People shouting 'fallacy' when there is no fallacy

[Barber Shop]

I have noticed that people occasionally accuse others of using fallacies even when there was no real fallacy.

I was interested in what others had to say about their experiences with this, as well as what particular faux fallacies are commonly 'spotted'.

The one I see the most is the accusation of 'appeal to authority'. While there are certainly illegitimate appeals, citing a reliable and knowledgeable source is not a fallacy. It seems that someone people think anytime someone mentions a source or expert they have committed a logical fallacy.

[Wildy]

It happens when people don't actually understand what the fallacy is in question. I had that problem with the old ad hominem, but as I started to actually understand what it was I stopped accusing people of it. Mine was a somewhat common misunderstanding where you associate ad hom with just an insult in general, which is very, very, wrong.

The problem with the appeal to authority is exactly as you have described. People confuse "authority within a particular subject" with a more generalised form of "authority". If you want a good example of a person who uses the appeal to authority properly I suggest Kary Mullis. Some of his comments have suggested that as a Nobel Laureate we should take his comments without question, no matter how wrong they might be.

[Noztradamus]

Outraged feminists.

[gnome]

Tu Quoque:

Legitimate use:

"People shouldn't steal from others."
"But you steal!"

That's an actual fallacy. It doesn't affect the value of the proposition "People shouldn't steal".

In a contest with two contenders:
"I should win because Snerdley cheated."
"But you cheated too!"
"Sorry, tu quoque fallacy."

Not a fallacy here. It is very relevant to the proposition that one contestant is superior to another.

[Dave Rogers]

The one I see the most is the accusation of 'appeal to authority'. While there are certainly illegitimate appeals, citing a reliable and knowledgeable source is not a fallacy. It seems that someone people think anytime someone mentions a source or expert they have committed a logical fallacy.

In formal logic, though, even the legitimate appeal to authority is a fallacy. The appeal to authority is valid for forming provisional conclusions, but does not override conflicting evidence. So, to construct a hypothetical* example, if a chemist states that the presence of certain reaction products indicates that a specific chemical reaction has taken place, then it is reasonable to assume in the absence of conflicting evidence that his expertise makes him likely to be correct. If, however, he fails to account for the fact that the energy yield of that reaction is insufficiently high to account for the heat produced by the process, then this contradictory evidence over-rides his expertise. If his original conclusion is still defended on the basis of his expertise, then this is fallacious, even though still (in principle) a legitimate appeal to authority.

Dave

* OK, so it's not really hypothetical.

[MysterOnyx]

In formal logic, though, even the legitimate appeal to authority is a fallacy. The appeal to authority is valid for forming provisional conclusions, but does not override conflicting evidence. So, to construct a hypothetical* example, if a chemist states that the presence of certain reaction products indicates that a specific chemical reaction has taken place, then it is reasonable to assume in the absence of conflicting evidence that his expertise makes him likely to be correct. If, however, he fails to account for the fact that the energy yield of that reaction is insufficiently high to account for the heat produced by the process, then this contradictory evidence over-rides his expertise. If his original conclusion is still defended on the basis of his expertise, then this is fallacious, even though still (in principle) a legitimate appeal to authority.

Dave

* OK, so it's not really hypothetical.

Would it still be an appeal to authority if no contradictory evidence has been found?

[RLBaty]

Or as my "authority" puts it, in part:

> "The argument ad verecundiam is committed when someone
> argues that a proposition is true because an expert in a given field
> has said that it is true.
>
> An expert's judment constitutes no conclusive proof; experts
> disagree, and even when they are in agreement they may be
> wrong.
>
> However, reference to an authority in an area of competence
> may carry some weight, but it doesn't prove a conclusion.
>
> Ultimately, even experts need to rely upon empirical evidence
> and rational inference.
>
> The fallacy of the appeal to inappropriate authority arises
> when the appeal is made to parties who have no legitimate
> claim to authority in the matter at hand.
>
> In every instance, an argument must be judged upon its
> own merits."

[One Tenth]

I most commonly see false accusations of the following two fallacies:

1. Ad Hominem. Telling someone they are wrong because of x, y, and z, and that they are also an idiot, is not an ad hominem. It's a rebuttal and an insult.

2. Strawman. A lot of people say something but couch it in weasel words so they can call strawman later. They want to assert (or at least insinuate) something, but they don't want to actually have to back up the assertion.

[MysterOnyx]

Non sequiturs run rampant online. People frequently say things that have no connection to their conclusion.

[Dancing David]

Would it still be an appeal to authority if no contradictory evidence has been found?

It could be.

The actual appeal to authority is when someone says "It must be right because so and so says it is right."

An examination of the evidence is what critical thinking is about, in the case where confounding factors, alternative explanation and bias have not been explored then the conclusion drawn should be 'more examination is warranted".

Now ehen people talk about mainstream theories and complain about appeal to authority they are usually not going to talk about teh evidence.

[Dancing David]

I most commonly see false accusations of the following two fallacies:

1. Ad Hominem. Telling someone they are wrong because of x, y, and z, and that they are also an idiot, is not an ad hominem. It's a rebuttal and an insult.

2. Strawman. A lot of people say something but couch it in weasel words so they can call strawman later. They want to assert (or at least insinuate) something, but they don't want to actually have to back up the assertion.

The usual strawman attack is more blatant than that, mostly it is setting up a false dicotomy to knock it down, or drawing false conclusions and knocking them down.

[blutoski]

I have noticed that people occasionally accuse others of using fallacies even when there was no real fallacy.

I was interested in what others had to say about their experiences with this, as well as what particular faux fallacies are commonly 'spotted'.

The one I see the most is the accusation of 'appeal to authority'. While there are certainly illegitimate appeals, citing a reliable and knowledgeable source is not a fallacy. It seems that someone people think anytime someone mentions a source or expert they have committed a logical fallacy.

I actually blogged about this problem: [Skeptical MythConceptions – part 1 – Authority (http://blog.bcskeptics.info/?p=7)]

My impression is that skeptics may be interested in critical thinking, but are not very experienced in it, and no better at it than laypersons.

One of the consequences is the misperception that identifying fallacies is the definition of critical thinking - that a formal fallacy means a bad argument. This is untrue.

[not daSkeptic]

I have noticed that people occasionally accuse others of using fallacies even when there was no real fallacy.

Sometimes people are overly interested in proving others wrong.

[Piscivore]

The usual strawman attack is more blatant than that, mostly it is setting up a false dicotomy to knock it down, or drawing false conclusions and knocking them down.
That's what an legitimate strawman fallacy is. Altering or misstating an opponent's argument in order to render it easier to refute.

Most of the time "strawman" is actually used around here, it's just an emotional exclamation. Considering the context in which I most often see it employed I've determined the utterer apparently means by it "I don't have any facts, evidence, or even a coherent argument to refute what you just said, but you're wrong anyway!"

[RLBaty]

(S)keptics may be interested in critical thinking,
but are not very experienced in it,
and no better at it than laypersons.



I think my own, recent "critical thinking" exercise here provides a little empirical evidence that the above observation is justified.

http://forums.randi.org/showthread.php?t=208300

:blush: :blush: :blush:

[RLBaty]

I actually blogged about this problem: [Skeptical MythConceptions – part 1 – Authority (http://blog.bcskeptics.info/?p=7)]



Speaking of "SkeptiCamp", it just so happens that I happened to speak at the local one recently, participated in a panel discussion, and was able to attend a variety of other presentations!

[Dave Rogers]

One of the consequences is the misperception that identifying fallacies is the definition of critical thinking - that a formal fallacy means a bad argument. This is untrue.

I think the second half of that is a bit overstated. It's generally true that an informal fallacy means a bad argument, and it's often true that a formal fallacy means a bad argument. However, identifying a bad argument doesn't refute the conclusion of the argument. For example, if I say, "I believe aspirin is an effective pain killer because Tim Minchin says it is," that's an extremely bad argument because it's an appeal to an invalid authority, and if I say, "Aspirin is definitely an effective pain killer because my doctor told me so," that's still at best a weak argument even though the authority is a valid one. However, aspirin is still an effective pain killer.

Dave

[JoelKatz]

The one I see the most is the accusation of 'appeal to authority'. While there are certainly illegitimate appeals, citing a reliable and knowledgeable source is not a fallacy. It seems that someone people think anytime someone mentions a source or expert they have committed a logical fallacy.If they claim that what the source or expert says is true because that person is an expert, that is a fallacy. An expert can make a claim worth listening to. And you may provisionally accept the claims of an expert because they have a history of being true and it doesn't matter enough to invest the time to understand it yourself. You may accept the claims of an expert because you don't have the ability to work things out yourself. But ultimately, an expert is an expert because he makes arguments that are valid and has a repository of verifiable information, the reverse is only statistically likely, and only when there aren't other reasons to think it's not so.

The biggest problem with experts is the selection problem. A Christian, for example, is likely to consider prominent Christians to be authorities on religion. After all, they have a pattern of making claims and arguments that he considers true, whereas atheists and Muslims (regardless of their educational credentials) do not. Accepting experts can feedback loop you into a significant mental commitment to a house of cards. This is probably a large part of why people so rarely change their core religious or political views.

[JoelKatz]

I actually blogged about this problem: [Skeptical MythConceptions – part 1 – Authority (http://blog.bcskeptics.info/?p=7)]I think your blog post creates a false dichotomy between having to personally verify everything yourself and accepting something merely because an authority says so. Both of these positions are incorrect. For example, you say:

If I reject all arguments with this type of format, I may run into serious problems promoting skeptical views. How do I know that Homeopathy doesn’t work? Experiments show this. My experiments? Well, no. Some other people did them. I’m pretty sure the results are correct, because I’m pretty sure the scientists involved are competent. And I guess this is the case because other scientists say they are competent, and anyway, their findings were peer-reviewed by a panel of scientists whom the publisher selected for their competence (no doubt, based on the opinion of their peers).The point is, this is all investigation you personally did. Your conclusion is the personal synthesis or large amounts of information from multiple sources.

In an imaginary dialogue with a non-skeptic, every argument we bring forward will ultimately be exposed to be based on a statement by some authority, be that a famous skeptic or physicist, or a publication like JAMA or Cell or Science, or a textbook, or Wikipedia…Exactly. That's why we need more than one argument. Presenting the word of a single authority and claiming it is true because that expert says so is wrong. We must marshal evidence from multiple sources and independently synthesize them to a conclusion.

You can use any input you want, but you must think for yourself.

If we were to decide that we no longer accept scientific publications we have not personally vetted to the last detail, what are we going to use when making decisions?Nobody, so far as I know, argues that. "Homeopathy is a farce because X says so" is nonsense regardless of who X is. But "Homeopathy is a farce because homeopaths themselves say X and Y says X is invalid because of Z, and these studies A, B, and C have been done by Q and R, and ...." That's fine.

You can take input from any source you want, and you can weigh the reliability of sources. But you cannot take the conclusion from a source. You must reach the conclusion on the basis of the evidence. You needn't do anything literally personally but reach the final conclusion.

[sphenisc]

... if I say, "I believe aspirin is an effective pain killer because Tim Minchin says it is," that's an extremely bad argument ...

No, "Aspirin is an effective pain killer because Tim Minchin says it is", is an extremely bad argument. Your statement isn't an argument at all, (which I suppose makes it an extremely bad argument)... I think I'll go out and come back in again. :o

[Meadmaker]

The "no true scotsman" fallacy. It occurs when someone asserts that a non-definitive characteristic is actually a definitive characteristic. In its original formulation, a violent act was committed by a Scotsman, but the arguer asserted that it could not be a "true" Scotsman because Scotsmen don't do that. On JREF, any example of "no true X" will bring up a characterization of the "No true Scotsman" fallacy, regardless of whether or not it applies.

The "slippery slope" fallacy. Sometimes, changing a law or custom really will lead to a change in people's behavior. It isn't a fallacy to point that out. Sometimes, slopes really are slippery.

The "straw man". Sometimes, someone misunderstands an argument. Not every mischaracterization is a straw man. Straw men exist only when someone deliberately distorts an argument for the purpose of appearing to refute the original. Most accusations of "straw men" on JREF are not really straw men.

[epepke]

Non sequiturs run rampant online. People frequently say things that have no connection to their conclusion.

Fish!

[Walter Wayne]

Fish!
That just a red herring.

[Barber Shop]

I have not had a chance to revisit this thread in a while and I am happy to see some interesting comments.

I did want to response to a few posts because, based on the responses, I may not have made my point very clear.


I know that any appeal to authority, even a reliable and valid authority, does not MAKE something true. It is true because it is true, regardless of what Professor Smartypants or the respected Journal of Whatever says.

But, I don't think that you have committed a fallacy by simply mentioning a reliable source. You commit a fallacy when you think that this source MAKES your statement true or your argument valid.

My real point in mentioning the "appeal to authority" is that some people, including some folks on this very forum, seem to actually think you WEAKEN your argument by citing reliable sources. They seem to think that you have lost the argument because you reference a knowledgeable person.


@Dave Rogers --- I suppose that everything you say is true, but not necessarily related to what I am trying to say. I am not saying that appealing to an authority overrides contradictory evidence. Nobody's credentials override reality.

Joel Katz stated: "If they claim that what the source or expert says is true because that person is an expert, that is a fallacy. " Yes, of course it is a fallacy. Who would disagree with that?

Again...perhaps this is my fault for not making my point clear...but I wanted to point out that there is nothing WRONG with citing a reliable source. You have not lost the argument by doing so.

[JoeBentley]

One problem is the difference between a logical fallacy and a properly used rhetorical device can often be subtle. There is not always a huge difference between a perfectly valid hypothetical situation or "taking things to their logical extremes" and a fallacious slippery slope.

[gnome]

Most false accusations of slippery slope or bad analogies are failing to acknowledge that the person is using the analogy or extension to refute a proposed (or implied) maxim, rather than equating the significance, or suggesting the extension is likely, in practice. Refusing the point generally leads to special pleading.

For example, someone says "I don't like that serial killer, killing is always wrong."
Someone else responds, "But this war hero killed a lot of people too."
The first says "Seriously? You're likening this war hero to a serial killer?"

No, they're not. They're just refuting the rule that killing is always wrong.

[JoeBentley]

Indeed. And I think most people realize that for the most part all rhetorical devices are inherently imperfect. Metaphors, similes, analogies, comparisons, parables, etc are not meant to define the topic, only to put them in certain perspectives.

Bob: "X is like Y"
Ted: "X is not the same as Y!"
Bob: "I know. It's like Y, not equal to Y. 'Like' is a statement of comparison, not one of definition or implied equality."

That and practically all rhetorical devices only work to a certain level and then break down and you have to abandon them. This doesn't mean they are unreasonable to use though.

[JoeTheJuggler]

1. Ad Hominem. Telling someone they are wrong because of x, y, and z, and that they are also an idiot, is not an ad hominem. It's a rebuttal and an insult.
I understand the distinction you (and others) are trying to make, but I think you miss the reason why ad hominems are fallacious. They are fallacious inasmuch as they are irrelevant. That is, the ad hominem fallacy is a type of red herring or irrelevant fallacy.

If you're in the context of a debate and one person insults the other, it is a fallacy, as is ANY statement put in place of an argument that is irrelevant to the issue at hand.


2. Strawman. A lot of people say something but couch it in weasel words so they can call strawman later. They want to assert (or at least insinuate) something, but they don't want to actually have to back up the assertion.
I've seen that technique. I think some people reject that implicit statements are real. It's not usually difficult to show that the words they actually used would have no meaning if they didn't mean to imply what they're later claiming to be a straw man proposition.

On the other hand, I've seen people wrongly deny the accusation of making a straw man fallacy on the basis that someone somewhere did indeed voice that position. It can be fallacious to ignore the position held by the people you're debating and the majority of people who share an opinion by focusing on an extreme minority expression of that position. This site (http://www.fallacyfiles.org/strawman.html) refers to this subtype of strawman fallacy as the "extreme man" fallacy.

[JoeTheJuggler]

The "slippery slope" fallacy. Sometimes, changing a law or custom really will lead to a change in people's behavior. It isn't a fallacy to point that out. Sometimes, slopes really are slippery.

I think you're bringing up an important point about informal fallacies. They're not really fallacious in the same way formal fallacies are. Formal fallacies really are strictly about whether or not a conclusion can be made from the premises--whether it logically follows. Informal fallacies have a lot more to do with whether or not a way of thinking is useful or. . . I dunno "good". (Begging the question, for example, isn't technically a fallacy. It's just asserting a conclusion as a premise. In a way, all question begging arguments are validating arguments, and must be valid. But that doesn't prove that the premise is true. Just that if it's true, it's true is valid logic!)

A lot of arguments to the consequences are used all the time. In discussing whether a law is a good or bad idea, or whether a court decision should go one way or the other, we hope reasonable people will consider the consequences of the decision. But technically, those are all slippery slope arguments and don't necessarily talk about the inherent merits of the proposition in front of you.

[JoeBentley]

I understand the distinction you (and others) are trying to make, but I think you miss the reason why ad hominems are fallacious. They are fallacious inasmuch as they are irrelevant. That is, the ad hominem fallacy is a type of red herring or irrelevant fallacy.

Errr I don't know. I certainly understand the "play the ball, not the man" rule for both basic politeness reasons and intellectual standards.

But I think we've all been in conversations with someone that is obviously bringing outside personal issues into the discussion. Bias, ulterior motives, personal emotional issues and baggage... these do all exist and sometimes do get brought into conversations and do influence them and that influence can be observed. I don't think its fair to expect everyone to just pretend to not notice this because of some idea that a standard exists that we can't even bring the person's possible motivations into the conversation.

So if someone is arguing that people should buy more widgets, and I learn the person is CEO of the #1 widget company in the world, I think there has to be some level were I can bring up the possibility that he is not looking at the topic objectively without it simply be dismissed as an ad hominem.

[JoeTheJuggler]

But I think we've all been in conversations with someone that is obviously bringing outside personal issues into the discussion.

Yes, and inasmuch as those outside personal issues are irrelevant, they are fallacious.

I like this Taxonomy of Fallacies (http://www.fallacyfiles.org/taxonomy.html). I think it sheds light on the fact that the reason ad hominems are fallacious is that they are a subtype of the irrelevancy fallacy.

Now, if the issue being discussed it the character of a person (as might come up in a political race or in a criminal trial), then remarks about the person are not irrelevant and not fallacious (at least not simply because they are remarks to the person).

Also, if the conversation in question isn't a debate of some sort, then personal attacks are just insults because they're not being made in the place of an argument. Of course, there are all kinds of conversations that are somewhere in the middle or a mixture of just conversation and engaging in reasoned argumentation. I think the rule of thumb is, again, whether or not the statement is relevant and whether or not it is a statement made where you would expect an argument or part of an argument.

That's why Dan Akryod's old, "Jane, you ignorant slut" was such an over the top ad hominem. It was made as his opening remark in a piece set up as a semi-formal debate called Point/Counterpoint. He never did say, "Your position is wrong because you are an ignorant slut", but it was a fallacy nonetheless because it was irrelevant to the debate topic and it occurred exactly where you'd expect an argument.

[JoeTheJuggler]

So if someone is arguing that people should buy more widgets, and I learn the person is CEO of the #1 widget company in the world, I think there has to be some level were I can bring up the possibility that he is not looking at the topic objectively without it simply be dismissed as an ad hominem.

Sorry--I guess I should have read your entire post before replying.

Yes, you are correct, and this fits with what I've been saying. If there is a reason for personal observations to be relevant, they're not fallacious.

But saying you should consider a conflict of interest or some such doesn't refute any argument that that person may have made. Those arguments stand or fall on their own merits. So even if you observe that person A might have a conflict of interest or something that means he may not be looking at things objectively, that observation does not refute any argument made by person A.

[Mr. Purple]

One of the better threads I have seen. Thanks all!

Though I vaguely remember a class called phil 102 or something in college (I don't recollect much from then...) I only really got "into" fallacies recently. I wanted to add my thoughts as they relate to JREF specifically-

1) It has been stated ad nauseum, but I will echo: An insult is not an ad hom. Saying X is true because Person A was wrong about Y, is an ad hom. As a pseudo-hypothetical, I often see this one in the economics subforum-
"Don't believe member A, he thinks we should go back on the gold standard". This is a blatant example of an ad hom. It irritates me, because I am trying to understand the world of finance better, and these statements do nothing to advance the discussion. Forget what else he/she says about the gold standard- why are they wrong on "this"?

These also relate to #3 (below), appeals to authority.

2) Strawman - Easy. Be the first to shout this in any argument and you win the thread. What? That's not how it works? :boxedin:

In all seriousness, I think this is one of the more important fallacies to recognize as it is an invasive little trend. Watch any political press conference thread in the politics subforum, and they run rampant. "Death Panels" is one example, some of the "Class Warfare" sentiments on the other side are another example.

3) Appeal to Authority - This is a very trickly one. I find myself battling with this one often. As a good jumping off point:
Errr I don't know. I certainly understand the "play the ball, not the man" rule for both basic politeness reasons and intellectual standards.

But I think we've all been in conversations with someone that is obviously bringing outside personal issues into the discussion. Bias, ulterior motives, personal emotional issues and baggage... these do all exist and sometimes do get brought into conversations and do influence them and that influence can be observed. I don't think its fair to expect everyone to just pretend to not notice this because of some idea that a standard exists that we can't even bring the person's possible motivations into the conversation.

So if someone is arguing that people should buy more widgets, and I learn the person is CEO of the #1 widget company in the world, I think there has to be some level were I can bring up the possibility that he is not looking at the topic objectively without it simply be dismissed as an ad hominem.

Yes, Joe (if I a many be so familiar :) ), there is good reason to apply scrutiny to this CEO's opinion. Regardless of what it is- it has no bearing on the truth.

To me, I break down the appeal to authority / ad hom problem this way: It doesn't matter what anyone thinks- the truth is the truth. While it may be instructive or helpful to analyze the opinion of those with direct information/expertise, it should be a starting point and not the end.

I think I have personally "grown" (how friggin lame does that sound? I am going to have to pound two more beers than normal to make up for any potential "maturity") the most by understanding the "be kind to collegues and ruthless with theories" approach that I first read about in Bakker's Dinosaur Heresies. I think the resistance to it (even subconciously) is that in the end, you may find that the most compelling argument is not personally palatable for you. Alternatively, somebody unpalatable to you may be making the best argument. In either case, the "better skeptic" weighs the evidence, and sides with the best argument- not the best people.

Should you listen to expert A rather than layperson B? I think reasonable minds would rightly say "yes". But in no way does that prove the truth of a situation.

So here is where it gets hard:
I know next to nothing about physics or astronomy (compared to the sharp crayons here). I have no reasonable way to look at the math in some peer reviewed journal and understand it in any way. I have to look towards the opinions of other posters. You see the problem....

I try my best to look at the discussion "structurally". No, I don't have any idea what those equations meant. But, I can understand when someone is dodging questions, proposing strawmen, etc. Knowing that it is entirely possible that I/they will be wrong, I can cautiously pick a side.

[Mr. Purple]

Taxonomy of Fallacies (http://www.fallacyfiles.org/taxonomy.html).

I just got a nerd-on. Thanks for that.

[JoeBentley]

Yes, and inasmuch as those outside personal issues are irrelevant, they are fallacious.

I like this Taxonomy of Fallacies (http://www.fallacyfiles.org/taxonomy.html). I think it sheds light on the fact that the reason ad hominems are fallacious is that they are a subtype of the irrelevancy fallacy.

True but if you think about it the defining characteristic of any logical fallacy is it's irrelevantness. I'd had a hard time considering anything which was relevant to the topic to be bad argumentative form.

That's why Dan Akryod's old, "Jane, you ignorant slut" was such an over the top ad hominem. It was made as his opening remark in a piece set up as a semi-formal debate called Point/Counterpoint. He never did say, "Your position is wrong because you are an ignorant slut", but it was a fallacy nonetheless because it was irrelevant to the debate topic and it occurred exactly where you'd expect an argument.

Indeed and there's a difference between bringing up a personal possible angle to a person's stance on an issue and simply attacking them outright.

But saying you should consider a conflict of interest or some such doesn't refute any argument that that person may have made. Those arguments stand or fall on their own merits.

To be sure which is why I grew tired of the old "You made a logical fallacy ergo I won the argument!" style of debate. You can use every logical fallacy in the book and still be 100% correct. There's an annoying trend amoungst what I mentally refer to as "Coffee Shop Intellectuals" online to simply quote a person's entire argument nothing but the name of the intellectual fallacy they feel they the person made, a style of argument I could not find more tiresome.

If in the course of a debate you feel a person does make a logical fallacy, there's nothing wrong with pointing it out. But don't just start your victory lap over it. Bring up the issue to the person, get their point of view on it (because as we've already established on person's logical fallacy is another person's valid rhetorical device. This ain't exactly a perfect science.), correct for any deviation it caused in the conversation and move on.

[JoeBentley]

3) Appeal to Authority - This is a very trickly one.

Yeah the Appeal to Authority is another one I'm really... uncomfortable with at times. I understand the dangers with any kind of blind following of any source of information... but on the other hand I accept the reality that for most subjects people exist that know more about the topic then I do. Recognizing the fact that someone is more likely to be correct on a given topic then the general layperson via is not a logical fallacy. It's the very concept of someone having a body of knowledge.

I've heard it said, I can't recall by whom at the moment, something to the effect that science is so effective because "It only has experts, not authorities." Or as Sam Harris put it during his TED speech. "Whenever we are talking about facts certain opinions must be excluded. That is what it is to have a domain of expertise. That is what it is for knowledge to count. "

So if I install some wiring in my house and I ask a licensed electrical contractor to double check my work I'm not making an "Appeal to Authority" I'm simply recognizing the fact that their exists people who's job it is to know more then I do about a this particular topic.

To me, I break down the appeal to authority / ad hom problem this way: It doesn't matter what anyone thinks- the truth is the truth. While it may be instructive or helpful to analyze the opinion of those with direct information/expertise, it should be a starting point and not the end.

Oh I totally agree. The truth exists in a vacuum. Sadly our disussions don't always.

[JoeTheJuggler]

True but if you think about it the defining characteristic of any logical fallacy is it's irrelevantness. I'd had a hard time considering anything which was relevant to the topic to be bad argumentative form.
I think you're confusing relevance for valid logic in general. There are many categories of fallacies that aren't based on irrelevance. All of the formal fallacies are fallacious for reasons other than irrelevance, for example.

Put another way, relevance alone doesn't guarantee a valid argument.



You can use every logical fallacy in the book and still be 100% correct.
Well, your conclusion could be true, but by definition if your argument is fallacious it is not a valid argument, and therefore not "correct".

ETA: Rather than "correct" I would use the terms "valid" and "sound". To be valid is the same as saying the argument has no fallacies, which is the same thing as saying the conclusion follows logically from the premises. (Being valid doesn't say anything about the truth value of the premises or conclusion.) To be sound, the argument must be valid, and the premises (and therefore the conclusion) must be true.


If in the course of a debate you feel a person does make a logical fallacy, there's nothing wrong with pointing it out. But don't just start your victory lap over it.
I agree wholeheartedly. In fact, just refuting an argument does nothing to establish a contrary position. I hate to mention it, but asserting that it does is yet another named fallacy! :D

[Andrew Wiggin]

That would be the 'making up a fallacy and hoping no one will call you on it' fallacy. I just invented it.

[HansMustermann]

I have noticed that people occasionally accuse others of using fallacies even when there was no real fallacy.

I was interested in what others had to say about their experiences with this, as well as what particular faux fallacies are commonly 'spotted'.

The one I see the most is the accusation of 'appeal to authority'. While there are certainly illegitimate appeals, citing a reliable and knowledgeable source is not a fallacy. It seems that someone people think anytime someone mentions a source or expert they have committed a logical fallacy.

But conversely, I too often see the opposite: people doing broken logic and then arguing why it's not really a fallacy.

The appeal to authority you mention is such a complex case, and it's just as often misunderstood by both sides.

Ultimately a conclusion must follow from true premises and valid logic. It doesn't actually matter who said it, anyone can be wrong. Einstein was wrong about the cosmological constant in his own theory, for example.

So in formal logic, an appeal to authority is always a fallacy. Sorry, there is no exception. Even Einstein about GR is not exempt. The conclusion must be true because of the supporting evidence, not because of who said it.

In informal logic, it is basically allowed to shortcut that, and evaluate how much you trust a certain source with just taking their word for the conclusion. E.g., if you just want to know whether you should take tamiflu for your viral infection, you don't necessarily need to go all the way to the clinical trial data, you can start with just trusting some medical authority to know about that.

But here's what a lot of people seem to miss: even in informal logic it's not what makes a conclusion true. That conclusion must still follow from premises. You're just allowed to take someone's word that it does, if you need to work from there.

Basically even for tamiflu, it's not that it works because Dr Some Smart Guy says so, but it works because it has the clinical trial data to say so. The latter is what actually makes it true or false. The allowance in informal logic is to just trust that Dr Some Smart Guy knows that stuff so you don't have to.

But that can't be used in a talk where the whole point is whether the conclusion follows from the conclusion. If the whole point is whether the clinical trial data adds up to the claimed confidence level, you can't insist that it's true just because some authority says so. Doing that is still an appeal to authority.

[HansMustermann]

To be sure which is why I grew tired of the old "You made a logical fallacy ergo I won the argument!" style of debate. You can use every logical fallacy in the book and still be 100% correct.

No. If you arrive at a conclusion by way of a fallacy, by definition you're not 100% correct. Sorry.

The conclusion may still be true or false regardless of that -- which is why we also have an argument from fallacy (http://en.wikipedia.org/wiki/Argument_from_fallacy) fallacy -- but a fallacious reasoning still is not correct, and hasn't supported anything.

E.g., I could say that

P1: All students wear backpacks.
P2: Jack's grandfather wears a backpack,
C: therefore Jack's grandfather is a student.

It's a classic fallacy (fallacy of the undistributed middle, to be precise), and therefore by definition not correct. But nevertheless, it could be that said grandfather re-enrolled in university to get a degree in his old age. But the argument for that is still broken.

If that is the only argument offered for the fact that Jack's grandfather is a student, then really the person reading it hasn't seen any logical support for that. As far as they know, there is no evidence that Jack's grandfather went back to school. So don't be surprised if someone asks for some actual evidence there.

[Dave Rogers]

That would be the 'making up a fallacy and hoping no one will call you on it' fallacy. I just invented it.

Various 9/11 truthers beat you to it. There's one using it right now.

Dave

[MarkCorrigan]

No True Scotsman is the one I find most annoyingly misused. Any group has at least one probably more defining characteristics that they must share to be part of the group, for example Scotsmen must be human and either have Scottish parentage or have been born in Scotland. This is irrefutable, because otherwise the term Scotsman is meaningless, and can be applied to literally anything at all. However, if you claim that a group must do or think something that is an absolutely intrinsic part of being a member of that group, someone here will shout "No True Scotsman!" at you.

[gnome]

I've always thought of the True Scotsman and the Straw Man as flip sides of each other...the Straw Man fallaciously attributes an undesirable characteristic, whereas the True Scotsman fallaciously disavows an undesirable characteristic.

[sphenisc]

No True Scotsman is the one I find most annoyingly misused. Any group has at least one probably more defining characteristics that they must share to be part of the group, for example Scotsmen must be human and either have Scottish parentage or have been born in Scotland. This is irrefutable, because otherwise the term Scotsman is meaningless, and can be applied to literally anything at all. However, if you claim that a group must do or think something that is an absolutely intrinsic part of being a member of that group, someone here will shout "No True Scotsman!" at you.

I've always thought of the True Scotsman and the Straw Man as flip sides of each other...the Straw Man fallaciously attributes an undesirable characteristic, whereas the True Scotsman fallaciously disavows an undesirable characteristic.

No True Strawman argument would fallaciously attribute an undesirable characteristic...

Damn, you're right Mark.

[JoeTheJuggler]

No True Scotsman is the one I find most annoyingly misused. Any group has at least one probably more defining characteristics that they must share to be part of the group, for example Scotsmen must be human and either have Scottish parentage or have been born in Scotland. This is irrefutable, because otherwise the term Scotsman is meaningless, and can be applied to literally anything at all. However, if you claim that a group must do or think something that is an absolutely intrinsic part of being a member of that group, someone here will shout "No True Scotsman!" at you.

Interestingly, there is no entry for the No True Scotsman fallacy in the Fallacy Files.

I think when used properly, it's the same as (or at least a subcategory of) Moving the Goalposts. I guess it's a bit more than that since it involves excluding from the definition of the class exactly the characteristic being debated. (In that way it's more similar to the flaw in the Ontological Proof of the existence of God--which predicates existence, or makes that a characteristic included in the definition when that is the point being debated--as such it is just question begging.)

I think your criticism, Mark, points out that the common misuse of it is the commission of the Division or Composition Fallacy (thinking the parts should all have the same characteristics as the whole and vice versa, respectively).

[stilicho]

(In that way it's more similar to the flaw in the Ontological Proof of the existence of God--which predicates existence, or makes that a characteristic included in the definition when that is the point being debated--as such it is just question begging.)

1. All gods are liars.
2. I am a liar.
3. Therefore I am god.

Simple!COUNT_THE_FALLACIES

[JoeTheJuggler]

1. All gods are liars.
2. I am a liar.
3. Therefore I am god.

Simple!COUNT_THE_FALLACIES

I only see one fallacy (remember, false premises and question begging are not fallacies). That one is affirming the antecedent. Note this argument takes the same form as

P1. if P then Q
P2. Q
C. Therefore P

(P="One is a god" Q="one is a liar")

ETA: If your example was meant to lead to a conclusion about the existence of god--and it doesn't mention anything about existence--then I would say it is question begging. For it to conclude anything about the existence of gods, liars or even "I" (which is just an example of a possible member of either of those classes), it must first hypothesize existence in a premise. For example, you could get from "I am a god" (using that conclusion as a premise) with the second premise "I exist" to reach the conclusion "a god exists". That argument is valid, but it is merely question begging. You're assuming a god exists in order to reach the conclusion that a god exists. It's a validating argument in that you could put anything there in place of god--dragons, unicorns, etc.--and it would still be valid, but not necessarily sound.

[stilicho]

I only see one fallacy (remember, false premises and question begging are not fallacies). That one is affirming the antecedent. Note this argument takes the same form as

P1. if P then Q
P2. Q
C. Therefore P

(P="One is a god" Q="one is a liar")

ETA: If your example was meant to lead to a conclusion about the existence of god--and it doesn't mention anything about existence--then I would say it is question begging. For it to conclude anything about the existence of gods, liars or even "I" (which is just an example of a possible member of either of those classes), it must first hypothesize existence in a premise. For example, you could get from "I am a god" (using that conclusion as a premise) with the second premise "I exist" to reach the conclusion "a god exists". That argument is valid, but it is merely question begging. You're assuming a god exists in order to reach the conclusion that a god exists. It's a validating argument in that you could put anything there in place of god--dragons, unicorns, etc.--and it would still be valid, but not necessarily sound.

I try to be internally consistent if not necessarily sound all the time so that might be the key. Performance review time is coming up so I'll keep it mind.

[Dave Rogers]

Interestingly, there is no entry for the No True Scotsman fallacy in the Fallacy Files.

I think when used properly, it's the same as (or at least a subcategory of) Moving the Goalposts. I guess it's a bit more than that since it involves excluding from the definition of the class exactly the characteristic being debated.

The odd thing about the No True Scotsman fallacy is that it isn't, properly speaking, a fallacious means of arriving at a specific conclusion from a given set of premises, and therefore isn't really a fallacy at all. The fallacy of moving the goalposts is a little different in that it involves setting a threshold of proof for a claim, then redefining that threshold if it appears that the orignal one will be achieved. The No True Scotsman fallacy, contrastingly, changes the conclusion to a different one than was originally proposed.

The classic formulation is something like:
Person A: I know a Scotsman who puts sugar on his porridge.
B: That's impossible. Scotsmen don't put sugar on their porridge.
A: Well, my friend Angus McTavish, who was born in Scotland, puts sugar on his porridge.
B: Then he isn't a true Scotsman, because no true Scotsman puts sugar on his porridge.

What's interesting here is that B is no longer disputing A's claim that he knows a Scotsman who puts sugar on his porridge; rather, he is claiming that Angus McTavish, though possibly a Scotsman, is not a member of a the newly-defined set of True Scotsmen, defined as the subset of Scotsmen who never put sugar on their porridge. Since it's already stipulated that Angus McTavish is not a member of that set, then the conclusion is in fact correct within the definition specified - in fact, as the definition is tautological, it cannot be incorrect.

So the No True Scotsman fallacy isn't, I don't think, a fallacy at all. Rather, it's a means of derailing an argument, in which the person employing it goes off at a tangent to prove a tautology that has no relevance to the original claim.

Which, I suspect, makes all that I've written above an extremely complex and roundabout version of the No True Scotsman fallacy.

Dave

[gnome]

I must disagree... here's another way of looking at it... it's equivocating on Scotsman, changing the meaning from someone who was born in Scotland, to someone who was born in Scotland and conforms to ad-hoc norms. So it could be a subset of fallacy of equivocation.

It's clear in the example (especially since it is claimed to be 'impossible') that the porridge Nazi wants the meaning of Scotsman initially to be taken only as "Born in Scotland" without further examination, and then backpedals when given a counterexample.

[JoeTheJuggler]

The odd thing about the No True Scotsman fallacy is that it isn't, properly speaking, a fallacious means of arriving at a specific conclusion from a given set of premises, and therefore isn't really a fallacy at all. The fallacy of moving the goalposts is a little different in that it involves setting a threshold of proof for a claim, then redefining that threshold if it appears that the orignal one will be achieved. The No True Scotsman fallacy, contrastingly, changes the conclusion to a different one than was originally proposed.

The classic formulation is something like:
Person A: I know a Scotsman who puts sugar on his porridge.
B: That's impossible. Scotsmen don't put sugar on their porridge.
A: Well, my friend Angus McTavish, who was born in Scotland, puts sugar on his porridge.
B: Then he isn't a true Scotsman, because no true Scotsman puts sugar on his porridge.

What's interesting here is that B is no longer disputing A's claim that he knows a Scotsman who puts sugar on his porridge; rather, he is claiming that Angus McTavish, though possibly a Scotsman, is not a member of a the newly-defined set of True Scotsmen, defined as the subset of Scotsmen who never put sugar on their porridge. Since it's already stipulated that Angus McTavish is not a member of that set, then the conclusion is in fact correct within the definition specified - in fact, as the definition is tautological, it cannot be incorrect.

So the No True Scotsman fallacy isn't, I don't think, a fallacy at all. Rather, it's a means of derailing an argument, in which the person employing it goes off at a tangent to prove a tautology that has no relevance to the original claim.

Which, I suspect, makes all that I've written above an extremely complex and roundabout version of the No True Scotsman fallacy.

Dave

I agree that it's not technically a fallacy, and I think saying it's changing the conclusion (or changing the question, really) is accurate too.

I do think it's akin to the problem in the ontological argument (though as an exclusion in the definition rather than an inclusion). As you say, it's a tautology which is just question begging.

If you define God as something that must have the characteristic of existence (to be perfect), then concluding that God exists is logical, but meaningless. It's just if P, then P.

As you point out, the Scotsman issue is slightly more complicated because it's unclear what conclusion we're after, but there is still the circular reasoning (if you accept that part of the definition of Scotsman is having the characteristic "never does x").

Also, as you correctly point out, Moving the Goalposts isn't actually a fallacy either--in the sense of resulting in an invalid argument. It's just changing the burden of evidence after the original one has been satisfied. I liken the No True Scotsman "fallacy" to it as changing the question after the original one has been proven.

[JoeTheJuggler]

I must disagree... here's another way of looking at it... it's equivocating on Scotsman, changing the meaning from someone who was born in Scotland, to someone who was born in Scotland and conforms to ad-hoc norms. So it could be a subset of fallacy of equivocation.

It's clear in the example (especially since it is claimed to be 'impossible') that the porridge Nazi wants the meaning of Scotsman initially to be taken only as "Born in Scotland" without further examination, and then backpedals when given a counterexample.

I'm not sure I'd go along with that simply because the No True Scotsman argument is so specialized and does involve changing around the structure of the argument entirely.

[ETA: Also equivocation involves ambiguous words. I think redefining a term in an unconventional way is something else. The term "Scotsman" doesn't seem to me ambiguous. The term "true Scotsman" is an explicit attempt to redefine the term or, rather, to change the question to refer to a new term which is defined as a subset of the term in the original question.]

Would you consider the flaw in the Ontological Argument for the existence of God to be equivocation as well? (Starting with a conventional definition of God, and then changing the definition to include the characteristic of "must exist" in order to beg the question?) The conventional criticism is to point out that you can't predicate existence, but I would generalize that to say you can't predicate a characteristic if that characteristic is the question. And you logically can, but the argument is then simply begging the question.

Another point I've heard raised is that complaining that an argument is circular is really a matter of judgement or degree, since, at some level the conclusion really is already present in the premises in a valid argument. But there really does seem to be something different about arguments like the ontological argument, such that even someone not schooled in logic and reason, upon first hearing it, senses that someone's trying to pull a fast one!

[Fredrik]

For example, someone says "I don't like that serial killer, killing is always wrong."
Someone else responds, "But this war hero killed a lot of people too."
The first says "Seriously? You're likening this war hero to a serial killer?"

No, they're not. They're just refuting the rule that killing is always wrong.
This one is so annoying. It makes it impossible to argue with some people. They always make some absurd argument, and whatever you say to prove that the argument is flawed, they get outraged over the "comparison". It drives me crazy.

[Bram Kaandorp]

This one is so annoying. It makes it impossible to argue with some people. They always make some absurd argument, and whatever you say to prove that the argument is flawed, they get outraged over the "comparison". It drives me crazy.

Same here. It's so hard to make a comparison without someone who assumes that the comparison is meant to be seen over the whole spectrum.

When I compare embryo sex selection to the Chinese one child per family program, I don't mean that there will be many deaths, and I certainly don't trivialise the Chinese situation. I merely point out that it will have adverse effects.

And just try to explain that in a conversation without bringing it to a grinding halt. It will always sound like justifying yourself, which seems to give the person you're speaking with more confidence in the conversation.

I would love to just sit with someone, and just talk all the time, take as much time as needed to clarify something, and not be interrupted because of something which I was going to address any way.

[JoeTheJuggler]

This one is so annoying. It makes it impossible to argue with some people. They always make some absurd argument, and whatever you say to prove that the argument is flawed, they get outraged over the "comparison". It drives me crazy.

I've got two more good examples of this.

"Abortion stops a beating heart."
"So does slaughtering a cow to make beef."
"You can't compare a child to a cow!"
"I wasn't. Try this: some kinds of open heart surgery stop a beating human heart, does it follow from that that it's morally wrong?"

And worse was the fallout when one of the skeptical magazines said exactly what I'd been saying about putting the 9/11 terrorism attacks into perspective: more Americans are killed in traffic fatalities about every 6 weeks than were killed on 9/11. I thought this was especially important when tens of billions of dollars went to Homeland Security to keep us safe from terrorism.

People right away--even the subscribers to the skeptical magazine--complained that you can't compare these deaths. My question is, is a death in the family due to an auto accident less tragic than a death from a huge newsworthy terrorism attack? Are the loss of those lives more important or significant than the loss of other lives? More importantly, does any of that somehow alter the cost/benefit analysis of what we spend our federal dollars on?

[gnome]

This one is so annoying. It makes it impossible to argue with some people. They always make some absurd argument, and whatever you say to prove that the argument is flawed, they get outraged over the "comparison". It drives me crazy.

Two possibilities:

1. Some people don't have the cognitive capacity to assimilate a hypothetical or an analogy. Abstract thinking, you know. Or they lose this ability when their emotions get involved. "Ain't nobody gonna suppose no hurt to George!"

2. They most certainly do, but rather than take back an overly broad statement, they take refuge in the confusion that can occur when you stop arguing about the subject and drift into a meta-argument about the analogy itself.

[Fredrik]

Two possibilities:

1. Some people don't have the cognitive capacity to assimilate a hypothetical or an analogy. Abstract thinking, you know. Or they lose this ability when their emotions get involved. "Ain't nobody gonna suppose no hurt to George!"

2. They most certainly do, but rather than take back an overly broad statement, they take refuge in the confusion that can occur when you stop arguing about the subject and drift into a meta-argument about the analogy itself.
Yes, I think that pretty much covers it. Option 2 is what you will get from a politician who chooses to throw the argument back in the opponent's face rather than try to refute it rationally. In a private conversation, I think we're almost always dealing with option 1. They always get emotional, but it's hard to tell if they stopped thinking rationally because they got emotional, or got emotional because they don't know how to think rationally.

I have some examples too. This is the best one from my personal experiences: I had been out to dinner with a friend, his girlfriend and some of their friends, and was walking to the subway with a couple that had been at the dinner. Somehow the conversation touched on the subject of prostitution, and the female half of the couple said something to the effect that paying for sex is an evil, abusive thing to do because these women don't really want to do what they're doing. I tried to explain that the first part of that doesn't really follow from the second, by saying that e.g. a cleaning lady may not like to do what she's doing for money either. By her logic, it would be morally wrong to hire a cleaning lady. She went completely ******* because I had "compared" having a boring job with being raped for money, and started accusing me of being a horrible, evil, rape-supporting *******. It was really unpleasant.

Here's an example from Real Time with Bill Maher: This was one of the first episodes of Real Time that I saw. Andrew Sullivan was one of the three in the Panel when Bill interviewed Noam Chomsky. Chomsky wasn't in the studio, he just appeared on a screen. When the interview was over, Sullivan went nuts because Chomsky had "compared" the Bush administration to Nazis. What he had actually done was to say that if we take the Nuremberg trials as the definition of what we mean by a "war crime", then these guys are war criminals. That's very different from saying that the Bush guys are as bad as the nazis. OK, there's certainly a hint about that in the argument, but if you have a problem with that, and you feel that the claim itself is wrong, then you should start by refuting the argument rationally and then say that Chomsky is an ******* for using a flawed argument that involves nazis just to make the viewers think of Hitler when someone mentions Bush.

[HansMustermann]

Actually, way I learned it, the No True Scotsman fallacy is a special form of circular reasoning.

1."All X are Y"
2."Well, Z is an X and not an Y"
3."Then Z is not a true X"

It's circular because the conclusion is turned into a premise to dismiss a counter-example, thus supporting the conclusion.

Basically the key there is that X=>Y is equivalent to !Y=>!X. The underlying argument in #3 above is basically "!Y=>!X" If Angus McTavish doesn't eat his porridge without sugar, then he isn't a True Scotsman. But that is equivalent to the original claim that "X=>Y", so basically the claim is used to support itself.

However it's not a fallacy when there is some other reason to dismiss the objection that Z is an X. E.g.,

"No eel can discharge electric shocks"
"Well, the electric eel can."
"Sure, but the 'electric ell' is not a true eel, but a species of knifefish."

... is not a No True Scotsman fallacy. The 'electric eel's' status as a true eel doesn't depend on conforming to the conclusion, but is simply how marine biologists classify them.

Or to use the porridge example, the following is not a No True Scotsman fallacy:

"No Scotsman puts sugar on his porridge."
"Well, I'm in the same WoW guild with some guy called McChirton who does put sugar on his porridge."
"Heh, I know that guy. He just calls himself McChirton online, but is a third generation American of Polish descent. I'm pretty sure that playing a dwarf with a Scottish accent doesn't qualify him as a true Scotsman."

[JoeBentley]

While it might not be the "According to Hoyle" definition, to me the "One True Scotsman" fallacy is when you adjust the definition of a group based on how you need to define it to win whatever argument you are in at the moment.

[gnome]

The One True Scotsman is Ian McIan the Redundant.

[Beelzebuddy]

Actually, way I learned it, the No True Scotsman fallacy is a special form of circular reasoning.

Wikipedia defines it (http://en.wikipedia.org/wiki/No_true_scotsman) as a special form of begging the question, exactly as others have been using it. Which is also the way I learned it, plus, hey, Wikipedia.

Also from that page:
For the practice of wearing a kilt without undergarments, see True Scotsman. (http://en.wikipedia.org/wiki/True_Scotsman)

[HansMustermann]

Well, begging the question is circular reasoning, so it would seem we're in agreement there.

[gnome]

Nuh-uh... TRUE circular reasoning is religious in nature.

[Taarkin]

"Ghengis Kahn was a Scotsman."
No he wasn't.
"GRAAH you used the no true Scotsman fallacy!"
No I didn't that's a strawman!

edit- here's an almost foolproof plan to test whether your post is a strawman: does it start with "So"? ex:
"Hamlet is overrated. I think The Tempest and Othello better capture the angst of a man torn between love and duty"
"So basically you're saying you're a gay pedophile who likes to kill babies with your best friend Adolph Hitler??!?"

[gnome]

Here's another one for straw men... based on Taarkin's idea (which is an interesting rule of thumb)

Sometimes a straw man argument is made from an extension of what the other person said... the debator knows that isn't what the other person intends, but challenges them to qualify their argument in order to avoid the straw argument.

An example:

A: "People who don't pay off their loans should go to jail until they're paid."
B: "So for people that have no way of raising more money, you're saying they should stay in jail forever?"
A: "Strawman! I never said that!"

It isn't really a straw man argument--it's B trying to get A to qualify their proposition. Or, sometimes, B isn't sure how far A wants to take it...

A: "All Muslims should be deported!"
B: "All, like every single one?"
A: "Yep, all of them."
B: "So you want to deport American citizens if they're Muslims?"
A: "What? When did I say that? Straw Man!"

[HansMustermann]

Well, yeah, things are more complex. Isn't that usually the case?

But I think that one should also be aware of extensional vs intensional contexts.

A) In an extensional or referentially-transparent context, you can substitute identicals: if X=Y and P(X) then P(Y). Best known as Leibniz's Law. E.g., if Batman is Bruce Wayne and Bruce Wayne has a butler named Alfred, then Batman has a butler named Alfred.

B) In an intensional or referentially-opaque context, Leibniz's Law doesn't apply. In fact, that's pretty much the definition of such a context. Usually just about anything that involves personal opinions, or stuff like "knows", "believes", "says", is intensional. So is just about any modal context.

E.g., Clark Kent is Superman, Lois Lane says she knows where Clark Kent lives, therefore Lois Lane says she knows where Superman lives... is quite false, actually. That's not at all what she's saying. As far as she knows or says, she doesn't know who Superman is or where he lives.

E.g., to use a proper masked man ;) Don Diego de la Vega is Zorro, Sgt. Pedro Gonzales is an enemy of Zorro, therefore Sgt. Pedro Gonzales is an enemy of Don Diego de la Vega... again is actually quite false. The sergeant is actually a friend of Diego, not having any idea that he's Zorro.

While that's not strictly a strawman -- it's actually the Masked Man Fallacy (http://www.fallacyfiles.org/illisubs.html) -- it sure looks that way to someone who doesn't even know, or thought it through that the two entities are identicals or even intersect.

So, yeah, if you're saying "so you're saying that", it's usually a good sign that it's a fallacy. Unless that person explicitly said exactly that.

I do understand that at times it's a very useful device to make someone realize the enormity implied by what they're saying, but strictly speaking it is a fallacy.

[daenku32]

Recent one I saw was in topic to acceptance of AGW. The author stated that IPCC report is sufficient to accept global warming, and a critical yelled (to his keyboard) that "argument from authority".

I don't think referring to peer review conclusion is a fallacy.

[Bram Kaandorp]

Recent one I saw was in topic to acceptance of AGW. The author stated that IPCC report is sufficient to accept global warming, and a critical yelled (to his keyboard) that "argument from authority".

I don't think referring to peer review conclusion is a fallacy.

In fact, saying that "it's an argument from authority" is in itself a logical fallacy. At least, if the person shouting logical fallacy is referring to the fact that it's peer reviewed.

[HansMustermann]

It's an argument from authority, if something is supposed to be true because of who said it. And peer review is cute as support of that fallacy, but still doesn't make it anything else than a fallacy. In fact, lots of peer reviewed authorities are wrong.

The geocentric epicycles system had had some two thousand years of peer review and consensus at the time of Galileo and it was wrong. Einstein's cosmological constant, at least how he originally set it, was coming from one heck of an expert and was peer reviewed, but even he himself later described it as his biggest mistake. Psycholanalysis of the Freudian kind had a heck of a lot of peer review and experts and universities, but now we know that Freud just made crap up, and that memories just don't work that way, and generally it all doesn't work that way.

Chiropractic has a heck of a lot of peer review, and even peer review boards consisting of other chiropractors, but it's still bogus. Acupuncture is subject to peer review by other acupuncture experts and authorities, but is still bogus. Astrology books are sometimes (though not always) reviewed by other astrology experts first, but we know that astrology is bogus anyway. Even ID has at least a dozen papers that have gone through a thorough peer review... by other ID-ers, and, make no mistake, there is no doubt among ID-ers that the irreducible complexity or NFL "theories" are correct and evolution is wrong. Too bad it's bogus too.

Ultimately whether something is true or false must follow from data and logic or maths, not from which authority said it and not from what authority reviewed it. Period.

In informal logic you are allowed to shortcircuit the whole checking the evidence, and just evaluate the credibility of the evidence and decide whose conclusion you'll just trust. But that doesn't mean it actually makes the conclusion true or false. It just means you're allowed to do a guess about whether you believe it to be right or wrong, which isn't the same thing as it actually being right or wrong.

[JoeTheJuggler]

Well. . . an argument from authority is inappropriate in certain circumstances:
When it's not necessary (when there is direct evidence one way or the other, an expert's opinion isn't despositive).
When the authority isn't an authority on the subject at hand (and for some subjects there can be no authority).
When the authority is not disinterested.

(This is taken straight from the Fallacy Files (http://www.fallacyfiles.org/authorit.html).)

I would add that the primary role of an authority is to summarize or characterize the result of a lot of inductive reasoning (experiments, studies, the literature in a field). It really is more like weighing evidence than a deductive logical proof.

So an appeal to misleading authority is inappropriate or "fallacious" in a different way than most of the fallacies we've been talking about. It does speak to the truth value of propositions (again in inductive reasoning) rather than merely point out that the conclusion doesn't validly flow from the premises.

[JoeTheJuggler]

RE: Strawman Fallacy

So, yeah, if you're saying "so you're saying that", it's usually a good sign that it's a fallacy. Unless that person explicitly said exactly that.
Amen. A stronger sign is when you invent quotations (words no one said) and put them in your opponent's mouth. Even worse is when you act out both sides of an argument in order to get your opponent to say things in just a certain way.

I point that this is pretty strong evidence you're arguing against a straw man over on the Politics subforums with alarming regularity. (Search the US Politics one using "I'll say it again" because I try to preface my comment with "I've said it before and I'll say it again" whenever I point this out.)

Despite how often we discuss and explicate fallacious reasoning on these forums, it's amazing that people still do this here, and relatively frequently. (I've got one person who seems to think if he just writes up the fake dialogue but doesn't use quotation marks that it's not arguing against a straw man.)

[Taarkin]

A lot of this can be avoided by just asking them if X is good instead of accusing them "So you think X is good?".
Even worse is preemptive strawmanning, where you ascribe an opinion to someone before they even say anything. Watch, gnome is going to come in here and say preemptive strawmanning is a good thing. :rolleyes:

Even worse is when you act out both sides of an argument in order to get your opponent to say things in just a certain way.
http://i.imgur.com/YnLdO.gif

[gnome]

I saw what you did there... :)

Really I'm not... your comment was the first time actually that I've thought about the wording in such situations. It is useful to consider if a different wording would encourage a more constructive response.

[ZirconBlue]

I have some examples too. This is the best one from my personal experiences: I had been out to dinner with a friend, his girlfriend and some of their friends, and was walking to the subway with a couple that had been at the dinner. Somehow the conversation touched on the subject of prostitution, and the female half of the couple said something to the effect that paying for sex is an evil, abusive thing to do because these women don't really want to do what they're doing. I tried to explain that the first part of that doesn't really follow from the second, by saying that e.g. a cleaning lady may not like to do what she's doing for money either. By her logic, it would be morally wrong to hire a cleaning lady. She went completely ******* because I had "compared" having a boring job with being raped for money, and started accusing me of being a horrible, evil, rape-supporting *******. It was really unpleasant.

So, where did you hide the body?

[Dave Rogers]

So, yeah, if you're saying "so you're saying that", it's usually a good sign that it's a fallacy. Unless that person explicitly said exactly that.

I do understand that at times it's a very useful device to make someone realize the enormity implied by what they're saying, but strictly speaking it is a fallacy.

It can be a good way of exposing a logical contradiction in someone else's argument, however. To take Gnome's example:


A: "All Muslims should be deported!"
B: "All, like every single one?"
A: "Yep, all of them."



A reasonable continuation might be to ask whether this includes Muslims who are US citizens, and, if not, to point out the contradiction and ask for a resolution.

Dave

[gnome]

In a sense that a strawman argument will take the form of attributing an argument to someone, "so you are saying..." definitely qualifies.

But to be a straw man it also has to be intentionally inaccurate. It being used as a legitimate inquiry, and giving the person the opportunity to distinguish their opinion from it, need not be straw man.

[JoeTheJuggler]

But to be a straw man it also has to be intentionally inaccurate.

I disagree. The intention is irrelevant.

ETA: If it were relevant, then "honest" stupidity and ignorance would be a defense against ever committing a strawman. (Also, if one unintentionally mischaracterizes a position, and then refutes that position, that refutation can't stand as a refutation against the actual position just because there was no intention to create a strawman.) And I put "honest" in scare quotes because it's a blurry line between laziness and intentional ignorance. But, logically, it's plain that intention is irrelevant. If you refute a mischaracterization of a position, the actual position is not refuted. This is the basic point of the strawman "fallacy".

[gnome]

You know, I think you're right. So I'll drop the word "intentionally" from my above statement. But I believe "So you are saying that..." need not automatically be suspected of strawman except that only arguments attributed to others could be a strawman.

It still needs to be actually a mischaracterization, and that phrasing doesn't really clue you in as to whether it is or isn't.

[JoeTheJuggler]

You know, I think you're right. So I'll drop the word "intentionally" from my above statement. But I believe "So you are saying that..." need not automatically be suspected of strawman except that only arguments attributed to others could be a strawman.

It still needs to be actually a mischaracterization, and that phrasing doesn't really clue you in as to whether it is or isn't.

I agree there. Asking for clarification is not arguing against a strawman.

Well. . not in itself. I've seen people proceed to purport to refute their made up version of an opponent's position in the guise of "Just Asking Questions".

But I stand by the observation that if you're making up fake quotes and putting those words into your opponent's mouth, it's pretty strong evidence that you're probably arguing against a strawman.

Asking for clarification is sometimes an excuse for doing just that. (Or even for attempting to trap your opponent into agreeing to a particular wording of their position that they didn't originally make just because that wording contains a weakness that the original wording did not.)

ETA: I guess that means I don't agree so much, because I think it *is* reasonable to be suspicious when someone says, "So you're saying that. . ." rather than responding to what the person actually said. If what they actually said is vague or ambiguous, pointing out that the wording is vague or ambiguous is a legitimate refutation of what they actually said. That person might then re-state their position without those problems, and then you can respond to that.

[ladmo]

I've got two more good examples of this.

"Abortion stops a beating heart."
"So does slaughtering a cow to make beef."
"You can't compare a child to a cow!"
"I wasn't. Try this: some kinds of open heart surgery stop a beating human heart, does it follow from that that it's morally wrong?"

And worse was the fallout when one of the skeptical magazines said exactly what I'd been saying about putting the 9/11 terrorism attacks into perspective: more Americans are killed in traffic fatalities about every 6 weeks than were killed on 9/11. I thought this was especially important when tens of billions of dollars went to Homeland Security to keep us safe from terrorism.

People right away--even the subscribers to the skeptical magazine--complained that you can't compare these deaths. My question is, is a death in the family due to an auto accident less tragic than a death from a huge newsworthy terrorism attack? Are the loss of those lives more important or significant than the loss of other lives? More importantly, does any of that somehow alter the cost/benefit analysis of what we spend our federal dollars on?
Excellent comments but it may not get play as it makes perfect sense.

[Taarkin]

PS: Analogies are NOT strawmen for ****'s sake.

[HansMustermann]

PS: Analogies are NOT strawmen for ****'s sake.

They can be, when they distort what was actually being said or weren't even much of an analogy to anything actually being said in the first place. I've seen the "it wasn't a strawman, I was just giving an analogy" defense before, and actually just about invariably it was an actual strawman.

[JoeTheJuggler]

Speaking of analogies, recently I've been accused of making an argument by faulty analogy (on two separate occasions) when I wasn't making an analogy of any kind. I was actually applying the principle my opponent expressed to an example of exactly what the principle was meant to apply to. (It was actually reductio ad absurdum--proving that a premise is false by showing that it logically leads to an absurdly false conclusion. And someone else accused me of committing the argumentum ad absurdum fallacy!)