Debate:Falsifiability/Archive1

This is a Debate page. Feel free to add your own spin on the story. Please keep it civil! This debate was created by signed users.

Ungtss

Moved from RationalWiki talk:What is going on at ASK?

Science hater Ungtss declares materialists use science to prove the non-existence of god. The usual "science can't explain free will" and other creationist arguments pop up. And that science can't test history! Glad to know creationists are redefining science in order to point out its shortcomings to other creationists.

I guess if you think falsificationism is a "creationist redefinition of science," you haven't troubled yourself to read Popper. Try it, you'll like it. Ungtss 16:41, 3 May 2009 (UTC)

What has your quote got to do with falsification? --Kels 16:48, 3 May 2009 (UTC)

The article I wrote defined science as experimentally falsifiable. The RW response quoted above accused me of "redefining science." Ungtss 16:52, 3 May 2009 (UTC)

And the relevance to the quote you placed above is...? --Kels 17:51, 3 May 2009 (UTC)

The last sentence of the quote is "Glad to know creationists are redefining science ..." I then responded to that quote by saying, in essence, "I don't see how falsificationism is a redefinition of science." Ungtss 17:59, 3 May 2009 (UTC)

Falsificationism isn't, as far as I know. It's the rest of it that seems to be the problem. --Kels 18:01, 3 May 2009 (UTC)

There are a number of materialists that argue that science falsifies (or at least makes highly unreasonable) the existence of God. I'd support that assertion with [this] link. The second sentence provides no rebuttal, so I don't know what to say. As to whether science can or cannot test history, what repeatable experiments can be performed to falsify our historical understanding of the Trojan war? Ungtss 18:18, 3 May 2009 (UTC)

You really don't get what "science" is, do you? History is scientific so far as it is empirical, ie. tests hypotheses against extant evidence. Falsification occurs when new-found data contradicts the hypothesis. It is no more or less repeatable than astronomy. 18:40, 3 May 2009 (UTC)

I'm a little curious how Ungtss thinks we verify dates, check veracity of documents, learn how undocumented artifacts were constructed, and preserve all of those things without science. I also don't buy that a few individuals say much about a large movement existing to falsify God, nor that it has much to say about the foundations of science or its modern mainstream. --Kels 18:50, 3 May 2009 (UTC)

Now who's redefining science? Falsificationism requires a repeatable experiment that could disprove a hypothesis. e.g., throw something up in the air and see if it doesn't come down. History doesn't work that way. That's why we have disputes in history -- because our understanding of it cannot be experimentally tested. Was Stonehenge an astronomical calendar or a temple? Who wrote the Gospels? Did the Vikings find the continent of North America? We can't use science for that. For history, we are dependent on error-prone historical accounts and error-prone interpretation of the evidence. Science is not dependent on those, because an experiment can be repeated.

It's naive to say that all empirical methods of study are "scientific." Science is falsifiable. Not all empirical methods of study are falsifiable, although they may well lead to reliable conclusions.

But if you'd like to redefine science, go ahead. If science doesn't require repeatable, falsifiable experiment, then what does it require? Ungtss 18:58, 3 May 2009 (UTC)

I would say you are. By seizing on the word "falsification" and claiming that it means solely experimentation and nothing else. And further, by claiming that science cannot and should not study things that have been the purview of religion such as the existence of the soul (at one point, I believe, it was assumed to be physical, was it not?). So by artificially restricting the definition of science, you are, in fact, redefining it. --Kels 19:09, 3 May 2009 (UTC)

Oh, and before I forget, archeology uses many scientific techniques, so yes, your examples about Stonehenge and the Vikings would apply. As to the Gospels, all of the ancient writings that make up the Bible have been subject to ongoing scientific examination, including the hated C-14 dating, DNA testing, chemical analysis of ink and paper, and a host of other tests, all of which are unquestionably scientific in nature. --Kels 19:21, 3 May 2009 (UTC)

Seizing upon the word "experimentation?" Experiment is the heart of the scientific method. What model of the scientific method lacks "experiment" as a key component? As to studying the unfalsifiable, I am not saying it "cannot and should not." I am saying it "cannot, but other methods can." History is a very valuable realm of inquiry, and there are many innovative means to perform it. But unless experiment can be performed, it is not science. Ungtss 19:19, 3 May 2009 (UTC)

Please read again, thanks. --Kels 19:21, 3 May 2009 (UTC)

You haven't answered my question. If science does not require experiment and falsification, then what does it require? Ungtss 19:29, 3 May 2009 (UTC)

your examples about Stonehenge and the Vikings would apply.

How? What experiment can we run to determine whether Stonehenge was a temple or a highly advanced observatory? Ungtss 19:33, 3 May 2009 (UTC)

As to the Gospels, all of the ancient writings that make up the Bible have been subject to ongoing scientific examination, including the hated C-14 dating, DNA testing, chemical analysis of ink and paper, and a host of other tests, all of which are unquestionably scientific in nature.

But none of those methods can answer the question, "Who wrote them?" Ungtss 19:33, 3 May 2009 (UTC)

Indeed. It's religion that claims to have all the answers, not science. That's sort of the point, science is about looking for them. --Kels 19:36, 3 May 2009 (UTC)

--Kels 19:21, 3 May 2009 (UTC)

The point I'm making is that science is about looking for answers in one specific way, which is highly reliable, but only capable of answering questions which can be tested by experiment. Ungtss 19:40, 3 May 2009 (UTC)

And you're full of shit, because science has always used observation alongside experiment. Copernicus, Kepler, and Galileo's astronomy was not experimental, yet it was still science. Newtonian gravity was not experimental in the celestial realm, yet it was still science. Geology is rarely experimental, yet it is still science. Francis Bacon himself wrote extensively on the application of empirical study to civil history, as well as countless other fields. There's a reason why what we call "science" was long "natural history" 19:43, 3 May 2009 (UTC)

Kepler's science was experimental in nature. His models could be tested against the condition of the skies on any given night. Hypothesis -- "The Earth revolves in X path." Experiment -- "Observe celestial objects and document locations." Falsification -- "If observations do not comport with predictions of the model, model is incorrect." It is repeatable and testable. Was then, is now. If you want to define science to include beliefs that don't require experiment, then please be a dear and tell me what you need to do to perform science so we can discuss your redefinition of the term. Ungtss 19:55, 3 May 2009 (UTC)

Uhoh, you've just defined "experiment" as observation, the construction of a hypothesis from extent data, tested with more data as you accumulate it, which is exactly how I described the empirical historical method. One is simply more easily quantized. 20:15, 3 May 2009 (UTC)

I know it's the evil WP, but a read of this would be a good start. Lots of good references at the bottom. Me, I'm very much a layperson (more of a fan of science than anything), so there's only so far I can take you. --Kels 19:49, 3 May 2009 (UTC)

I think that article supports my argument pretty well. The scientific method requires experiment. No experiment, no scientific method. No scientific method, no science. Ungtss 19:55, 3 May 2009 (UTC)

How exactly does "Using controlled methods, scientists collect data in the form of observations, record observable physical evidence of natural phenomena, and analyze this information to construct theoretical explanations of how things work." support what you just said? --Kels 20:02, 3 May 2009 (UTC)

The "Controlled method" mentioned is defined two sentences later: "The methods of scientific research include the generation of hypotheses about how phenomena work, and experimentation that tests these hypotheses under controlled conditions." Ungtss 20:08, 3 May 2009 (UTC)

Split hairs, much? What's the difference between an observation based on an experiment and an observation not based on an experiment? Sterile^{walkie-talkie} 20:12, 3 May 2009 (UTC)

A direct observation is an experiment. But a historical inference about an event that cannot be directly observed is not. "The human heart has four chambers" is a hypothesis that can be experimentally tested anytime. "Socrates had a mole on his cheek" is not. Ungtss 20:19, 3 May 2009 (UTC)

I do like how you've managed to redefine "include" to mean "is limited to". Well done. --Kels 20:14, 3 May 2009 (UTC)

You asked me to explain how the article supported my view that science requires experimentation. I did. You still haven't defined what the "Scientific method without experiment" looks like. Ungtss 20:19, 3 May 2009 (UTC)

That's true. I asked you how it supported your view. You responded by dishonestly ignoring and redefining the terms you found there. Congratulations, you make an excellent Creationist. --Kels 20:29, 3 May 2009 (UTC)

I just quoted the article's one-sentence definition of the methods of science, which included experiment. You still haven't explained how the scientific method works without experiment. How is anyone supposed to take your views seriously when they are completely unsupported? Ungtss 20:38, 3 May 2009 (UTC)

Included. Yes. Do you know what "included" means? --Kels 20:44, 3 May 2009 (UTC)

"Included" can be read in an inclusive (includes plus more) or exclusive (includes only) sense. If you want to read it in an inclusive sense, then go ahead and include more things. How does the scientific method work without experiment? What are the criteria? What's your scientific method without experiment? Ungtss 20:47, 3 May 2009 (UTC)

(Actually, more to the point, do you understand the difference between observation, hypothesis, law and theory?) Sterile^{walkie-talkie} 20:16, 3 May 2009 (UTC)

Yes, I do. Can anyone explain to me what the "Scientific method without experiment" looks like? Ungtss 20:19, 3 May 2009 (UTC)

Yes, one can use an experiment to verify a hypothesis. But that doesn't mean you can't use observations that were not bourne out of an hypothesis.

Suppose, for example that we wanted to understand the color of a certain group of compounds are related to structure. Must we have a hypothesis before we look at the complexes? Sterile^{walkie-talkie} 20:24, 3 May 2009 (UTC)

I never said you need a hypothesis. But you need experimental observation that could falsify the assertion. In that case, anyone could replicate your method and see if the compounds turn the color you say they did. Testable, falsifiable science. Ungtss 20:27, 3 May 2009 (UTC)

Then your beloved creation science is a waste of time because we can never observe the creation. But evolution is science because it is observed all the time. Sterile^{walkie-talkie} 20:29, 3 May 2009 (UTC)

That's where you set up your double standard. Common descent is the point at issue, not variation and natural selection. And common descent is not observed all the time, because all the relevant events occurred in the past and are not experimentally repeatable. It's not that creationism is science. It's that common descent is not. Ungtss 20:31, 3 May 2009 (UTC)

One could say that common descent is a hypothesis from natural selection and variation. Speciation has been observed. There is no reason to think it hasn't occurred in the past as well. As for baraminological distinctions, they are worthless. They aren't even defined well. Sterile^{walkie-talkie} 20:38, 3 May 2009 (UTC)

It's not a hypothesis, it's an inference. It goes like this, "Species vary naturally, therefore all the complexity in life is due to natural variation diverging from a single protocell." It takes a limited observation and extrapolates it to explain things that have not and cannot be experimentally observed or tested. You may or may not think it's a reasonable inference, but it's not experimentally testable, observable, and falsifiable, so it isn't science. Unless you can explain how science works without experiment. Ungtss 20:43, 3 May 2009 (UTC)

Um, all hypotheses and all science involve inferential reasoning. Speciation has been observed and implies common descent. Let's take chemistry again. That atoms exist is inferential thinking. No one has ever observed an atom with an eye. Yes, they have detected them through indirect methods, usually spectroscopy. Does that mean that atoms don't exist?

When you take a crystal structure of DNA, we don't observe directly how it was formed. Does that mean we can make any inference--any hypothesis--about how DNA comes to be based on the structure? Can we assume that replication worked the same way in the past as it did in the present even though we don't have the same DNA strand or sequence? That's an inference after all. Sterile^{walkie-talkie} 20:52, 3 May 2009 (UTC)

The assertion "Atoms exist" could be falsified if any of the spectroscopy methods detected phenomena inconsistent with their existence, so it is science. The assertion "DNA has always worked the same way as it does in the present" is indeed an inference, not science. If common descent is true, DNA has come through a lot of evolution to get where it is today, and has not always worked as it has. If creationism is true, on the other hand, DNA probably has always worked the same way. This one can't be tested, and therefore isn't science. Ungtss 20:59, 3 May 2009 (UTC)

Science is inductive ("inferential") reasoning. It is not deductive thinking. Why are you making this distinction? Sterile^{walkie-talkie} 21:03, 3 May 2009 (UTC)

Science is a subset of inductive reasoning -- it is knowledge that is derived from controlled, repeatable experiments and observations. That's why it is reliable. But there are many inferences that are not scientific, because they are not experimentally based. Ungtss 21:16, 3 May 2009 (UTC)

Science is knowledge and does not always require experiment. Often a simple measurement is sufficient to prove something. In some branches of science like physics or chemistry, many experiments are nothing more than just measurements although sometimes there is a "what-if" factor. In history, archaeology provides the science by measurement of dates (C-14 or dendrochronology for example), physical location (through direct or inferred measurement - geophysics can show the outlines of ancient buildings where all physical remains have vanished) or DNA analysis of ancient human remains. Troy was thought to be a myth until it was discovered by an archaeological dig. Scientific analysis showed that there were nine phases to the city, ancient myths only told of a single episode. Science provides the skeleton for historians to hang a skin on. What you cannot do is make the skeleton fit the skin, the skin has to fit the skeleton. Psychology has shown that human memory is fallible and that people often see what they want to see, eye-witnesses to a crime or accident often give conflicting accounts and repeated copying and translation of documents introduces differences from the original and those who record the details may have their own biases which corrupt the facts. So historical documents always have to be treated with some scepticism. Of course it is entirely feasible to do experiments in history/archaeology. Many experiments have been done to see how the stones that were used to create Stonehenge could have been moved and erected. Генгис^{RationalWiki GOLD member} 21:04, 3 May 2009 (UTC)

You say "often a simple measurement is sufficient to prove something." Yes, and if the conditions leading to that measurement can be repeated, then it is an experiment. Experiment required and supplied. You say "archaeology provides the science by measurement of days, physical location, etc." Again, insofar as the location of the objects can be observed, it is experimental, because the locations of the objects can be verified by observation. But insofar as archaologists infer things that cannot be observed or tested, they are not doing science -- they are speculating. You say "Many experiments have been done to see how the stones that were used to create Stongehenge could have been moved and erected." Yes, but those experiments cannot show how it was moved and erected -- only how it could have been. There is no way to scientifically answer how it was erected. Ungtss 21:16, 3 May 2009 (UTC)

Edit break

So the experiments to replicate the RNA world have no meaning because... Sterile^{walkie-talkie} 21:08, 3 May 2009 (UTC)

It's not that they have no meaning. It's that they cannot make the scientific claim that "life originated this way," because that hypothesis cannot be tested by experiment. There are a lot of things in life that are not science but yet have meaning. The mistake of our generation is equating "true" with "science." No, science is a reliable method of observation, but things may be true and not science, and things come to through science may or may not be true. Ungtss 21:16, 3 May 2009 (UTC)

So if I observe that it is raining today, but I didn't look up the pressure and temperature conditions yesterday, I can't make any assumptions about how the unique, distinct rain system of today formed? Sterile^{walkie-talkie} 21:18, 3 May 2009 (UTC)

Of course you can make assumptions. They just wouldn't be "science." Ever notice why the weather guys are so often completely wrong? It's because they can't do a controlled experiment on the weather. Ungtss 21:20, 3 May 2009 (UTC)

But you can bracket the pressure and temperature ranges and assign a probability that they are correct, no? And they aren't as far off as you would imply. Sterile^{walkie-talkie} 21:30, 3 May 2009 (UTC)

Yes, you can bracket temp/humidity parameters, and develop probabilities. And probabilities are very valuable tools. But it would only be scientific insofar as you drew a hypothesis about tomorrow's weather from a model, and tested that model against reality. But the weatherman predicted mostly cloudy and gusty in southern Turkiye today, and we got sunny with a gentle breeze. Ungtss 21:40, 3 May 2009 (UTC)

And BTW, science can only be validated. Truth is something else. Sterile^{walkie-talkie} 21:33, 3 May 2009 (UTC)

Agreed. Things may be true and not science, and things may be science and not true. Ungtss 21:40, 3 May 2009 (UTC)

...but the nice thing about science, unlike philosophical "truths," is that we can test it to find out whether or not it's true. --Phentari 21:47, 3 May 2009 (UTC)

Exactly. The problem is that people run around calling a lot of things "science" that can't be experimentally tested -- like common descent. Ungtss 21:51, 3 May 2009 (UTC)

...except that it can, in precisely the same way that orbital mechanics can be tested. We have never directly observed Pluto orbiting the sun. We have never traveled to, say, Betelgeuse to find out how far away it is. Much of science is impossible to DIRECTLY test experimentally.

However, what a valid theory will do is generate predictions--and those PREDICTIONS can be tested. If the predictions turn out to be true, then it is likely that the theory is true; the more predictions are proven true, the more likely that the theory is correct.

Common descent (which is YOUR goalpost, Mr. "The scientists are imposing a double standard") has a high inductive probability of being correct based on an overwhelming amount of observation, particularly biological, but also geological, chemical, etc. It comes directly from the observed principles of genetic drift and natural selection. The "ark model" has a much lower inductive probability of being correct based on the same evidence. It comes from one source, the Bible. Sterile^{walkie-talkie} 21:57, 3 May 2009 (UTC)

You keep asserting without support that science can proceed without experiment and giving me concrete examples of things you think "are science," and therefore support your conclusion. However, you have not defined this "scientific method without experiment," and I can find no basis for it or description of it in philosophy of science, anywhere. I'm not going to debate concretes with you until you go to the trouble of defining your term. What are the criteria of a scientific method without repeatable, testable experiment?? Ungtss 03:36, 4 May 2009 (UTC)

New member here; please forgive me for any violations of wiki etiquette and for my ignorance of wiki editing protocols. Question: why is it that everyone (creationist or otherwise) looks to Popper for a definition of science? I'm not aware of anyone in contemporary philosophy of science who uses falsification as a solution to the demarcation problem. Falsification is a good thing to have in a theory, but it's neither necessary nor sufficient to make something science. So the claim that common descent is not falsifiable (which is itself questionable) and hence is not science is a non sequitur. Common descent is a consequence of the models used in evolutionary biology. These models are well-confirmed, and that gives us good reason to believe that all extant organisms on Earth descended from a small number of common ancestors.TallMan 02:42, 4 May 2009 (UTC)

The difference is that it is possible, in the logical sense, to disprove the theory of evolution (if, for example, Noah's Ark turned up on an archaeological dig), while a theory like Last Thursdayism is known not to be falsifiable. ListenerX^TalkerX 03:24, 4 May 2009 (UTC)

The issue here isn't really falsification, though; evolution is susceptible to empirical investigation, whereas Last Thursdayism isn't. Last Thursdayism is really a kind of skeptical hypothesis, and as such it's designed to be compatible with any piece of evidence you might try to provide (compare: I might be dreaming/deceived by an evil demon/a brain in a vat/hallucinating/imagining everything). LTism can't be falsified, which is a problem; falsifiability is a virtue in a theory. But the real problem is that LTism doesn't provide a model of the world at all.TallMan 03:49, 4 May 2009 (UTC)

TallMan -- if you don't think falsificationism is a good criteria for the demarcation, which one do you propose? Ungtss 03:36, 4 May 2009 (UTC)

I've been watching this all day. Prediction. A good test of a hypothesis is accurate prediction. ħuman 03:45, 4 May 2009 (UTC)

Okay, here's my "theory." What goes up will come down, because little fairies in the sky pee invisible piss on everything you throw, because Pan told them to. Accurate prediction: What goes up does in fact come down. Boom! Accurate prediction! Science! Ungtss 05:04, 4 May 2009 (UTC)

Wow, talk about a stupid response. ħuman 05:47, 4 May 2009 (UTC)

This is why I prefer to make models central to demarcation. "Little fairies pee on things" isn't a model until you flesh it out. What properties do these fairies have? What properties does their piss have? In what way does their piss interact with objects? How do the fairies interact with objects? Give me some numbers here, show me how they entail that the fairies piss on the appropriate objects in the appropriate amounts, and then you'll be doing science. You'll be doing bad science, because your theory will not be parsimonious and will be ad hoc. Your theory will be false, because there are no invisible fairies, and once you've specified their properties it will probably be easy to show that this is the case. But that doesn't mean you're not doing science.TallMan 05:33, 4 May 2009 (UTC)

Now there are some good criteria for science. Much better than "accurate predictions!" The thing about models, parsimony, and "not-ad hoc" criteria are all good criteria for reasoning, and are therefore things to strive for in science. However, while they are all necessary for science, none of them are sufficient conditions for science, and therefore do not make for good demarcation. Regarding the "model" criterion: theology has an extremely complex, detailed, nuanced model of God and theodicy. But none of it's falsifiable, so it's not science. Regarding the "parsimony" criterion -- parsimony is probabilistic, not experimental -- that is to say, it tells us to prefer simpler theories because they're more likely true, but the fact that the doctrine of humors is "simpler" than modern molecular biology does not mean it's "science," because is has been falsified. As to the "not ad hoc" criterion, again, it is necessary for science, but not sufficient. "There is no God" is certainly "not ad hoc." But it isn't science, either. Ungtss 07:03, 4 May 2009 (UTC)

Theology doesn't have a model for God, since it doesn't have any story about the mechanisms by which God operates in the world. Similarly, it doesn't assign operational/functional properties to God. So while theologians might maintain that God is necessary, for instance, this doesn't tell us anything about how God interacts with other objects. In fact, many of the properties traditionally assigned to God seem to be chosen to make God undetectable (such as God's not being in space or time).

Theology could have such a model -- all it would have to do is make one up. Watch. "God is an organic being, living physically outside the observable universe, of such massive size that the physical universe is a giant Crystal Ball. He has three hearts, four lungs, and wee, beedy little eyes. He interacts with the universe by sending beings much smaller than he is, called "messengers." He watches events on Earth in a giant flatscreen TV that constantly changes channels to different locations, and is fed by tiny little video cameras put in place by the messengers." Etc. Etc. You can come up with all the models you want, but unless they stem from observation + experiment, they really hold no meaning ... Ungtss 16:13, 4 May 2009 (UTC)

More importantly, I note a tension between your claims here and your claims below. You write:

Falsifiability is a necessary, but not sufficient condition for a claim being scientific. A claim must be both falsifiable and unfalsified to be science. A claim that is falsifiable and falsified is "false."

First, false claims can still be scientific claims. The caloric theory of heat is a scientific theory, albeit a false theory, and if we found someone carrying out experiments to test it they'd be doing science (they'd also be wasting their time, but productivity isn't what makes something science). We know that Newtonian physics don't accurately describe the world, but when high school students test the theory in physics class they're doing science. Second, you're criticizing me for offering necessary but insufficient demarcation criteria, while you do just that. So you'll need to either find a new line of criticism or offer responses to my counterexamples below. Note that I've offered responses to your counterexamples. My proposal is intended to give necessary and sufficient conditions for a proposal's being a scientific theory, and as it stands I think your purported counterexamples fail to demonstrate insufficiency.

There's a big difference between testing a false theory through science and calling a false theory "scientific." Naturally, one does science when one tests ... say ... the ptolemaic system. But who would claim that the ptolemaic system, which has been falsified, is "science?" Or that one is being "scientific" when one espouses it? While conclusions arrived at through the scientific method are not necessarily true, the word "scientific" does carry a connotation to the effect that the idea has not been proven false, don't you agree?

Two clarificatory notes: 1) the model criterion isn't a revision or expansion of my initial proposal for demarcation. The requirement that scientific theories include a model has been part of the criteria from the beginning, though I probably should have defined "model" explicitly in the initial formulation. 2) The parsimony and not-ad-hoc criteria aren't supposed to be demarcation criteria. They're standards for evaluating proposed explanations, and that's it.TallMan 13:23, 4 May 2009 (UTC)

Is "model" a necessary or sufficient condition? If necessary, what model is there in the scientific observation, "Humans have a four chambered heart." If sufficient, then if I come up with a speculative model of who God is and what God does, would that be science? Ungtss 16:13, 4 May 2009 (UTC)

A fair question, and a hard one. The demarcation problem is a tough nut. My intuition is that science is our best epistemic practices, whatever those turn out to be. That, however, is an uninformative answer. So here's a second approach: science is the practice of making models of the world. Those models that reflect the way the world is are good ones; those that don't are bad. A scientific theory is a model or collection of models that are susceptible to repeatable investigations from multiple disparate sources. Caveat: this doesn't account for the way in which scientific theories build on their predecessors.TallMan 03:49, 4 May 2009 (UTC)

One other note: I'd recommend looking at the SEP's page on science and pseudo-science. I'm actually fond of Lakatos's more advanced approach to falsification, but I think he ignores the structure of scientific theories.TallMan 03:52, 4 May 2009 (UTC)

Alright, let's go with "model or collection of models that are suspectible to repeatable investigations." There's no criterion here, though, for the type and/or quality of those investigations. I can investigate psychology by bringing in patients and speculating about how their dreams reflect "world-soul" archetypes a la Jung. Is that science? I can investigate the origin of the Earth by asking every dimestore prophet in town what God says happened (multiple lines of repeatable investigation). Is that science? I don't think so -- I think the investigations must be of a particular type or quality in order to be "science." Would you agree? If not, why? If so, what type of investigations would qualify as science?

Second, I don't think "multiple lines" are necessary. Sometimes one line is all you need -- for instance, "to determine how many chambers a frog's heart has, dissect + observe."

Short story: I think that "investigation" is not an adequate criterion for science, because it says nothing of the quality of the investigation. I also don't think "multiple" is necessary, because in many cases, one can adequately address the issue. The common thread, of course, is falsifiable experiment. When that quality is present, one line of investigation may be sufficient. When that quality is absent, no amount of investigation constitutes "science." Ungtss 04:55, 4 May 2009 (UTC)

1) I've already got restrictions on the sort of investigations in place. I'm not familiar with Jung, but from what I understand his inquiries into psychology aren't repeatable and can't be independently verified and hence isn't science. The same will go for any attempt to use revelations as a source of information. Revelations might be accurate, but if they're not assessible and accessible by anyone (at least in principle) they won't count. It's also worth noting that Jungian psychology could be science; as I said, it's not my speciality. If it is, it isn't good science. Doing science isn't hard; doing good science is the trick.

Okay, so we've added another qualification -- "accessible to anyone." Agreed. But what needs to be accessible? In the case of Jung and the dimestore prophet, the interview of the subject is accessible to anyone, but the event in question (the dream) cannot be repeated, and is not accessible to anyone. Would you agree that for a claim about a phenomenon to be scientific, the phenomenon itself (and not merely circumstantial evidence which can be interpretted multiple ways) must be repeatable and accessible to anyone? If not, why? If so, how is the phenomenon "Alexander conquered Persia" both repeatable and accessible to anyone? Ungtss 06:39, 4 May 2009 (UTC)

This isn't another qualification. I'm explaining what is meant by "multiple sources of investigation." But that's not central. The claim is not that any given event needs to be repeatable. Events are particulars and thus are necessarily not repeatable. Rather, what needs to be repeatable is the investigation. The prophet's method of investigating the world is revelation; revelations aren't universally accessible. If there was a way to reliably induce revelations and revelations reliably tracked the truth, then prophecy would be a legitimate means of investigating the world. But there isn't, and they don't, so prophecy isn't science. The Jungian psychoanalyst's method of investigation seems to be making shit up, and while anyone can make shit up you're not going to get consistent results from this method. That is, when you present two investigators with the same data (the same patient, in this case), and they perform the investigation using the same methods (the methods of Jungian psychoanalysis, whatever they are), you're not likely to get the same result except by coincidence.TallMan 14:02, 4 May 2009 (UTC)

So the investigation must be repeatable and yield consistent results. Fair enough, I'll grant that it's necessary -- but not sufficient. Can we agree? Ungtss 16:13, 4 May 2009 (UTC)

2) I don't know where you're getting "multiple lines" from. My claim is that any investigation needs to a) be repeatable and b) not dependant on a particular observer in a privileged position. When it comes to frog hearts, what matters is that anyone with the proper skills and equipment can cut open a frog and examine the heart.

I misunderstood what you meant by "multiple lines." I thought you meant "different ways of answering the same question." My bad. Ungtss 06:39, 4 May 2009 (UTC)

3) There are counterexamples to the falsifiability thesis. I've been told that Jungian psychology is one of them; Freudian psychology is another. Freud's views are falsifiable; in fact, they've largely been falsified. His method was unscientific not because it was unfalsifiable but because it wasn't empirically adequate, i.e. it did not accurately model how the mind works. Astrology is arguably another counterexample. The claims of astrologers are falsifiable. Like Freudian psychology, astrology has been falsified. If Popper was right, then astrology would be a science. But it isn't, and one way to see why is to ask several different astrologers to perform a reading. You won't get the same results, which suggests that their investigations can't be repeated in the right way.TallMan 05:19, 4 May 2009 (UTC)

You're confusing necessary and sufficient conditions. Falsifiability is a necessary, but not sufficient condition for a claim being scientific. A claim must be both falsifiable and unfalsified to be science. A claim that is falsifiable and falsified is "false." A claim that is unfalsifiable and unfalsified is "Not science." Ungtss 06:39, 4 May 2009 (UTC)

Very nicely put, TallMan. ħuman 05:57, 4 May 2009 (UTC)

Ungtss, you are confusing your "understanding" of science, limited by Popper - and your interpretation of his work, with "real science", as done by "real scientists" ħuman 08:49, 4 May 2009 (UTC)

There's no confusion here. Popper initially offered up falsifiability as providing necessary and sufficient conditions for science. My counterexamples show it's insufficient (and I could add a few more, e.g. phrenology and palmistry). And while they're more controversial, there are counterexamples to be had for the necessity claim as well. If we're willing to make the necessary adjustments in our beliefs, any proposal can be made unfalsifiable. As Lakatos pointed out, any theory is surrounded by a "protective belt" of ancillary hypotheses that can be adjusted to resolve apparent falsifications. There are, I am told, still a few defenders of the aether theory, which was purportedly falsified by the Michelson-Morley experiment. But it turns out you can develop the aether theory in such a way as to be consistent with these experiments by introducing the Lorentz transformation. If you're willing to make significant enough changes in your ancillary hypotheses, aether theory can be made consistent with just about any collection of experimental data. So aether theory is unfalsifiable. But intuitively it's still science—unless, of course, you want to say Poincaré and Lorentz weren't scientists.TallMan 14:02, 4 May 2009 (UTC)

I agree with you, empirically, that complex models can be shifted this way and that to get out of an evidentiary problem. However, I'd argue that not all scientific claims suffer from this fault. The gas law, for instance, is falsifiable. So are claims about the rate of acceleration due to gravity. Those are not so buried in ancillary hypotheses as to be totally malleable. My argument is that it is those (and only those) claims that can claim the full credibility of the "scientific method," because they are the only ones that arise from it.

Other realms of inquiry, such as historical inquiries, are so surrounded by assumptions that they can be revamped at all. In these situations, it's much like a litigated trial. There are a limited number of facts in evidence which can be interpretted any of a number of ways. Each side develops a "theory of the case," in which they try to account for all the facts in a way consistent with their theory. Throw an inconsistent fact on the pile, and a creative attorney can adjust his theory of the case to account for it without admitting his client's guilt. My argument, though, is that this is not science, because it is not experimental or observation. It depends on interpretation, and, as you said, is subject to adjustment to account for inconvenient facts. That's not to say that this method of evaluating the evidence is not relatively reliable (depending on the quality of the evidence). It is only to say that it is not in the same category as the gas law or Chem 1 experiments demonstrating Molarity, because the ultimate issue of the case is not subject to falsification or experiment.

The evidentiary approach of the court-room and the archaeology pit is reliable, but it's not nearly as reliable as the approach of the laboratory science. Equating the two forms of inquiry and granting them equal status gives undue credibility to the former, I think. Ungtss 16:13, 4 May 2009 (UTC)

I'm going to respond to all your points down here rather than in line, in the hopes that we can get the entire argument in one place. But first, let me restate my position: on my view, a scientific theory is a model or collection of models of the world that are susceptible to investigation. These investigations must be repeatable by independent observations. A model is a description of (some part of) the world from which it is possible to derive, in a non-trivial and more-or-less precise way, claims about how the world is (using a tenseless 'is'). In this way models get at what's important about prediction in science. Models must be consistent. These conditions are individually necessary and jointly sufficient.

Now, on to your remarks. I see four distinct lines of argumentation. First, there's a question about what counts as a model and a proposed counterexample involving God. Second, there's the question of false scientific theories. Third, there's the issue of falsification. Finally, there's the border between science and related but distinct modes of investigation. I will take these issues in order.

1) You've claimed that theologians could offer up a model of God. This is true. However, the story about God given above (to wit, "God is an organic being, living physically outside the observable universe, of such massive size that the physical universe is a giant Crystal Ball. He has three hearts, four lungs, and wee, beedy little eyes. He interacts with the universe by sending beings much smaller than he is, called "messengers."") isn't a model on my view, though it might become a model if it were sufficiently detailed. More information is necessary. If God is larger than the observable universe, as suggested above, he can't be organic; organic structures don't scale up that way. The model also lacks enough detail to allow us to derive non-trivial claims about the world. Finally, the model as presented can't be investigated. If it were developed sufficiently and made consistent, it could become a model in the sense which is of interest here, in which case it could be part of a scientific theory. Science isn't limited to any particular domain, and if theologians made their claims susceptible to investigation then they'd be (potentially) doing science. They'd also discover that their claims are false/unwarranted, which is probably why the claims don't get tested.

2) The status of a theory as scientific is independent of its truth-value. The Ptolemaic model of the solar system is a scientific theory. It has been superseded by better theories, but it's still scientific. If someone were attempting to test the Ptolemaic theory, they'd be doing science. They'd also be wasting their time, since the structure of the solar system is well-understood, but science need not be productive. I've given an example; students who test Newtonian mechanics are doing science even though Newtonian mechanics has been superseded. This strikes me as highly plausible. Researchers studying cold fusion are doing science; they, too, are probably wasting their time, since cold fusion hasn't been demonstrated and no promising lines of inquiry remain, but so be it. Most lines of investigation in science don't pan out, and those people who were unlucky enough to follow the lines that failed were still scientists.

3) You've claimed that some claims, such as the ideal gas law, do admit of simple falsification. This is false. We observe physical systems (seeming to) violate the ideal gas law all the time (in fact, since there are no ideal gases, I suspect it's exceedingly rare to find a system that obeys the law). But we don't reject the law, because we instead reject one of our ancillary hypotheses (for instance, that our measurements of pressure, volume, etc. are accurate). This is what happens when we get rid of outliers in statistics, as well. If the ideal gas law were falsified as easily as you suggest, then we should think it false. But we don't, and it isn't. As a philosophical point, any claim at all can be held come-what-may in the face of whatever evidence we might have, if we're willing to alter our other beliefs to accommodate the recalcitrant data. I'd recommend looking at Quine's "Two Dogmas of Empiricism" for more on this point.

4) You've also suggested that my view on scientific theories and demarcation can't adequately distinguish between science and the investigations performed in archaeology or in the courtroom. I have three responses. First, the distinction is not so clear as you seem to suggest. You've given no reason to think that there is such a distinction at all. Furthermore, science is a method for investigating the world, and it isn't domain-specific. Many scientific investigations are narrowly tailored to very particular questions. So there's no in-principle problem with limiting the investigation to matters relating to a courtroom case or archaeological site. Second, the intuitive differences between science and the courtroom can be accounted for by my view. Lawyers in the courtroom might develop a "theory of the case," but this won't often be a model in the sense given above. Archaeologists may or may not develop a model in the relevant sense; I'm not familiar with their methods and so can't comment. Third, you're confusing two distinct questions. The first is "is this science?" and the second is "what evidentiary standards are in play here?" Evidentiary standards vary within the sciences; what counts as "proof" for a chemist will likely differ from what counts as "proof" for a mathematician. So the archaeologist or lawyer might be doing science, but doing science under lower evidentiary standards than are customary in, say, physics or math.

So, what of the issue that started the debate? Common descent is entailed by the models of evolutionary biology. These models admit of repeatable independent investigations (I'd recommend Elisabeth Lloyd's "The Structure and Confirmation of Evolutionary Biology" for more here). I'd note, also, that common descent is falsifiable, or at least is as falsifiable as any other claim. If we found a number of organisms that can't be placed in a phylogenetic tree with other organisms, and perhaps if they had alien morphology, I suspect most evolutionary biologists would reject (universal) common descent. So we have ample reason to count common descent as a scientific claim, and evolutionary biology as science.TallMan 04:15, 6 May 2009 (UTC)

Of course, after writing that I think of one more point. One common criticism of Popper is that he didn't attend to how science is actually practiced. This is one reason why Popper is more-or-less roundly rejected by contemporary philosophers of science, or at least those who are interested in the demarcation problem. My view, in contrast, attempts to be sensitive to how scientists actually act, and in this it shares much with the so-called "semantic conception" of scientific theories. I'd recommend Fred Suppe's "The Structure of Scientific Theories" here, but it's 1) well over 700 pages and 2) deadly dull.TallMan 04:19, 6 May 2009 (UTC)

Another edit boobie

The goal here is to define "science," because a scientist does a lot of things, some of which are science and some of which are not. I think scientists do science when they directly observe phenomenon, and when they develop experimental ways to test hypotheses. Perhaps you have another, useful definition? Or do you believe that whatever they do or say is "science" just because they do or say it? Ungtss 08:59, 4 May 2009 (UTC)

I don't "believe", for starts. I also don't think you are the honoree who gets to define "science" for now and for evermore. Especially since you deny some of its results. Why does your religion hate science and why do you think you have to fight it? ħuman 09:14, 4 May 2009 (UTC)

Evading the topic lets everybody know you don't know what you're talking about. If you know what science is, let a poor ignorant fundamentalist rube in on the big secret! Or do you not know? Do you think something's "science" if it uses fancy tools or if somebody in a lab coat tells you it's true? Ungtss 09:19, 4 May 2009 (UTC)

I have not evaded the topic. You have decided that your interpretation of Popper's "falsifiability" is the sine qua non of "Science". I'm sorry, but you're wrong. Predictive ability of a hypothesis is my test of a hypothesis. Oh, and, hey, rube, question: what do you consider to be "scientific fields of endeavor" in this modern age? ħuman 09:29, 4 May 2009 (UTC)

I showed how your definition doesn't work, but you didn't respond. Perhaps you'd like to point to a prominent philosopher of science who holds your view that predictive ability alone is the hallmark of a good hypothesis? As to your second question, in my opinion each field makes certain claims that are science, and certain claims that are not. Your question overgeneralizes. Ungtss 09:44, 4 May 2009 (UTC)

Why do we need to point out philosophies of science to show the workings of science? Popper's ideas helped reshape issues of testing and falsifying hypotheses, but his declarations are not established and wholly accepted. To decide that science must follow Popper's methods or be wrong is your opinion; and given that you have no real scientific background that I'm aware, I doubt your opinion matters to anyone in the science field.

We need philosophy of science to know what science is. Otherwise we're left with nothing but vague, meaningless generalizations and ad hominem arguments like yours. Ungtss 18:07, 4 May 2009 (UTC)

You do realize that ad hominem means to attack without the substance of an argument. That I provided substance to why Popper-philosophy is not the end-all declaration of what science needs to do to be right, in that his arguments are not wholly accepted, is far from an ad hominem. That I point out your lack of scientific background to showcase that your opinion of what science must do to be right is also not an ad hominem. If you have a science background, provide your credentials; if not, admit it's simply your uneducated opinion. --Irrational Atheist 19:00, 4 May 2009 (UTC)

The "accepted" or "non-accepted" nature of a philosophy is purely ad hominem, because it relies on the people accepting the philosophy, rather than the merits of the philosophy itself. It's really sad to see, over and over, how many people have stopped thinking for themselves and instead punt to "well everybody believes it!" What an unscientific attitude. Science questions, challenges, and demands. It is never satisfied with "well everybody says so!" That's what makes it such a powerful tool. I can't fathom why those who claim to advocate science are running around taking a poll on what everybody thinks. Sad. Ungtss 05:13, 5 May 2009 (UTC)

How do you argue that? The accepted or non-accepted nature of a philosophy depends on whether it is accepted. Its merits are a major aspect of whether it is accepted. If it is without or lacking merit, why should people accept it? You don't seem to tie together your arguments too well here.

There's a key difference between the logical merits of an idea and its status as "accepted." 85% of people in the United States believe in God. It's "generally accepted." But it doesn't make it true, it doesn't make it science, and it doesn't make it reasonable. Until the 19th century, the medical community believed that a fever was caused by having too much blood. That didn't make it right. For 1500 years, the ptolemaic system was "generally accepted" among astronomers. It was also wrong. It is completely irrelevant what "most people think." Most people are wrong, most of the time. What matters is what can be demonstrated. And when you stop responding to my arguments on the merit of the issue, and punt to "But everybody agrees with me!" it's meaningless to a person like me, who actually thinks about the issue itself. It's the refrain of every dying idea in history -- its proponents can't defend it on any basis except popularity, and so it is about to sink. Ungtss 03:08, 6 May 2009 (UTC)

Again, you mix up belief with acceptance of fact. And your number seems pretty high. According to Wikipedia, "There are also stark regional differences within the United States with respect to a belief in God. Only 59% of Americans living in Western states report a belief in God, yet in the South (the 'Bible Belt') the figure is as high as 86%." But belief in the Christian God is not acceptance there is one. The former is all there is with religion; the latter requires evidence of fact.

So tell me, if "belief" and "acceptance of fact" are different, how is a person to tell when they (or someone else) "accept a fact" versus merely "believing?" What's the difference? Ungtss 13:14, 6 May 2009 (UTC)

Your further argument is again an appeal to ignorance. What science accepted long ago is not evidence of how wrong it is today. Our technology and understanding is far superior to what it was a hundred years ago. You also fail to realize the irony of trying to paint rational thinking, science, as wrong (most people are wrong, most of the time?), and that what is correct can be demonstrated; yet you edit and support a site where the Biblical worldview is taken as truth. None of it can be demonstrated, so by your argument, it doesn't matter.

You've lost coherence here. The point is that just because everybody "accepts something as fact" doesn't make it fact. People have been wrong before and they'll be wrong again. The only thing that keeps us on track is science, reason, and morality. Not popularity contests for ideas. Ungtss 13:14, 6 May 2009 (UTC)

And you change the game yet again. I didn't stop responding to your arguments, but pointed out how ludicrous they were. Scientific understanding can be defended on the basis of evidence, repeatability, predictability, inter-discipline agreement, etc. It does take some understanding of science to comprehend it; demanding "definitions" then arguing the definitions do not explain how the science works, is you being on the losing end of the discussion from the start. You can't demand one thing, expect another, and when you don't get what you want, declare victory. That's what grade school kids do to win arguments. That's not what us adults do. --Irrational Atheist 12:59, 6 May 2009 (UTC)

You originally stated that "the merits of an argument are a major component of whether it's accepted." That's just plain stupid. Wrong things are accepted all the time. Were the "merits" of eugenics a major component of why it was accepted? Or was it racism? Hmm. But I guess things are different now, eh? We're no longer subject to the biases and errors of the past? Scientists are perfectly neutral and detached from the baser instincts? But it seems you have some magic formula for differentiating between "belief" and "acceptance of a fact." Lemme know how that works if you would. Ungtss 13:14, 6 May 2009 (UTC)

Science isn't about beliefs. The scientific attitude is asking questions and seeking answers. But not creationist questions that are setup to be gotcha questions (how very grade school of them, no?), but serious inquiries into the workings of the world. No creationist wants to seek the answers, but just assume the answers from the Bible, etc. I also find your projection of people who have stopped thinking for themselves onto those of us living in reality, when ASK, like CP, is full of people who can't tolerate critical thinking. --Irrational Atheist 21:05, 5 May 2009 (UTC)

Again, that definition is nice and inspiring, but it is not rigorous enough to be useful. So a person is "asking questions and seeking answers." How are they doing this? Are they asking questions and seeking answers through prayer? If so, is that science? I don't think so. I think that in order to be doing science, you must be asking questions and seeking answers through falsifiable experiment. Otherwise you're just being curious. And it's good to be curious. Shit, to be a good scientist you have to be curious. But being curious certainly doesn't make you a scientist. Do you understand why "asking questions and seeking answers" this isn't a rigorous, useful definition? Under that definition, "Prayer" could be defined as "scientific," and I think we can both agree prayer isn't part of the scientific method. Science must be more than mere curiosity. Ungtss 03:14, 6 May 2009 (UTC)

You are unable to grasp the difference between a definition and an explanation. Why is that?

And we're back to your definition of what science should be is just your opinion; your lack of scientific background means your opinion is irrelevant. Why? Because what you declare science should be is already wrong on its face. You say to do science, one "must be asking questions and seeking answers through falsifiable experiment(s)." So no one is doing science by sending out probes to study far away planets? No one is doing science by excavating fossils from rock? No one is doing science by tracking the spread of the H1N1 virus? These aren't falsifiable experiments, after all.

For now, I'll just be done with your inanity, until you learn the difference between a definition and an explanation. --Irrational Atheist 12:59, 6 May 2009 (UTC)

My point still stands. You have shown yourself to be a science hater, because our scientific understanding of the world begins to unravel the infallibility of the Bible. No great flood was possible. The world wasn't made in six days. The earth cannot stand still. It is impossible to turn water into wine by willing it. Need I go on? --Irrational Atheist 17:13, 4 May 2009 (UTC)

Perhaps a little philosophy would help you here. In order to support a statement that I've "shown myself to be a science hater," you need to present evidence that supports your conclusion. I haven't said anything about any of those topics. What you're doing is called "irrelevant rambling." That's why you need philosophy. Ungtss 18:07, 4 May 2009 (UTC)

Science - mankind's understanding and knowledge of the workings, history and existence of the natural world. Would anyone disagree? --Irrational Atheist 17:07, 4 May 2009 (UTC)

Agree, but your definition says nothing of the means of discovering that knowledge, and the criteria we use to differentiate that knowledge from speculations. Ungtss 18:07, 4 May 2009 (UTC)

You didn't ask for that. You wanted a definition of science. Now when you're done moving goalposts, we can return to the current discussion. --Irrational Atheist 18:54, 4 May 2009 (UTC)

Those are all part of the definition of science. Science is not only a scope of inquiry, but also a method and a set of assertions about the world. Unless you can define the method and define what differentiates "scientific assertions" from "non-scientific assertions," you haven't defined the term. Ungtss 19:00, 4 May 2009 (UTC)

A definition of a term or idea does not require its etymology to be correct; a definition of science does not require the encompassing field of studies that comprise science to be correct. You said we needed to define science. I gave a definition. You agreed with the definition then switched to an argument that what is also necessary is how to obtain everything. Thus, moving the goalpost. If you want a discussion of how man understands the world, that's different than saying we need to define science. Is this not sinking in? --Irrational Atheist 19:17, 4 May 2009 (UTC)

I'm looking for a definition of science that can differentiate between "science" and "non-science." You haven't provided that, despite repeated requests. So I'm not wasting any more time with you. Ungtss 05:09, 5 May 2009 (UTC)

Evolution = Science

Creation = Non-Science

Do I win? Ace McWicked^{TK, TK, TK, TK, TK, TK, TK, TK, TK, TK ad infinitum} 05:11, 5 May 2009 (UTC)

Depends what game you're playing. If you're playing the "make myself laugh" game, I'm guessing you won. If you're playing the "actually address the issue" game, not so much. Ungtss 05:15, 5 May 2009 (UTC)

I gave you the definition of science. You agreed with it. What's non-science? That which isn't science.

But what you have so much trouble with is the English language. You keep saying you want definitions, and when you get them, argue that they don't explain how science works. That's not what a definition is. That's what an explanation does. That's what a discussion uncovers.

When you keep looking for definitions and find a lack of explanations, that would be your own fault for not saying what you mean. --Irrational Atheist 21:05, 5 May 2009 (UTC)

As I explain in the section below, I'm looking for a "rigorous" definition -- the kind they use in science, math, philosophy, and law. The kind you can actually use, instead of the kind that gets you a passing grade on the test in Jr. High. Ungtss 03:08, 6 May 2009 (UTC)

Is it just me?

Or is this all starting to sound like trolling with a better vocabulary? I'm learning a lot from the responses though, so it's not exactly a waste of time. --Kels 19:45, 4 May 2009 (UTC)

Perchance we are seeing how much people do not understand about scientific thinking and clarifying our own ideas. Sterile^{walkie-talkie} 20:16, 4 May 2009 (UTC)

Or maybe none of us understand science as well as we think we do, but instead of developing and articulating clear answers, you're attacking the person who's asking the questions. Ungtss 05:10, 5 May 2009 (UTC)

Dear Argumentative Guy, I skimmed through this debate and caught a whiff of: "Common descent isn't science because it isn't observable." While I generally agree with your speculation, I hope you don't take that as any support for Creationism. Essentially, if you accept natural selection as the process by which evolution occurs, then what possible hypothesis could explain the origin of life? At this juncture I would invoke Occam's Razor and simply assert that common descent is the simplest explanation for the origin of life on Earth. If you have a hypothesis that better explains the origin of life I would love to hear it. Jorge 05:31, 5 May 2009 (UTC)

Perhaps it's that "whiff" that keeps people from discussing the issue I actually want to discuss. I started this because I was accused of "redefining science" when I said that science must be subject to repeatable, falsifiable experiment. That's it. Never said I was a creationist, or that creationism was science, or that "Y'all need to be SAVED now y'hear?" I never said creationism was science. I'm only arguing that (under any meaningful definition of science), common descent isn't either. Ungtss 05:39, 5 May 2009 (UTC)

Wait, are you arguing "science must be subject to repeatable, falsifiable experiment" or "common descent isn't [science]"? I do note a tendency on your part to float back and forth between what you have to say that is relatively intelligent and your end goal, which is to argue for ID. ħuman 08:45, 5 May 2009 (UTC)

I'm primarily interested in the first. The other is a corrollary. As to my "end goal," why not assume a scientific attitude and refrain from making conclusions until you have evidence to support them? I don't give a fuck what you think about the begining of the world. That's your business. I'm just tired of the smart people (and y'all are the smart people) being intellectually lazy regarding the definition of science, and sitting around like the two old dudes in the muppets, making fun of the ignorant. Get off your asses and think scientifically. Thanks. Ungtss 09:02, 5 May 2009 (UTC)

I'm not assuming anything. The one part of science you always come back to, while you attempt to impose your definitions on it, is common descent. Strange how that is. Science pretty simple: Observe. Hypothesize. Predict based on hypothesis. Test prediction via more observation or experiment. Reject or refine hypothesis. Lather, rinse, repeat. ħuman 09:07, 5 May 2009 (UTC)

I'm using common descent as an example, not an agenda. Pick anything historical. Who wrote the Gospels is another good, controversial one. Same problem. Science is powerless to help, so everybody speculates in the name of science. Ungtss 09:13, 5 May 2009 (UTC)

To show you how your thinking isn't getting you anywhere, I'll give you an example based on a topic you just used. "Who wrote the Gospels" could give you the answer of who wrote them, but can't give you the details of whether those stories are accurate. "Did John write the book of John" cannot tell you whether or not Jesus returned to life and escaped the tomb after his crucifixion. Like above, when you ask for a definition of a term, then argue that the definition doesn't explain how that subject works and its limitations, you're asking for one thing and demanding something else in return. Understand yet? --Irrational Atheist 21:08, 5 May 2009 (UTC)

Got it. You don't understand what a "Definition" is in science, math, philosophy, and law. In science and law, when you are defining something, you're not just answering the basic question -- "What does it mean?" You're developing a set of criteria that can be applied to place things "inside" or "outside" the category. For instance, the definition of "Plant" uses criteria like "cellulose cell wall," "photosynthesis," "multicellular," "eukaryotic," etc. It's very precise and rigorous. If somebody comes along and says "Well plants have roots," you'd say, "Well some do, but not all do. Moss, for instance, doesn't. So that's not a good definition." It's the same thing in philosophy. Socrates spent a lifetime tearing up people's vague definitions of key concepts like Justice and Freedom, by pointing out how their criteria didn't really match their intuitions, and hence they didn't really know what they were talking about. In law, assault is defined as "an overt act placing another in apprehension of immediate physical harm." If someone came along and defined assault as "taking a swing," they'd have a poor definition. The reason these definitions are so important is because they allow us to structure our thinking -- to know what we're talking about with precision. That's the sort of definition I'm asking for here. Not a sixth grade definition like "science is everything we know and how we figure stuff out!" That's useless beyond middle school. We need something we can work with in figuring out what is science and what is not. Get it? Ungtss 02:57, 6 May 2009 (UTC)

Jorge

At this juncture I would invoke Occam's Razor and simply assert that common descent is the simplest explanation for the origin of life on Earth. If you have a hypothesis that better explains the origin of life I would love to hear it.

First, let's point out that we're outside the realm of "falsifiable science" and in the realm of "interpretation of evidence," like in the courtroom. Each player has a "theory of the case," and interpret the facts in light of their theory.

Now let's go on to "which answer is correct?" An answer is not correct simply because it is simple. In fact, the correct answer is usually quite complicated. Occam's razor only says that an answer should not be more complicated than necessary to adequately account for the facts. But in order to be legitimate, an answer must explain the facts.

RNA world hasn't been demonstrated. Metabolism world hasn't been demonstrated. The huge changes required to go from animal to plant, bacteria to fungi, etc, have not been demonstrated. Those are speculations that have not been demonstrated or comprehensively explained, and therefore do not explain the facts. Occam's razor doesn't apply.

Personally, I'm open to all potential explanations. But I'm a pretty conservative guy -- I go with the explanation that's closest to being demonstrated. I've seen genetic engineering. I know scientists have built viruses from scratch. I see no reason not to consider the possibility that life originated via genetic engineering. Not only does it seem plausible, it seems like the most plausible alternative at the moment. Does that mean it's true? Certainly not. But the best guess we have, I reckon. Ungtss 05:51, 5 May 2009 (UTC)

So you do you have a plausible alternate hypothesis? Jorge 06:12, 5 May 2009 (UTC)

Again, we're not looking at "hypothesis" here -- we're looking at "theory of the case." The best, most reasonable, most parsimonious way to interpret the evidence. And in every trial, there's disagreement among the jurors, because interpretation depends on what the individual thinks is most "reasonable." So what happened? I don't know! But I'm interested in finding out. Problem is, battle lines are drawn so tightly that nobody will consider the valuable clues raised by the other side. Everybody's so focused on the theory of the case that they're first reaction to a new fact is not "What does this mean?" but "How can I reconcile this with my theory?" I think evolutionists make excellent points against the self-contradictory supernaturalism of the creationists. I think creationists make excellent points against common descent. Based on the evidence I've heard so far, I'm inclined toward a naturalistic intelligent design. Ungtss 08:49, 5 May 2009 (UTC)

Occam's razor applies immediately when you put a "genetic engineer" into this theory.

Not unless there is another theory that can fully account for the facts There is no such theory at the moment. Metabolism world and RNA world are just speculation. We don't even know how they would work fully, much less can we demonstrate how they might have happened. Unless there's a fully explanatory alternative, Occam has nothing to say. Ungtss 08:49, 5 May 2009 (UTC)

I think you are absolutely right that we can't say that common descent is a fact, until science won't have other (more direct) means to test it, although I would argue that it cannot be falsified. If we accept evolution as a fact (and I believe you do) then the best plausible explanation is, by our knowledge, common descent. This won't make it true, but all the evidence suggests that it is. Falsifiability in this case means that it is possible to either prove it could not have happened that way, or science would gather other evidence contradicting this theory (or reevaluate the present evidence). The theory of common descent allows both to be possible, and if it's false, both of them should happen sooner or later. In our present however, I can only repeat myself, the best plausible explanation is common descent.

Again, falsifiability doesn't work in this context, because we are dealing with "Theories of the case," not lab science. A creative lawyer can always interpret the evidence to support his client's case -- but that doesn't mean he's following a scientific approach. It means he's defending the theory he's being paid to defend. As many have pointed out, common descent is such an all-encompassing theory that virtually any evidence can be interpretted within its framework. The only question is whether the interpretation is reasonable -- but reasonableness is a matter of perception. Is common descent the most reasonable explanation? Depends on how you see the world. Ungtss 08:49, 5 May 2009 (UTC)

Common descent is like Russel's teapot, but in the opposite way. We can't say for sure that there is no teapot orbiting... whatever, as we can't say that it is a 100% that common descent is true. Yet again, it would be stupid of us pretending that there is a teapot nebula somewhere so far, we can't directly test it. Of course the probability of the two is "slightly" different, but I think it proves my point that the lack of direct testing shouldn't mean it can't be considered true.

That's where history comes in. In order to claim that there's no evidence of incredibly powerful beings in early human history that created us, ruled us, and interbred with us, you have to ignore virtually every history of the world written in any culture in the world. Timaeus, Critias, Homer, Genesis, the Annals, Herodotus, virtually anything by the Chinese, basically anything ever written before Europe in the 1850s or so. You name it -- virtually every ancient history records direct interaction between us and our corporeal creators. Are we to give those histories any credibility? That's a question of individual interpretation. You might think they were written by ignorant savages living in caves. Another might be less willing to ignore virtually every single piece of ancient history ever written.

On the topic of what is considered scientific or not, I tend to agree that falsifiability is a criteria of a scientific theory, but I certainly need more reading on that topic before going into further conjectures. --Kangaxx 08:15, 5 May 2009 (UTC)

"The courtroom" is a very poor analogy for how science works. Court sessions end, formally and declare judgment based on available evidence and arguments. Science is always open to - and looking for - new evidence, and new arguments. ħuman 08:48, 5 May 2009 (UTC)

You missed the point, Human. The point is that courts, and the historical "sciences," are so dependant on and surrounded by ancillary assumptions, biases, and interpretations that they are not nearly as reliable as the repeatable, falsifiable laboratory science that has given us the modern world. The airplane is built on a hundred thousand mistakes that could be observed, analyzed, and corrected. We don't have that luxury with abiogenesis, because we can't watch it happen. All we can do is make a guess and stick with it. Abiogenesis is the "simplest" answer to the question of origins -- we observe that we are here, and we observe no God, therefore we came to be on our own. Now let's make up some way that it might have happened, and call it science. Ungtss 08:52, 5 May 2009 (UTC)

What does abiognesis have to do with the courtroom analogy? Or airplanes to do with anything? ħuman 09:04, 5 May 2009 (UTC)

Okay, let me make this simple. If I'm wrong about aerodynamics, what do I do? I build an airplane and see if it works. If it doesn't work, I watch what happened, study it, analyze it, come up with another guess, and test it. That's why airplanes improve. But we can't do that with abiogenesis. You can't build "abiogenesis," see if it works, and if it doesn't work, try another guess. We can't test it, because we're not there to see it. That's why it's shitty "science." Because we can't DO what science does best -- learn from our mistakes. We can just make the mistake of our choice and then stick with it. Ungtss 09:11, 5 May 2009 (UTC)

Your agenda is getting rather naked despite your protests. And your analogies are getting sillier. Your conclusion, however, is tiresomely consistent. ħuman 09:13, 5 May 2009 (UTC)

Okay, I made the mistake of thinking you were sincerely interested in discussing this topic, and I let you draw me back in to your trolling. Lather, rinse, repeat. I'm done with you now. Ungtss 09:14, 5 May 2009 (UTC)

So no alternative hypothesis? That's what I thought. It seems that you would reject common descent/ambiogenesis because there is no direct, testable evidence to support it. However, if you hold this belief my question is: What do you replace it with? Did life come from nothing?

My point was not that there is no alternative to CD. My point was that there is no alternative hypothesis, because a hypothesis must be experimentally testable, and none of the alternative origin speculations are testable. "God exists," for example, is not a valid hypothesis, because there's no way to test it. There are plenty of alternative possibilities for the origin of life tho -- the strongest one, I think, is genetic engineering by intelligent, corporeal beings. It's supported by history, consistent with the evidence, and conceptually sound. Not a hypothesis though. Just a theory of the case. Ungtss 18:52, 5 May 2009 (UTC)

Let me draw an analogy from everyday experience. You come home one day and there is a cake sitting on your coffee table. In the kitchen there is an empty box of cake mix. In the trash there are some eggshells. The oven is still warm. What logical explanation could there be? A rational person might hypothesize that their roommate made the cake (for arguments sake you have a roommate). There is some evidence for this, but since you cannot go back in time and directly observe this you cannot know for sure. However, you chose to reject this hypothesis based on the lack of rigorous scientific proof.

I choose the possibility that my roommate cooked the cake (rather than it materializing out of the air), but not on the basis of the scientific method. I choose it based on common sense. I cannot test any hypothesis. Therefore I cannot perform the scientific method. I am outside science, and well inside the realm of "how do things normally work in the world?" Unless I can experiment, I can't do science. But that doesn't mean I have no means to weed out stupid ideas. When my brother lies to me, I can tell he's lying based on all sorts of things, including what I know about how he normally behaves, the look in his eyes, and what seems most reasonable to me. But I can't use science. So it is with the origin of life. We cannot test a hypothesis, therefore we cannot do "science." We can only use our common sense of how the world works. But unfortunately, views diverge much more sharply on the issues of origins than they do on the issue of cakes baking themselves. Ungtss 18:52, 5 May 2009 (UTC)

Now, Ungtss, perhaps you believe that magical elves came into your house and made the cake? Perhaps you would prefer to believe that the cake and the accessories materialized out of thin air (there is, after all, a non-zero probability of this occurring). But regardless of your willingness to believe in any particular explanation, the most logical one remains.

Again, that's based on my reason, not science. Science has nothing to say about whether the cake baked itself. It doesn't have the tools, because you can't experiment. Ungtss 18:52, 5 May 2009 (UTC)

You can reject common descent all day long, but it still remains the best hypothesis based on the available evidence; and until you offer a reasonable alternative hypothesis, your argument will remain unpersuasive. Jorge 18:08, 5 May 2009 (UTC)

Now that we're out of the realm of science and into the realm of "reason" and "interpretation of evidence," your opinion that "common descent is the best hypothesis based on the available evidence" is only your opinion. Your opinion may well be right or wrong, because it is not science. Every jury member sees the evidence differently. Dammit, they have a right to. All you need is one year in law school to realize that perfectly reasonable people can interpret the exact same evidence in completely opposite ways. Our persistent failure to recognize this fact is one of the tragedies of human history. Ungtss 18:52, 5 May 2009 (UTC)

Please indulge me, how do you interpret the evidence? Jorge 19:36, 5 May 2009 (UTC)

My interpretation of the evidence is not really significant. It changes daily, as I learn and grow. The real important point is that we agree on the rules of the game -- what science is and isn't, what it can and can't do. And how we should proceed in learning about the topic. Do we agree on the rules of the game? If not, further discussion is just as useless as playing chess with two different sets of rules. If so, I'd love to discuss the topic with your further, as I find it fascinating. Ungtss 19:47, 5 May 2009 (UTC)

For purposes of this debate your interpretation is indeed significant. Anyone can plead ignorance to a philosophical postulate. Just as it is impossible to convince someone they are not living in a Cartesian dream world, it is impossible to debate with someone who is not willing to share their opinion on the topic being debated. To use your analogy, this is akin to playing chess with someone who refuses to make a move. Jorge 20:40, 5 May 2009 (UTC)

You say I'm not willing to share my opinion on the matter being discussed. I think we might be debating different issues. I'm not discussing the matter of "how did life originate?" If you track this discussion all the way back to its beginning, I was defending my statemet that experiment and falsifiability are necessary elements of the scientific method, and that because neither CD nor creationism is experimentally falsifiable, neither is science. That is what I'm here to discuss, not "how did life begin?" I'm well aware that if we get sidetracked into the "which interpretation is best" discussion before we've agreed on the ground rules of epistemology, we'll go round and round in circles until one of us either quits or reaches Godwin's point. But if you'll tell me we agree on the rules, I'm more than willing to move my first piece. Ungtss 02:39, 6 May 2009 (UTC)

But if you keep using your pet "CD (or CS?) isn't science" example, you sidetrack the discussion of "what is science?" Sticking to "neutral" or even "trivial" examples and analogies is how philosophy usually proceeds. ħuman 02:53, 6 May 2009 (UTC)

Fair enough. I tried it a few times but nobody seems interested. Let's go with "Who wrote the gospels?" Ungtss 03:19, 6 May 2009 (UTC)

Ungtss' Analogy

...neither CD nor creationism is experimentally falsifiable, neither is science.

If that is your argument then you are simply making an incorrect analogy, let me map it out for you: If one uses natural selection as our premise-which we both agree is a scientific principle-one may draw the logical conclusion that life originated from simpler forms that evolved over time. If we follow this line of reason, we regress until we are left with the hypothesis that life originated from some basic chemical form. While this is not 'science' according to the strict definition of falsifiability, I assert that this is the Best possible explanation based on the available scientific evidence. I believe, therefore, that I have drawn a logical, reasonable conclusion. Since you agree that this conclusion is unfalsifiable, you cannot possibly present any evidence that would disprove my assertion. Therefore, if you disagree with my conclusion, what alternative do you offer?

In short, your analogy is false because common descent is a logically derived belief based on scientific evidence, while Creationism is a belief based on ...? Is it something other than scientific evidence?

Do you still disagree? Please, fill in the blank. If you believe otherwise then tell me what it is and give me some evidence that supports your claim. However, if you insist on continuing your game of sophistry than I have no reason to spend any more time on this discussion. Jorge 03:26, 6 May 2009 (UTC)

Yes, but your extrapolative argument does not constitute "science" -- it constitutes "reason." And as we all know, reasonable minds can interpret things differently. You see that extrapolation as extremely strong. Others see extrapolations to that scale as being silly, noting that while there are no demonstrable hurdles to variation on a small scale, there are huge hurdles to variation on a large scale (most notably the origin of the first protocell itself) and therefore the extrapolation is naive. Does that mean you're wrong? No. But it means you're in the realm of reason, not science -- and reasonable minds disagree. Ungtss 03:34, 6 May 2009 (UTC)

Nice job, Ungsst, you called Jorge "silly" and "naive" in one short reply. I would disagree with your trivializing of his point, especially if you want to equate an old story in a copy of some old books with his reasoning - to me, that is truly "silly" and "naive". But I hope that isn't your point (to defend CS as equally reasonable as CD). ħuman 04:47, 6 May 2009 (UTC)

Indeed we disagree. Yet you remain unwilling to offer any reason why. Jorge 03:38, 6 May 2009 (UTC)

Just tell me if you agree that we're in "reason-land," not "science-land," and I'm more than willing to get into the meat of the issue with you. Ungtss 03:40, 6 May 2009 (UTC)

Unfortunately, as science and reason are intrinsically tied, I could not make such a concession. If the issue you raise does have any meat, I remain hungry. Jorge 03:50, 6 May 2009 (UTC)

Alright, let's try something else. Let's define our criteria for a "reasonable alternate view." I propose the following:

1) Not self-contradictory;

2) Not directly contradicted by physical evidence;

3) Not a tautology;

4) Based on well-established, reasonable inferences from well-established, reasonable forms of evidence;

Things that are not required:

1) Falsifiability, as falsifiability is impossible in the context of unrepeatable events and phenomena that cannot be directly observed;

2) Acceptance by any particular group of people, as most people are wrong most of the time;

3) Any particular motive for belief in its adherents. That is to say, a person may believe a reasonable thing for an unreasonable reason, and yet the thing they believe be both reasonable and true.

Perhaps we can come up with more as we go, but that seems a good start. I propose limiting our discussion to those criteria (and not applying the non-criteria). What do you think? Ungtss 04:36, 6 May 2009 (UTC)

I know you're not directly asking me, but those criteria sound "reasonable" to me. It's a good start, anyway. ħuman 04:51, 6 May 2009 (UTC)

Like Human said... Jorge 05:30, 6 May 2009 (UTC)

Reasonable alternative

Okay. So we have our goal and our criteria. Now let's get some preliminaries out of the way: Things that aren't reasonable.

Supernatural creationism is not a reasonable alternative view, because it violates criteria 1,3, and 4. It is self-contradictory because by definition something "beyond nature" cannot causally interact with nature. If something is causally linked to nature, then it is, for all intents and purposes "part" of nature. It is a tautology, because everything can be explained by "God did it." It is not based on well-established, reasonable inferences from well-established reasonable forms of evidence, because it is based on the allegedly divine authority of a single book.

Agree? Ungtss 05:49, 6 May 2009 (UTC)

Still waiting... Jorge 16:28, 6 May 2009 (UTC)

Do you agree? Ungtss 19:49, 6 May 2009 (UTC)

Definitions

I'm really surprised no one has done this already. Ungtss, could you offer definitions of "evidence", "observation", and "experiment"? OneForLogic 03:07, 6 May 2009 (UTC)

Good question. Here's a first swipe that will probably need to be refined.

Evidence -- anything that can be used to determine the truth or falsehood of an assertion;

Observation -- receiving data through the senses, or recording data with scientific instruments

Experiment -- This one's harder. An approach to investigating a phenomenon in which circumstances are controlled so as to allow a hypothesis to be tested. In controlled experiments, this means isolating a single potential fact or cause, and observing whether the expected fact or effect is observed. Ungtss 03:29, 6 May 2009 (UTC)

Those all sound reasonable. Next, while I could attempt to extract this from what's been written so far, it would be much better if I let you do it. What is your preferred definition of "science"? OneForLogic 04:08, 6 May 2009 (UTC)

Cool. Science is both a method, and a body of knowledge extracted using that method. The method is well known. 1) Define the question. 2) Gather information and resources (observe). 3) Form hypothesis. 4) Perform experiment and collect data. 5) Analyze data. 6) Interpret data and draw conclusions that serve as a starting point for new hypothesis. Conclusions that stem directly from an experiment testing a hypothesis are "scientific knowledge." That's science. Nothing else is science. Everything else falls into other categories, like "reason" or "philosophy" or "religion" or "speculation" or "madness." Ungtss 04:28, 6 May 2009 (UTC)

Human's comment on "4."

I would comment on "4". Sometimes just collecting "more" data is essentially the same as "experiment [with controls]" - like the observation of gravitational lensing that strengthened the ole relativity thing way back when. (Or, when archeologists dig up more cool stuff and clarify what historians claim, for better or for worse.) ħuman 04:54, 6 May 2009 (UTC)

This is a goldmine. I disagree with you emphatically. Additional evidence which is interpretted as being consistent with a preexisting belief is not equivalent to an experiment, because it does not test anything. Without a test, there is nothing eliminating alternate possible explanations for the result. The number one enemy here is "confirmation bias." You already believe something to be true, so all future learning is seen as confirmation of your pre-existing belief. Science cannot allow for that. No experiment, no science. Ungtss 05:01, 6 May 2009 (UTC)

You misinterpret me. Say one says "we think a city existed here from 800-900 CE due to our interpretation of other evidence". So we dig there to an appropriate layer of detritus - the prediction was we would find "x" type of artifacts and evidence. If we don't, the hypothesis is weakened (maybe demolished). If we do find them, it is strengthened. "Belief" has nothing to do with it. Such a dig is a test. If alternate explanations are not affected, they still exist. Or may be likewise weakened or strengthened by the evidence uncovered. Ideally, each hypothesis has made its predictions, so the weakening and strengthening has similar weight on some or all hypotheses.

"You already believe something to be true, so all future learning is seen as confirmation of your pre-existing belief. Science cannot allow for that. No experiment, no science." That's bullshit, and has nothing to do with what I said. Often, we explore, dig, observe with the intent of disproving our hypotheses (looking for the falsifying evidence?). There's more to evaluating predictions than controlled experiments, although, of course, tightly controlled experiments are the sine qua non of the so-called "hard" sciences. ħuman 05:15, 6 May 2009 (UTC)

When you start with the hypothesis "A city existed here from 800-900ce," you've got a compound hypothesis -- a bad one. You have to break it into two hypotheses: "There used to be a city here" and "the city was here from 800-900ce." You can't answer the second question until you answer the first. You experiment on the first by digging. Digging can prove the hypothesis true (by finding a city), but it cannot prove the hypothesis false (because just because you don't find a city there doesn't mean there didn't used to be one there that got completely razed clean, or that you dug in exactly the right spot, or that you missed key clues). Therefore, finding a city and saying "there was a city here" is science. Not finding a city and saying "there was never a city here" is not science.

Then you get onto the second question. How long ago was the city there? And for that, you need a dating method. That's much tougher. But the issue is the same. You have to find a dating method that establishes that a particular date is true, and no other date is true. Not easy. Ungtss 05:26, 6 May 2009 (UTC)

Ungtss, sometimes I wonder if you know anything about how historians and archeologists work. There are literally hundreds of ways to date an excavation - and guess what? If none are applicable (we found some cool stuff, we have no idea how old it is), they say so. But usually, there are indicators, ranging from C14 dating (in this case), artifacts of comparable style with known dated sites, documents found referring to astronomic phenomena, documents referring to events identifiable with other places (wars, trade, etc.). I didn't mean to sound rude in my first sentence, but this stuff is rather basic, really. ħuman 05:42, 6 May 2009 (UTC)

PS, if no evidence was found on that dig, of course we know we didn't prove or disprove anything. We keep looking if the reasons for our hypothesis are strong, or we modify or abandon the hypothesis if the basis was weak. Modify might mean "look elsewhere, reinterpret sources", abandon might mean "reinterpret sources, were they just a myth"? It's not perfect, like photographing subatomic particles in a cloud chamber, or analysing fossil DNA, or measuring planetary orbits, but it's pretty darn good science. The only tough part is the "absence of evidence" part. ħuman 05:47, 6 May 2009 (UTC)

C14 dating does not date a city. It dates an artifact. I live in Turkiye, and I have 2500 y/o castle ruins just 10 min down the road. Go into one of these castles and you'll see artifacts from the entire 2500 year spread. You have old blocks, replacement blocks, the wood used as a bathroom, the urine stains on the wall outside, old bones from people and animals that used it as a shelter during Ottomon times, coins dating from 2008ad to 200 bc, the 30 y/o rug of the homeless guy who lives in the old kitchen, and spraypaint from the kids who vandalized it last week. Dating an artifact does not date a city scientifically. But many archaeologists, unfortunately, don't understand that. Same thing with "style." Look in my house, and you'll see artifacts that are in the "style" of the 19th century SW USA, 17th century Turkey, colonial England, 21st century US, etc. The existence of things in the "style" of an era does not necessitate a date for the city. Many archaeologists (and you, apparently) don't understand that either. It's hard to do science in the realm of archaeology, because the evidence is scant and very very difficult to test. Many archaeologists use their credibility as an alternative.

Secondly, the scientific method does not allow for "strong" or "weak" hypotheses. Tested hypotheses are true or false, yes or no. Reason allows from "strong" and "weak" beliefs. But reason is in the eye of the beholder. Ungtss 05:57, 6 May 2009 (UTC)

Firstly, how do you know how old those artifacts are? Secondly, yes, it does. The strongest hypotheses become known as "theories" and the weakest end up being known as "discredited". Science is not "off/on", it's "more or less probably". Thirdly, can you drop the [Many archaeologists (and you, apparently) don't understand that either] random ad homs against all those you are smarter than? ħuman 06:33, 6 May 2009 (UTC)

Firstly, I don't know how old the artifacts are scientifically, I'm just estimating based on common sense. That's the point, and the problem. It's naive to find a few artifacts that date to a certain era and "scientifically date" the city to that era. You can say, "I think based on these things," but you can't say "science says." Period. Secondly, "probably and less probably" is a subjective determination, not an objective one. What's probable to one person is improbable to another. Science, on the other hand, is objective. That's what makes it so powerful, and so beautiful. It doesn't depend on what people think is probable or improbable. It depends on experiments and observations that can be repeated by anybody, anytime. That's why I love science. By calling subjective things "science," you are cheapening science by throwing a lot of bullshit in with the golden stuff. Thirdly, it's not a random ad hom -- it's an assertion. An ad hom is an argument for a conclusion which depends on an individual's lack of credibility. I'm not arguing that anything is true or false based on your credibility or lack thereof. I'm simply observing that many archaeologists (and you) don't appear to understand how hard it is to apply to full rigor of the scientific method to archaeological ruins. You might find RW's article on ad hominem to be helpful here. Especially the section entitled "strict usage," which talks about how "some people like to throw around ad hominem because it sounds cool." Ungtss 07:25, 6 May 2009 (UTC)

We either date them or we don't. We try to - based on whatever evidence we can "dig up". Some evidence is strong, other is not so strong. Some is ironclad, and I think you might be missing that - some old stuff is really well credentialled.

"I'm simply observing that many archaeologists (and you) don't appear to understand how hard it is to apply to full rigor of the scientific method to archaeological ruins" How many (name one?) archeologists are you arguing don't understand the scientific method? Oh, and once again, you insult me. I don't care, but you take the trouble to do so. ħuman 07:54, 6 May 2009 (UTC)

The point is that a determination of what evidence is "strong" and "not-strong" is a subjective determination, because it does not come out of the scientific method. There is no "strong" or "not strong" evidence in the scientific method. There is "Experiment showed hypothesis to be true" and "Experiment showed hypothesis to be false." What you're arguing for is "science" without the scientific method. I have no use for your science without method, thanks. Ungtss 08:02, 6 May 2009 (UTC)

Then apparently you have no idea how it works. For you it has to be black and white? No science is that clear, sorry. ħuman 08:39, 6 May 2009 (UTC)

You claim that "no science is black and white." Let me falsify your statement with some black and white science: Rate of acceleration due to gravity. Organelles in the cell. Observed Speed of light. Stages of mitosis. Chemical composition of water. Function of radio waves. Hormone-caused aggression. The function of the hypocampus in memory. The stages of decomposition. The function of genes. The number of chambers in the heart. The structure of a volcano. On and on and on. There is an enormous amount of black and white science out there. That's the good stuff. And the better we get at developing experiments, the more stuff we'll be able to make black + white. In physics, the TOE. In biology, the origin of life. Etc. etc. etc. See how that works? You said something wrong, I showed you how it was wrong with evidence. Now you need to infer a new hypothesis: "Some ideas I call science are black and white. Some are not." Now you analyze, "Why do I call 'fuzzy' things science? What's the objective method and criteria for that?" If you come up with something, you present it for discussion and evaluation. If you can't come up with anything, then you conclude, "If I don't have a reason to clal those things science, maybe I should call them something else. I wonder what I should call them." Get it? Ungtss 09:18, 6 May 2009 (UTC)

From the top...

The rate of acceleration due to gravity - black or white - hardly. First of all what you're talking about is the Gravitational Constant.

Grey area #1 - is it really a constant and, if not, what are the implications.

Grey area #2 - what exactly is gravity?

I'm not enough of a scientist to cover all the points but all science is about pushing back the grey areas. The only thing that's black and white is Cartesian existence. Silver Sloth 13:24, 6 May 2009 (UTC)

1 -- Is there any evidence it's not? If not, it's black and white until there is.

2 -- That's not part of the black and white science I was referring to. So far it's unknown. Maybe someday we'll have a scientific explanation.

3 -- Agree that all science is about pushing back the grey areas. But cartesian existence is certainly not black and white. Relativity killed that one a long time ago. Ungtss 13:28, 6 May 2009 (UTC)

Come on - you say you've read Popper - Something is black and white until it's disproven. Err No! Try looking at the problems of empiricism. Nothing is ever proven true, only false. It's a good working hypothesis that the Gravitational Constant is just that but, as the Creationist are fond of saying, it's only a theory - it's no more black and white than the Theory of Evolution. Silver Sloth 13:36, 6 May 2009 (UTC)

Additionally, by Cartesian existence I was referring to Cognito, ergo sum which is not affected by relativity. I know that something, which I choose to call I is doing some thinking and therefore must exist. All else is guesswork. Silver Sloth 13:43, 6 May 2009 (UTC)

The observed rate of gravitationial acceleration is a lot more black and white than common descent. It can be directly observed, while common descent cannot be subjected to repeatable, falsifiable experiment. By "black and white" I don't mean "immutably true." I mean "No reason to doubt." New evidence might always create reason to doubt, but until then, we have an "answer" to the question that is correct within our scope of experience. Is Cogito immutably true? Certainly. But other things (like science) are very, very reliable because they can be tested. Ungtss 13:50, 6 May 2009 (UTC)

(Undent)The key word there is more - The Gravitational Constant may be a darker shade of grey but it's still not black and white. In both cases the observed evidence supports the theory. It's easier to observe evidence for the GC than for common descent but it's only a matter of scale. In either case, as soon as there's evidence against the theory, the theory falls. Silver Sloth 13:55, 6 May 2009 (UTC)

And, if you want falsifiability for Common Descent - show me a significant life form which doesn't use DNA/RNA - and the theory takes a major hit. I'm not up enough to say that it's sunk but it would definitely be holed below the water line. Silver Sloth 14:13, 6 May 2009 (UTC)

But there's one fundamental difference -- in a (good) controlled experiment, you've eliminated the alternative possibilities by the way you set it up. In an untestable extrapolation, you can't do that. Observe the effect of a mutation on a fruitfly in a controlled environment, and you can control for all other effects, and do some science. Infer that a certain unknown mutation took place in an unknown location in an unknown ape-like ancestor to distinguish the lines of descent, and you can't eliminate any number of possible mutations, or the possibility that the species don't share descent. With all those other possibilities hanging around, it's just not science IMO. Ungtss 19:03, 6 May 2009 (UTC)

No, Ungtss, science is not always absoloute. Science makes conclusions based on the best available hypothesis based on the evidence. But, first let me demonstrate how some of your examples also fail to be absolute. (1)Scientists believe that universal constants are constant. That's a hypothesis supported by the available evidence. But since we cannot measure the speed of light or the rate of gravity with perfect precision, we can only estimate them based on observations we have made. We may be 99.99999% certain that these fall within a certain range, but this is not absolute.

You're attacking a strawman. I'm not saying that science is laid in stone truth for all time. I'm saying that Science is black and white to the degree to which we can test it. Of course our knowledge is grey beyond that. But that POINT is the point at which science stops being testable, and therefore stops being science. Ungtss 18:40, 6 May 2009 (UTC)

(2) Cells, organelles, mitosis, etc. We believe these based on some set observations. Even if every observation fits our model, we can never prove that there will never be exceptions. Just because every swan I have seen is white does not prove that all swans are white.

Again the strawman. I'm not saying science is truth for all time. I'm saying it's truth so far as we can test. But beyond what we can test, there's only "educated guesses," not science. Ungtss 18:40, 6 May 2009 (UTC)

(3) Molecular composition of water. This depends on how you phrase the argument If you say all observed water is composed of two hydrogen atoms and one oxygen, then since that is the definition of water, your assertion is circular and redundant. Likewise with volcanoes. If every volcano is only what fits your definition of a volcano, then you are making the No True Scotsman claim.

Now that's a silly argument. We knew what water was long before we knew what it was made of. That discovery was actually revolutionary, not circular and redundant. Water has all sorts of uniquely identifiable qualities apart from its composition. Therefore it's not even a little bit circular. Ungtss 18:40, 6 May 2009 (UTC)

(4) Not all hearts have four chambers.

Again you're attacking a strawman. Nobody said all hearts have the same number of chambers. But given a species, we can predict (apart from the occasional broken specimin) the number of chambers its' heart has. That's pretty remarkable. Ungtss 18:40, 6 May 2009 (UTC)

Lastly, I would like to address your claim about archaeology and the age of cities. Of course artifacts date over a range of time; cities and civilization persist over time. If I find a set of artifacts dating from 4000-3000 BCE at a particular site, I might argue that the best possible explanation is that people lived there during that time and they made those artifacts. An alternative hypothesis might be that a medieval collector found artifacts from that specific range and buried them there. A third explanation might be that aliens did it. But since the first one is the most likely, reasonable explanation, we stick with that in the absense of further evidence.

Well, you'll have to be more specific. What artifacts are we talking about? Because you can only carbon date certain things -- e.g., dead stuff. And dead stuff pops up all over the place for any number of reasons. So what kind of "artifacts" are you talking about in your hypothetical? Ungtss 18:40, 6 May 2009 (UTC)

Have you ever read a technical scientific journal? They are filled with 'we believe that', 'most likely caused by', and 'probably due to'. Statistical hypothesis testing is the foundation of science. Jorge 17:01, 6 May 2009 (UTC)

Statistical hypothesis testing is the foundation of bad science. Ungtss 18:40, 6 May 2009 (UTC)

Please provide evidence to support your claim. Can you provide an example of a scientific discipline that does not use statistical hypothesis testing? Though I generally prefer not to be condescending, your persistence in this argument seems to reveal your ignorance of the scientific method. Jorge 18:52, 6 May 2009 (UTC)

My claim is not that some disciplines don't use statistical hypothesis testing. My claim is that no matter where it's conducted, it's a lazy, half-ass approach to science that doesn't meaningfully and scientifically answer the questions it seeks to answer. Ungtss 19:06, 6 May 2009 (UTC)

<sigh>This, once again, begs the question, "What approach does meaningfully and scientifically answer the questions it seeks to answer? Come on, don't be shy, give me an example. Jorge 19:37, 6 May 2009 (UTC)

Out of curiosity, are you reading back through the thread when I respond? I'm responding to each paragraph individually, but you never seem to respond to any but the last. I gave you a long list. The chemical composition of water. The flow of blood. The structure of a volcano. Etc. Those are things we can sink our teeth into. They're useful, reliable, and (in my definition) scientific. Compare this to "this drug dencreases illness in 60% of cases, and increases it in 20%." Bullshit. What good is that? Figure out what the drug DOES, dammit. Ungtss 19:43, 6 May 2009 (UTC)

I did offer a refutation of how those claims do not constitute "falsifiable scientific assertions". Lets take the chemical structure of water as an example:

The claim "all water molecules are composed of H2O" is merely a definition, it is reflexive and unfalsifiable. Yes, it is a fact that we all can agree upon, but one bourne out of scientific observations using empirical statistical hypothesis testing. Let us, for example observe a molecule x, and through some experimental test, we conclude with a reasonable degree of certainty that it is in fact a molecule of H2O. We could never prove this to be the case because of the possibility of experimental error.

Lastly, if you do not agree that hypothesis testing is part of the scientific method, then for purposes of this argument, you are wrong. Feel free to invent your own method of inquiry, but please don't confuse us by calling it science. Jorge 20:09, 6 May 2009 (UTC)

Again, did you read the thread above? I responded to your water argument immediately afterward, but you're not addressing my response. We knew what water was before we knew what it was made of. Discovering the composition of water was not "circular." It was "revolutionary." We finally knew what the mysterious substance we'd been dependent on since life began was composed of. As to your last point, I have been arguing for hypothesis testing in science, and (unless I'm mistaken) you have been arguing that it is not necessary. Ungtss 20:33, 6 May 2009 (UTC)

Joke time

Cue one of my favourite jokes:-

Three university professors are travelling together by train to a conference in Inverness. The professor of Astronomy looks out of the window and sees a black sheep.

"Look guys" he says. "In Scotland all sheep are black."

"Nonsense," says the Physics professor. "All you can say is that in Scotland there are some black sheep."

"Nonsense to you." says the Maths professor. "All you can say is that, in Scotland there exists at least one sheep of which at least one side is black."

So, what's the point. Well, in Astronomy evidence is hard to find and Astronomers are often forced to make vast leaps of generalisation from very little evidence. Physicists, in general, have to hold their evidence to far tougher rules but neither have the rigour that Maths demands. Does this make physicists or astronomers bad scientists - I don't think so. There's a lot more to science than black and white statements and falsifiability and the demands made of the evidence differ according to the discipline. What doesn't change is that the scientific method is evidence based and about testing the hypothesis - or maybe I just wanted a chance to tell my favourite joke ;-) Silver Sloth 18:07, 6 May 2009 (UTC)

I agree with you 98% of the way. My only objection is that I don't think those "best guess" answers are really of any use. If we can't test it, can't use it, can't manipulate it, can't know it with falsifiable certainty, then why, really, do we need to know what the scientists' "best guess" is? I think scientists should shut up about an issue until they develop a method to falsifiably answer the question. You know -- actually do science. "Scientists think this is how stars are born" doesn't mean anything to me. Show me data, and explain how they falsify the alternative interpretations of the evidence. Nothing else is particularly useful, because it's not reliable. Ungtss 18:48, 6 May 2009 (UTC)

The point you seem to be missing is that nothing is absolutely certain. Science is the method of arriving at the best possible guess. If you can't handle that, then I suggest that the aliens who are manipulating the brain in the jar which believes itself to be a resident of the non-existent country of Turkey (it is merely a figment of your imagination) just pull the plug because the illusion of reality can never be reasonably proven for subject#x4143u935uy. Jorge 19:01, 6 May 2009 (UTC)

We're not talking about absolutely certain. We're talking about falsifiably certain -- which is not absolutely certain, but a helluvalot closer than "I think so ..." Ungtss 19:07, 6 May 2009 (UTC)

No scientific assertion can be absolute because a valid scientific assertion must be falsifiable. If a claim is falsifiable this necessarily means there is some non-zero probability of it being false, therefore it is not certain; if a claim is unfalsifiable it is either untestable or logically impossible to be false. Tell me then, where do you draw the line? 95% certainty, 99%, 99.9999999%? While we can agree that some scientific assertions are more certain than others, they all belong to the field of science because they employ the scientific method. Jorge 19:32, 6 May 2009 (UTC)

See logical discussion currently occuring immediately below with 1forLogic on exactly this topic. Ungtss 07:11, 7 May 2009 (UTC)

One For Logic on experiments

Below is Ungtss' preferred definition of science, for reference.

Science is both a method, and a body of knowledge extracted using that method. The method is well known. 1) Define the question. 2) Gather information and resources (observe). 3) Form hypothesis. 4) Perform experiment and collect data. 5) Analyze data. 6) Interpret data and draw conclusions that serve as a starting point for new hypothesis. Conclusions that stem directly from an experiment testing a hypothesis are "scientific knowledge." That's science. Nothing else is science. Everything else falls into other categories, like "reason" or "philosophy" or "religion" or "speculation" or "madness." —Ungtss

This definition again seems generally reasonable to me. Obviously, anyone who disagrees with it will object to the claim that "Nothing else is science". That sentiment would seem to the central point of this debate.

I agree that point #4 could use refinement. I think the many examples of data collection in history and archaeology and such that have been discussed here have made it clear that it is possible to draw scientifically valid conclusions from data acquired by means other than strictly controlled experiments. I think it's worth noting again (it's been said in various forms already) that determining exactly how probably various conclusions fit the data gathered on a dig to find an ancient city (for example) is not trivial.

I agree with Ungtss that confirmation bias needs to be countered systematically and that we should work to eliminate possible alternate explanations for phenomena, but I would argue that evidence confirming preexisting theories does indeed have significant value. There is also the practical problem that eliminating every possible alternate explanation for some given phenomenon could become an infinitely long process. It is far more feasible to take an existing theory, figure out all of its possible measurable implications (theoretical physicists and mathematicians spend lots of time doing this), and test them all to see if reality is consistent with all of these implications. Also, there are quite a few phenomena out there that we have yet to explain in even one way, so falsifying alternate explanations is, at this point, simply out of the question.

I do believe that there are steps the scientific community needs to take to self-regulate more effectively, particularly in their public relations. I'm sure there are ways to further reduce the magnitude of problems like confirmation bias, and we need to work on finding and implementing them. However, I am entirely confident that, right now, overall, we're honestly doing the best we can. OneForLogic 18:36, 6 May 2009 (UTC)

I like what you say. One quibble. Can you clarify what criteria you are applying for "scientifically valid conclusions" when you say "I think the many examples of data collection in history and archaeology and such that have been discussed here have made it clear that it is possible to draw scientifically valid conclusions from data acquired by means other than strictly controlled experiments."? There's no doubt we can gather useful information by other methods. But by what definition of "scientifically valid" does that sort of information meet the criteria? Ungtss 18:43, 6 May 2009 (UTC)

My first instinct is to throw the question back at you: "Well, what does it mean to be "scientifically valid", anyway?" That's a hint to me that I'm about to do this wrong and that I need to think more before answering. I'll try again later this evening. OneForLogic 18:53, 6 May 2009 (UTC)

Your handle is not only inspiring, but has also thusfar proven highly descriptive of its object. Ungtss 18:55, 6 May 2009 (UTC)

Thanks. I do my best.

I think there's quite a bit of depth to the notion of "scientific validity". Go deep enough, and the issue does cross over into the realm of personal opinions and social and philosophical considerations that readily become prohibitively subtle to debate.

I'll start a first attempt at defining "scientific validity" by discussing how science treats errors and uncertainty. It's already been pointed out that all measurements have some non-zero uncertainty. Any scientist who attempts to report a measurement without convincingly explaining what the error of their apparatus or technique is will be ignored. What constitutes "convincingly explaining"? That's one of the points where we might start to cross over, here going into the social art of persuasion. How do you convince people you're reported the right error? I've been studying physics for years and not found a satisfactory general answer to that, so I leave it as an exercise to the reader to propose one if they're interested.

Ok, so everything has an error. In controlled experiments, reporting a convincing value for the error of any given measurement is relatively easy; this is the whole point of “controlled experiments”. In the examples about history and archaeology that keep coming up, the various errors in the measurements that can be taken need to be discussed. Also, since we don’t necessarily control it, context needs to be discussed. Such a discussion is possible, though, and can even be quantified, though it might not be trivial to do so. One can record the position, carbon dating results, material analysis, structural analysis, etc., of the artifacts found while looking for this ancient city (I kind of like this example), all with associated error. I would consider all of this data, with errors, to be “scientifically valid”. I would also believe that, in archaeology, there are models and theories that can be used to systematically analyze this data and describe the ancient city. I’m not an archaeologist, so I unfortunately don’t know the details of such theories.

So, bottom line: A conclusion is scientifically valid if its uncertainty, or some similar expression for its probability of being true, can be quantified. This can be done readily when the conclusion is based on data from controlled experiments. I also believe it can be done while analyzing historical artifacts, geological data, and other “past” information. OneForLogic 21:34, 6 May 2009 (UTC)

I'd agree completely. That's the beautiful thing about science: you eliminate virtually all subjective sources of error (like confirmation bias and alternative causes for the phenomenon, ideological bias, etc.) so you can quantify the error -- which will constitute imperfect performance of the experiment, inaccurate measurement, etc. "Ideological preference" is not an error we have to worry about in a truly scientific experiment, because the method of a proper experiment eliminates it.

Now I'd like to apply the definition, with which we both agree. What's the quantified uncertainty of the scientific conclusion: "Life originated via RNA world."? What about the scientific conclusion: "Life originated spontaneously"? Or the quantified uncertainty of the conclusion: "Sodom and Gomorra were actual cities, destroyed by fire + brimstone?" (FYI I'm flying back to the States today to go be with my wife as she delivers our first baby, w00t -- but consequently I'll be out of the loop for a couple days. Ungtss 07:07, 7 May 2009 (UTC)

Congratulations. Good reason to be out of the loop. I think the probability that Sodom and Gomorra actually existed and actually were destroyed by God is somewhere between 0% and 100%; I'm not interested in or capable of narrowing it down any more than that. My understanding is that the conclusion that "Life originated spontaneously" is much more likely to be true, but I don't have a number. I'm not a molecular biologist, so I'm afraid that's the best I can do there. They might not generally agree on a number for how certain they are on that point yet. There really haven't been too many experiments that I know of studying the spontaneous generation of biological molecules, mostly just theoretical work. And I'm afraid my understanding of RNA is even more sketchy, so there's even less I can say about that one. Anyway, let all this go for a couple days and have a good trip. OneForLogic 14:12, 7 May 2009 (UTC)

Thanks! You remind me of a good friend from a few years back at WP. You aren't also an attorney, are you? I agree with you -- we don't have any experiments that can quantify the error in the statement "Life arose spontaneously." That's the very limited claim I'm making here -- if a scientific conclusion must have quantifiable uncertainty in order to be considered science, and the uncertainty of the statement "Life arose spontaneously" cannot be quantified because its likelihood is so subjective and relative to our personal biases about how the world operates, then the statement "Life arose spontaneously" does not meet our criterion for being a "scientific statement." Does that mean it's false? Of course not. Could well be true, and could well be believed on the basis of reason, rational consideration of the options, or (in the alternative) blind faith and fanaticism. But either way, it's not science, because there's no quantifiable uncertainty. If someone can point to an experiment or experiments that quantifies uncertainty through reasonable and reliable methods, I'm willing to call it "science." In my opinion, no other category of statements is afforded the luxury of hijacking science's remarkable credibility. Ungtss 12:25, 9 May 2009 (UTC)

No, I'm a student of physics and mechanical engineering. I have no doubt I'd do fine in law school if I wanted, but that's not the plan at this point. I think you might have read slightly more into my claim about "life [arising] spontaneously" than I meant to say. I would assert that it is possible to define "life", "arising", and "spontaneously" with sufficient precision to study the issue in a scientific way. I know that many molecular biologists are trying to do that, but, not being an expert on this myself, I don't know how if there is widespread agreement on such definitions yet. I also know that they are trying to design models of how life might have arose spontaneously; the "RNA world" model you mentioned is among these models. Some experimental biologists are trying to figure out how to perform experiments to verify the plausibility of the chemical mechanisms of these models. So far, it's my understanding that the experiments that have been done haven't collected enough data to declare any one particular model clearly more plausible than any of the others. I agree with Kels below that some proposed models will certainly have been proven impossible based on the data we've gathered. Also, I hope your trip to the States has been good. I guess the fact that you're back means your wife and kid are doing well? OneForLogic 20:24, 10 May 2009 (UTC)

I agree with you that it's possible to investigate the possibility of life arising spontaneously -- my only contention is that the claim "Life arose spontaneously" is not (yet) a "reliable conclusion of science," because the hypothesis has not been tested. Could be true, one might even reasonably be convinced by the evidence it is true. But in my opinion, it's not yet a "reasonable conclusion of science." So far everybody's well -- baby's still in utero tho, but she's got some legs on her, kicking away:). Ungtss 21:16, 10 May 2009 (UTC)

I think we've reached an understanding. "not yet a reliable conclusion" seems like a fair assessment of the set of hypotheses that suggest that "life arose spontaneously". There are a lot of very capable people working on this, so I'm extremely confident we'll have a much better idea of how plausible all of this is in a decade or two.

I apologize for the hostility you're getting from some of the others and glad to see that it hasn't kept you from coming back. Best of luck to you and your family. OneForLogic 23:16, 10 May 2009 (UTC)

Thanks man -- enjoyed the thoughtful, nuanced thought. Best to you. Ungtss 23:20, 10 May 2009 (UTC)

There are controlled experiments testing the feasibility of various abiogenesis mechanisms under various conditions, and the data from those impacts the feasibility of the overall hypothesis. For instance, if life is thought to have started in a certain way, but it's found via experimental testing that some of the mechanisms don't actually work under the hypothesized conditions, then the hypothesis has to be reworked, because now it's far less feasible. Alternatively, if the mechanism is shown to be feasible, that strengthens the hypothesis and may make it more probable than another. That's good science. --Kels 15:58, 7 May 2009 (UTC)

All essentially true as far as I know, and I agree that this is good science. It's my understanding, though, that there isn't enough data to declare that one hypothesis is clearly more feasible than any other yet. OneForLogic 20:04, 7 May 2009 (UTC)

No idea. But I'd bet there are a few hypotheses that have been ruled out along the way because the mechanisms aren't possible. "Magic man done it" among them. --Kels 20:09, 7 May 2009 (UTC)

Ungtss, Make Up Your Mind

Do you claim you support or oppose hypothesis testing?

As to your last point, I have been arguing for hypothesis testing in science, and (unless I'm mistaken) you have been arguing that it is not necessary.

Clearly, I'm in favor of it, you however:

Statistical hypothesis testing is the foundation of bad science.

My claim is not that some disciplines don't use statistical hypothesis testing. My claim is that no matter where it's conducted, it's a lazy, half-ass approach to science that doesn't meaningfully and scientifically answer the questions it seeks to answer.

Perhaps you don't like statistics? Unfortunately, mathematical data analysis is used in every scientific field. I actually studied statistics in college, so perhaps I might illuminate you as to how it is indispensable to modern science. All those famous scientist that brought you to those understandings of the molecular composition of water, volcanoes, gravity, etc. Yes, they used the maths too.

Your cannot believe that the scientific method is valid whenever you chose. Your insistence on this argument leads me to believe that you either don't understand the scientific method or (as I suspect based on your interaction with other users) you just like to hear yourself argue. Jorge 22:08, 6 May 2009 (UTC)

Why are you conflating "math" and "statistical hypothesis testing?" Of course the sciences use math. I'm objecting to "science" that fails to isolate all the variables, such that the experiment leads to split results: "80% A, 20% B." All a result like that means is that you haven't conducted a controlled experiment. If you have an adequate, controlled experiment, you get 100% A, 0% B, because you've done your job and eliminated the other causes that lead to the unpredictable results. Do sciences do half-ass experiments? Sure. But they're still shitty. Ungtss 06:37, 7 May 2009 (UTC)

So an experiment to determine the half life of a radio active element which fails to eliminate the randomness with which radioactive particles decay is shitty science - what total bollocks! Silver Sloth 11:40, 7 May 2009 (UTC)

I think we all feel a line between the probability that water is composed of 2 hydrogen and 1 oxygen atoms and the probability of a certain painkiller will have a positive effect on a patient. In the latter case there are just too many variables to account to, too much holes in our understanding of the human body, especially when you add human psychology into the equation. After we conducted all the necessary trials hopefully we come up with a probability higher than 95% and we'll acknowledge that the drug has a positive effect. Thats a whole lot different than declaring that the Sun is "fueled" by hydrogen fusion or that fire is a phenomena caused by the addition of oxygen into combustible materials.
The difference I see is the lack of an absolutely thorough explanation of the effects of the drug. We have explanations, we don't start drug trials randomly with every chemical compound and combination, but after the drug is in a human body we can't track every molecule of it through every chemical reaction, we can't say for sure that it won't cross-react with something unexpected, or yet unknown. So we use clinical trials to gather a much more simple data of the effects and use statistical analysis to determine if it will generally have the desired effect or not.
I agree that this is bad science in a way that it settles for a (relatively) enormous margin of error, that it relies on a not complete understanding of how things work. However, it is science. It does everything we could do, it uses all our knowledge of the particular segment of science. It's a valid method of increasing our understanding of the world and all in it, and it won't stop till we get the absolute truth (or more probably until black holes dominate the known universe). --Kangaxx 07:26, 7 May 2009 (UTC)

Agree. Science, but half-ass. What really turns me on are efforts to push that margin of error back. Sure, we can't track the molecules through the body. But what do those 20% have in common and what do the 80% have in common? Develop a hypothesis. Isolate a control population without the trait and a population with the trait. See if the error margin is smaller. It's beautiful, really. Scientists really need to evaluate the evidence -- figure out why different people react differently to the same drug. That way when a doctor's considering prescribing the drug, he can ask you "do you have X protein in your system? Because if you do, you're going to have severe vomiting." That's good science. It's just a fairly rare thing, because you can get published (and avoid perishing) with a half-ass job, and the FDA only requires half-ass jobs. So why put in the extra effort? Why? Because there's nothing better than good science. Ungtss 08:09, 7 May 2009 (UTC)

All you're saying is that you have no understanding of medical statistics. We don't know, and for the foreseeable future we probably won't know, why different people have different outcomes from the same treatments. Medicine is a minefield of unknowns and impossible variables - attend any meeting of The Royal Statistical Society and see what I mean. Does that mean that Medicine is bad science - or, as you so insultingly put it, half-ass science. No, absolutely no. Guesswork and experimentation using the scientific method have dramatically increased life expectations and this has been the direct result of the life work of dedicated scientists. Silver Sloth 10:36, 7 May 2009 (UTC)

Just to put some background to my rant.

In 1960, when my sister died of leukemia survival rates were around 15%. Nowadays we still don't have a cure, we still have huge areas of unknowns but survival rates are now 50%. Science did that. Good science done by good scientists. If you ever find yourself in the nightmare where someone close to you has contracted leukemia you will realise just how much you owe to those scientists. Then you will understand the difference between 15% and 50% and how good science can still work when the data isn't black and white but needs careful statistical analysis. Then you will learn to value those who you so haughtily dismiss as half-ass. Oh, and the person who dedicated his life to analysing the results of this half-ass science may well have been my father who specialised in medical statistics for most of his working life. His speciality - how to measure the missing data. Don't ask me how it's done, it goes way over my head, but the RSS gave him a gold medal for it so those in the know valued his contribution. So yes, I feel pretty strongly about those who dismiss experiments that have an 80% - 20% outcome as half-ass. Be careful who you insult - they might just be the ones who will save the life of someone you love, and they will have done it with science. Silver Sloth 12:03, 7 May 2009 (UTC)

Well I would feel pretty strongly about anybody who said that too -- especially since I didn't, and wouldn't. A toast to the scientists who are doing real science, not accepting 80/20 and pushing toward 100%. And a pox on those who claim that anything less is good enough. Ungtss 13:28, 7 May 2009 (UTC)

Could you please specify who (in this conversation or elsewhere) who are claiming "good enough"? Quotes would be nice. --Kels 14:02, 7 May 2009 (UTC)

Don't bother, Kels, he won't listen to anything but his own prejudices. Silver Sloth 14:09, 7 May 2009 (UTC)

I don't understand what you mean by this 80-20 thing. The world is far more complex then A then B. Sterile^{walkie-talkie} 15:09, 7 May 2009 (UTC)

Ungtss seems to believe that experiments that don't have clear yes/no black/white outcomes are unscientific or, in his words half ass. He states that experiments which yield such results, in his words, means is that you haven't conducted a controlled experiment. As such he repudiates as unscientific the vast majority of, for example, clinical trials and suggests that those who publish such results are, again, I quote, because you can get published (and avoid perishing) with a half-ass job, and the FDA only requires half-ass jobs. So, according to the world of Ungtss, most medical research is half ass science published by those who can't be bothered to do better because that's all that's required by the FDA. Enough said. Silver Sloth 15:25, 7 May 2009 (UTC)

I think you said what I meant. In chemistry, for example, there are times when there are fuzzy grays, and how much more concrete, hard, physical science can you get? Sterile^{walkie-talkie} 02:02, 8 May 2009 (UTC)