Question on scientific method, evolution, creationism.

[Rob J]

Sometimes science is straightforward - you can just measure data to support your case eg toss a coin and count it and you can show head/tails is roughly 50/50.

 

Sometimes it is about having a theory that explains the data and slowly building on this case so that the scientific community accepts it as the most likely explanation. If your case it strong enough it almost becomes accepted as 'fact'.

 

An example is evolution - the fossil record, mutations in genes and observation of microbial evolution all support it as the most likely explanation.

 

However Creationists would say that God fits the same purpose, a single unifying theory that explains all the observations.

 

Philosophically how can you differentiate between these two?

[ajb]

A theory should be able to predict outcomes of some experiment/observation.

 

If god was controlling everything then we would lose the above assertion. That is, we would have no predictability as god could influence any experiment we undertook.

[the tree]

One is falsifiable, the other isn't.

[Rob J]

A theory should be able to predict outcomes of some experiment/observation.

 

You can't observe macroevolution.

[swansont]

You can't observe macroevolution.

 

Define "observe"

[Phi for All]

It's not really about the scientific community's acceptance. It *is* about the most likely explanation. You state it almost perfectly when you say

If your case it strong enough it almost becomes accepted as 'fact'.

Even with overwhelming evidence to support it, evolution remains a theory and so it waits for a more logical, better supported explanation. Remembering that science is interested only in observable phenomena, any suggestions that an inherently unobservable deity broke physical laws to instantaneously create matter that appears to be billions of years old will not fall within science's purview. Bring it up and you invoke faith so science simply has no further function in the discussion.

You can't observe macroevolution.

What?! Common descent is just one aspect of macroevolution that has had thousands of empirically tested predictions based on millions of observations. This type of testing does not assume a truth a priori the way creationism does. Observations lead to predictions and tests for those predictions. Common descent, transitional forms evidence and even molecular evidence help support evolutionary theory at the macro level.

 

It sounds as though you've listened to a few detractors talking about evolution rather than learn about evolution itself. Not good, not rigorous enough, not fair to yourself. You should study evolution to see what it really is, rather than listen to other people who also haven't studied it tell you what it's not.

[insane_alien]

You can't observe macroevolution.

 

well, if you are talking about species with long lifespans then we cannot DIRECTLY observe evolution, though, by looking at fossils and other remains we can collect indirect evidence. this is still classed as observing it by science.

 

for species with a short generation time, say bacteria, we CAN and HAVE observed evolution.

 

also, micro and macro evolution are very illdefined concepts imposed by the creationists.

[swansont]

Creationists would say that God fits the same purpose, a single unifying theory that explains all the observations.

 

Philosophically how can you differentiate between these two?

 

Ran across something that sums a lot of this up nicely, why while "Goddidit" is the simplest answer, it isn't a scientifically acceptable answer from the point of view of Occam's razor

 

http://skeptico.blogs.com/skeptico/2007/09/occams-razor.html

 

"Suppose I have a cat. One night, I leave out a saucer of milk, and in the morning the milk has gone. No one saw who or what drank the milk. Lets say there are two possibilities:

 

1. The cat drank it

or

2. The milk fairy drank it

 

Occam tells us to reject option 2. This is because option 2 requires us to invent an unnecessary entity - the milk fairy."

 

———

 

IOW, there is no good reason to hypothesize God from observing evolution.

[Rob J]

What?! Common descent is just one aspect of macroevolution that has had thousands of empirically tested predictions based on millions of observations. This type of testing does not assume a truth a priori the way creationism does. Observations lead to predictions and tests for those predictions. Common descent, transitional forms evidence and even molecular evidence help support evolutionary theory at the macro level.

 

Could you clarify what you are saying here here?

In the most rigorous form of scientific method a theory makes a prediction that you can then test.

 

Are you using the term 'prediction' to refer to predicting what pre-existing 'evidence' you are going to find?

 

In other words, an example of the usual definition of prediction in scientific method is tossing a coin and predicting 50/50 (ie theory->prediction->test prediction)

 

Are you using 'prediction' to refer to, for instance, what kind of fossil you are going to find?

 

This is better than nothing but ideally science should make prospective predictions not retrospective - that makes is easier to introduce bias because you will tend to find and notice the evidence you want, and ignore the rest.

 

It's not really about the scientific community's acceptance. It *is* about the most likely explanation.

 

But someone has to decide what 'the most likely explanation' is. If you can come up with a fantastic predictive experiment you can 'prove' something in one paper. Most things involve slowly building up the mass of evidence until the scientific community accepts it as the most likely explanation. There is a probability level that can a paper published. There is no fixed probablility level for when a theory becomes part of scientific cannon, that comes from majority consensus.

 

1. The cat drank it

or

2. The milk fairy drank it

 

Occam tells us to reject option 2. This is because option 2 requires us to invent an unnecessary entity -the milk fairy."

 

I read the blog. However:

The difference is we have seen the cat and haven't seen the milk fairy, so yes the milk fairy involves inventing/creating a new entity.

 

However both evolution and God have to be 'created' to explain the data.

[iNow]

To clarify, an example of the usual definition of prediction in scientific method is tossing a coin and predicting 50/50 (ie theory->prediction->test prediction)

Just to be clear, a coin actually has three sides, it's just incredibly improbable that it will land and stay on it's edge.

[Rob J]

Just to be clear, a coin actually has three sides, it's just incredibly improbable that it will land and stay on it's edge.

 

Sure, and it also won't be exactly 50/50.

[Paralith]

Could you clarify what you are saying here here?

In the most rigorous form of scientific method a theory makes a prediction that you can then test.

 

Are you using the term 'prediction' to refer to predicting what pre-existing 'evidence' you are going to find?

 

In other words, an example of the usual definition of prediction in scientific method is tossing a coin and predicting 50/50 (ie theory->prediction->test prediction)

 

Are you using 'prediction' to refer to, for instance, what kind of fossil you are going to find?

 

This is better than nothing but ideally science should make prospective predictions not retrospective - that makes is easier to introduce bias because you will tend to find and notice the evidence you want, and ignore the rest.

 

I think all observable phenomena (both directly and indirectly observable) would count as "pre-existing evidence." If you hypothesize that a plant will grow better if you give it fertilizer than if you don't, the plant on fertilizer does grow better because the mechanisms to take advantage of the fertilizer already existed. If you are testing a process, like evolution, one way to test it is to look at instances where the process already happened, look at what it started with, and see if your hypothesis can explain what it ended with, or vice versa. ANY type of scientific testing can be subject to bias. I don't think it's fair to say that testing against a fossil record is somehow more susceptible to bias than testing against other evidence.

[Rob J]

If you hypothesize that a plant will grow better if you give it fertilizer than if you don't, the plant on fertilizer does grow better because the mechanisms to take advantage of the fertilizer already existed

 

The mechanisms pre-existed but the phenomenon of the plant growing better didn't.

 

An example of the difference between prospective/ retrospective analysis:

 

If I make a prediction that wearing a red hat will win me the lottery tomorrow and I win the lottery, that's pretty impressive. Plus, if it's a genuine correlation I can repeat it again and again.

A 1/10 million lottery repeated 3 times has a (1/10 million)cubed probability

 

If I look back at the evidence and find that a lot of people with red hats have won the lottery, even if I can show it's statistically significant, it's not so impressive. To increase the significance all you can do increase the size of your dataset.

[Paralith]

The mechanisms pre-existed but the phenomenon of the plant growing better didn't.

 

An example of the difference between prospective/ retrospective analysis:

 

If I make a prediction that wearing a red hat will win me the lottery tomorrow and I win the lottery, that's pretty impressive. Plus, if it's a genuine correlation I can repeat it again and again.

 

If I look back at the evidence and find that a lot of people with red hats have won the lottery, even if I can show it's statistically significant, it's not so impressive.

 

I see. So timing is all that matters? No matter what way you look at it, the results are the exactly the same: a statistically significant correlation of red hat wearers winning the lottery. A correlation by itself is hardly means anything, no matter how you arrived at it. You need other types of supporting evidence from multiple viewpoints and data that all continually support the red hat hypothesis.

 

This is the problem with your criticism of fossil evidence. It might hold more water if fossil evidence was the only line of evidence used to support evolution. But it isn't. Fossil evidence is just one of many different lines of evidence that all align to support evolutionary theory. By itself fossils may not have as much strength, but as yet another of many supporting elements, it can be very valuable.

[Rob J]

IA correlation by itself is hardly means anything, no matter how you arrived at it. You need other types of supporting evidence from multiple viewpoints and data that all continually support the red hat hypothesis.

.

 

Not true. If I was to repeat the red hat experiment twice with a probability of

(1/10 million)squared the whole world would take notice (if it wasn't such an absurd example and so unbelievable). If it was a reasonably authoritative body doing the experiment it might even change the scientific cannon immediately.

It would be much harder to get this reaction with a retrospective analysis.

 

Prospective analysis is always more potent than retrospective analysis. A lot of pseudoscience (psychics etc) is reliant on this. People will talk about coincidences in their past and claim psychic ability. They don't predict winning lottery numbers.

 

Btw, by analogy with your plant example, the mechanism causing the red hat to increase chance of winning the lottery pre-exists but not the the phenomenon of winning the lottery.

[foodchain]

The mechanisms pre-existed but the phenomenon of the plant growing better didn't.

 

An example of the difference between prospective/ retrospective analysis:

 

If I make a prediction that wearing a red hat will win me the lottery tomorrow and I win the lottery, that's pretty impressive. Plus, if it's a genuine correlation I can repeat it again and again.

A 1/10 million lottery repeated 3 times has a (1/10 million)cubed probability

 

If I look back at the evidence and find that a lot of people with red hats have won the lottery, even if I can show it's statistically significant, it's not so impressive. To increase the significance all you can do increase the size of your dataset.

 

You are going in all kinds of directions really. If what you are looking for is science making a goat breed in a few seconds into a boar or something, no, science cannot do that as of yet...

 

Macro and micro evolution get confused a lot with the simple term bauplan. I think thats what most of the creationist arguments go into in that regard. The bauplan is not that differentiated along evolutionary lineages though. Such as our skeleton as humans compared to other primates, or the bones for that matter which make up your hand in comparison. More evidence on fossil records in action can be found with whales, as they still retain in some degree vestigial bones which serve absolutely no purpose at all, such as having bones that would relate to a leg, or legs, or being on all four limbs. Micro and macro evolution truly can be viewed in regards to advantageous traits in a sense of time. Such as in a human sense going directly to work and not college could be advantageous in the short term but in the long term not work out to well for the individual, such as in successive generations.

 

Evolution also has evidence in the form of homologies from a genetic, or molecular sense. IN which evolution used as the backbone of such studies has offered in the ability to gain more understanding of biology, if evolution were not anything such endeavors would produce null results.

 

What science can do is force evolution to occur in a laboratory. In which for ethical reasons bacteria is used. IN which the same population of bacteria exposed to a medium which is not nice to the health of the bacteria will induce the bacteria to change via natural selection from mutations. In order mutations that favor survival in this medium become more prominent while other traits which are not as successful reduce in appearance. This can lead to easily visible changes in the appearance of the bacteria also. Its also unpredictable really, in that amount of mutants, type of mutations, and many other variables will be different on each test group, some groups don’t even make it. Also, genes, the parts in which lead to genotypes and phenotypes that get worked on by natural selection can have varying degrees of conservation, and or rate that mutations impact them. Also how gene flow in species or populations works can vary. The bottom line is that natural selection can be shown in a lab, its just science cant make a temporal laboratory that can run say 50 million years of evolution in a few seconds.

 

The last point I would like to make is that you really have to know about evolution and biology in general and all its nifty fields before you can really digest creationist arguments. Its a simple fact that they lie and distort data to fit a mold. They have been caught doing this trying to debunk how old the earth is, and famous ones in there ranks will say certain bodies of data don’t work or exist even while books are being published on the subject on a regular basis. They hold a great deal of bias, and should not be looked to really in order to engage science on some debate why something might be wrong or not completely right.

 

A basic example would be evolution as a math equation, in which not all the variables are known yet. Science in regards to evolution goes about trying to figure out what it is, those variables. Creationist means would basically insert god into the unknown variables and call it good. Its hardly science, its hardly honest and basically its a lie. ITs like the entropy argument and plants using photosynthesis, personally I don’t think they even go outside really.

[Paralith]

Not true. If I was to repeat the red hat experiment twice with a probability of

(1/10 million)squared the whole world would take notice (if it wasn't such an absurd example and so unbelievable). If it was a reasonably authoritative body doing the experiment it might even change the scientific cannon immediately.

It would be much harder to get this reaction with a retrospective analysis.

I disagree. Before the scientific cannon changed you'd need further studies to at least prove that there wasn't any other factor at play - for example, maybe all those red hat people chose similar numbers, and it was in fact the numbers chosen that affected who won, and not the color of hats worn. It just seemed that way, since people who pick similar numbers also have similar tastes in other things, such as clothes. These kinds of things in the least need to be ruled out. Additional corroborating studies are required for this. (As you say, if this lottery situation wasn't the made-up example that it is.)

 

Prospective analysis is always more potent than retrospective analysis. A lot of pseudoscience (psychics etc) is reliant on this. People will talk about coincidences in their past and claim psychic ability. They don't predict winning lottery numbers.

Psychics etc. will only pick the instances their predictions came true to use as evidence. The same use of bias can be done with prospective analysis - continuing with the example above, you chose only to look at hat color vs. lottery winnings, and chose to not make any other predictions and see how they came out as well. If we accurately assessed every prediction the psychic ever made and saw how many of them actually came true, the truth of the matter would be easy enough to see.

 

Btw, by analogy with your plant example, the mechanism causing the red hat to increase chance of winning the lottery pre-exists but not the the phenomenon of winning the lottery.

 

That seems kind of silly to me. So a phenomenon is only real if it never happened before but happens when you set up the right experiment? The theory of gravity is only valid because when a human came along and predicted that a dropped pencil would fall to the ground, it fell? All the other times that pencils fell in the past means nothing?

[swansont]

I read the blog. However:

The difference is we have seen the cat and haven't seen the milk fairy, so yes the milk fairy involves inventing/creating a new entity.

 

However both evolution and God have to be 'created' to explain the data.

 

That may have been true in the past, but DNA, mutation, selection, gene drift, etc. are now known. Nothing has to be created. Otherwise, logically, God is as good an explanation as gravity for why we don't fly off the earth and why the planets stay in orbit.

[PhDP]

Evolutionary biology produces predictions and tests, both at the so-called "macro" and "micro" levels. Most of the "hard" proofs of "macro"-evolution come from either palaeontology or molecular evolutionary biology, and most predictions are formed by theoretical population genetics/molecular evolutionary genetics/quantitative genetics, and are tested within the framework of molecular biology, it’s just that creationists don't bother to read about evolutionary biology before attacking it. If you want proofs and predictions, just read a book about molecular evolution, the books by Wen-Hsiung Li are generally very good.

[pioneer]

A theory should be able to predict outcomes of some experiment/observation.

 

According to evolutionary theory what is the next form of humans?

 

I believe in evolution, but I don't buy into much of the theory used to support it because it say lots without having to say anything. It is too empirical and doesn't follow logical principles. It sort of depends on the gods of random and chaos instead of the gods of order. The creationists use the gods of order but don't have the science to back it up. The evolutionist use the gods of chaos to help support their science.

 

I did this example before but I will do it again to show a possible problem with the current assumptions behind evolution. Say you kept a journal of a child ,from birth to 21 years old. Each day you make an entry. At the end, we try to simulate the discontinuous data of evolution, by asking someone to randomly pick 100 days from that 21 years, and using only that data, tell us why that young man is who is, on his 21st birthday.

 

The 100 random data points are sort of hit and miss with respect to picking what could and may have been pivotal times in the childs life. The theory that one may come up with, may fit the chosen data perfectly, but it may not reflect the reality of why the child is who he is. For example, on the 200th day of his 17th year, he was in a automoble accident. If we didn't pick this day or didn't have this data, the entire analysis may be consistent with the chosen data, but still be moot. Based on our data, we may conclude nothing in his environment caused this change. Therefore we conclude, it must be in his genetics. Genetics is thrown around like a trump card, in an areas of science based on gambling statistics.

 

That is my beef. The discontinuous data may have gaps such that this data, of itself, may not reflect the reality of the modern situaiton. So the push should be a logical system that doesn't depend on broken data put should be able to explain it, as part of the logical progression. For example, going from single to multicellular should be logical. Selective advantage, mutations or some type of gambling theory is pitiful. This is so basic and fundamental to life, yet it is treated with a gambling ritual. The god of chaos needs one to sacrifce reason to worship him properly. This is why religion is trying to step in, since the god of choas is an illusion that only works because the discontinuous data helps support this illusion.

 

I do not support Creationism in the literal sense. But I do support the basic idea that an orderred sequence of events occurred. Evolution was not a long chain of gaming tables, but more like a walk up spiral stairs.

[PhDP]

I'm curious, pioneer, have you read a book about molecular evolution and population genetics or you're just bashing what you don't understand ?

 

It's clear that, following the discovery of molecular genetics, we've understood that "randomness" (drift,mutations,et cetera...) was much more important than we initially thought. But it's something that is based on powerful mathematical models and empirical studies both at the level of the genome and by studying the paleontological record, it's certainly not a "god".

 

I think it's also clear, reading the comments made here by nonbiologists, that many people are just not going to accept "randomness".

[foodchain]

I'm curious, pioneer, have you read a book about molecular evolution and population genetics or you're just bashing what you don't understand ?

 

It's clear that, following the discovery of molecular genetics, we've understood that "randomness" (drift,mutations,et cetera...) was much more important than we initially thought. But it's something that is based on powerful mathematical models and empirical studies both at the level of the genome and by studying the paleontological record, it's certainly not a "god".

 

I think it's also clear, reading the comments made here by nonbiologists, that many people are just not going to accept "randomness".

 

I agree. I think its because of the time thing again in evolution for one and lacking any real ability to transmute if you will what the randomness means. You cant study evolution in terms of phylogenics and say hey, it was bacteria to intelligent species in one single series. If not for the extinction of the dinosaurs for instance I doubt mammalian evolution would be anything that it is today for instance. You cant combine natural selection and artificial selection into natural selection, for organisms to pick and choose or have visibility say for a bacteria to know how to evolve somehow inside of it into anything would be artificial selection, and it would be the decider, obviously this is not the case I would hope. If things were not random every mutation would be in a micro and a macro sense then nothing but beneficial. So would be the relationship in the genotype/phenotype area. Evolution by the reality of it implies that life does not know how to for instance always save itself from extinction, or such again would not exist, evolution would be perfect and I doubt for it to have taken so long to even occur. People with understanding of such can conduct artificial selection, but that is artificial selection and even in that such is not perfect because you cannot foretell perfectly what a mutation will be and or where it will occur within absolute precision. Its akin to QM really in that regard, or an uncertainty principle really. Naturally then whatever mutations in time that conduct advantageous traits over time survive longer in a particular gene pool, but whatever those happen to be really is naturally selected. I don’t know if its a good example of convergent evolution really but the duckbill platypus I think is a good example of traits that are beneficial being selected naturally.

[Paralith]

That is my beef. The discontinuous data may have gaps such that this data, of itself, may not reflect the reality of the modern situaiton. So the push should be a logical system that doesn't depend on broken data put should be able to explain it, as part of the logical progression. For example, going from single to multicellular should be logical. Selective advantage, mutations or some type of gambling theory is pitiful. This is so basic and fundamental to life, yet it is treated with a gambling ritual. The god of chaos needs one to sacrifce reason to worship him properly. This is why religion is trying to step in, since the god of choas is an illusion that only works because the discontinuous data helps support this illusion.

 

Mm-hmm. Oh yes, it's all about the roll of the dice. It's quite random that animals living in cold climates have fur that insulates them from the cold and that the ones without fur died from exposure. If you actually understood evolutionary theory, you'd understand that selective advantage is a far cry from gambling. Selective advantage allows that which works best in a given environment to carry on, while that which does not work does not. The variety is generated randomly, but the selection of what stays and what goes from within that variety is not.

 

Besides, how do you suppose we test this logical progression of yours, my friend? Because after all, the only thing we can test it on is this "broken data" that consists of all the real-world observations ever made about biological systems. Incomplete knowledge of the incredibly complex world around us is a fact of life, and whatever theories you "logically" come up with won't do anyone any good unless we know they reflect reality, which can't be known unless it is tested against the empirical data that you so despise. And that there's no getting away from.

[CDarwin]

I've laid this out before and I will again because people seem to get hung up on 'randomness.'

 

Evolution occurs in two stages: The first the origin of variation. That's mostly random. The second is the elimination of that variation. Sometimes that too is due to random events (like rock slides killing gazelle who just happened to be walking near by), but often it's non-random selection. So evolution has both 'random' and 'non-random' components.