Analogies are poor debate tools

[Subterraneanus]

@Left_Hand_of_Dorkness

Subterraneus, thanks for a very interesting series of posts!

I do think it’s a good question MfM asked, slightly modified: can you give an example of a strong non-trivial argument that is wholly reliant on analogy?

You can’t frame a hypothesis without assuming that the causal events to be explained must share certain testable implications with similar causal events we have already observed and explained to our satisfaction.

Some examples of how analogical reasoning works in science are here:

http://plato.stanford.edu/entries/reasoning-analogy/#Exa

[Little_Nemo]

@John_Mace

Analogies are like naked bodies. Some are damn near perfect, and others are not even worth imagining.

And some are so distracting they make it impossible for you to follow what people are saying.

[Little_Nemo]

@Measure_for_Measure

Show me a non-trivial valid argument wholly reliant upon analogy. (Hey, maybe I’m wrong.)

No argument is wholly reliant upon analogy. An analogy is used when you’re making what you consider to be a valid argument but the person you’re making it to doesn’t see its point. So you apply the same argument to a different situation in hopes that he will comprehend the point you’re making and then be able to apply it back to the original issue.

[you_with_the_face]

Anyone else need an analogy to understand what Subterraneanus is arguing?

Just me? Ok.

[Irishman]

@The_Joker_and_the_Thief

I generally agree with you, by the way, but analogies when done in a less adversarial setting can be very useful.

Acknowledged in my second paragraph.

@Fotheringay-Phipps

The ultimate purpose of an analogy is to present the underlying principles or logic in a form which strips out extraneous factors that may be present in the specific case being discussed. Of course, whether those other factors are truly “extraneous” will frequently be subject to disagreement.

But that’s the point - too often the debate gets derailed to argue about whether the analogy is valid instead of trying to understand the point the analogy is trying to make.

Witness the whole “same sex marriage is like marrying a washing machine” fiasco.

@Morgenstern

This thread is about as exciting as dancing with your sister.

So, you don’t have to dance with my sister.

@Ravenman

I think the time and effort spent to debunk analogies is the far greater disservice to debate, rather than an analogy being raised in the first place.

That’s what I’m talking about. That hijack wouldn’t happen without the analogy. The point of any analogy is to try to show the likenesses, but the setting of a debate makes accepting someone else’s argument a challenge. They have a strong motivation to find the “flaws” in the analogy.

I have to constantly remind myself when I post here to cut back on analogies even for really basic explanations of things, because I get so tired of people extrapolating analogies so far beyond what they were intended to illustrate.

Bingo!

@The_Second_Stone

So yes, your point is absolutely correct, your description is never the same as the real thing. But it breaks down in that analogies are effective rhetorical tools because they allow an audience with no background to get a glimpse like: experience of quantum physics to understand that the cat is alive and dead at the same time and that they just do not understand it, but they understand the outlines of what they do and don’t understand by analogy.

But this brings up a related point, that analogies can be poor tools in educational settings, too.

I haven’t read the book, but Edwards points out in his review that Hofstadter and Sander bring up this very point.

In Surfaces and Essences, Hofstadter and Sander argued that all thinking is analogical and naive analogies poison a lot of thoughts. The authors stated that when schoolchildren are taught to view division as sharing rather than as measurement, it makes it harder for students to complete more complex mathematics later on.

I don’t know the validity of that specific conjecture, but the Schrödinger’s Cat one is a great example. As mentioned, Schrödinger didn’t intend his analogy to explain QM, he was trying to demonstrate by argumentum ad absurdum the ramifications of the Copenhagen interpretation, and how preposterous it is.

Similarly, when trying to study the idea of the Big Bang and the expansion of the universe while the universe has no physical center, two analogies are often used: the balloon and the muffin.

The balloon analogy says the universe is like a big balloon with dots drawn on the surface. Inflate the balloon more, the dots spread further apart, but there’s no physical center on the surface of the balloon. That’s great, except it relies on curvature in a third dimension. Does that mean the universe is curved through a different dimension?

The muffin example says blueberries in the muffin, when the muffin bakes it expands, but the blueberries move apart. Except a muffin has a physical center.

I have run into those examples many times, and typically the person doesn’t get the point of the analogy, they get hung up on the points I mentioned.

@Left_Hand_of_Dorkness

The key point in working with an analogy is determining what features are relevant and what features are irrelevant.

And if you’re trying to craft an analogy, a lot of care should be put into this element.

I think that the value of an analogy is not in ending a disagreement; rather, it gives an arena in which to hash out the disagreement. A good analogy provides a good arena, a bad analogy a bad arena.

Okay, that’s an interesting way to approach it. Too many debates use bad analogies and hammers to belittle their opponents rather than as maps to show the way. Or something like that.

@Subterraneanus

There’s a fair amount of confusion in this thread about analogical reasoning. Arguments from analogy are inductive arguments, not deductive arguments. Deductive arguments are valid or invalid: if the presumed truth of the premises guarantees the truth of the conclusion, then they are valid; otherwise, they are invalid. A valid argument with premises that are in fact true is sound.

[snip]

So arguments from analogy are perfectly good inductive arguments when the premises refer to shared properties that are relevant to the inference one is trying to make. But they are inductive arguments, and so they at best make the conclusion likely to be true. They are not deductive arguments, and so their premises do not guarantee the truth of a conclusion, even if they are in fact true.

@Subterraneanus

Again: analogical arguments are inductive, not deductive, so they are neither valid nor invalid. Rather, they are strong or weak, depending on how well their premises support their conclusions probabilistically.

Thank you.

@Mijin

Well firstly I think you’re misusing the term “inherent”. If analogies are only flawed when used incorrectly, then that is not an inherent property.

No, I think they are always flawed, it’s just sometimes we overlook the flaws and sometimes we get caught up by them. I’m using “flawed” to mean “not completely the same” - because with analogies the value is in the similarities and the differences. In lawman’s terms, we get caught up talking about valid analogies and invalid analogies. Subterraneanus points out that strictly speaking that is the wrong evaluation to make of analogies. Instead, you evaluate an analogy on how strong it is. A “good” analogy is one that the premises between the points of interest are the same, a “bad” analogy is one where the premises differ. The difficulty is identifying which premises are relevant and which can be ignored.

All analogies have holes, it’s just how big the holes are. If the holes are tiny compared to the issues, the argument can’t leak through. If the holes are big, the argument falls out the bottom.

OK, so why is that not useful in a debate?
Debates are not some separate realm of the universe which is always adversarial.

In theory at least people are trying to get a point across to each other and that is a form of education – both parties are trying to explain something (usually…I will concede that some of the times a debate can be merely cheering one side and booing the other). So, why not use analogies to help in that aim, if, as you say, they are an effective learning tool?

Because the value in analogies is in connecting the premises of one situation with the premises of the other situation. But debate situations (especially here, where “debate” often means “piss on each other until it’s too deep to breathe”) are not built on an atmosphere of learning from the other person, but rather an atmosphere of telling them why they are wrong. So instead of accepting an analogy for what it is trying to show, the response is typically to try to overextend the analogy to show that it is flawed, and thus the point the person was trying to make with the analogy is, therefore, invalid. Or they change the analogy to replace it with one that highlights what they want to say rather than look at the analogy given and accept what it shows.

@FXMastermind

Excellent thoughts, but it seems like you want opponents to argue in good faith, and with logic and reason, rather than rhetoric, and any other weapon available. Whilch is another topic of course.

Well, it certainly would be nice.

Yes, but you are simply describing what happens, that isn’t an argument against the tactic.

The argument against the tactic is that the conversation gets hijacked into arguing over whether or not washing machines can enter contracts.

Exactly, that is exactly right. But your argument that it is a poor debate tool is just pointing out that it is an awful tactic, it doesn’t actually convince that it should not be used.

Um, it’s an awful tactic doesn’t mean it shouldn’t be used?

It’s like (and yeah, I know exactly what I am about to do), it’s like your opponent in a fight just hit you in the nutsack with a frozen herring, which really hurt, and the crowd is cheering, and you point out that a frozen herring, especially with combined with an illegal groin hit, is a poor tool to use in a fight. The problem with this, is you are saying it through clenched teeth, while on the ground in pain.

And you just introduced a flawed analogy to make your point. So I can spend time telling you why the analogy is flawed, and thereby do exactly what I say is the problem with analogies - they distract from the conversation.*

@Frylock

In my view (though it seems to be controversial) science itself is built on arguments from analogy. You can’t do science unless you assume that certain similarities between things will make other similarities between those things more probable.

That gets back to the argument presented by Hofstadter and Sander.

@you_with_the_face

Anyone else need an analogy to understand what Subterraneanus is arguing?

He’s saying that analogies can’t be measured by being “valid” or “invalid”. Rather, they can only be measured for being “strong” or “weak”. A strong inductive argument has conclusions that are likely based upon the premises, and a weak inductive argument has conclusions that are less likely based upon the premises.

Similarly, a strong analogy is one where more of the relevant issues are similar between the two items being compared, and a weak analogy has fewer similar issues taken into account, or some of the relevant issues are ignored.

In other words, all analogies are flawed, it’s just a matter of how important the flaws are to the point being made. Is it a hairline scratch in the paint or is the car totaled?

(Yes, I am using analogies. Sue me.)

---

• What kind of a fight is it? A boxing match with structured rules? A duel? An attack by a mugger (wielding a fish)? I would say unless it’s a structured competition, any tool that gets the job done is a good tool. A frozen herring wouldn’t be an ideal tool, but if one is handy it’s certainly got merits. A poor tool would be a wet spaghetti noodle - it’s limp, doesn’t have much mass, and poor structural integrity. Hit me in the nuts with it all you want, it’s not dropping me to the floor gasping. And anything that’s not a structured competition, there’s no such thing as an illegal groin hit. If it’s a structured competition, then the opponent who used the groin hit is going to face sanctions, up to and including being disqualified. Getting yourself disqualified is a horrible tool for winning.

And if you’re having a clown fight, then a frozen herring is a great tool.

[FXMastermind]

@Irishman

Um, it’s an awful tactic doesn’t mean it shouldn’t be used?

Well that’s the issue, isn’t it? It’s why I used the example of a red herring as a tool in a fight. If it works, it isn’t awful for the winner. Hell, just disrupting a debate is considered a win for somebody who doesn’t want an issue discussed.

@Irishman

I would say unless it’s a structured competition, any tool that gets the job done is a good tool.

Yes, and since the internet isn’t structured, even if you show with little doubt an analogy is an awful weapon, that doesn’t mean anyone will stop using one.

@Irishman

And if you’re having a clown fight, then a frozen herring is a great tool.

You just described the situation.

[FXMastermind]

@Irishman

• What kind of a fight is it? A boxing match with structured rules?

I was thinking of internet debates, where there actually aren’t any rules of debate at all.

While it’s actually a simile, the meme of Pigeon Chess illustrates the problem with internet debates.