# Statistical Argument Against Evolution

## Recommended Posts

A creationist once told me that the probability that evolution occured according to scientific theory is very low. There is a low probability that the steps of naturalistic processes that occur could happen that way.

But I've been thinking. If you take a coin and flip it a billion times, whatever result you get, the probability of obtaining that result is extremely low, yet is that evidence that what you just did (flip the coin a billion times) didn't occur?

• Replies 70
• Created

#### Popular Days

thats right. the chances of someone winning the lottery are very low, but that doesnt stop it from happening on a regular basis.

##### Share on other sites

Really, the problem is that we just don't know what sparks life in the first place. For all we know, it's easy, just not somethng we've yet learned to understand, and probably won't until we discover more life on other worlds. Let's say we find fossil or hibernating life on mars, and then more living in the Europan seas. Suddenly, a whole new equation is necessary.

##### Share on other sites

That's because of this retarded list of "probabilities" that certain creationists circulate' date=' which is basically a stack of nonsense "prerequisites" for life emerging, each of them tagged with the arbitrary probability of one in a billion. It's tripe.

[/quote']

Besides, the probability of an event occurring is 1 if it has already occurred. You can't use probability to go back and prove that the lottery winner didn't actually win the lottery.

##### Share on other sites

the one that i really like is when they ask me "what? and we just HAPPEN to be on the planet that life sprung up on? out of all the billions of planets in the universe it just magically takes place on OUR planet? what are the odds of that?"

/chuckle

(almost gives you a headache to even read it, doesnt it?)

##### Share on other sites

Probabilities are a way for us to model our own uncertainty, but the actual behavior of the thing being modeled may not be random or probabilistic at all. For instance, coin tosses are often regarded as random with 50/50 odds. But is the behavior of a coin truly random? You can write equations of physics which describe the behavior of the flipped coin to the letter (granting any assumptions the equations make), so it really isn't random at all.

Likewise, there are rules/equations which describe organic chemical reactions exactly. Proteins will form a certain way under certain conditions; if this were not the case and protein formation were truly random, we would not exist as our bodies would be cranking out random proteins. There may be some uncertainty that we would have to model with probability, but this stems from our own inability as observers to account for all possible factors that affect the reaction, not from the reaction itself being random in any sense of that term.

##### Share on other sites

Probabilities are a way for us to model our own uncertainty

No.

Uncertainty will obey binary logic.

either you are uncertain or you aren't uncertain

probability can take on a range of values between zero and one.

I can prove to you what I say, with the following experiment designed to demonstrate that probability is not a measure of belief.

I give you a bag with 100 things in it, and I tell you that if a thing is in the bag then the thing is a swan, and if a thing isn't in the bag, then the thing isn't a swan. You now know there are exactly 100 swans in existence, even though you have never seen a swan, nor has any definition been given.

You then take out your first swan, and note that it is white. I then ask you what is the probability that all swans are white, and without thinking you just perform the following calculation P(all swans are white) = 1/100. You then map this number to your uncertainty.

You then pull out 49 more swans, and each of them was white. Again I ask you what is the probability that all swans are white, and without thinking you perform the following calculation:

P(all swans are white) = 50/100 = 1/2

Currently, you neither doubt, nor believe all swans are white, but you respond to me by saying 1/2.

You pull out 49 more swans and note that all of them are white. Again I ask my now famous question, "What is the probability that all swans are white?" And without thinking you respond 99/100. You then map this number to your own uncertainty. You now respond to me by saying, "I currently believe that all swans are white, and doubt that there is at least one swan that isn't white."

You now pull the last swan out of the bag, and note that it is black, are you surprised?

Regards

##### Share on other sites

everything in that post seems arbitrary. why, after pulling 50 white swans out of a bag would you assume that half of all swans are white? thats just totally illogical. i would expect a child to reach that conclusion but not anyone with any knowledge of statistics. any educated person, after pulling 50 white swans in a row would be of the belief that much closer to 100% of swans are white.

but besides all that, how does that counter what he said? the fact is, as he stated, some things are not random chance, they are determined by physics, probabilities are our way of expressing the fact that we dont have the facts to determine which way it will go. if we knew each and every one of the forces acting on the coin we could determine which way it will land, but we dont. so despite the fact that the coin WILL land a certain way, we dont know which way, it is our lack of knowledge that makes for the 50/50 chance.

##### Share on other sites

No.

Uncertainty will obey binary logic.

either you are uncertain or you aren't uncertain

What if I am uncertain about whether I am uncertain?

As I understand it, the whole concept of low probabilities surrounding evolution rests on too many assumptions to be a valid argument. First we must assume what we know as 'life' is the only form capable of existence - big guess, and a bit arrogant, in my opinion.

I've seen this debate stretch from the origin of proteins to the affects of gravity itself. Our solar system, with relative distances from the Sun and Jupiter helping us out, is especially suited to promote life on Earth (and possibly Mars). This seemingly random structuring is the end result of gravitational forces, so not really that random, or improbable, after all...

The response? "gravity operating as it does to produce the result that it has is statistically unlikely." Actual quote.

Personally, I find it cute when religious fundamentalists try and use science to prove their beliefs. Kind of like a bald man trying to use a comb...

-JC

##### Share on other sites

No.

Uncertainty will obey binary logic.

either you are uncertain or you aren't uncertain

probability can take on a range of values between zero and one.

It is possible to quantify the level or degree of certainty or uncertainty.

##### Share on other sites

Johnny5 appears to be viewing it this way:

If our expectation of an event is certain then probability is not an issue. If we are uncertain then, and only then, is that uncertainty expressed as a probability.

Johnny5? Seems a little non-standard?

##### Share on other sites

everything in that post seems arbitrary. why, after pulling 50 white swans out of a bag would you assume that half of all swans are white?

I didn't say you would assume any such thing.

any educated person' date=' after pulling 50 white swans in a row would be of the belief that much closer to 100% of swans are white. [/quote']

Why do you say this?

but besides all that' date=' how does that counter what he said? the fact is, as he stated, some things are not random chance, they are determined by physics, probabilities are our way of expressing the fact that we dont have the facts to determine which way it will go. [/quote']

I was only responding to one of many things he said. The one thing which I responded to, was his statement that probability is a way for us to measure our own uncertainty.

The problem I gave illustrates why you cannot quantify uncertainty using probability. If you really did map those numbers onto your own uncertainty, you will have foolish beliefs and/or foolish doubts, prior to seeing the 100th swan.

A perfect reasoning agent would neither believe all swans are white, nor doubt all swans are white, until he actually saw a non-white swan. If all swans are white then he would be uncertain as to the truth value of the statement, up until he saw his 100th swan, at which point he would be certain as to the truth value of the statement, and not one moment before.

if we knew each and every one of the forces acting on the coin we could determine which way it will land' date=' but we dont. so despite the fact that the coin WILL land a certain way, we dont know which way, it is our lack of knowledge that makes for the 50/50 chance.

[/quote']

This is a separate issue, and one which I did not address with the swan problem.

##### Share on other sites

It is possible to quantify the level or degree of certainty or uncertainty.

How?

PS: Only if your system of probability has two values, 0,1.

How?
Experience
##### Share on other sites

Why do you say this?

because thats the best guess, statistically.

I was only responding to one of many things he said. The one thing which I responded to, was his statement that probability is a way for us to measure our own uncertainty.

The problem I gave illustrates why you cannot quantify uncertainty using probability. If you really did map those numbers onto your own uncertainty, you will have foolish beliefs and/or foolish doubts, prior to seeing the 100th swan.

after seeing one white swan out of a hundred you would probably not be very confident about the color of the remaining 99. after seeing 50 you would probably be expecting most, if not all swans to be white, but not enough to be sure. after seeing 99 white swans you would probably be fairly certain that the last one is white as well, but not absolutely sure. hence, varying degrees of certainty. the fact that the last swan is not actually white does not change the fact that after seeing 99 white swans you are fairly certain that the last one is too. being wrong doesnt mean you didnt have a greater degree of certainty that you did at the beginning.

i believe another problem lies in how you are quantifying it. after seeing 2 white swans the probablity isnt 2/100. if all other swans were black the probability of getting 2 white ones in a row is 2/100 X 1/99, which is 2/9900. which makes this a pretty unlikely scenario. if 50 swans are white then the probability is 50/100 X 49/99, which is 2450/9900, making it far more likely that there are 50 white swans in the bag. the problem here (that i cant figure out how to quantify) is that given this logic the most likely possibility, after drawing ANY number of swans, is that they are all white, which probably isnt the assumption most people would make. does anyone know how to work this into the equation?

##### Share on other sites

How?

PS: Only if your system of probability has two values' date=' 0,1.[/quote']

No. It depends on the system, but statistics lets you determine confidence levels.

##### Share on other sites

No. It depends on the system, but statistics lets you determine confidence levels.

You say it depends upon the system, I'm not sure what you mean.

As far as what are called "confidence levels" go, in the computation of a confidence level, you choose alpha, and beta, before you use the charts, so you can fudge. I know you can fudge, I've taken statistics, I know how to do it (fudge). So statistical analysis doesn't provide genuine certainty. But I realize that's not what you are saying, you are asserting that you can break uncertainty into levels, and statistics proves it. And I repeat, "No, the issue is binary. You can't be .7 of certain."

Using statistics, you can conclude that smoking causes cancer. You could also use it to determine that drinking water causes cancer too.

But the issue raised, is about genuine certainty of an individual. I think its also a very understandable, and personal issue. Can the certainty level of an individual be gauged using numbers or not? What I will say, is that the rules of uncertainty/certainty apply to all individuals, equally.

So that my position is clear, I am emphatically saying no, you cannot measure uncertainty (in any non-trival sense), because at any moment in time, on any specific issue, either an individual is certain of something or not. If not, then the individual is uncertain. To say that person X is half as uncertain as person Y, on whether or not statement S is true, is bad epistemology.

This doesn't mean that statistical analysis isn't useful, it's just that you cannot apply it to human certainty, and I think that is what we are talking about here. The certainty/uncertainty of a human reasoning agent. You have to interpret statistics another way, you cannot use it to break human certainty into degrees. I think I've made my point clear.

##### Share on other sites

because thats the best guess' date=' statistically.

after seeing one white swan out of a hundred you would probably not be very confident about the color of the remaining 99. after seeing 50 you would probably be expecting most, if not all swans to be white, but not enough to be sure. after seeing 99 white swans you would probably be fairly certain that the last one is white as well, but not absolutely sure. hence, varying degrees of certainty. the fact that the last swan is not actually white does not change the fact that after seeing 99 white swans you are fairly certain that the last one is too. being wrong doesnt mean you didnt have a greater degree of certainty that you did at the beginning.

[/quote']

This is exactly the kind of reasoning which a stochastic reasoning agent will demonstrate. And it is wrong.

After seeing 99 white swans you should not be fairly confident that the last swan is white. You should be totally uncertain, and for the following reason...

Suppose instead of giving you a bag with 100 swans in it, I gave you a bag with only one swan in it. You have never seen a swan before in your life, and you know that the bag contains the only swan in existence. I ask you, what is the probability that all swans are white. Well you haven't looked in the bag yet, you have no clue, you are TOTALLY uncertain.

Well this problem with N=1, is equivalent to the other problem with N=100, and you just took out 99 white swans. So why would you be totally uncertain in one case, and partially certain in the other case, when they are one and the same case.

##### Share on other sites

the one that i really like is when they ask me "what? and we just HAPPEN to be on the planet that life sprung up on? out of all the billions of planets in the universe it just magically takes place on OUR planet? what are the odds of that?"

/chuckle

lol; what i always say in reply to this is:

immagine that you are a timless entity. you watch a planet coalesc out of space-dust and achieve a stable orbit around the nearest sun, you see its climate stabalise, mountian ranges rise and fall, oceans form, freze, thaw, you see the tectonic plates move about, and eventually you see the sun expand as it reaches the end of its fuel supply and you see the planet burned back into space-dust: all without life spontaniously occournig.

you now choose another planet, just coalescing out of space-dust, and you watch this one. again, after millions of years, it is distroyed without ever having spawned life.

you find another newly-forming planet...

...you have just watched your 9,999,999,999th planet be distroyed without forming life. you start to watch a new planet, coalescing from space-dust, you see its climate stabalise, mountains rise and fall, continents shift around on the changing oceans, and then, after failing to happen so many times, you finally see life emerge, and gain the capability of inteligent thought and language. then one of them goes "ooh, isnt it unlightly that we just randomly happened? the odds must be one-billion-to-one!"

people generaly still wont understand. i think that people who dont understand the logic just plain and simple dont want to understand the logic, because they percieve it to contradict their religion.

##### Share on other sites

so you have a bag with 100 swans. you pull out 99, and they are all white.

two hypothesis can fit the fact that 99 swans are known to be white, vis:

H1: 99 swans are white and 1 swan is non-white

H2: 100 swans are white.

the probability that the first 99 swans out of the bag will be white is the P(1st swan is white)*P(2nd swan is white)*P(3rd swan is white)...*P(99th swan is white)

the probability that a white swan will be chosen = the number of white swans/the total number of swans.

if H1 is true, then the P(first 99 swans are all white) = (99/100)*(98/99)*(97/98)...*(1/2), 1/2 being the 99th choice, where according to the hypothesys there would be two swans left, one of which is white and one of which is not. this gives a probability of 1/100 (logical, concidering the probibility that the non-white swan would be picked last is equivelant to the probability that the non-white swan would be picked 1st = 1/100)

so there would be a 1/100 chance of getting the observed results if H1 is the case.

If H2 is the case, then there would be a 100/100 chance of getting the observed results (if all 100 swans are white, the chance of the first 99 swans being picked would be 1).

not entirely sure where to go from here, but i think that the following is logically/mathematicaly true...

probability of obtaining the observed result if the given hypothesis is true:

H1: 1/100

H2: 100/100

therefore, H2 is 100 times more lightly to yield the observed results than H1, ie P(H1 is corect) = 1/101, P(H2 is correct) = 100/101

or in other words, H2 has a 99.00990099(reccuring)% chance of being correct,

accept H2 (at 99% confidence interval)

so not certain that the last swan would be white, but would quite highly expect it to be. have also quantified my uncertanty with the confidence interval, which is essentialy the P(my assumption is correct)

would i be surprised if it were black? yes and no. yes because there is only a 1/101 chance that it would be non-white. no because 1/101 chance is not the same as 0 chance. im sure that things with only a 1/101 chance of happening fail to happen just over 99% of the time, but occasionaly they do happen

##### Share on other sites

As far as what are called "confidence levels" go' date=' in the computation of a confidence level, you choose alpha, and beta, before you use the charts, so you can fudge. I know you can fudge, I've taken statistics, I know how to do it (fudge). So statistical analysis doesn't provide genuine certainty. But I realize that's not what you are saying, you are asserting that you can break uncertainty into levels, and statistics proves it. And I repeat, "No, the issue is binary. You can't be .7 of certain."

[/quote']

Sure you can. The issue is not binary. If it were, and there was only uncertainty, then you could not evaluate one possibility over another.

If you have your bag with 100 swans in it, and you know one is black and 99 white, you cannot be certain if you will draw a black or white one from the bag. According to you, you have no clue, so you must conclude that it is a 50-50 chance. Which is wrong.

##### Share on other sites

This is exactly the kind of reasoning which a stochastic reasoning agent will demonstrate. And it is wrong.

After seeing 99 white swans you should not be fairly confident that the last swan is white. You should be totally uncertain' date=' and for the following reason...

Suppose instead of giving you a bag with 100 swans in it, I gave you a bag with only one swan in it. You have never seen a swan before in your life, and you know that the bag contains the only swan in existence. I ask you, what is the probability that all swans are white. Well you haven't looked in the bag yet, you have no clue, you are TOTALLY uncertain.

Well this problem with N=1, is equivalent to the other problem with N=100, and you just took out 99 white swans. So why would you be totally uncertain in one case, and partially certain in the other case, when they are one and the same case.[/quote']

thats just not the way people work. while the actual odds are correct it doesnt really fit the problem.

you have spent your entire life studying dolphins and every dolphin you have ever seen is grey, after seeing millions of them. one day you come across a purple dolphin. are you surprised? YES. should you be? absolutely.

it is not the same scenario as the very first time you ever see a dolphin. you have experience, with experience come expectations and a level of certainty.

it works the other way too:

technically the probability of giving birth to a boy after giving birth to 9 other boys is exactly 50/50, how many people are expecting a 10th boy to come out?

we have a certain level of knowledge about our world, with that we can predict they way things SHOULD happen. whether the technical odds are the same or not.

##### Share on other sites

This doesn't mean that statistical analysis isn't useful, it's just that you cannot apply it to human certainty, and I think that is what we are talking about here. The certainty/uncertainty of a human reasoning agent. You have to interpret statistics another way, you cannot use it to break human certainty into degrees. I think I've made my point clear.
Yes, you have made your point clear, but it is a highly individual point and I am certain it constitutes an opinion not an absolute, objective reflection of reality.

I fully accept that when you are assessing the external world that your certainty is indeed binary. This is neither good nor bad, right or wrong, it is simply different from how some others deal with uncertainty. I have no difficulty assigning a degree of uncertainty to either the probability of events occuring, or to a particular hypothesis being accurate or inaccurate. This is not a binary condition.

I am certain that I have made my point clear to some readers.

##### Share on other sites

Sure you can. The issue is not binary. If it were' date=' and there was only uncertainty, then you could not evaluate one possibility over another.

[/quote']

I don't see how this applies.

If you have your bag with 100 swans in it' date=' and you know one is black and 99 white, you cannot be certain if you will draw a black or white one from the bag. According to you, you have no clue, so you must conclude that it is a 50-50 chance. Which is wrong.[/quote']

You do not know if there is a non-white swan in the bag a priori. Interesting problem huh?

## Create an account or sign in to comment

You need to be a member in order to leave a comment

## Create an account

Sign up for a new account in our community. It's easy!

Register a new account