Philosophy of Linguistics -- Damning with Faint Praise

I recently came upon that video where the commentator mentioned an idiom I hadn't heard before.

He was comparing two people as you saw and mentioned the term:

"Damning with faint praise"

~~~~~~~~~

I instantly looked it up:

Fig. to criticize someone or something indirectly by not praising enthusiastically. The critic did not say that he disliked the play, but he damned it with faint praise. Mrs. Brown is very proud of her son's achievements, but damns her daughter's with faint praise.

~~~~~~~~

Once I figured out what it meant, it was just such a colorful way to say something, and as someone who enjoys writing, I love the pictures that are invoked when you say you're damning someone with faint praise.

I felt inspiration, and I actually saw one person practicing this faint praise(or really it should be thought of lack of praise) on another.

You really see the powers idioms have on our discussions, when they invoke so much emotion and so much color into our conversation.

They, in essence, bring into existence incarnations of the feelings and meanings deep inside our words.

They are manifestations of our true intentions and beliefs. Maybe that's why we memorize or are familiar with hundreds or thousands of them.

Philosophy of Causation / An eye into Statistics

I was watching this video, and read a comment that I wanted to discuss.

The comment is as follows:

"The three states that had abortion laws three years earlier had crime rates that decreased three years earlier and Ben thinks it's meaningless. lol"

My response:

Statistically it is.(It is meaningless)

If you can't isolate a situation, any number of factors could be the cause of it.

People who don't know statistics can't understand this point.

Try googling Correlation versus causation. The two things are not the same.

Ben went to Columbia university, and although he's annoying at times, it's why he was right and you were wrong.

~~~~~~~~~~

I didn't mean to come off as too cruel, but it's true that I think Ben's understanding of statistics easily lead him to his conclusion; again thanks to his upper class education.

Such a conclusion would probably elude a person whom had not taken statistics and would not be able to grasp the reason why they would be wrong.

The reason is simple, a correlation of things is not a causation because there may be hidden causes that you are unaware of.

As the old saying goes, sometimes a coincidence is just a coincidence, especially when you're comparing state wide stats; something enormously complicated.

~~~~~~~~~~~~

Example:

You're looking at a graph, and for some reason it seems to show that the more firefighters there are the more damage occurs to a home.

You look at the chart and for homes that had 20 firefighters the damage that occurred was much more than in homes that had 18 19 or 15.

It's a trend and the line beautifully goes up and to the right.

So you automatically assume, more firefighters means more damage, and therefore you want to limit how many firefighters go to homes.

Although you'd be wrong, because you just let correlation become causation.

You missed a simple fact, that the more a home is on fire and the bigger the fire is, the more firefighters will be there. Also since the fire is much bigger it's bound to do more damage than a smaller fire.

It's not the firefighters causing the damage, it's the fire, and the bigger it is, the more firefighters you need to stop it.

By limiting the number of firefighters going to homes, you would actually be increasing the damage and allowing the fire to burn for longer.

That's why correlation, is not causation.
The p-value test is a good safety net, for the problem of lurking variables.