Join 3,501 readers in helping fund MetaFilter (Hide)


You are wrong because you use the fallacy wrong
March 6, 2011 3:53 AM   Subscribe

Across the internet, over the shouts of "First!" can be heard thecries of "Ad hominem!" shamefully and ignorantly used against disagreeing attacks most who call this out fall victim of the dreaded Ad Hominem Fallacy Fallacy

Coming soon: The ad hominem fallacy fallacy fallacy

Seriously though, if you do this your point is invalid.
posted by AndrewKemendo (107 comments total) 22 users marked this as a favorite

 
Yeah, well, you're ugly.
posted by Faint of Butt at 4:09 AM on March 6, 2011 [4 favorites]


Ad Hominem phallus-y fallacy occurs when your opinions are discounted simply because you're the kind of guy who thinks with his dick.
posted by tigrefacile at 4:16 AM on March 6, 2011 [1 favorite]


"Yeah, well, you're ugly."

Oh yeah, well, you're a frigging weasel.
posted by tomswift at 4:22 AM on March 6, 2011


So the entire thing comes down to taking offense at having one's abusive behavior incorrectly labeled? Boy, winning that argument is a tiny, tiny victory if ever there was one.
posted by Linda_Holmes at 4:49 AM on March 6, 2011 [12 favorites]


It's not about winning arguments; it's about logical accuracy. And that's something that really, really matters if you want to have a discussion about a difficult topic.
posted by meese at 4:53 AM on March 6, 2011 [5 favorites]


This all begs the question of what the author would like us to do with the information.
posted by TypographicalError at 4:57 AM on March 6, 2011 [22 favorites]


The author's smartass style is really off-putting for me. I scanned a few of the other articles from the index, and yeah, the guy's pretty well full of himself. Sorta reminds me of one or two people here at Mefi, hate to say.
posted by flapjax at midnite at 4:58 AM on March 6, 2011 [3 favorites]


Never argue with an idiot. They pull you down to their level, then beat you with experience.
posted by netbros at 5:06 AM on March 6, 2011 [19 favorites]


Just sayin flapjax, but both doctors who have operated on me were really, really smug. The kinda guy who would be off-putting at a party. But I'm not sure I want the couched, meek, and/or shy person who went to school for 24 years putting a knife in me to be anything less than absurdly confident. Sometimes it's ok for people to be smug, if they really know their stuff, to me.
posted by efalk at 5:12 AM on March 6, 2011 [2 favorites]


Yeah, well, you're ugly.
posted by Faint of Butt at 12:09 PM on March 6


Yeah, well you walk so funny it's like you're on a slippery slope. Also, you smell of red herrings.
posted by Decani at 5:16 AM on March 6, 2011 [6 favorites]


Yeah, well, you're uglier than a person who also advanced a similar argument.
posted by StickyCarpet at 5:21 AM on March 6, 2011 [9 favorites]


This rant is a perfect example of how someone can be so engaged with an absolute truth that they disconnect entirely from functional reality. Clearly, ad hominem fallacies are not difficult to understand, and are used frequently. Especially on mefi. In fact, although I know it is a faux pas these days to reference self-reference, it would not be a huge leap of faith to consider the argument that I am making right now an ad hominem fallacy.
posted by flyinghamster at 5:21 AM on March 6, 2011


It's not about winning arguments; it's about logical accuracy. And that's something that really, really matters if you want to have a discussion about a difficult topic.

Understood. Agreed. They're not technically ad hominem arguments. They're technically abusive personal attacks. I'm just saying that to me, "I'm not illogical; I'm just rude and personally abusive" is a very tiny victory. That's all.
posted by Linda_Holmes at 5:24 AM on March 6, 2011 [1 favorite]


"I'm not illogical; I'm just rude and personally abusive"

Let's put itching powder in his surgical gloves.
posted by StickyCarpet at 5:27 AM on March 6, 2011


You can make a perfectly reasoned argument and still add, as an aside, that somebody is a dick. Sometimes people are dicks.
posted by Astro Zombie at 5:31 AM on March 6, 2011 [11 favorites]


Oh yeah? Well, meta!
posted by iamkimiam at 5:31 AM on March 6, 2011


I wonder how long it will be before we need a webpage on the Ad Hominem Fallacy Fallacy Fallacy.
posted by Dr. Eigenvariable at 5:34 AM on March 6, 2011


also, I have to add for the sake of sanity -

pointing out that someone is making an ad hominem fallacy is not an ad hominem fallacy (unless you make it one)


and to you logicians who still think there is a nice logical pot of gold at the end of the rainbow, I recommend you look up the Münchhausen Trilemma
posted by flyinghamster at 5:34 AM on March 6, 2011


A lot of times, "That's an ad hominem attack!" is used to shut down an argument. If you say that someone's giving an ad hominem, then you don't have to engage with what they're actually saying. And then, if you want to support your argument against the complaint, you get sidetracked into a discussion about ad hominemity. Calling out ad hominems when they actually are ad hominems is perfectly appropriate, yeah, but mistaken call-outs are just flashy distractions.

The reason ad hominems are bad is that they move attention away from the matter actually at hand. "You're a drunk, so you're wrong about thermodynamics." Suddenly, we're talking about alcoholism rather than thermodynamics. This is pointing out that incorrect accusations of making ad hominems can have the same effect. "Psh, that's just an ad hominem, so obviously you didn't say anything important." Equally toxic to a conversation.

It's not cool to be personally abusive, either. And if you're making a point in such a way that it can easily be confused for an ad hominem, you're failing at reasonable discourse. But, then again, there've been plenty of times in Mefi history where saying, to quote the article, "I've been following this thread for a while, and I hate to say it, but you're being an asshole. You're really taking this whole thing too personally, and seriously misconstruing everyone else's arguments" is perhaps the most appropriate response.
posted by meese at 5:36 AM on March 6, 2011 [11 favorites]


So in other words, using the cry of ad hominem is a way to undermine the arguments of another person by discrediting that person instead of their arguments. If only there were a word to describe this situation...
posted by Mister_A at 5:40 AM on March 6, 2011 [2 favorites]


I recommend you look up the Münchhausen Trilemma
posted by flyinghamster


You, ,sir, have shown yourself to be an infantile advocate of coherentism by accepting the "circular" horn of the trilemma.
posted by StickyCarpet at 5:41 AM on March 6, 2011


The author's smartass style is really off-putting for me.

Me, too. Also, I began reading with interest, thinking that I might find out that mislabelled ad hominem attacks are more common than I thought or that I misunderstood what "ad hominem" means. They're not and I didn't.

Maybe it's a big issue elsewhere on the internet, but as I wander around MeFi, in general, when a serious argument gets called "ad hominem," it actually is. The "trained zoologist" example struck me as the most familiar.

So I know he wants us to get it right, but I'm not sure where he see people going so horribly, horribly wrong. We call each other assholes here at times, but I'm not sure that most would say those are "ad hominem arguments." I think most people already get it. It's a lot of effort to go to for something that's not really a widespread problem.
posted by Miko at 5:45 AM on March 6, 2011


B: "Well, you're a rodent and a weasel, so there goes your argument."

Well, you're Mama's a rodent and a weasel, so there goes your argument.
posted by StickyCarpet at 5:46 AM on March 6, 2011


I think Metafilter tends to have a larger problem correctly identifying straw men, rather than ad hominems..... But it could just be that I have clear memories of, like, two occasions where someone claimed something was a straw man when it totally wasn't.
posted by meese at 5:48 AM on March 6, 2011


His example is merely an supported counter argument followed by a direct attack on the reasoning and training in logic of the original poster. It is an ad hominmem attck because the assertion boils down to weasels are too mammals. Having demolished this ridiculous claim let's critique his views on online dating profiles and humanities faculty, and reading list, not to mention his web design skills..
posted by humanfont at 5:49 AM on March 6, 2011 [1 favorite]


And if you're making a point in such a way that it can easily be confused for an ad hominem, you're failing at reasonable discourse.

This is a better statement of what I was trying to say.

mistaken call-outs are just flashy distractions

Absolutely, they are. As are most of the comments those mistaken call-outs are aimed at, I find. I do completely get and agree with the bit about "calling something ad hominem is a way of not dealing with the argument." But I struggle to be sympathetic, because if you really want people to engage your argument and you're really trying to advance the discussion, it's a lot easier to leave out the "fuck you, you're clearly an ignorant moron" part than it is to try to talk people into addressing the substance of what you're saying and ignoring the personal insult on the basis that it isn't part of your argument.
posted by Linda_Holmes at 5:50 AM on March 6, 2011 [1 favorite]


What I want to know, is does this post, or at least the comments in it which evaluate the science by the quality of the journal that publishes it, commit the ad hominem fallacy? Or, if you consider the social nature of science, in which we rely on peer review to weed out the bad stuff, is judging the quality of the publication an article appears in a valid form of criticism?
posted by Obscure Reference at 5:55 AM on March 6, 2011 [1 favorite]


Having demolished this ridiculous claim let's critique his views on online dating profiles and humanities faculty, and reading list, not to mention his web design skills.

Hooooooo-eeeeeee. Also, don't miss fat whiners and politeness,
posted by Miko at 5:59 AM on March 6, 2011 [1 favorite]


I read this article as far as the first few non-examples, and the thought I kept having was, "Yeah, you would say that."
posted by Kirth Gerson at 6:00 AM on March 6, 2011 [1 favorite]


Christ, what an asshole.
posted by Emperor SnooKloze at 6:04 AM on March 6, 2011 [1 favorite]


Has anybody ever tried to market jars of pickled red herrings, cos that's be cool.
posted by jonmc at 6:05 AM on March 6, 2011


Has anybody ever tried to market jars of pickled red herrings, cos that's be cool.

Well, it's a band name, so there's that.

and "that's be cool" is my new favorite phrase
posted by flapjax at midnite at 6:09 AM on March 6, 2011


Has anybody ever tried to market jars of pickled red herrings, cos that's be cool.
posted by jonmc at 2:05 PM on March 6


I don't think they'd be popular, therefore your argument is false.
posted by Decani at 6:09 AM on March 6, 2011


I'm not sure I want the couched, meek, and/or shy person who went to school for 24 years putting a knife in me to be anything less than absurdly confident.

Absurd confidence is a sign of extreme ignorance, since it discounts two unequivocal truths:
  1. Fate is a bitch
  2. There is always someone better than you
Give me quiet confidence—or better, honest humility—any day of the week.

If you're a doctor, I already take for granted the decades of school. If you've got a diploma from a good university on your wall, I naturally assume you have a modicum of kick-ass inside you, because good medical schools are exceptionally difficult to get into. You must have studied your ass off, and/or had a smattering of natural talent. And the obvious, the fact that you're still practicing, means you most likely still have your license to practice, which means (probably) no horrendous mistakes resulting in egregious loss of life or the creation of hideous monstrosities that now roam the hospital catacombs. You don't need to be an outrageously overconfident asshole to prove anything.
posted by Civil_Disobedient at 6:10 AM on March 6, 2011 [8 favorites]


What I want to know, is does this post, or at least the comments in it which evaluate the science by the quality of the journal that publishes it, commit the ad hominem fallacy? Or, if you consider the social nature of science, in which we rely on peer review to weed out the bad stuff, is judging the quality of the publication an article appears in a valid form of criticism?

Interesting question and maybe I'll take a stab at that though I'm no great logician. I would say yes, I think it's an example of poisoning the well, which is a version of an ad hominem argument. It's like,

All articles published in the journal Science Now! are crap. Dr. Hoover's article was published in Science Now!, therefore Dr. Hoover's article is crap.

Like the ad hominem, the argument doesn't engage the substance of Dr. Hoover's article.

As I think about, I think this is actually why peer review works - because it doesn't use this kind of argument. If Science Now! got a new editor and suddenly started practicing strong peer review, the reviewers would engage the argument in Dr. Hoover's article, instead of saying "pbbbffft, Science Now! Anything in there is crap."

But in practical terms, knowing that an informed peer community is not reviewing Dr. Hoover's article is a perfectly fine reason to remain unconvinced of Dr. Hoover's conclusion until such review happens, there or elsewhere.
posted by Miko at 6:10 AM on March 6, 2011


I thought ad hominem and Godwin were just waiting for gay marriage to be legalized so they could hook up.
posted by localroger at 6:15 AM on March 6, 2011 [1 favorite]


it was a typo, flapjax, but go bananas.
posted by jonmc at 6:21 AM on March 6, 2011


It's not about winning arguments; it's about logical accuracy.

No, it's about winning arguments.
posted by LordSludge at 6:24 AM on March 6, 2011 [2 favorites]


it was a typo, flapjax, but go bananas.

Oh, I had assumed as much, jonmc, with the S located right next to the D and all. And I am definitely going bananas.
posted by flapjax at midnite at 6:25 AM on March 6, 2011


No, it's about winning arguments.

Unless you're Charlie Sheen, and then it's just about winning.
posted by flapjax at midnite at 6:26 AM on March 6, 2011


Well, whatever you think of ad hominem, this post is a personal attack on grammar.
posted by Malor at 6:26 AM on March 6, 2011 [1 favorite]


I am so fucking sick of Charlie Sheen.
posted by jonmc at 6:27 AM on March 6, 2011 [2 favorites]


this post is a personal attack on grammar.

Levae family out of this.
posted by jonmc at 6:27 AM on March 6, 2011


Linda_Holmes, it appears you and I are in perfect agreement!

The Dr. Holmes stuff could be seen as poisoning the well, or it could be seen as the genetic fallacy ("That's the same journal that published [something I now forget but which was discredited]! BAH!").

Really, a lot of the terms for different types of fallacies are so closely associated, and depend so much on slight nuances in reading, that multiple identifications are usually possible. The different names all have distinct meanings, yes, but they're very close.
posted by meese at 6:30 AM on March 6, 2011


And that, by the way, is why weak-willed teachers hate grading fallacy-identification tests so, so much.
posted by meese at 6:31 AM on March 6, 2011


First!
posted by Artw at 6:31 AM on March 6, 2011 [2 favorites]


Levae family out of this.

Hey, what have you got against the Levae family?

let me assure you that riffing on your typos is NOT an ad hominem attack
posted by flapjax at midnite at 6:34 AM on March 6, 2011


riffing on your typos is NOT an ad hominem attack

no, it's an ad himonem atack
posted by jonmc at 6:37 AM on March 6, 2011 [4 favorites]


ad himonem atack

jonmc, if anyone attacks you, just let me know and I'll get my dog and sic himonem.
posted by flapjax at midnite at 6:41 AM on March 6, 2011 [7 favorites]


I just came in here to add homonyms.
posted by Devils Rancher at 6:45 AM on March 6, 2011 [5 favorites]


Yo' Gramma is a straw man.
posted by StickyCarpet at 6:49 AM on March 6, 2011 [1 favorite]


Awful typo. Help requested.

Aw, come on, it's in the spirit of recent thread developments! Perfect, really! I mean, prefect.
posted by flapjax at midnite at 6:55 AM on March 6, 2011


Are there really that many ad hominem arguments on the internet? I think it is just confirmation bias.
posted by TedW at 6:58 AM on March 6, 2011 [1 favorite]


The author's smartass style is really off-putting for me. I scanned a few of the other articles from the index, and yeah, the guy's pretty well full of himself. Sorta reminds me of one or two people here at Mefi, hate to say.

This thread is so meta.
posted by yourcelf at 7:07 AM on March 6, 2011 [2 favorites]


Ralph Kramden is here to ad hominemhominemhominem....
posted by jonmc at 7:07 AM on March 6, 2011 [4 favorites]


This definition of the ad hominem fallacy relies on knowing the intent of the user. On the internet, this is very difficult. In any case, attacking the character of the person you're arguing with is an ad hominem attack, just not an ad hominem fallacy. "Ad hominem" means "on the person".

Pointing out that you're attacking someone when you should be attacking their arguments strikes me as a fine thing to do.
posted by LogicalDash at 7:11 AM on March 6, 2011 [1 favorite]


It's not about winning arguments; it's about logical accuracy.

Almost nothing interesting that people talk about is or can usefully be constructed as a strict logical form. We're all limited by time, limited by training and access to raw data to verify empirical claims, and limited by the (non)existence of data which can unambiguously answer questions.
Even if we were discussing an interesting mathematical proof, very few people would be able to follow it closely enough to know if it contained an error. Strictly, it's not correct to say that something published in JoC is bunk without reading. It's just a very useful approximation, especially given that we would rarely if ever be able to say that claims in any paper are strictly true.

Just sayin flapjax, but both doctors who have operated on me were really, really smug. The kinda guy who would be off-putting at a party. But I'm not sure I want the couched, meek, and/or shy person who went to school for 24 years putting a knife in me to be anything less than absurdly confident. Sometimes it's ok for people to be smug, if they really know their stuff, to me.

There's an extent to which surgery makes you that way (I just took somebody apart and put them back together), but many non-insane surgeons put on that air to reassure patients and staff.
posted by a robot made out of meat at 7:30 AM on March 6, 2011 [1 favorite]


Sometimes it's ok for people to be smug

Confident, sure. Smug? No way. It's one of the most repellent personality traits.
posted by jonmc at 7:32 AM on March 6, 2011 [1 favorite]


There's nothing wrong with questioning a scientific conclusion based on where it was published. Logical fallacies only really come into play if you're arguing about whether a conclusion logically follows from given premises. If you personally have no ability to check the premises (true of most science papers) or to check the logical steps in reaching the conclusion (true of most math papers) then you have to decide whether to believe the conclusion based on how much you trust the people who did check.
posted by jhc at 7:33 AM on March 6, 2011


Yes, it is strictly speaking invalid to argue that the science is bad based on the journal that published it. But that doesn't mean such reasoning isn't a useful heuristic.

(We have no valid argument that the sun will rise tomorrow. That doesn't mean it won't.)
posted by voltairemodern at 7:42 AM on March 6, 2011 [2 favorites]


Hooooooo-eeeeeee. Also, don't miss fat whiners and politeness,

Wow. that 'fat whiners' article is. Wow.
posted by device55 at 7:51 AM on March 6, 2011


Almost nothing interesting that people talk about is or can usefully be constructed as a strict logical form. We're all limited by time, limited by training and access to raw data to verify empirical claims, and limited by the (non)existence of data which can unambiguously answer questions.

And not only that, but the powers of logic alone are fairly narrow. Logic can't point the way to social solutions or dictate human behavior. It's just a useful tool for carefully examining and evaluating certain kinds of statements to see whether they support a proposed conclusion or not. It can't tell you how to apply conclusions or whether it's a good idea to do so. Most of our argumentation here isn't that systematic and doesn't stick to pure logic, and that's fine. The style of discussion people use should relate somehow to the outcome you want from the discussion, and sometimes that's emotional and persuasive rather than logical and rationale. And sometimes, yes, it's calling an asshole an asshole.
posted by Miko at 7:56 AM on March 6, 2011 [1 favorite]


What I want to know, is does this post, or at least the comments in it which evaluate the science by the quality of the journal that publishes it, commit the ad hominem fallacy? Or, if you consider the social nature of science, in which we rely on peer review to weed out the bad stuff, is judging the quality of the publication an article appears in a valid form of criticism?

If you say "the article is wrong, solely because it's in a non-peer-reviewed journal," that's ad hominem.

However, if you say "I remain unconvinced of the article's conclusions, because it's in a non-peer-reviewed journal," I think that's OK.
posted by DevilsAdvocate at 8:01 AM on March 6, 2011 [1 favorite]


However, if you say "I remain unconvinced of the article's conclusions, because it's in a non-peer-reviewed journal," I think that's OK.

That's right - and it's OK to remain unconvinced, because it doesn't logically follow. If you can remain unconvinced, there's a problem in the argument that sought to convince you - in this case, the ad hominem nature of the statement.

I totally agree that of course, it's useful shorthand to be able to make sweeping characterizations of content based on reputation. But that shorthand is based on a set of assumptions. I'm not critiquing it by saying so - this is generally how people operate, since it's impossible to verify everything - we have some degree of trust that reputations are supported. And you have to assume something to start any argument, even if it's just that observed phenomena are real, or whatever.
posted by Miko at 8:21 AM on March 6, 2011


I think there's another issue at play here. While ad hominem does not logically refute a claim, it can be (and sometimes should be) correct.

Consider the following:

A: The Lurgi Institute shows in their latest paper that global warming is unlikely to have a negative effect on humanity for at least the next 100 years, contrary to other published research.
B: The Lurgi Institute gets funding exclusively from oil companies and has been criticized heavily in the past for sloppy methodology and just plain dishonest research.

B's response is (a) ad hominem and (b) right on target. No, it does not logically refute (or even reference) A's argument, but in the real world I'm not going to waste time on someone who has proven to be dishonest in the past and has a likely bias. Yes, an actual attack on the paper would be better, but I live in the real world and I don't have an infinite amount of time and, seriously, the Lurgi Institute? Buncha hacks.
posted by It's Never Lurgi at 8:22 AM on March 6, 2011


Of course, no one has that kind of time, which is where the assumptions and acceptances of probability/likelihood come in - and we do that every day. But a "real world" argument doesn't make the ad hominem attack "correct" in a logical sense. In other words, it might be right, but it's not logical argument. This is one of those instances I was talking about where logic may not point the way to a morally appropriate conclusion.

To refute the argument with logic, you have to refute A, "global warming is unlikely to have a negative effect for 100 years," with a response that shows global warming IS likely to have a negative effect. Anything else doesn't logically dismiss the argument they've made.
posted by Miko at 8:25 AM on March 6, 2011 [2 favorites]


Yeah, oftentimes, the distinction between a good argument and a bad argument really depends just on intention.

So, "This was published in journal X, so I don't buy it." Can be interpreted in (at least) two ways:
1) "This was published in journal X, and journal X has a history of publishing poorly-reviewed, flawed science. Therefore, this too must be wrong." That's the genetic fallacy.
2) "This was published in journal X, and journal X has a history of publishing poorly-reviewed, flawed science. Therefore, I'm withholding assent to this until I can see some better analysis of its evidence." That's just inductive reasoning leading to an appropriate and appropriately humble conclusion.

This is why the principle of charity is so important when trying to cash out arguments. But, at the same time, the principle of charity, itself, can be taken to an unrealistic extreme.
posted by meese at 8:32 AM on March 6, 2011 [1 favorite]


I think the thing is that the ad hominem attack in my example does at least give some reason for doubting the validity of the argument (even if, again, it's not remotely a logical refutation).

Consider:

A: The Lurgi Institute believes that global warming is not a problem
B: The Lurgi Institute? They are funded entirely by oil money. They are full of crap.

A: The Lurgi Institute believes that global warming is not a problem
B: The Lurgi Institute? Didn't the head of that get charged with massive bank fraud? They are full of crap.

Both ad hominem, but I don't think that in the real world you'd treat them equally. The first one, well... you have to admit that there is a point there. The second one is just abuse. Anyway, I was acquitted.
posted by It's Never Lurgi at 8:33 AM on March 6, 2011 [1 favorite]


It is possible to make an absolutely sound argument whilst also quite correctly pointing out that one's opponent has a complexion like a plate of dried maize kernels. This is known as an ad hominy attack.
posted by Decani at 8:40 AM on March 6, 2011 [5 favorites]


No, It's Never Lurgi, your first example probably shouldn't be considered an ad hominem. The point, there, is "Without evidence to the contrary, there's good reason to assume that someone funded by oil money would make biased claims about global warming. Since the Lurgi institute is funded by oil money, and since there is no evidence to the contrary, it is reasonable to assume they are full of crap."

Of course, here, I'm appealing to a charitable reading of what you put. You could've meant, "My father was insulted by an oil man once, so I frikken' distrust anyone associated with oil money. The Lurgi Institute can go screw itself with its studies!" In which case, yeah, there's ad hominem all over that.

The real problem is that arguments presented in ordinary English almost always have implicit premises, where the inference being made depends on something not explicitly spelled out. We generally assume that the people we're talking to can follow what we mean without everything being so exactly spelled-out, but the result that the way arguments are actually expressed verbally or in writing only rarely contain all the information intended to be used in the inference.

... And, really, this is another reason that calling out an interlocutor for fallacious reasoning can be an under-handed move. It's very, very easy to make accusations about fallacies if one's determined to be an uncharitable jackass.
posted by meese at 8:47 AM on March 6, 2011 [1 favorite]


n the real world you'd treat them equally.

I don't know that I percieve much difference, but what I am trying to say is that "in the real world," not all conclusions are logical conclusions. In your examples, it seems like the conclusions you reach are smart and fine. But the underlying structure is corrupt - if you changed the premise to a pro-gay or pro-progressive statement, for instance, the same sorts of ad hominem attacks could be used and also wouldn't lead to logical conclusions.

A: The Foundation for Equal Rights has issued a report stating that children of gay marriages grow up unharmed by same-sex parenting.

B: The Foundation for Equal Rights? They are funded by a bunch of donations from wingnut liberals. They are full of crap.

or

B: The Foundation for Equal Rights? Isn't the head of that a former abusing priest? They are full of crap.

Many would leap to point out the logical ad hominem fallacies in these arguments. They just aren't good logical arguments, they don't address the argument in A, no matter what the terms of the premise. Can they still be right in moral principle? Yes, but it's fair to say that the conclusions, then, are best treated as tentative pending stronger arguments.
posted by Miko at 8:48 AM on March 6, 2011


This is known as an ad hominy attack.

We try not to get that gritty around here.
posted by jonmc at 8:48 AM on March 6, 2011 [1 favorite]


"That's so arrogant."
posted by 0xdeadc0de at 8:50 AM on March 6, 2011


I don't know about it's application to the interweb, but on the positive side, this may prove helpful in explaining to my four year old why "I hate you" is not a cogent response to "I'm turning off The Phantom Menace because it's your bedtime".
posted by TheShadowKnows at 8:52 AM on March 6, 2011


That guy writes like Hitler.
posted by Meatbomb at 9:02 AM on March 6, 2011 [5 favorites]


The fallacy the post describes sounds a lot like the generalized version of The Tone Argument - while his examples show the person being an abusive ass, there's plenty of cases where someone isn't being abusive and basically being hit with the ad hominem of being accused of an ad hominem.
posted by yeloson at 9:05 AM on March 6, 2011 [2 favorites]


The reason ad hominems are bad is that they move attention away from the matter actually at hand.

Agreed.

And while I think it's worthwhile to know the formal definition of "ad hominem," in real-life conversations, true ad hominems and faux one tend to have the same effect:

A: All men are moral; Socrates is a man; therefor socrates is immortal.
B: Well, since you're an asshole instead of a man, you have no way of knowing anything about the subject!"
A: Hey! YOU'RE the asshole!

----

A: All men are moral; Socrates is a man; therefor socrates is immortal.
B: Your logic is faulty. The conclusion that follows is "Socrates is MORTAL," you asshole!
A: Hey! YOU'RE the asshole!

-- People tend to get defensive when they're called names.
-- Everyone knows people tend to get defensive when they're called names.
-- When people get defensive, they rush to defend themselves rather than continuing to discuss the topic.
-- Everyone knows that when people get defensive, the rush to defend themselves rather than continuing to discuss the topic.
-- Therefor, by calling someone a name -- whether in the form of an ad hominem or not -- you are purposefully trying get them to stop talking about the topic.

The reason PERSONAL ATTACKS OF ANY KIND [my edit] are bad is that they move attention away from the matter actually at hand.
posted by grumblebee at 9:09 AM on March 6, 2011 [1 favorite]


I'm assuming it's intentionally ironic that he starts right out by invidiously characterizing the people who like to perceive ad hominem arguments and quickly moves on to characterizing their speech. ("Those who are quick to squeal "ad hominem".)
posted by BibiRose at 9:10 AM on March 6, 2011


QI [here in the U of K] kinda covered this ground with Reductio Ad Hitlerum which, in turn, derives from Godwin's Law
posted by Incubus.exe at 9:40 AM on March 6, 2011


I just came in here to add homonyms.
posted by Devils Rancher


I thought you came in her to add homunculi.
posted by StickyCarpet at 9:48 AM on March 6, 2011


Anyway, I thought at this point I'd just make the observation that some people use "Godwin" in the way this guy complains about the use of ad hom - to simply close down an argument without addressing the substance of it, or to falsely imply it can have no substance simply because Hitler has been mentioned.

Analogies featuring Hitler and/or Nazis can be perfectly sound and powerful ways of making a valid logical point.

And now, back to the puns and funs!
posted by Decani at 9:57 AM on March 6, 2011 [1 favorite]


He did ignore the less-common usage of "ad hominem fallacy," which is usually discounted as an appeal to emotions now. But that's what "to the man" originally applied to, when you're making arguments that follow the "You care about justice, right? or similar form.

But yeah, my general experience is that people who link to those fallacy pages in arguments tend to be pedantic dicks.
posted by klangklangston at 9:58 AM on March 6, 2011 [1 favorite]


The reason PERSONAL ATTACKS OF ANY KIND [my edit] are bad is that they move attention away from the matter actually at hand.
posted by grumblebee at 5:09 PM on March 6


This is perfectly true, assuming that your only aim is to concentrate wholly on the matter in hand as opposed to, say, causing merriment and guffaws in onlookers.
posted by Decani at 9:58 AM on March 6, 2011


I don't know about it's application to the interweb, but on the positive side, this may prove helpful in explaining to my four year old why "I hate you" is not a cogent response to "I'm turning off The Phantom Menace because it's your bedtime".

The kind of person who would let his children watch the prequels would say that.
posted by naoko at 10:16 AM on March 6, 2011


New term for attacks that are personal but not precisely ad hominem in nature? Ad hominy.
posted by dixiecupdrinking at 10:34 AM on March 6, 2011


A lot of the stuff on this guy's site is distasteful at best, and stupid and hateful at worst, but I did enjoy his take on Ender's Game.
posted by dersins at 11:05 AM on March 6, 2011 [2 favorites]


You're all silly billies.
posted by biochemist at 11:09 AM on March 6, 2011


We are silly billies, but that does not have any bearing on the validity of our arguments.
posted by Mister_A at 11:20 AM on March 6, 2011 [2 favorites]


Analogies featuring Hitler and/or Nazis can be perfectly sound and powerful ways of making a valid logical point.

And sometimes evoking Hitler is the ONLY way to make the point. (Well, you could probably use Pol Pot, Stalin or Darth Vader, but you have to use some Hitler-like figure.)

This usually happens when someone makes an ethical or pragmatic claim about how people should act or be treated. If we're going to discuss this claim in a rational way, it's vital that we understand how universally the claimer intends it to be used. We need to test it in both mundane and extreme cases. It still may be a great rule, even if it only applies in mundane cases (though we need a clear understanding of what situations are "mundane"), but knowing the rule's boundaries is key to knowing if it's a good rule at all.

This is very similar to testing a computer program or function. If I write a function called double(), which supposedly multiplies every number by two, and I test it out by feeding it 10 and getting 20 and then I say, "See, it works!," I shouldn't accuse someone who wants to feed it zero as an extremist. If he feeds it zero and gets one, which is not zero times two, I need to admit that either my function doesn't work, or I need to rename it something like doubleNumbersGreaterThanZero().

If I say "Stealing is NEVER okay," it's impossible to understand what I mean without probing me with questions like, "What if you're starving and the only way you can survive is by stealing food?"

If I tell someone who asks that, "You can take anything to extremes!" I'm crippling the conversation, at least if it's a conversation ABOUT my ethical rule and not just "shooting the shit" or venting. I'M the one who claimed that it's "NEVER" okay to steal. Do I really mean never or do I mean it's generally not okay? If there are exceptions to my rule, what are they? Are they obvious exceptions or are they really complex?

Sometimes I make ethical claims that I'm pretty sure are sound, but I haven't fully worked out all their implications. And I may be totally unaware that I haven't worked out all their implications. So I may confidently say "It's NEVER okay to steal," really believing that to be true, even though it's not.

But if someone then hits me with "what if you're starving?" and if I care at all about rational discourse, I owe it to the conversation to respond by saying, "Even in cases like that, it's not okay to steal, because..." or "Okay, you're right. There are some extremes cases where it's okay to steal, even if it's generally not okay. The exceptions are..." or "Hm. I guess I haven't worked out my ethics as well as I thought I did. While I still feel that it's generally not okay to steal, there probably are some extreme cases where it is okay. Since I haven't thought this through, I can't give specifics."

Of course, it's possible to "bring up extremes" or Godwin just for shock value, bringing Hitler into the mix because you know he upsets people. But Hitler really does serve an important (and not easily replaceable) purpose in discussions about ethics. Since Hitler clearly CAN happen in this world, our ethical stances have to be tested against Hitler.

And it's often not enough to JUST test against Hitler. If we really want to understand the applicability of a rule, we may have to work backwards from Hitler:

"I claim ethical rule X should ALWAYS be followed!"

"Isn't that what Hitler did?"

"Okay, that's an extreme. But you're right. In places like Nazi Germany, my rules doesn't work."

"Well, what makes it not work in Nazi Germany?"

"Nazi Germany was a country rules by fear. There were horrible consequences if you didn't obey the leader's orders. My rule doesn't apply in cases like that, but it does in countries where citizens have relative freedom."

"So, does it apply in schools? What if the school paddles kids who don't obey? Do they still have to abide by your rule? What about in corporations were people will get fired if they don't obey?"

Etc.

At some point, the person who posited the rule may get tired of being grilled. That's fair enough, as long as he acknowledges that he no longer wants to discuss it. At that point, we still don't fully understand the rule, because we haven't tested all its use cases. That's fine, as long as we don't pretend that we do understand it -- and as long as the rule-creator doesn't cry "Godwin," just because he is tired of talking about it or defensive because when he said the rule in the first place, it was really just a way of venting, joking or making an off-the-cuff remark -- not a attempt at rigorous ethical thought. (Off-the-cuff remarks are fine, as long as you don't later let defensiveness make you pretend they were something more than that.)

This, I think, is often the source of discord in conversations. Person A makes a strong claim, but the strength of it is more based more on strong feelings (he's venting) than having fully-thought-out his claim's implications.

Someone who is, in this particular situation, more removed (or someone who is naturally a "logical thinker"), takes the claim literally (and to be fair to him, it's phrased in a way that makes a literal interpretation possible) and asks for all sorts of details that the claim-maker can't give.

The claim-maker then feels on-the-spot. He probably should say, "Look, I have really strong feelings about this but I haven't really through the implications of my claim," but he feels this will minimize his role in the conversation (or make him look stupid or whatever). So instead of saying, "I hadn't really thought that through," he gets defensive and shouts "Godwin" or "You can take anything to extremes!."

Often, on sites like Metafilter, a thread is full of some people who are trying to be hyper-rational and other people who are trying to "share" or "vent." Neither of these ways of speaking are problematic in and of themselves, but they don't work well together, especially when they use the same sorts of phrasing, so that one can be confused for the other. And when you mix defensiveness and devotion-to-logic in, the results can be explosive.
posted by grumblebee at 11:39 AM on March 6, 2011 [2 favorites]


(We have no valid argument that the sun will rise tomorrow. That doesn't mean it won't.)

It does day after day. Never a miscommunication. So therefore,... wait, what were we talking about?

Oh, yeah, logic. This is a frequently made, glib assertion. There are a couple of logical ways of supporting this presumption. Neither says the sun will rise tomorrow with 100% certainty, just with a very, very high probability. And if probabilities are high enough we behave as if events are certain.

The first argument is purely a Bayesian argument that the prior probability is extremely high based on past experience and as long as no new data have been collected to modify that prior, the probability remains high. This isn't syllogistic, but probabilistic logic. It is essentially built into our biology. If we didn't behave in this (usually very conservative) Bayesian way, our lives would be impossible. Purely based on our prior experience, we assume that if we don't open the door before attempting passage through it, we will smack into the door and it will hurt. We did this long before we had such things as scientific theories and mathematics. Nonetheless, it is a logically valid way to approach our behavior.

The second is based on the conceptual model of our physical solar system, which is a logically built (using syllogisms to build that model via testing and failing to falsify the model to within stated accuracy of the data and model: if the event doesn't occur, then the model is wrong. The event doesn't occur. Therefore the model is wrong.). Of course, this model allows the sun to fail to come up due to any number of cataclysmic events that could occur, but the model also places very small probabilities on those events. Now you can argue until you're blue in the face that the above syllogism doesn't prove the model is correct, but that doesn't mean the model or our use of it is "illogical" in any sense of the word.
posted by Mental Wimp at 12:07 PM on March 6, 2011 [3 favorites]


As a rhetorical device ad hominem attacks can be effective at swaying an audience. They are not part of logic, but people are emotional emotional and tribal creatures who make decisions based on social connections and mores.
posted by humanfont at 12:46 PM on March 6, 2011


The first argument is purely a Bayesian argument that the prior probability is extremely high based on past experience and as long as no new data have been collected to modify that prior, the probability remains high. This isn't syllogistic, but probabilistic logic.

I'm not a logician, but isn't it both?

The syllogism would go something like this:

1. If an event has always been observed in the past under these conditions, it will occur in the future under the same circumstances.

2. It's "the future," and those same circumstances are present.

3. Therefor, the event will occur.

=====

Point 1 may be false, but if it's true, doesn't the conclusion follow via the syllogism?

Syllogistic logic is also used with models. WITHIN your model, you can prove, via a syllogism, that the sun will always rise. You can REALLY prove it, because, since a model is a kind of fiction, you (it's author) get to say what its givens are. In the world of "Star Trek," if you set your phaser on kill, fire it at someone and him him, you WILL kill him. It's stupid for someone (who isn't the author) to say, "the premise is false." It's true because you say it's true.

Of course, models are only useful if they map onto reality well enough to allow us to use them as prediction-making tools. But couldn't we construct a syllogism like this:

1. In a fictional world, the sun always rises because...
2. our sun behaves just like the sun in the fictional world.
3. therefor, our sun will always rise.

These syllogism have BIG Achilles' Heels, because, even if they are internally sound, they may be founded on faulty premises (maybe the model DOESN'T map onto reality!) But this is true of all logic, as all logic rests on axioms (such as causation) that aren't provable. At some level, we have to take some stuff on faith or we can't get any work done at all.
posted by grumblebee at 12:48 PM on March 6, 2011


As a rhetorical device ad hominem attacks can be effective at swaying an audience. They are not part of logic, but people are emotional emotional and tribal creatures who make decisions based on social connections and mores.

I am not sure what you're saying other than "rhetoric works," which I doubt anyone disputes. Did you intend to go further and claim that it's morally okay (in certain circumstances) to use rhetoric to sway people? When is it okay and when is it not okay?

Is it okay to use rhetoric that masquerades as logic (Sophistry), which is a kind of lying?
posted by grumblebee at 12:52 PM on March 6, 2011


Arguments that involve probability (even EXTREMELY HIGH probability) are not valid. "Valid" is a technical term used to describe a certain type of argument, and arguments involving probability are not those types of arguments. Now, that doesn't in any way mean that arguments about probability are illogical. Not at all. It's just that saying an argument is valid is not in any way the same as saying the argument is good.
posted by meese at 1:27 PM on March 6, 2011


I found the article very clear, but I'm untrained in logic.
posted by Lovecraft In Brooklyn at 1:30 PM on March 6, 2011


Arguments that involve probability (even EXTREMELY HIGH probability) are not valid.

I'm not sure where you're getting this definition of validity. Outside of theoretical arguments, i.e., mathematical arguments, you have have to deal with probabilities other than 1 and 0. See, e.g., probabilistic logic. The problem with insisting on 100% probability is that it doesn't exist in physical reality. It is merely an ideal construct. I think the concern expressed regarding logical fallacies is not concerned with mathematical proof, but with arguments concerning the physical world.
posted by Mental Wimp at 1:47 PM on March 6, 2011


dersins: "86A lot of the stuff on this guy's site is distasteful at best, and stupid and hateful at worst, but I did enjoy his take on Ender's Game"

"But then again, if you get your jollies from self-pity, you are sick and need help. The same goes for geek revenge fantasies. This is the real problem with Ender's Game: it's not just porn, it's sick porn."

This is really great. (I loved Ender's Game and I loved his take on it. I think *I'm* sick and need help.)
posted by iamkimiam at 1:49 PM on March 6, 2011


"I think the concern expressed regarding logical fallacies is not concerned with mathematical proof, but with arguments concerning the physical world."

Not really, or at least, I don't think of it that way. Probabilistic logic is inductive logic; Hume was very much concerned with formal proofs when arguing against induction (or rather, when arguing the limits of induction).

And that's why simply setting up the syllogism doesn't work — if any of the premises are false or unprovable, the syllogism is false (even if the conclusion is true).
posted by klangklangston at 2:12 PM on March 6, 2011


Hume was very much concerned with formal proofs when arguing against induction (or rather, when arguing the limits of induction).

I agree that much of the earlier historic discussion of logic in argument focused on "pure" reason, but it certainly does not apply to most argumentation on the Internets. Outside of the realm of mathematics, you have to worry about probabilities. As soon as you introduce the real world, there is no way to hew to pure syllogism. Full stop.
posted by Mental Wimp at 2:34 PM on March 6, 2011


When we're talking about logic, validity is defined as follows: if the premises are true, the conclusion must be true.

So, the quintessential valid argument: All men are mortal, Socrates is a man, therefore Socrates is mortal. If it is true that all men are mortal, and if it is true that Socrates is a man, it is therefore impossible for Socrates to not be mortal.

Saying that an argument is valid really isn't telling you that much about its real-world worth. Consider the following argument: "John is not at home. Therefore, John is not at home." That counts as a valid argument. Why? Because, if it is true that John is not at home, then it is absolutely impossible for it to be false that John is not at home. However, if ever you're in a position where you need to prove that someone is not home, you'd best find some other means of supporting your position than that.

Klangklangston is right, that probabilistic logic is inductive in nature. And inductive arguments, by definition, cannot be valid (because inductive arguments are defined as arguments meant to show that it is very likely that the conclusion is true. And if it is very likely the conclusion is true, that means there's still some room -- however unlikely -- that the conclusion will be false. Hence, it is not the case that, if the premises are true, the conclusion must be true).

Again, this doesn't mean probabilistic logic isn't good. Inductive arguments can be terribly strong, and there are countless instances where you are acting appropriately by being convinced by inductive reasoning. But, it's not valid. Inductive reasoning isn't trying to be valid, so it's not really something to get worried over (unless you're reading about the problem of induction).

Now, out in the wild, "valid" is often used interchangeably with "good." So, people say, "now, that's a valid point!" They're not using the term the way I am, the way it is defined as a technical term within the study of logic. This usually leads new students of logic through serious bouts of confusion before they can get the non-technical and the technical definitions separated out from one another.
posted by meese at 2:35 PM on March 6, 2011


Yeah, well, there are plenty of assholes ITT. Some of us can't eat corn and with you all dancing about with your hominy this and hominy that, it is completely insensitive to those of us digestivally challenged.

You know who else was completely insensitive to the digestivally challenged?
posted by Samizdata at 3:22 PM on March 6, 2011


"Clearly, ad hominem fallacies are not difficult to understand, and are used frequently. Especially on mefi."

Argumentum ad forum universalia?

(Sorry about the lame "discussion board" concoction, but I don't actually know Latin.)
posted by sneebler at 3:36 PM on March 6, 2011


It's unfortunate that they teach formal logic (generally first order logic) with these types of 'real world' syllogisms, because it really has very little applicability to the real world, where virtually all premises and most implications are uncertain. Statistical techniques and Bayesian inference are really much better methods for arriving at real world conclusions.

Let's take the premises of the quintessential argument:
1. All men are mortal - very likely true, but not formally proven
2. Socrates is a man - we can't be absolutely certain Socrates even existed, let alone that he was a man (it's not logically impossible for Socrates to have been a woman, or even a space-alien for that matter)
posted by Pyry at 4:46 PM on March 6, 2011


Ad Hominem gets used a lot here on mefi - when I saw it used I assumed that this place was fully of learned, tweedy ivy- leaguers who knew latin and stuff.
posted by sgt.serenity at 5:58 PM on March 6, 2011


He introduces an ad hominem attack of his own in the second sentence, claiming the personal delicacy of certain persons makes them prone to misusing the term

One of the most widely misused terms on the Net is "ad hominem". It is most often introduced into a discussion by certain delicate types, delicate of personality and mind,
posted by canoehead at 6:07 PM on March 6, 2011


When we're talking about logic, validity is defined as follows: if the premises are true, the conclusion must be true.

Yes, I understand what a valid syllogism is, but I'm arguing that that restricted definition of "valid" is fairly trivial and not of much use for argumentation, since the truth depends upon the state of the components of the syllogism, which can NEVER be determined with certainty when they refer to anything other than axioms in mathematics. I mean, all men are mortal? How does one determine that? Is that assumed or is there a syllogism that led to that? Socrates is a man? Does he have man pants on? Did you check his genitals? How do you define a man for this purpose?

If you look at what people are really arguing about in any real discussion, it rarely, if ever comes down to the incorrect use of syllogisms. Well, for some rather retarded people it might, but for this purpose I am ignoring them. The real nut of any problem worth arguing about comes down to things like determining what assumptions are being held as true and what the real desiderata are. These are what people disagree about. These problems of formal logic are not the interesting part at all. Trying to elucidate each side's assumptions and goals is where the work and the fun are. YMMV.
posted by Mental Wimp at 9:31 PM on March 6, 2011


Klangklangston is right, that probabilistic logic is inductive in nature. And inductive arguments, by definition, cannot be valid (because inductive arguments are defined as arguments meant to show that it is very likely that the conclusion is true. And if it is very likely the conclusion is true, that means there's still some room -- however unlikely -- that the conclusion will be false. Hence, it is not the case that, if the premises are true, the conclusion must be true).

Oh, and in my understanding this is a gross mischaracterization of probabilistic logic. Induction is only part of it. Note that it's part of the classic syllogism as well, because general premises (e.g., "All men are mortal") must be built somehow and you're kind of left with induction to build them. And the conclusions regarding specific events addressed by classic logic aren't determined by probabilistic logic to be either true or false, but rather to have specific probabilities attached to them. The conclusions are about probabilities of events, not in general their truth (probability =1) or falsity (probability =0).
posted by Mental Wimp at 9:37 PM on March 6, 2011


« Older Unless you are in an extremely remote location, y...  |  "When [700] hundred years old ... Newer »


This thread has been archived and is closed to new comments