|
Register | Sign In |
|
QuickSearch
Thread ▼ Details |
|
Thread Info
|
|
|
Author | Topic: Induction and Science | |||||||||||||||||||||||
Modulous Member Posts: 7801 From: Manchester, UK Joined: |
According to Rudolph Carnap, the degree to which scientific statements such as Newton's laws are supported by inductive evidence is precisely zero. Do you have a reference that shows otherwise? quote: When I was talking about induction, I was talking about the above. Are you saying the above is wrong? Do you think the above doesn't happen? Would you be fine calling it schminduction and accepting that scientists do it?
quote: I'm just arguing that the more modern view that has emerged in which the statistical and the inductive have been merged.
|
|||||||||||||||||||||||
Modulous Member Posts: 7801 From: Manchester, UK Joined: |
Aren't calculations the main difference between inductive logic and statistical reasoning?
No - I don't think so: Stastical reasoning can be used to make inductive inferrences and can be used to justify the confidence levels of our inductions. See the article I linke to in Message 76:
quote:
|
|||||||||||||||||||||||
Modulous Member Posts: 7801 From: Manchester, UK Joined: |
In Message 200 you seemed to imply that Newton's third law was a case of induction. I'm challenging you to provide actual evidence to support that. My evidence is that there is evidence to support Newton's laws. In science, when we check our laws and theories against the evidence we grow confident that the laws and theories are at least somewhat true. As the evidence builds, so too does our confidence. Despite the fact that we don't actually know deductively that Newton's laws are a true description of the areas of motion and impulse they are treated as tentatively true. It is as if we make a less-than-certain inference that they are true, with enough confidence to justify strapping some people into a rocket fueled machine and shooting them at the moon.
However, statistical inference is largely deductive, and it gives confidence intervals. Scientific laws such as Newton's do not give confidence intervals, so cannot have resulted from that kind of statistical deduction. They are also far more precise than anything that you would ever get from statistical inference. But our evidence that Newton's laws describe motion is acquired with confidence intervals, and the inference that Newton's laws are true enough is uncertain and has potentially calculable confidence intervals. As the article I raised earlier puts it:
quote:
|
|||||||||||||||||||||||
Modulous Member Posts: 7801 From: Manchester, UK Joined: |
That's pretty weak. Creationists are often making similar assertions about evidence supporting their Adam and Eve beliefs and their flood beliefs. And you don't let them get away with it. Not really. I didn't think the claim would be disputed.
However, I'll weaken my request to make it easier. Can you provide evidence, or citations of evidence, explicitly supporting Newton's third law? Sure - I offer 'rockets have worked as we have expected them to when we assume it is approximately true'.
That's not induction. That's pragmatism. You are accepting what works. I agree that science is very pragmatic. What I am questioning is the use of induction, not the use of pragmatism. Induction is pragmatic. It allows novice birdwatchers to discount black and brown birds if they are looking for a swan. Pragmatism is a possible justification for using induction, but I wasn't doing that. I was talking about growing confidence in a hypothesis based on increasing supporting evidence. This is induction in science, and it is a pragmatic decision to allow it.
|
|||||||||||||||||||||||
Modulous Member Posts: 7801 From: Manchester, UK Joined: |
There's nothing there that explicitly supports Newton's third law. I've no idea what your qualifier 'explicitly' means in this context and how it differs from plain old scientific support.
Maybe, though I think that's a stretch. Induction can often lead to wrong conclusions, and that does not seem very pragmatic to me. That's just wordplay. We use it, not because it is perfect but because it works often enough to justify using it. That's pragmatics. And it is far more pragmatic to be occasionally wrong (as long as you admit it ahead of time (principle of fallibilism)), than to not be able to make any general scientific statements based on limited evidence. It also ignores the methods used by science to minimise the frequency of erroneous inductions: statistical reasoning. (Again, a very pragmatic thing to do)
However, induction is usually described as a method for producing general statements from a collection of specific observations. see Newton's laws as a package of procedures used to make predictions. The package is tested as a whole, and accepted based on pragmatic principles (how well does it work for making predictions). There's nothing in that pragmatic acceptance that has to do with forming general statements on the basis of specific observations. Fine - but to deny that they are general statements would be foolish. And in science - we only have specific evidences to support those general statements. If we have 1,000 predictions that have gone positively, and 0 that have gone negatively, we have a 100% success record so far. It is an induction if we say that we have discovered a law or theory that applies to 100% of cases when in fact it only applies to 100% of tested cases - where the tested cases are fewer than the actual possible cases (as is mostly the case, in practice). As you said, we could be wrong. Just like we might stumble upon a black swan - It could turn out that in some small number of cases of macro objects in low energy conditions things work completely differently and our next rocket will accidentally escape the Moon-Earth gravitational influence. This possibility leads us to falsification. Indeed - every test from tomorrow onwards could completely contradict the relationships Newton described, or at least show them to only cover a small subset of cases. While it is something we can have more confidence in:
quote: is unfortunately useless as far as science is concerned. Maybe this is what you mean by 'explicit'? Maybe you'd say the evidence explicitly supports this law and implicitly supports Newtons law as he worded it.
And, secondly, nothing could be more obvious than that Newton's third law is not any kind of generalization from specific observations. But we can accept that it is a generalisation that science justifies by supporting it with a smaller set of specific cases. Edited by Modulous, : No reason given.
|
|||||||||||||||||||||||
Modulous Member Posts: 7801 From: Manchester, UK Joined: |
Ah, sigh. That sounds like a good old apologetics argument to me. I would prefer actual evidence. If you'd like evidence of what pragmatism means I suppose you could look it up. I know you have the capacity to do that, so I didn't want to patronise you too much.
I have never denied that Newton's laws were general statements. But that they are general statements does not demonstrate that they were arrived at by means of induction from specific observations. Fantastic - I'm not claiming that because they are general therefore they were arrived at by means of induction. Newton could have seen them in a dream for all I care for the point I'm making.
Oh, bullshit. How many specific evidences to you need to support the general statement "There are 100 centimetres in a metre"? That's not an empirical claim, its tautologous, so is not relevant to a discussion on science.
But we can accept that it is a generalisation that science justifies by supporting it with a smaller set of specific cases.
So are we supposed to accept that on faith, without any actual evidence? Unless you claim that scientists take the third law of motion on faith, I don't think I need to. But the methodology of science strongly suggests that before a law can be accepted, it must have evidence of being true.
|
|||||||||||||||||||||||
Modulous Member Posts: 7801 From: Manchester, UK Joined: |
I'm calling foul on that. The particular discussion was on inductive support for Newton's third law. You are twisting it to pretend that it was about pragmatism. The quoted section which you called apologetics and demanded evidence was about induction and pragmatics.
That's not induction. That's pragmatism. You are accepting what works. I agree that science is very pragmatic. What I am questioning is the use of induction, not the use of pragmatism.
Induction is pragmatic.
Maybe, though I think that's a stretch. Induction can often lead to wrong conclusions, and that does not seem very pragmatic to me. That's just wordplay. We use it, not because it is perfect but because it works often enough to justify using it. That's pragmatics. And it is far more pragmatic to be occasionally wrong (as long as you admit it ahead of time (principle of fallibilism)), than to not be able to make any general scientific statements based on limited evidence. It also ignores the methods used by science to minimise the frequency of erroneous inductions: statistical reasoning. (Again, a very pragmatic thing to do)
Ah, sigh. That sounds like a god old apologetics argument to me. I would prefer actual evidence. Next time you ask for evidence - try specifying what it is you are asking evidence for, it'll help avoid these issues.
You still don't get it, do you. Newton's third law is a convention, an empirical principle. It is prior to any evidence. Apparently I do get it. As I said, Newton could have derived them from a dream, or indeed a bag of scrambled letters and numbers. The proof of the pudding is in the eating, and that's where the science gets done. In science as it is actually done we make use of evidence. Whether the evidence leads to the formulation of the law, or the law is supported by the evidence afterwards does not matter. It is the evidence comparison to the law that involves the inductive leap whichever order that comparison happens to occur in. Deriving a law can be done empirically in science - whether or not you believe Newton's laws were not.
|
|||||||||||||||||||||||
Modulous Member Posts: 7801 From: Manchester, UK Joined: |
Which of those forces is being measured by the scale? How do you independently measure the other one so as to get an actual observation that could be used in the alleged induction? Newton used pendulums, rather than scales:
quote:{From Principia} As for the source of his Laws - Newton cites Galileo's observations of projectiles and pendulums as well as Christopher Wren, Dr Wallis and Mr Huygens. As Newton said at the start of the section, the partial quote above comes from, "Sir Christopher Wren discovered the truth of the thing before the Royal Society by the experiment of pendulums". Discovered the truth, by experiment? Sounds like induction being employed in science as its done, no? Edited by Modulous, : deleted "From what I recall" since I looked it up.
|
|||||||||||||||||||||||
Modulous Member Posts: 7801 From: Manchester, UK Joined: |
However, it was still really about concepts and not about facts. Most of what we take to be forces today, would not have been considered forces before Newton's time. Newton had to persuade people to use his concept of what to consider a force, and his laws are all about defining that concept. I propose that part of his persuasion involved pointing at the experiments that supported his position and asking people to make the inductive leap from something like 'the pendulums at my house over the past few years' to something like 'every object past, present and future.' This is what people are saying when they say science uses induction. They are talking about actual science as carried out by actual scientists. They aren't talking about some philosophical idealistic concepts that are nothing to do with facts. Generally they talk about taking the results of some experiments (looking at swans for example) and coming up with a law (all swans are white) and then testing that (by examining more swans, swan feathers, swan photos, swan videos all over the world) to increase confidence in the law. And that's where the induction is. You can, if you like, try to tell us that Newton wasn't talking about facts and was just engaged in some philosophical conceptual idealism. You can try and say his laws are just a self-consistent framework of concepts that he was defining (rather than describing (limited) observed facts mathematically) - but if that's the case, it wasn't science so is off topic.
|
|||||||||||||||||||||||
Modulous Member Posts: 7801 From: Manchester, UK Joined: |
quote: quote: What was really about concepts and not about facts? Why is 'it' relevant to a discussion about Newton and science and induction (which is very much about facts)? You initially stated that
quote: I said Newton did the measuring using pendulums. You retorted "it" was really about concepts not facts and then 1 post later tell me that Newton was all about facts. I think my 'gross misunderstanding' of such a confusing thread of discussion is entirely warranted.
However, we very much disagree on how science works. We very much disagree on "fact". At this stage, we should probably just agree to disagree. We'll have to if you insist on remaining so mysterious. Could you for example, explain what evidence is for in science and how it isn't employed with induction in the same way that 'all swans are white' is (based on the evidence that all known swans are white)?
|
|||||||||||||||||||||||
Modulous Member Posts: 7801 From: Manchester, UK Joined: |
Facts (as representations) do not exist independent of humans. If science worked by just picking up facts that existed independent of us, it would not work. I still have no idea what 'it' was that was not about facts but about concepts or why it is relevant. Or why you decided to start talking about 'facts' and 'concepts' in the first place. I just pointed out that Newton used pendulums as evidential support for his laws - could you explain what your point here is? Assuming it isn't really relevant (just to type something else as a response), do you accept that Newton's Laws have empirical support? That this support relies on a sequence of observations that are in accord with the laws? Do you agree that to suggest that "All the swans I've ever seen are white" supports (with some degree of uncertainty) "All swans are white"? Do you agree that 'all the pendulum experiments Newton had ever seen" supports (with some degree of uncertainty) "Every reaction has an equal and opposite reaction"'?
Whether facts can be said to exist independent of humans, depends on whether you are a Platonist or a nominalist. Great. But we're talking about using pendulums. What relevance does the nature of facts or concepts have to do with anything?
|
|||||||||||||||||||||||
Modulous Member Posts: 7801 From: Manchester, UK Joined: |
In that text, he appears to be only talking about something broad and imprecise, not about his precise mathematical laws. I had been talking about that 'broad and imprecise' thing that Newton called 'experimental philosophy' throughout this thread - we call it science today. Have you been talking about mathematics all this time? That would explain a lot.
|
|||||||||||||||||||||||
Modulous Member Posts: 7801 From: Manchester, UK Joined: |
I'm going to assume that what you just said made sense. We can worry about the rough edges later.
What I have been consistently saying in our discussion of Newton's laws, is how they are being used as a reconfiguration of level 2. Somehow, the people debating with me have been completely oblivious to that, probably because they are strongly committed to the conventional wisdom. We're not being completely oblivious nwr, you are just being a patronising asshole that believes we're being oblivious and translating what we are saying to help preserve the myth that we're being oblivious.. Actually, we were just using normal English. Translating into nwr philosophical bibble babble: Newton 'reconfigures' level 2 (observes stuff and has neural impulses about those observations), then moves to level 3 to put that reconfiguration to use. That use is to make universal statements about level 1.
definition, induction is something done exclusively at level 3. So if science is acting at level 2, then what it does there cannot be inductive. Scientists actually use the data, so they must use level 3, and since the entire point of science is "to learn more about level 1.", I think we've established your deduction that science cannot be inductive is invalid. If you are only doing stuff at level 2, you aren't doing science. Edited by Modulous, : No reason given.
|
|||||||||||||||||||||||
Modulous Member Posts: 7801 From: Manchester, UK Joined: |
You seem to have missed the word "there" in what I wrote. I have emboldened it this time, to make it easier for you to find. I missed nothing, but nice attempt to avoid the issue again.
So if science is acting at level 2, then what it does there cannot be inductive. Not.. "When science" but 'if science'. Would you like to deal with my argument? You know the part where science is never just operating at level 2. That to be called science you need to include Levels 1 and 3. And it is at this point induction comes in. You can keep pointing at parts of science where you think induction doesn't take place - but other than as a case of confirmation bias it isn't really useful; the rest of us are talking about the part where it does and you are steadfastly refusing to address it. Let's look at your comments again about level 2:
quote: The conventional wisdom is that the important bits are treating the three levels in a certain way. Do it one way, it's science. Do it another, it's not science.
quote: We don't disagree. But, once again by focussing on level 2, by your own admission you are necessarily ignoring the parts where induction might come in. Let's focus on 3 - which is the topic of the debate, yes? Have you wondered why we've been giving you examples of USING the data? Like using the data of observed swans to inductively conclude that all swans are white.
quote: But your clause 'if science is acting at level 2', is false. Science is not acting at level 2 - science is a process that utilises all three levels, only doing 1 of them and we're not talking about science. Your dogged insistence of evading this concept is quite startling - but isn't helping you beyond rhetoric. And we're talking about the stuff that connects level 2 to level 1 by way of level 3...where induction comes into play. And you keep insisting we should talk about level 2 and not about 1 and 3. Let's talk about level 1 and 3.
quote: They maybe a 'reconfiguration of level 2' (incidentally aren't you the guy that said "It is how philosophers claim that science works. But most philosophers do not actually do science, so have no basis for making such assertions."? Newton was a philosopher that did science and he disagrees with you - will you not take the hint?) but when Newton went beyond that reconfiguration and said that this his reconfiguration is a universal principle that applies to level 1 - was he not applying level 3 thinking to the level 2 stuff to draw a universal conclusion about level 1?
|
|||||||||||||||||||||||
Modulous Member Posts: 7801 From: Manchester, UK Joined: |
What's to deal with? I never claimed that science is only operating at level 2. You're still avoiding it, the gist of my post was about the importance of level 3 in science and yet you still focus on the bits about level 2. I am perfectly comfortable that you think science operates outside of level 2 - but when I try and discuss those parts of science, the parts where induction is used - you avoid it and focus on something about level 2 - like you just did. So let's talk about how science uses levels 1 and 3, I've been doing it for some time now but you keep repeating over and over again that at level 2 there is no induction. This is true by definition, so let's move on from that, neh?
Can't you tell that the alleged "swans" induction is a made up story? When was "all swans are white" ever a part of science? Since when did I claim it was part of science? I was using it as an example of induction. I tried talking about pendulum interactions but you carefully avoided that line of thought. If you're ready to address any of the issues I have raised about a small set of observed pendulum interactions being used to generate a mathematical relationship (level 2) which is inductively reasoned (level 3) to apply universally to novel circumstances in the realm of objective reality (level 1), until proven otherwise - let me know. You gave up last time I tried. Edited by Modulous, : No reason given.
|
|
|
Do Nothing Button
Copyright 2001-2023 by EvC Forum, All Rights Reserved
Version 4.2
Innovative software from Qwixotic © 2024