consider how many people who are inclined to complain about logical fallacies do not understand these things, and therefore misapply the labels)
Not really. They help for the same reason memorizing formulas in math help - they save the work and time needed to derive the formula again and again. That matters an awful lot in real time discussion. Also, if you know what fallacies exist it is much easier to recognize and encounter them compared to when you have to process the fallacy for the first time. You can even think them through in advance and prepare the argumets.
The rules surrounding syllogisms are so simple that I cannot believe anyone would use the word "derive" to describe the process of recognizing a problem with the logic with something like that (the syllogistic fallacies). I'm not sure most people even need to think consciously about the logic, let alone the taxonomy of possible problems with the logic.
On the other hand, informal fallacies, or inductive fallacies... you've got a point there in some cases, I suppose.
Arguments aren't presented mechanically step-by-step with a citation of rule at each step; they elide and summarize information, and skip logical steps. The fallacies show you where to look for insufficencies, to avoid being misled by invalid shorthand arguments.
(I don't remember thinking about the individual fallacies by name very much at all except when I was learning logic, where they were used as examples. Once you're a little more experienced, it would be akin to saying to oneself "okay, now I am using the present tense" while writing.)
> I'd like to see more books about how people actually argue in life and how to change their view.
I’m confused. You say normal people follow no rules in arguments, but then you say you’d like to learn more about how normal people argue. The two ideas seem mutually exclusive to me.
> I've always wondered about the usefulness of books that point out argument fallacies.
Presumably, the author would like people to approach arguments more methodically all around.
1. If the outcome of an argument matters, it matters in the sense that we are emotionally invested in the outcome. The word argument itself implies an inextricably emotional component.
2. Most apparently fallacious arguments boil down to abuses of analogy. But analogy is also, fundamentally, how induction works. Unless the premises of the argument are completely unambiguous (and they rarely are), "rationality" only gets us so far.
IMO, the most important part about learning these "fallacies" is learning when they aren't fallacies at all, but actually the best tools we have for grappling with a complex problem, and hopefully arriving at some unexpected clarification.
It has a lot on how people make ethical arguments in particular which I found quite eye opening (if a bit bleak in terms of its consequences for civic society).
Rule #0 for changing someone's view is that they have to actually be open to changing their view. It's surprisingly common that people will explicitly admit, if you ask them, that there is no realistic evidence or argument that could possibly change the side that they're on.
http://i1.kym-cdn.com/photos/images/original/000/531/718/f03...
Strange to see no mention of confirmation bias on the logical fallacies poster. That would have been the first on my list.
This is both logical (because it follows that in order to "think", one must "exist") and known.
There are things that are logical but not known, and others that are known but not logical. Gravity, for example, is pretty well known. Does it follow as a matter of logic? Not that I'm aware of. We believe it to be true because it works.
The basis for knowledge is a complicated subject. One framework is that to truly know something, you must have justified true belief. i.e. You must believe the statement, you must be justified in believing it (so a fortune that turns out to be right doesn't count as knowledge), and it must be true. Naturally, thousands of years of debate continue as to what the definition of "true" is.
Has anyone built a similar collection for "bad arguments" that are logically correct but completely unpersuasive? I'm often frustrated with these coming from people I agree with, because a bad argument for something is often worse than no argument at all.
Kind of like when they logically explain you that x should be better then y, but you tried both and still remember x being a hell. May sound logical, but you know something is wrong.
Perhaps the idea is that they are morally bad arguments. Still I would rather call it a collection of illogical arguments.
On the other hand, the practical value of logic in debates is subject to serious inquiry. If we readily accept that logic is the correct way to "win" an argument, then why are we so categorically bad at applying it?
“Never argue with stupid people, they will drag you down to their level and then beat you with experience.”
Otherwise - if they're stupid and they're right, they're not stupid.
For more information, please contact JasperCollins Publishers, 99 St Marks Pl New York, NY 94105.
Another common problem is that we use different associations or interpretations of a word.
E.g. I find discussing politics with Americans quite difficult because Americans have redefined the meaning of conservative, liberal, socialism and capitalism to a large degree.
Socialism e.g. seems to imply command economy and dictatorship, while capitalism seems to automatically imply freedom and democracy rather than merely describing an economic system.
This applies to many things e.g. when discussing programming languages, advocates of OOP programming tend to assume that encapsulation and polymorphism is uniquely OO concepts.
As I've been exploring various topics of logic, argument, epistemology, propaganda, "fake news", and more, over the past several years, I've turned up a few concepts I'm finding particularly useful.
Dialectics vs. rhetoric. There are numerous modes of communication. In exploring areas of disagreement or uncertainty, two principle modes are dialectic discussion, in which all parties are engaged in identifying the truth of a matter, regardless of their initial position, and rhetorical discussion, in which at least one party is engaged in promoting their initial position, regardless of the truth.
There's a heated debate on the merits of both methods, going back to Plato (he had some very unfriendly things to say of the Sophists, who practiced rhetoric).
https://en.m.wikipedia.org/wiki/Dialectic
https://en.m.wikipedia.org/wiki/Rhetoric
The useful takeaway for me is to recognise when I'm in a dialectic or rhetorical discussion. My own default is to engage dialetically (more chances to learn), though, as noted above, if any one participant isn't playing the same game, they tend to spoil things for the rest. I'm not aware of any way of ensuring that all are engaged in didactic discussion, though again, recognising the game you're in is quite useful.
There are a number of diversionary tactics which are surprisingly effetive at derailing discussion. Being aware of these, calling them out, and if at all possible, eliminating those who are engaged in such practices from discussions you're hoping to maintain as productive, is particularly effective. Several sets of common tactics (one attributed, though not confirmed, to Karl Rove) are listed here:
https://www.reddit.com/r/dredmorbius/comments/2d0r1d/the_rea...
There's another long line of tactics adopted by bullshitters here:
https://www.reddit.com/r/dredmorbius/comments/28ge14/on_nons...
For those interested in epistemology and criteria of truth, I recommend Wikipedia's "Criteria of Truth" or the Stanford Encyclopedia of Philosophy's "Truth" pages:
https://en.m.wikipedia.org/wiki/Criteria_of_truth
https://plato.stanford.edu/entries/truth/
This isn't a greenfield. There are authorities to consult.
Otherwise, this old summary stands:
https://www.reddit.com/r/dredmorbius/wiki/faq#wiki_so.2C_red...
When infotainment gets called out, it spins up a distracting, angering crisis and ignores the problem.
And by then it's too late: the headlines have been seen by the masses, and they're asking their friends, "Hey did you hear about...?", and they're adding it to their mental models of the world. The media source doesn't suffer; in fact they have already profited from the clickbait, so there's virtually no incentive to not publish false or barely researched stories, even if they later put two sentences of "Note:" at the bottom, which they actively discourage people from reading to anyway.
If you accept that fox is right wing, where is the 'centrist' media and what's leftist? In international terms their are no left wing media outlets, except insofar as not constantly expounding propaganda.
PS: The US democratic party is moderate right/centrist, the democrats are right, left is outside of the discourse.
PS: The US democratic party is moderate right/centrist, the republicans are right, left is outside of the discourse.
>only picks Fox News
You see you're part of the problem right?