## What's that one logical fallacy called?

For the serious discussion of weighty matters and worldly issues. No off-topic posts allowed.

Moderators: Azrael, Moderators General, Prelates

DaPwnzlord
Posts: 23
Joined: Mon Jan 18, 2010 5:22 pm UTC

### What's that one logical fallacy called?

What's it called when you try to point to the occurrence of an unlikely thing happening as evidence that this event was directed by something else? I know I'm mangling it, but more specifically I'm asking about Paley's watch argument for the existence of God. He would say that the fact that the universe occurred the way it did against all the odds is a an argument in favor of his ideas. But I know that's a logical fallacy, I just can't remember the name or how to describe it.

Capt. Obvious
Posts: 36
Joined: Wed Aug 19, 2009 5:55 am UTC

### Re: What's that one logical fallacy called?

What's it called when you try to point to the occurrence of an unlikely thing happening as evidence that this event was directed by something else? I know I'm mangling it, but more specifically I'm asking about Paley's watch argument for the existence of God. He would say that the fact that the universe occurred the way it did against all the odds is a an argument in favor of his ideas. But I know that's a logical fallacy, I just can't remember the name or how to describe it.

Well, there are two problems you have, nomenclature and description. I'll start with your description. The fact that something is unlikely to happen via one method is evidence that another method is responsible. To wit, the gravitational constant may be randomly reset every 45 seconds, and it just happens to pick the same number each time. Another theory is that the gravitational constant doesn't change. (If the gravitational constant changes over time, based on some complex formula, I apologize in advance for my example.) If a watch were to appear in a box that you locked while empty, it's possible that matter changed in such a way such as to coalesce into a watch. But it is far more likely that there is a hidden door, or the lock was picked, or some other way a watch was slipped in there.

The fallacy you're describing is that while it is exceedingly unlikely that human life would have evolved if any conditions were slightly different, the conditions influenced human life. That is, if the earth were slightly colder, we all may simply have more insulation and die should the average temperatures be as hot as they are now, and if the universe was adverse to all forms of life then human life wouldn't have been around to discover the universe. That is, if you shoot an arrow, then draw the target around where the arrow lands, all those bullseyes don't make you are an exceptional archer.

As for the name, it might be called an outcome bias ( http://en.wikipedia.org/wiki/Outcome_bias ) or a selection bias (see wikipedia as well). There's also http://en.wikipedia.org/wiki/Texas_sharpshooter_fallacy
There's too many names for any given fallacy, unless it's one of the big ones, to be authoritative.

EmptySet
Posts: 1196
Joined: Thu Jul 24, 2008 6:33 am UTC

### Re: What's that one logical fallacy called?

Capt. Obvious wrote:The fact that something is unlikely to happen via one method is evidence that another method is responsible.

Not as such. It's unlikely that if I flip a coin twice it will come up heads both times. However, if it does, that's not really evidence that aliens used tractor beams to manipulate the coin. If there are finite possible outcomes which are mutually exclusive, ruling out one might increase the likelihood of the others - for example, if I know that if I roll a die, it must come up showing of the numbers. If it can be demonstrated that one of them is less likely to show, the others must be more likely. Note, however, the use of relative terms - this still doesn't mean an outcome is likely or "true".

For example, say there are twenty possible explanations and I prove that 19 of them are unlikely, with a 3% chance of any given one occurring. This would imply that the remaining explanation had a 43% chance of being correct, ten times more than any other explanation, assuming that the explanations provided are exhaustive. However, note that if you assert this explanation is true, it is more likely than not that you are wrong, even though it is the most likely explanation by far. And of course, in the real world, you generally don't have an exhaustive list of every possibility with associated probabilities. If I can show that Alice didn't commit the murder, that doesn't mean I can turn around and say conclusively that Bob must have done it, even though there is no evidence that this is so. Maybe it was some other person, or maybe it was a suicide, or a really weird accident.

Anyway, I would say that the fallacy the OP is thinking of is the false dichotomy ("Evolution is unlikely, so what I say must be true! Just ignore all other explanations, they totally don't exist. Honest.") Regardless of the fallacy, Paley's argument rests on a questionable premise, and also a flawed understanding of probability and evolution.

Capt. Obvious
Posts: 36
Joined: Wed Aug 19, 2009 5:55 am UTC

### Re: What's that one logical fallacy called?

EmptySet wrote:It's unlikely that if I flip a coin twice it will come up heads both times. However, if it does, that's not really evidence that aliens used tractor beams to manipulate the coin.
Sure, but if I told you I was going to flip a coin 20 times, and it would come up in the pattern HHHHHHHHHHHHHHHHHHHH, then I did it, you would probably assume it was not a fair coin/fair toss. And it would be reasonable to assume so. However, if I tossed a coin 20 times, and it came up with any other pattern, it would be just as unlikely to have produced that pattern. However, the fact that I choose the pattern AFTER the tosses as opposed to before makes me far better at using the original fair coin/fair toss methods to come up with unlikely outcomes.

The basis of the scientific method is that you choose the answer that best fits the data. Could the sun revolve around the earth and it's a bunch of optical illusions that make us believe otherwise... I suppose. But it's not likely.

The breakdown in communication we are having is you are saying unlikely as in 3%, and I'm saying unlikely where the high end is 0.00001%, and the normal usage is much less likely than that.

EmptySet wrote:Anyway, I would say that the fallacy the OP is thinking of is the false dichotomy ("Evolution is unlikely, so what I say must be true! Just ignore all other explanations, they totally don't exist. Honest.") Regardless of the fallacy, Paley's argument rests on a questionable premise, and also a flawed understanding of probability and evolution.
But what OP is thinking of isn't the false dichotomy. He's specifically asking about the fallacy where the target is chosen after the fact.

Rackum
Posts: 132
Joined: Tue Jan 26, 2010 5:15 pm UTC

### Re: What's that one logical fallacy called?

I believe what you are referring to is the Texas Sharpshooter.

On second thought, this may be what you were referring to: Teleological Argument (or Argument from Design).

EmptySet
Posts: 1196
Joined: Thu Jul 24, 2008 6:33 am UTC

### Re: What's that one logical fallacy called?

Capt. Obvious wrote:
EmptySet wrote:It's unlikely that if I flip a coin twice it will come up heads both times. However, if it does, that's not really evidence that aliens used tractor beams to manipulate the coin.
Sure, but if I told you I was going to flip a coin 20 times, and it would come up in the pattern HHHHHHHHHHHHHHHHHHHH, then I did it, you would probably assume it was not a fair coin/fair toss. And it would be reasonable to assume so.

Maybe. However, I think "the coin toss is not fair" is a different sort of conclusion to the one presented in Paley's argument, because it is a more general criticism of the toss - "If the toss was fair, it would be unlikely that I would get 20 heads in a row; therefore it must not be fair". The equivalent statement would be "If mutations occurred and propagated through pure chance, it would be unlikely that you would end up with humans; therefore, there must be some other influence". However, watchmaker analogy jumps from the general "there's some kind of influence which hasn't been accounted for" to making specific claims about the nature of that influence without properly supporting them.

Capt. Obvious wrote:The breakdown in communication we are having is you are saying unlikely as in 3%, and I'm saying unlikely where the high end is 0.00001%, and the normal usage is much less likely than that.

As far as evolution goes, .00001% is higher than you might think, given the size of the universe. And, of course, the anthropic principle means our data collection is rather biased.

Earl Grey
Posts: 65
Joined: Tue Apr 29, 2008 4:50 am UTC

### Re: What's that one logical fallacy called?

Texas Sharpshooter seems to be what I was thinking you were after, though I've personally never run into 'Texas Sharpshooter' before. This thread has taught me something.

Another related point of consideration is tautology.

For example, if I dealt out a deck of cards to a group of people then examined everyone's hand, I could marvel at how unlikely it is that we've achieved this very particular arrangement of cards. It may be so rare, that even if you spent the next ten years dealing out the cards randomly, you wouldn't get that exact arrangement again. But... there's nothing remarkable about the fact that it happened, or at least, no more remarkable than every hand you could deal, because it happened. What happens, happens.

It's possible that there are an infinite amount of alternate universes, each with different physical laws, and we exist to think about this because we happen to be in one of them where the laws allow for matter to arrange itself into creatures that can think about this.

When some try to apply Paley's argument to evolution, it fails for a different reason, in that selection pressures weed out pure randomness, giving a non-conscious direction to things, but a direction nontheless.
"You're human, which is sort of like an ape that builds airplanes and gets depressed over stupid shit."

~Tac-Tics

Samsoneffect
Posts: 318
Joined: Sat Aug 02, 2008 4:50 am UTC
Location: Burlington, ON
Contact:

### Re: What's that one logical fallacy called?

DaPwnzlord wrote:What's it called when you try to point to the occurrence of an unlikely thing happening as evidence that this event was directed by something else? I know I'm mangling it, but more specifically I'm asking about Paley's watch argument for the existence of God. He would say that the fact that the universe occurred the way it did against all the odds is a an argument in favor of his ideas. But I know that's a logical fallacy, I just can't remember the name or how to describe it.

The one I've seen that strikes me as similar to that is called the False Lottery, described by AndromedasWake here. Paraphrasing Wake's description of it, it's a post-hoc assessment of probability for an event occurring when you only have one result, and no known probabilities of other possible outcomes, or even nothing to suggest that the other outcomes are possible.Best example I can give would be the Fine-Tuned Universe Argument, where physical constants that allow for life to form are so few and statistically improbable that it had to have been manipulated somehow. This makes an assumption that those physical constants can actually vary, which is not necessarily founded. It leads to postulating universes exist with different constants and that the life we have here can't have formed as a result, even though we only have a sample size of one universe. You'll also see this argument used by William Lane Craig, as shown in that video. It's similar to the Teleological Argument, but rather than just saying 'Design!' it says, "Really really really unlikely based on post-hoc probability, therefore design!"

I don't know if this is an actual fallacy that has a particular name other than what AW has called it, but it seemed like the most analogous.
Most people would post something witty, humourous, or perhaps offensive in their signature. Unfortunately, you won't find anything like this here.

Have_A_SnApple
Posts: 32
Joined: Thu Oct 07, 2010 3:24 am UTC

### Re: What's that one logical fallacy called?

What would this fallacy be called?
I heard this one during the Mosque debate, naturally it was on Fox News:

"Not all Muslims are terrorists, but all terrorists are Muslims."

Or, in formal form:

All X are not Y, but all Y are X.

phlip
Restorer of Worlds
Posts: 7573
Joined: Sat Sep 23, 2006 3:56 am UTC
Location: Australia
Contact:

### Re: What's that one logical fallacy called?

I don't think that's a fallacy, since there's no deduction there, it's only a matter-of-fact statement. It's just wrong.

A fallacy is when you start from premises and arrive at a conclusion that doesn't actually follow from those premises, because of dodgy reasoning. But that's not the only way a conclusion can be wrong... it's also possibly to just have false premises, in which case it's not a fallacy per se, you're just wrong.

As an aside, can I say how much I hate the "All A are not B" wording? It always seems to mean ∀x:Ax->~Bx, ie there is no A which is also B... rather than ~∀x:Ax->Bx, ie not all A's are also B's, or there exists an A which is not B. Just move one word, so it's "Not all A are B", and it's so much more clear.

Code: Select all

`enum ಠ_ಠ {°□°╰=1, °Д°╰, ಠ益ಠ╰};void ┻━┻︵​╰(ಠ_ಠ ⚠) {exit((int)⚠);}`
[he/him/his]

PeterCai
Posts: 865
Joined: Tue Feb 17, 2009 1:09 pm UTC

### Re: What's that one logical fallacy called?

i think it's:
some terrorists are muslims.
not all muslims are terrorists.
therefore, all terrorists are muslims.

faux logic ftw

Роберт
Posts: 4285
Joined: Wed May 14, 2008 1:56 am UTC

### Re: What's that one logical fallacy called?

Have_A_SnApple wrote:What would this fallacy be called?
I heard this one during the Mosque debate, naturally it was on Fox News:

"Not all Muslims are terrorists, but all terrorists are Muslims."

Or, in formal form:

All X are not Y, but all Y are X.

While that can be true, that can easily be proved false. Just find one Y that is not X. In this case, one terrorist that is not Muslim. Easy enough, since plenty of terrorists aren't. I have no idea what is wrong with some people.
The Great Hippo wrote:[T]he way we treat suspected terrorists genuinely terrifies me.

mmmcannibalism
Posts: 2150
Joined: Tue Jun 30, 2009 6:16 am UTC

### Re: What's that one logical fallacy called?

Have_A_SnApple wrote:What would this fallacy be called?
I heard this one during the Mosque debate, naturally it was on Fox News:

"Not all Muslims are terrorists, but all terrorists are Muslims."

Or, in formal form:

All X are not Y, but all Y are X.

As said above, that's only logically flawed if you use it as an argument like because of A(the stated fact) we should declare war on group x because a subset is dangerous.
Izawwlgood wrote:I for one would happily live on an island as a fuzzy seal-human.

Oregonaut wrote:Damn fetuses and their terroist plots.

jdig
Posts: 1
Joined: Sat Nov 13, 2010 3:34 pm UTC

### Re: What's that one logical fallacy called?

I believe the phrase you are looking for is "non sequitur"

Kain
Posts: 1140
Joined: Wed Aug 27, 2008 4:29 am UTC
Location: At the center of the observable universe.

### Re: What's that one logical fallacy called?

As for the OP's question, the blind watchmaker argument (I am assuming that is what you are referring to) contains several fallacies, some of which are already mentioned. I would immediately include a false dichotomy tag, as the parable or whatever you want to call it suggests that either we are here via a completely random arrangement of molecules, or via the guidance of some omnipotent being. I would also through in a selection bias tag, as if we didn't exist, or if life had only progressed to unicellular organisms without nuclei, or even less complex organisms, we would not be having this discussion.
I am sure there are more fallacies involved, but those two are more than enough for me to dismiss the argument.
Look, you know it's serious when a bunch of people in full armor and gear come charging in to fight a pond of chickens - Steax

Ghavrel
Posts: 168
Joined: Thu Jul 16, 2009 4:51 am UTC

### Re: What's that one logical fallacy called?

Have_A_SnApple wrote:What would this fallacy be called?
I heard this one during the Mosque debate, naturally it was on Fox News:

"Not all Muslims are terrorists, but all terrorists are Muslims."

Or, in formal form:

All X are not Y, but all Y are X.

Well, it isn't a formal syllogistic fallacy because nothing is being concluded; there are simply two terms (I think that's the right word) being stated. Of course, the second one isn't right, but truth values of individual terms aren't really relevant to formal fallacies. You could apply that to another situation and have it be perfectly acceptable: not all rhombi are squares, bit all squares are rhombi. For example.

Now, if the person were implying that the second term were somehow a consequence of the first, they'd be running into into tons of fallacies. I guess you could argue that this statement is kind of trying to imply that all Muslims really ARE terrorists, in which case I think it might be committing fallacy of the undistributed middle?
"Si ad naturam vives, numquam eris pauper; si ad opiniones, numquam eris dives."
Live rightly and you shall never be poor; live for fame and you shall never have wealth.
~Epicurus, via Seneca

Mabus_Zero
Posts: 245
Joined: Fri Apr 18, 2008 6:30 am UTC

### Re: What's that one logical fallacy called?

To the OP:

You're talking about the Fallacy of Accident. http://www.fallacyfiles.org/accident.html

Specialization is for insects.

-Robert A. Heinlein

Posts: 5654
Joined: Wed Jun 11, 2008 11:03 am UTC
Location: The Netherlands

### Re: What's that one logical fallacy called?

I don't think Paley's watch-maker argument is a logical fallancy at all. The argument "If something is irreducibly complex than it can not have formed via evolution" is entirely sound. Darwin actually makea that very observation himself in The Origin of Species.

It's just that no such things have been found so far.
It's one of those irregular verbs, isn't it? I have an independent mind, you are an eccentric, he is round the twist
- Bernard Woolley in Yes, Prime Minister

Zamfir
I built a novelty castle, the irony was lost on some.
Posts: 7605
Joined: Wed Aug 27, 2008 2:43 pm UTC
Location: Nederland

### Re: What's that one logical fallacy called?

Diadem wrote:I don't think Paley's watch-maker argument is a logical fallancy at all. The argument "If something is irreducibly complex than it can not have formed via evolution" is entirely sound. Darwin actually makea that very observation himself in The Origin of Species.

It's just that no such things have been found so far.

True, but in practice few people make such a moderate use of the argument. Most go immediately on to proving their own personal religious view of creation, including the son of a carpenter who was not really the son of a carpenter and died on the cross but didn't really die.

In fact, you might need an alternative view on creation just to give the argument power. Without an alternative, the choice will always be between "Evolution is disproved by object X" and "we have to think a bit longer about object X before we are sure it is irreducibly complex", and the latter will sound attractive in many situations. Negative facts are hard to prove.

You see this in biology all the time. When some new feature of an organism is discovered without a clear evolutionary history, no one concludes that evolution is now in doubt. Except people who do see a viable alternative in their religion.

phlip
Restorer of Worlds
Posts: 7573
Joined: Sat Sep 23, 2006 3:56 am UTC
Location: Australia
Contact:

### Re: What's that one logical fallacy called?

In my experience, though, the watchmaker argument isn't usually presented as "if something is irreducibly complex it can't have been made by evolution" (which is true, even if it currently doesn't apply to anything) but rather "if something is complex, it can't have been made by unrestricted randomness". Often backed up by the tornado-in-a-scrapheap-building-a-747 analogy. All of which stems from a misunderstanding of evolution, since it's not unrestricted randomness, there's natural selection in there too.

Code: Select all

`enum ಠ_ಠ {°□°╰=1, °Д°╰, ಠ益ಠ╰};void ┻━┻︵​╰(ಠ_ಠ ⚠) {exit((int)⚠);}`
[he/him/his]