Analogies?

[cosine]

The creation of this thread being rightly inspired by Pangloss:

http://www.scienceforums.net/forums/showpost.php?p=227050&postcount=42

 

I don't agree that analogies are attempts to find isomorphisms. I agree that that CAN happen, but the implication that they're always a logical fallacy is not fair. They can also be a valid tool for giving an example for one's position, without necessarily suggesting that the comparison is direct and perfect. (I'm interested in discussing that further, but we might want a separate thread for it. I think others might be interested in that discussion as well, if you don't mind starting it.)

 

I don't mind at all!

 

Here are the analogies that inspired this:

 

Pangloss's Grocer/President analogy:

http://www.scienceforums.net/forums/showpost.php?p=226783&postcount=36

 

My personal opinion is that the evidence of WMDs was marginal and circumstantial' date=' and the administration, upon realizing that the terrorism/9-11 angle was going to be insufficient to bring the country to war, decided that WMDs constituted a more powerful motivator. That's deceptive, but it's not the same thing as a "lie". We were sold a bag of goods, Madison Avenue style, and we bought it, hook, line, but no sinker (not everyone was convinced, but not enough people disagreed to stop it).

 

This is different from saying that we were LIED to. And frankly I think the problem here is that we as Americans are unwilling to own up to our own responsibility for this mess. It's too easy to dump it all in George Bush's lap. Who are the real fools here? [/quote']

 

With Mokele's Response:

http://www.scienceforums.net/forums/showpost.php?p=226798&postcount=37

 

And, while I would agree about owning up to the responsibility of having fallen for it, I'm not sure your analogy is valid; we have good reason to expect a street vendor to try to cheat us in order to get our money, but to expect *any* branch of government to spin (or otherwise present in a biased manner) information to lead us to war?

 

Then there was the second analogy:

http://www.scienceforums.net/forums/showpost.php?p=226909&postcount=39

 

[b']We went to the auto dealer to buy Ford Explorer, and came home in a Ferrari, and now the wife is standing in the doorway with the kids demanding to know what the hell we think we're doing. And all we can do is stare at the ground and mumble something about the color red.[/b]

 

And my subsequent response:

http://www.scienceforums.net/forums/showpost.php?p=226929&postcount=41

 

First of all, analogies are attempts to find isomorphisms... however they are frequently false isomorphisms. I really don't see how your analogies are applying or what they even mean.

 

 

So now that you're all caught up...

Analogies are attempts to compare two like things, to try to map the properties of one onto another. However when two systems have intrinsic differences, its important not to overapply the properties of one onto another. For example, in mathematics 1=1, and we usually apply that to metersticks, where the length of one meterstick is equal to the length of another. However if one meter stick is hurled at the sun and it burns into flames, and one kept on earth, then the one on earth is not equal to the one burnt up in the sun.

 

Gah, time for class, hope that was a good example, though I'm not sure if the connection with analogies is readily obvious. But before I go for now, I'd just like to say: An analogy to a situation is like getting a crayon-drawn stick figure after asking an artist for a portrait.

 

Edit: So yeah, I didn't explicitly state this: What are your thoughts on analogies? Please, discuss amongst yourselves.

 

AND OMG MY 100th POST!

[cosine]

Okay I'm back, and I had a thought to elaborate more:

 

My position is that while analogies may be used as a starting point to suggest further investigation, they can not be used to substantiate a point.

[Mokele]

Well, I'm personally very fond of analogies. However, some things need to be kept in mind when using and reading them.

 

First, they usually only apply to one or a few aspects of the system. "Life is like a box of chocolates" works quite well with the conclusion "you never know what you're gonna get" and less well with the alternative "it may be recalled by the State Health Department on account of an unacceptably high content of rat feces".

 

Second, analogies are not arguements, but rather illustrations. Sometimes you draw a picture to show what you're talking about, other times you use an analogy.

 

Third, I'd argue that analogies are only a fallacy if used as the arguement itself, rather than merely an illustration. A bad analogy not used as an arguement is really no worse than a bad picture used for illustration; it's distracting, confusing, and makes you look inept, but doesn't actually impact reasoning. Using an analogy for an arguement is *usually* fallacious, though, due to the differences in the systems. The only time I'm aware of it *not* being a fallacy is when the systems are extremely similar on account of working on the same basic principles (like comparing two types of jet engines, or two chemically similar reactions, or an evolutionary convergence); however, this could be considered inference rather than analogy.

 

Mokele

[ecoli]

http://www.scienceforums.net/forums/showpost.php?p=219933&postcount=26

 

Here's my favorite worse analogy on SFN.

 

You don't own the "space" above your house, would you like it if the FAA decided to install a large tarp to fully surround your house? Hell, we can make it out of tinfoil and you wont even have to worry about "wasting" energy heating.

 

This anlogy simply doesn't work. You're talking about two completely different systems that are uncomparable.

[Pangloss]

Thanks for setting that up, Cosine.

 

I think you should give us some examples (or perhaps analogies? (grin)) of how you feel that analogies can be used in error, how to spot them, and what we can do about them.

[Phi for All]

Gah, time for class, hope that was a good example, though I'm not sure if the connection with analogies is readily obvious.

Funny you should put it like this. I usually use analogies because I'm unsure if the connection I'm trying to make with my audience is readily obvious. If I'm explaining a position and get a blank stare, or if I'm writing about something and assume the readership will be composed of those who understand the position and those who won't (and I'm cursed with continually underestimating my readership), I will resort to analogy.

 

I'm not really trying to develop isomorphy between two things as much as clarifying the former by using the latter. It's like trying to explain evolution to a farmer by using the analogy of plowing and planting his fields in order to make it clearer.

 

Ooops! There I go again....

[ecoli]

I'm not really trying to develop isomorphy between two things as much as clarifying the former by using the latter. It's like trying to explain evolution to a farmer by using the analogy of plowing and planting his fields in order to make it clearer.

 

Ooops! There I go again....

 

That was a funny example, but you are right. Anologies should be used, because it helps someone learn a new fact in a way that is more in sync with flibit's understanding of the world. Bringing an alien sitution into their level of understanding can help flibit understand the new situation better. From there, you can go ahead and explain how things really are, in terms more specialized for the field. I like to think of this as a parallel analogy.

 

However, I believe the example I posted above, this did not apply to the above. I think is because Hailstorm did not succeed in making an analogy on a parallel plane. Instead he tried to explain a situation on a "perpendicular" and so it made almost no sense at all.