Confusion about reasoning in science [closed]

Question

Situation 1:

One puts the hypothesis (Call it hypothesis H1) that "there exists a force of attraction between every two masses". One obtains a high degree of belief in this hypothesis by testing it experimentally several times. Even though there exists objects which were not tested for gravity (I believe objects faraway from earth, perhaps in other galaxies were not tested to see if they satisfy the claim H1), I think it is commonly believed that H1 holds. So to summarize, every object in the universe was either tested for gravity and gave a positive result for the hypothesis H1 or is not known whether it satisfies the hypothesis H1 or not (because possibly it is not feasible to do that).

Situation 2:

One puts the hypothesis (Call it hypothesis H2) "every object was created". Some people obtain high belief in this statement by observing that many sophisticated objects were created by humans or perhaps sometimes other animals. People who claim hypothesis H2 will not be able determine the status of every object in the universe i.e. they will not be able to determine for every object whether it was created or not. So to Summarize, every object in the universe was either known to be created or is not known whether it satisfies the hypothesis H2 or not (because possibly it is not feasible to do that).

It appears to me that situations 1,2 are the same. Question 1: Do you agree ?

Yet, people wouldn't consider situation 2 scientific. It seems to me that the two are equally scientific (Yet I wouldn't consider situation 2 scientific, I am in fact confused). Question 2: Why is this the case ? Is it because I am having a wrong understanding of what scientific means ?

I just had an uncivil discussion :) with someone discussing the above two questions. I will put a list of some of the possible objections that arised during the discussion along with my reply to each objection.

Objection 1: Scientists believe in the assumption (Call this assumption A1) that the universe behaves the same way in all places (very vague in my opinion and could lead to contradictions). Using this assumption, we deduce that even masses that are far away from earth will behave like masses that were tested on earth.

Reply 1: The same applies to situation 2. All objects which we can determine whether they were created or not we find out that they were created using assumption A1 we are led to the believe in hypothesis H2

Objection 2: Created is vague. Define Created.

Reply 2: I will not be able to define all words I use, since that would lead to an infinite sequence of definitions and words but the set of English words is finite. Instead of doing that I will use certain words as atomic (meaning I will not give definitions for them). These words will be the building blocks for language. I will define every word using atomic words. I will also set the word "create" to be atomic. The same thing happens in situation 1, the words force/attraction if undefined will be treated atomically. Even if they will be defined, I believe any good definition would use the words "motion" or "displacement" and they most likely will be treated atomically.

Objection 3: Hypothesis 1 is falsifiable, Hypothesis 2 is not. This is because one can make an experiment where an object is let go and we observe the direction where it travels.

Reply 3: One can also get an object in our world and ask the question whether it was created or not

Thank you

---

Edit:

I would like to have a good non-artifical definition of "scientific reasoning" that will make situation 1 scientific but will make situation 2 non scientific. I can't come up with such a definition, can you come up with one ?

Answer 1

The two situations are essentially the same, as far as you've described them.

The difference, undescribed, lies elsewhere. The theory of gravitation has immense predictive power--you can calculate times of eclipses, send spacecraft to Saturn, etc. etc.. This doesn't mean that every attractive force is gravity, though (e.g. magnets!). But a lot are, and the model fits observation extremely well. (So much so that people infer the presence of "dark matter"--something that has a gravitational effect but isn't visible--to explain things like galaxy shape. Dark matter is not as certain as gravitation, but gravitation works so well that dark matter needs to be very seriously entertained as a hypothesis.)

On the other hand, noting that humans make complex objects is kind of like noting that gravity pulls things together. It's a pretty good bet that certain kinds of artifacts have been made (archeology would be really difficult otherwise!), but we now know of a process (evolution) that enables living objects to become more elaborate and complex in appropriate circumstances. And we know that appropriate circumstances are pretty common.

If you don't know about electromagnetism (or strong and weak nuclear forces), it's not entirely unreasonable to suggest that attractive forces are all gravitational, even if your hypothesis doesn't seem to account for magnets very well. Likewise, if you have no idea that a complex object can arise in any fashion other than creation by a human, postulating that everything was created by humans or human-like entities is not that unreasonable, even if that hypothesis also doesn't account for many things very well.

Once you find a superior hypothesis, holding onto the old not-very-satisfying best effort is not the scientific approach. And when your best effort doesn't fit the data all that well, doesn't provide all that much explanatory power, you should be skeptical of it.

When it gets down to it, any process that helps you explain observations or theories and has some robust path for you to discover if you're wrong can count as scientific reasoning. It's easy to get caught up in the sociology of how the scientific method is typically practiced now (e.g. peer review in journals), but that's not logically necessary, just a heuristic that works sufficiently well for us at the moment.

Answer 2

"every object was created"

A puddle of water evaporates, leaving crystals. Those crystals were not created, therefore I have falsified your hypothesis.

You can rescue your hypothesis by making it increasingly vague and untestable. You could say "the crystals were created by the water evaporating". If that's all that you mean by "created", then your hypothesis is trivial. You could also evade the issue by focusing on the origin of the atoms that make up the crystal (or perhaps, the matter/energy that makes up the crystal, since we know that the atoms can form inside of stars). If you take that path, then you are just negating the possibility of any test of your hypothesis. At that point, the hypothesis becomes complete speculation.

I think the distinction between science and wild speculation is that scientific theories are defined in a precise-enough manner that they can be meaningfully tested. Untestable speculation is not necessarily false, but it is useless and irrelevant to us.

Answer 3

You might enjoy reading E.T. Jaynes' Probability Theory: The Logic of Science, freely available for example here: http://bayes.wustl.edu/etj/prob/book.pdf

The basic idea is that there exists some vast range of hypotheses (most of which you haven't thought of yet) which can explain any given phenomenon, any of which could potentially be true. As you acquire observational evidence, you use it to reduce your estimate of the probability of the truth of some of those hypotheses, and increase your estimate of the probability of the truth of others.

You can also use some appropriate metric, such as Occam's razor, to distinguish between hypotheses that explain the data equally well; for example, you can decide to penalize the prior probability of any particular hypothesis being true by how complicated it is (which can be formally measured by the information content of the hypothesis). For example, "Every particle is gravitationally attracted to every other particle, in proportion to the mass and inverse square of the distance", is less detailed than "Every particle except that one neutron over there is gravitationally attracted to every other particle, in proportion to the mass and inverse square of the distance". Although no instruments are precise enough to distinguish between these two hypotheses, we can heavily penalize the latter one for its extra information content. In other words, we can be highly skeptical that that one neutron gets an exemption from gravity, even before doing a careful experiment, because universal gravitation is a much simpler theory that equally fits all of our data.

The goal here being to progress towards a state of beliefs that will be as accurate as possible, ie, we hope that through this process, the beliefs we consider "most probably true" are the ones that are actually true. And the advantage of having accurate beliefs is, of course, the ability to make accurate predictions.

So that's the basic philosophy of science and the scientific method, although not every scientist may even be aware of this formalization.

Let's look at your examples:

So to summarize, every object in the universe was either tested for gravity and gave a positive result for the hypothesis H1 or is not known whether it satisfies the hypothesis H1 or not (because possibly it is not feasible to do that).

Although it may not be known whether it satisfies the hypothesis H1, we can assign a much higher probability to H1 being true even for particles that we haven't specifically tested, because all the data that we have so far supports gravity being universal.

Of course, as soon as we observe convincing evidence that conflicts with our laws of gravity, we will have to start seriously considering more complicated theories.

Some people obtain high belief in this statement by observing that many sophisticated objects were created by humans or perhaps sometimes other animals. People who claim hypothesis H2 will not be able determine the status of every object in the universe i.e. they will not be able to determine for every object whether it was created or not. So to Summarize, every object in the universe was either known to be created or is not known whether it satisfies the hypothesis H2 or not (because possibly it is not feasible to do that).

In this case, all of the artificial objects serve as evidence that H2 is universally true. And if you were raised in an entirely artificial environment, you might convince yourself that everything that exists was made by humans (although you will have to consider heavily penalizing the probability of that hypothesis on account of its chicken-and-egg quality). However, those of us who weren't born in a laboratory will observe enough non-artificial objects that we can confidently discard H2.

If you mean "created" in a broader sense than "created by humans", then we need to determine the precise meaning of this hypothesis, not least of all in order to assess its complexity, but also just so that we can determine which observations count as evidence in its favour. As it stands, it is more than a little bit vague, and that is a major distinction between H2 and H1. You anticipated this objection by claiming that "created" can be considered 'atomic' and thus needs no further definition; however, this is something of a cop-out. If you want H2 to be meaningful, you need to be able to discern between things that were and were not created, which means you need some kind of logical circuitry that can take a 'thing' as an input and produce created or not created as an output. Otherwise, the claim that "all things were created" is equivalent to the claim that "all things are bjogl", where 'bjogl' is an atomic entity that needs no further explanation. It will clearly be a hard hypothesis to assess.

Currently, although you have claimed 'created' to be atomic, you are in practice relying on the operational knowledge of English-language speakers to infer its meaning. This is why modern scientific hypotheses are generally grounded at a basic level by mathematical expressions whose meanings are unambiguous, and whose truth is therefore possible to evaluate.

One can also get an object in our world and ask the question whether it was created or not

Yes, but this relies on our operational understanding of the word 'created', and it is even ambiguous enough that not everyone will agree on whether a smooth river stone was 'created'. (Is an intelligent, purpose-driven, living creator a necessary condition for an object to be considered created? Or does any natural optimization process count? If a 'thing' is an assemblage of components, do the components also have to be created? English leaves such interpretation open according to context, which is something that a good hypothesis does not do.)

The particular danger here is that, even if you establish a formal definition of 'created', the similarity to casual language can cause people to make mistakes when they substitute the everyday meaning for the formal one. You can formally define a test for the 'created' property, but it might not match up well with our intuitions. This kind of playing loose with words can mislead good scientists into 'proving' such absurdities as sentience being impossible: http://www.newscientist.com/article/dn25560-sentient-robots-not-possible-if-you-do-the-maths.html . Here, for a suitable definition of 'sentience', it is possible to prove that it is impossible; however, that may not be regarding the same concept that you and I think of when we see the word 'sentience'. If we substitute our own interpretation of the word, the proof no longer applies, and we are left with false beliefs.

Answer 4

I would like to have a good non-artifical definition of "scientific reasoning" that will make situation 1 scientific but will make situation 2 non-scientific.

I like challenges!

Having admitted as much, I fear you will be disappointed. You might get a definition. You might get a good definition. But you ain't gonna get a non-artificial (not ad hoc) definition. That's because any definition that will work for your particular case won't likely solve the demarcation problem in a descriptively satisfying way (assuming that it is a sound problem to begin with).

Yet, people wouldn't consider situation 2 scientific. It seems to me that the two are equally scientific (Yet I wouldn't consider situation 2 scientific, I am in fact confused). Question 2: Why is this the case ? Is it because I am having a wrong understanding of what scientific means ?

Probably. There are two common meanings of "scientific" that you might want to keep separate:

1. what is amenable to empirical testing
2. what is (actually) part of science

A lot of hypotheses which are scientific in the sense of (1) are not therefore part of (2).

Consider the hypothesis H that the earth was created approx. 6,000 years ago. It is certainly scientific in the sense of (1), but it is not scientific in the sense of (2). "Well", you might say, "that's because H is false!" And you would be correct. But that's not the only reason why hypotheses in (1) are not therefore part of (2).

Consider the hypothesis-schema that all persons born under X are less likely to display Y than persons born under Z (H2, where X, Z are astrological signs and Y is a behaviour). Again, H2 is certainly scientific in the sense of (1), but it is not scientific in the sense of (2). But this time not because we have established that H2 is false (arguably nobody took the effort to test it), but because it lacks contact, as it were, with the body of actual scientific theories (which, incidentally, is the reason why nobody tested it).

I submit that the hypothesis "every object was created" in your example is of the H2-kind.

And that brings me to the difference between (1) and (2). In the context of (1) we can speak of single hypotheses, e.g. empirical generalizations, as being "scientific" somewhat in isolation. But in the context of (2), hypotheses don't exist in isolation, they are scientific in virtue of being part of theories, larger structures connecting many hypotheses with observations, laws, initial conditions and other things.

In this sense your "objection 2" is crucial: "Created is vague. Define Created." Here the request should not be aiming to get a clear definition of "created", but for the other party to introduce and explain the 'theoretical pedigree', as it were, of the particular hypothesis, i.e. in what theory the hypothesis is embedded and how. (I would say that this actually gives you the meaning of "created", but this is a matter of dispute.)

Concluding, your uneasiness with "every object was created" is justified insofar as the the hypothesis lacks a theoretical framework in science, i.e. it is not scientific in the sense of (2). Your confusion stems from the - equally right - observation that it is scientific in the sense of (1).

Keep in mind, however, that it would be unwise to require a definition that excludes a priori the hypothesis "every object was created" for there's no principled reason why it couldn't be assimilated in a scientific theory in the future. Search for a definition that prevents a priori the hypothesis from being a part of science and you overstep your aim unnecessarily. Do not confuse the actual lack of a theoretical framework in science with a principled impossibility of such an embedding. The former suffices to justify your uneasiness.

Answer 5

Question 1

The two situations are similar insofar as they concern the same very general problem, i.e. the problem of induction (see http://plato.stanford.edu/entries/induction-problem/). In short, the problem of induction questions the justification we have for making (inductive) inferences to objects we don't have experience or knowledge of.

In the two situations above the generalizations you mention (H1 and H2) are inferred from a finite set of observations. However, the generalizations, in virtue of their universality, concern all objects, even those that are yet not observed. Consider the following example:

Premises: Swan 1 is white, Swan 2 is white, Swan x is white...
Conclusion: Therefore, all swans are white

The generalization that 'All swans are white.' is inferred from the fact that 'Some swans are white.' Inductions, contrary to deductions, aren't truth preserving. The conclusion doesn't necessarily follow from the premises. Inductions are ampliative inferences because their conclusions logically contain more information than their premises.

Question 2

The two situations are arguably different. The general problem at stake here is the "demarcation problem" (see http://plato.stanford.edu/entries/pseudo-science/): What are the criteria for distinguishing genuine scientific claims from non-scientific ones?

Various criteria have been proposed, but let me suggest a simple and pragmatic one that might be helpful for the case at hand: 'Science is what scientists do.' Using this criterion, we could at least distinguish Situation 1 and Situation 2 in the following manner. Firstly, H1 is a claim actually entertained by scientists while H2 isn't. H2 isn't part of any actual scientific discipline. Secondly, H1 is justified in virtue of evidence gathered by scientists while H2 isn't (e.g. casual observation isn't a methodology scientists follow). We could then conclude that H1 is a scientific claim while H2 isn't.

Answer 6

My first thought as to what the difference would be was "sampling bias". You can prove that some things were created, and you can't determine whether many other things were created. If you ignore everything not created by humans, your conclusion must be reduced to "everything created by humans was created" by your sampling bias.

Now, it's true that the gravity claim also has a sampling bias: we can only test it for things that are nearby (although "nearby" in this case means "within the known universe"). This means that we can really only be sure that gravity applies within the known universe - and there are likely some models of the universe being developed that predict that gravity doesn't apply in some region of reality. "Everything within the known universe is subject to gravity" is still a very strong and useful statement, however.

Answer 7

Gravity is commonly thought of as a modern invention; in fact it dates from antiquity. It was then understood (by Aristarchus) that the Earth was a sphere, and Aristotle divided the universe into a terrestial sphere, where things fall in a straight line, and the celestial sphere where thins move in a circular motion.

It was Newtons great achievement to show that this divided cosmological scheme can be united - hence universal gravitation; in fact his basic assumptions were all universal - that space & time was everywhere the same (and that time flowed regularly); this universality is still a key assumption in physics. But one should notice that it is a key metaphysical assumption, it is not a derived notion, and its utility is in its explanatory power in the construction of physical theories.

Scientists believe in the assumption (Call this assumption A1) that the universe behaves the same way in all places (very vague in my opinion and could lead to contradictions).

This is generally true, but isn't completely true - we've just been through a period of multiverse enthusiasm where it was believed/theorised by some top-flight physicists that the laws of physics could be very different in different universes; but there is a deeper sense where it still remains true in that some substrate of the theory must remain true across the whole of the multiverse.

It appears to me that situations 1,2 are the same. Question 1: Do you agree ?

Considering that the current consensus of the physical sciences is that the universe was 'created' at the Big Bang, one should say yes, they are. It isn't because there is an embargo on terms with theological import - its part of its sociology; this does not mean that scientists cannot hold religous views - plenty do; its simply not expressed in public scientific forums - journals, papers etc; they remainprivate or are expressed in public forums that explicitly engage in theology; but notably the Templeton Foundation amongst others are explicitly engaged in bridging this divide publicly.

Interestingly enough there is a theological perspective that considers that the universe is created from moment to moment - in this sense creation is continually engaged in.

Answer 8

I do appreciate the attempts to detail the differences in the other questions; and I do think that a lot of good points are covered. But I believe the answer to your question is fairly straightforward:

The simple fact of the matter is that Q1 and Q2 are addressing different types of subject matter (this is assuming that by 'created' you mean 'created' in the same sense that is usually intended when asking 'was the Universe created').

Q1 is an empirical question. You mention that—being a statement about all bodies—there is no way to fully test whether the law holds. This is true, but it is also a non-sequitur. Physical Science is concerned with finding a model of the Universe that most closely aligns with experimental, empirical, data; no general law will be fully testable. The point is that over time, a massive number of observations are taken, and over time the model is changed as theories are found that more accurately allow for both modeling of past events, and prediction of future ones.

Q2 is a metaphysical question. Is there something outside the Universe that created it? (If the question is a physical one, e.g. is there something inside the Universe that created everything within, then the answer is: current experimental data gives no reason to suggest such a thing.) As a metaphysical question, it is inherently non-scientific, as physical science is concerned with things that are within the Universe, rather than things outside and/or prior to it. If you have particular hypothesis about some physical condition that would hold as a result of "everything was created" being true, then you would have two things to deal with: the reasoning leading you from the statement to the hypothesis, which would be non-scientific; and the hypothesis itself, which could be tested against existing physical models to evaluate it. Note that the question of whether any particular hypothesis proves that the overall statement is still outside of science, as science is not geared towards making such determinations.

[edited: 2014-07-07]

As the question is more-or-less a summary of (or parellel to, depending on the scope one allows) the belief that 'creation science' is a scientific endeavor, a discussion in the comments led me to the following quote from the Wikipedia article on Creation Science , which I think summarizes things quite incisively:

The overwhelming consensus of the scientific community is that creation science is a religious, not a scientific view, and that creation science does not qualify as science because it lacks empirical support, supplies no tentative hypotheses, and resolves to describe natural history in terms of scientifically untestable supernatural causes. Creation science has been characterized as a pseudo-scientific attempt to map the Bible into scientific facts. According to a popular introductory philosophy of science text, "virtually all professional biologists regard creation science as a sham".

[end edit]

_{As an aside: it seems you think that somehow scientific concepts are ill-defined; they are not: everything is defined rigorously, science does not attempt to find the 'why' of things, just the 'what'. Force is mass times acceleration; acceleration is increase in velocity. Velocity is distance moved per time; and so forth, eventually coming to rest at defined measures that anyone can use (generally with easier to use alternatives, such as a ruler or stopwatch, but at heart defined upon some physical property measured as accurately as possible) thus making anyone that uses the terms able to run tests on any scientific hypothesis without any disagreement as to what force, etc. are.}

_{However, this does not hold for 'X was created'. Created is not a well-defined concept, nor is it measurable. Axiomatic definitions are fine, but they must be well-defined to be science. In a purely philosophical discussion, yes, definition of terms may sometimes lead to an infinite regress, and at some point you have to find common ground in your definitions without arguing them. And, although your second question definitely has long been a matter of philosophical discussion, turning it into a scientific discussion most certainly does require definition. (And, really, once you have done that, it's hardly even the same question.)}

Answer 9

You have made an a priori mistake in your logic. In the scientific method, a hypothesis is not the first step. The first step is an observer observes a natural phenomena - or fact - and sees if the natural phenomena can be seen (or tested) again. Having observed the phenomena multiple times (or tested), the observer then comes up with an hypothesis to explain the phenomena observed. The observer then tests the hypothesis by seeing if the hypothesis can predict the future outcome of the observable phenomena before it occurs. If the hypothesis does predict future behavior, then the observer develops a theory to explain the phenomena. If the theory developed can be supported by or enhances our understanding of other previously accepted theories, the theory is then generally accepted.

Your second scenario does not fit the scientific method as "every object is created" is not an observable fact. You have started with a hypothesis that is not relying on any observable fact. The observable 'fact' is that we observe the universe, but if it was 'created' is not an observable fact. If you wish to have a hypothesis that it was created then you need to be able to observe a 'creation'. Read "Chaos: Making a New Science" by James Gleick. The universe is really a wave function.

Answer 10

If one found two or more objects that hypothesis H1 predicted should be attracted to each other and could demonstrate that this attraction did not exist or did not conform to theory, then H1 would be disproved. [Note 1]

If one found one or more complex objects for which no creator could be identified, then hypothesis H2 would be disproved.

It is fairly easy to find complex object for which one cannot identify a "creator" -- unless one begs they question by avowing that there MUST be a creator for such a complex object. Who/what created a tree?

It is not that H2 is not "scientific" it is just that on the basis of scientific reason it is demonstrably untrue.

[Note 1] and in fact the Newtonian theory gravity suffered just such a fate -- hence relativity.

Answer 11

Situation 3:

One puts the hypothesis (Call it hypothesis H3) "John likes everyone". Some people obtain high belief in this statement by observing that many John likes many people. People who claim hypothesis H3 will not be able determine the status of everyone i.e. they will not be able to determine for every person whether John likes them or not. So to Summarize, every person in the universe is either liked by John or is not known whether it satisfies the hypothesis H3 or not (because possibly it is not feasible to do that).

Is this science?

To answer the question(s):

Question 1: Do you agree ?

No.

Question 2: Why is this the case ? Is it because I am having a wrong understanding of what scientific means ?

Yes. There does seem to be a gap in understanding here.

From the first paragraph of the wikipedia page on Science:

Science is a systematic enterprise that builds and organizes knowledge in the form of testable explanations and predictions about the universe.

Situation 1

In Situation 1, we have a theory which has been tested time and time again. That is we can test the gravity of objects, we have experiments to test them. It also great predictive power. Using what we know about gravity, we can predict where the planets will be in orbits, a considerable amount of time into the future.

Most importantly, Situation 1 is falsifiable. If we ever found 2 objects that didn't attract each other, we immediately disprove gravity (or at least, we would need to revise what we know about gravity).

Situation 2

Situation 2, while you've tried to give it the same linguistic form, is very different. I noticed that you've removed the word "experiment" from paragraph two. The big difference here is that we have objects that we can observe in great detail, that we have not been able to determine whether it's created or not. We have them. Right in our hands. Nobody has been able to come up with an experiment that can produce results that could go both ways. Mainly because "creation" has no real definition in our natural world. We deal with what we can naturally observe. That simply isn't creation.

I would like to have a good non-artifical definition of "scientific reasoning" that will make situation 1 scientific but will make situation 2 non scientific. I can't come up with such a definition, can you come up with one ?

What do you mean by non-artificial?

But the scientific method, deals with testing falsifiable hypotheses. That's the bare minimum. That also, by the way, requires a coherent hypothesis. Which means you should probably define "creation" a little bit better.

Answer 12

H1 Gravitational forces are given by mathematical equations, which have been--and continue to be--measured and tested for veracity. There is strong empirical evidence that these relations hold in the observable universe. Just as importantly, the existence of gravitational forces is supported by a generalized theory, upon which many other strongly supported scientific theories rest. Think of scientific reasoning as a crossword puzzle; if all the words surrounding an answer fit in with your proposed answer, there's a much better chance that your solution is correct. Science in this way can provide positive feedback and reinforce itself.

H2 There is no precise definition for "created" as you have pointed out, but more importantly, using intuition, this claim is easily falsifiable. Mountain ranges, the oceans, the topography of Earth in general cannot be easily defended as having been created by any sentient being without appeal to some sort of higher power. If this is a prelude to the Intelligent Design argument, then know that formally stated, the Intelligent Design arguments begs the question, i.e., it uses circular logic. It can be stated as follows:

• P1: Things appear to have been created.
• P2: Things that appear to be created must be created.
• C: Therefore things must be created.

In addition to it's other short-comings, H2 simply uses bad logic :)

EDIT

Towards a "non-artifical" definition of science: in Objection 3 you hint at a Popperian definition of science with which you are understandably unsatisfied. Popper is flouted by many fierce defenders of science, ironically, as Popper is essentially a skeptic of scientific knowledge. Popper's argument is loosely that scientific knowledge is characterized by falsifiability. However, with such being a necessary condition for knowledge, there really can be no knowledge since it necessarily can be disproven at any time.

This is a fine line of thought, if you set the highest possible bar for knowledge, viz. that it must be reasoned out deductively. But this of course is silly! Science is based in empirical observation and thus must rely on inference.

However, we can still be comfortable asserting certain scientific facts, as they extremely well supported by mounds of evidence over long periods of time. I personally wouldn't get on a plane if it weren't for well laid-out science, engineering, and a superb track record demonstrating that that plane is, in fact, going to stay in the air. This is not a house of cards, but rather a web of theory, practice, and observation woven together to solidify this scientific reasoning.

To make a long edit short: No, science is not a deductive enterprise, and this will always leave room for skeptics, but this does not change the fact that science is capable of explaining things extremely well (if it's good science, that is).

Aside

If you are interested learning more about the philosophy of science, and when and how it can be defended, and if you don't buy into the sloppy definition that "science must be falsifiable", then I recommend this book: http://www.amazon.com/Defending-Science-Between-Scientism-Cynicism/dp/1591024587

Answer 13

The way I understand scientific reasoning, you do not start with vague, ambiguous hypotheses like H2 (and H1, the way you present it.) You start with observations and measurements. Precise observations that can be quantified and expressed mathematically (at least that's the way its done in physiscs). Based on these observations you perform induction to formulate a precise hypothesis that makes predictions which can be checked and tested by further observations.

For example, you might notice that all rocks you drop seem to fall equally fast, i.e. with the same accelaration. You then perform experiments where you drop a lot of rocks and use the best measurement tools you have to check their accelaration. You might then try a hypothesis like "all rocks fall with that accelaration a." Of course you have not checked all rocks everywhere and everytime, thats clearly impossible, but there are no absolute and final truths in science - there are just theories that have so far have passed every test. Back to the example. You then drop some other objects, like metal balls, and notice they fall with the same accelaration as the rocks. So you change your hypothesis to "not just rocks, all objects fall with the same accelaration a." But then you drop a feather or a piece of paper and they do not. But you get the idea that might be caused by air resistance, and drop them in vacuum. Your hunch proves correct, in vacuum they do also fall with the same accelaration. So you change your hypothesis to "all objects IN VACUUM fall with the same accelaration a".

Then some colleague who has heard of your hypothesis performs similar experiments high in the mountains, using more sophisticated measuring tools than you have, and discovers that the acceleration there does not equal a, is it a little bit different, depending on the altitude. Even worse, astronauts report that in space objects do not fall at all, and on the surface of the moon they fall with an accelaration that is only one sixth of the one at earth. So now things have gotten very complicated. You might give up, but by being smart and persistent you might get to formulate a hypothesis that matches the experiments everywhere, and that would be Newton's equation of gravity. (Assuming for the example of course that you lived before Newton's time. Yeah I know then there wouldn't be astronauts on the moon.)

To summarise, you start with simple assumptions that makes precise predictions which can be checked and falsified by precise observations. You keep your assumptions as simple as possible (Ockham's razor), but only as long as they agree with observations; if not, you have to change them. Similarly, with Ockham's razor you keep it simple by generalizing ("The law of gravity is the same everywhere in the universe"), but you have to remember we obviously cannot measure everywhere in the universe, so what that really means is "We think the law of gravity is the same everywhere in the universe, because in those places where we made measurements it has always proven correct so far." And you have not only to repeat your experiments at other places, but also with more precise measurement tools, if those become available to you. That would lead you to discover that Newton's law of gravity is not 100 percent correct after all, its just an approximation of Einstein's General Relativity.

To address the question of your two hypotheses, as others have pointed out and you yourself suspected, in H2 the term "created" is left so vague and ambiguous that H2 is useless from the scientific point of view: It does not make precise predictions, cannot be falsified by experiments. Unless you really define "created" as "created by visible creatures like humans", and then H2 could be dismissed at once, because obviously plants can grow without humans or other observable beings interfering.

Maybe a creationist would point out that plants grow from seeds, which although they do not look like it have a just as complicated structure as the grown plant. What he has in mind is of course the famous watchmaker analogy, "if I find a watch in the desert, I know it was created by someone. So I conclude that all complicated things must have been created by someone even more complicated."

The way I see it, there are two objections which make that analogy useless:

A) It is assumed without saying it that the person in the desert knows a lot about how objects created by men (like watches) look like, and also how objects not created by men (like plants) look like, and of the typical difference. So what is really said is more like "No one I ever heard of has witnessed complicated metal objects like watches to come into existence without a observable creator creating them. Sure, for other complicated objects like plants it looks rather different, but lets not dwell to closely on that." He could argue "seeds are just as complicated as grown plants, so the complicated things just change state but do not get more complicated." But that seems rather questionable reasoning to me, especially if the plants multiply.

B) If all complicated objects must have been created by a creator, who created the creator? A creatonist would probably say something like "god was not created, he is eternal." But that is just a sleight of hand to distract from the fact that he tries to throw away the watchmaker analogy the moment it turns against him. Because we tend to be emotionally satisfied by the idea of a eternal god, we do not look too closely for that sleight of hand. If all complicated things seem to need a creator, I see four possible scenarios:

B1) The line of creators stops somewhere with a first creator which existed for ever.

B2) The line of creators stops somewhere with a first creator which just popped into existence in his finished state.

B3) There is an endless line of creators. Time had no beginning, it goes back like that to infinity.

B4) At one point, things popped into existence without a creator.

With the watchmaker analogy you cannot choose between these scenarios (or if you applied it very strict, it would force B3), so it is useless as an argument for the existence of god. What's more, of course we have a very plausible explanation of a way in which complicated things can come into existence out of less complicated things: Darwin's evolution.

Answer 14

If the gravity hypothesis and one sets up a Cavendish experiment then one will observe a (rotational) displacement of the masses.

If the created hypothesis is true then ... what?

I don't see how the created hypothesis, without additional explication about the term "created", leads to predictions/descriptions of any observations in any realizable way. This lack of predictive power, which is a consequence of the manner that you've responded to objections 2 and 3, is a major difference between the two hypotheses. Some might term this as a lack of falsifiability for the created hypothesis, but as I read the question I see it as an even more basic aspect of not being concrete enough to constitute a valid scientific hypothesis.

The idea that scientific hypothesis need to make specific observable statements about features of the world is an important part of the scientific method; although in itself, is (probably) not a complete description.

Answer 15

Your question includes several mistakes. The first mistake is that you talk about degrees of belief. Either an idea is right or it is wrong, your belief or lack thereof in an idea is irrelevant.

The second mistake is that you have proposed two false ideas neither of which bears any substantive resemblance to a worthwhile scientific theory. You write:

[T]here exists a force of attraction between every two masses.

This idea doesn't explain what constitutes a force, or how you would test for its existence or anything like that. What you write gets even worse:

So to summarize, every object in the universe was either tested for gravity and gave a positive result for the hypothesis H1 or is not known whether it satisfies the hypothesis H1 or not (because possibly it is not feasible to do that).

Really? There is lots of stuff we could test for this idea but hasn't been tested. There are a million dust motes in the room I'm sitting in right now, none of which have been tested although it might be technologically possible to do it if we tried. Nobody is doing it, not because it isn't feasible but because it is not relevant to anything important. Also your suggestion contradicts the prevailing theory of gravity: the general theory of relativity. GTR claims that gravitational attraction is not the result of a force at all: it is a result of curvature in spacetime. Non-inertial motion is required to resist this attraction, not to produce it.

Scientific knowledge is created by conjecture and criticism, not by measuring every object in the universe or anything like that. You spot a problem with your current ideas, propose solutions, criticise the proposed solutions until one is left and then look for a new problem. Experiments don't show any idea is true or probably true, they eliminate bad ideas. Quite aside from everything else, every experiment involves an interpretation of the results, an explanation of what is happening to produce the results that could easily be wrong. For example, everybody until Einstein interpreted gravity as a force of attraction because it pulls objects together but that was dead wrong and all of the observations that claimed to illustrate it were misinterpreted. There are more and deeper mistakes in your theory of knowledge, see "Realism and the Aim of Science" by Karl Popper, especially Chapter I.

You write:

One puts the hypothesis (Call it hypothesis H2) "every object was created". Some people obtain high belief in this statement by observing that many sophisticated objects were created by humans or perhaps sometimes other animals.

Their observations have no bearing on anything. First, the fact that they looked at a bunch of stuff just illustrates that they personally haven't refuted that idea, presumably because they're not trying very hard. I went outside and found a little pebble on the ground. That pebble is a result of geological forces acting on rocks over millions of years, nobody created it. See what happens when you look for problems instead of trying to entrench your ideas?

Second, the creationist hasn't explained how stuff gets created. He just said "somebody did it". I could solve a lot of scientific problems and get a lot of credit if finding stuff out was that easy. Why do multicellular organisms reproduce sexually? I did that. I made every large organism in the world reproduce sexually. Where's my Nobel Prize? And shouldn't you be thanking me for imposing sexual reproduction on you? No Nobel Prize? Why not? Oh right, because I haven't actually given an account of what happens in reality that results in the existence of sexual reproduction. I just made a wild assertion that doesn't solve any new problems and contradicts the only explanations anybody has ever given for how biological complexity is created: the theory of evolution.

Some steps you might want to take before discussing creationism again. First, learn epistemology, by which I mean critical rationalism. Second, learn about the theory of evolution. Stuff to read: "Realism and the Aim of Science", Chapter I, "Objective Knowledge: An Evolutionary Approach" Chapter 1 by Karl Popper, "The Selfish Gene" and "The Extended Phenotype" by Richard Dawkins, "The Fabric of Reality" and "The Beginning of Infinity" by David Deutsch.

Answer 16

Q1 - Yes, I do agree that theory H1 and H2 are both sort of correct - but not for the reasons you are thinking. Q2 - Theory H2 can be considered to be scientific, and your conclusions can be shown to be consistent with modern science but only if you accept much tigher definitions for "everything" and "created" as follows.

If you want to frame H2 in the creationist sense, that everything was created over a short period of time as suggested in the Bible, that can be done without breaking lots of well established physical rules if you simply remember that a 'day' was just an easily understood period of time until you had a Sun with an Earth to go around it. Now if you look more closely at Bishop Ussher's interpretation of Genisis says that the first 'day' was exactly 24 hours, which is how he arrived at the date of what he called 'creation' as about 6,000 years ago, there is no reason to think that his guess was any more accurate than the current estimate which says everything was created (in the big-bang) 13.8 billion years ago, and there is lots of evidence supporting the scientists. It is also means that you no longer have to ignore or excuse all the obvious things around you that would have taken more then 6,000 years to become the way they are.

Answer 17

The creationist hypothesis doesn't foster further questions as a scientific hypothesis would. On the contrary, it closes off any further questions. We might ask, for example, why are there so many similarities among living things? The creationist answer is that this is simply the way they were "created" by "The Creator." A scientific dead-end.

Answer 18

I wrote this answer to collect the ideas of the answer of Rex Kerr, the answer of Adam.r and the comments of Karolis Juodele in one place and to add some of my ideas to them. I think I came up with a good necessary condition that differentiates between situations 1,2.

Definition 1: The visual atomic words is the set of words: distance in x-direction, distance in y-direction, distance in z-direction in some coordinate system, colors, colors: red, yellow,..., ( I know that it is better to define colors using wavelengths, but please ignore that for the moment). The math atomic words are all mathematical terms (can be made precise if one knows that ZFC is the foundation of mathematics). I will call the union of the set of visual atomic words and the set of math atomic words the set of atomic words and denote it A.

One may use words from the set A without defining them, thus I call them the set of atomic words.

I will now define the terms "practical hypothesis" and "scientific hypothesis".

Definition 2: A practical hypothesis is a hypothesis that has all words defined using words from the set A. A theoretical hypothesis is a hypothesis that is not practical.

For example the hypothesis that the shortest path between any two points is a straight line is a practical hypothesis since all words it contains can be defined using words from set A. The hypothesis that the earth revolves around the sun is a practical hypothesis as well. "The hypothesis that objects denser than iron fall to the ground when let go" is a practical hypothesis as well (One may argue that there are implicit notions of mass and time in these hypotheses, but these can be defined again using the distance measurements of the equipment used to measure them).

Both of hypotheses 1,2 are theoretical hypotheses unless one gives a definition of the words "force", "create" using only words from the set A. (People may argue that force may be defined as mass*acceleration, but my opinion is that this does not take into account forces like normal contact force).

Necessary condition for a hypothesis to be scientific: It has to satisfy at least one of the following conditions:

1) It is a practical hypothesis

2) It is a theoretical hypothesis with good practical predictions (A term which I will define now).

Definition 3: A practical prediction is a prediction about the universe such that all words used to state the prediction are definable using words from the set A. A theoretical prediction is a prediction that is not a practical prediction.

Examples of practical predictions: Let's consider the scenario of a spring with a mass hanging from it. One can use newton's laws and hook's law which will be theoretical hypothesis) to find:

1) The maximum force (tension) that occurs in the spring (This will be a theoretical predictions since it is stated using words that are undefinable using words from set A)

2) The position of the tip of the spring as a function of time (This will be a practical prediction, since all words used in it can be defined using words from set A, time can be defined using words from set A as I explained earlier)

Theory of gravitation is a set of theoretical hypotheses that has good practical predictions. Hypothesis 2 is is a theoretical hypothesis that as far as I can see does not have practical predictions. It has theoretical predictions though, an example would be the prediction that mountains were created.

Remark: I am satisfied with these definitions because I believe a theory in physics that all of its predictions are theoretical and non practical (not visual) would be useless in describing the behaviour of the universe.