Scientific Assumptions

[spidergoat]

Are there unproven or unprovable assumptions on which all of science is based? Specifically the principles of causality and uniformity (of natural laws)?

[arauca]

Are there unproven or unprovable assumptions on which all of science is based? Specifically the principles of causality and uniformity (of natural laws)?

Finally you got off from behind 8 ball.

[spidergoat]

Thanks guys, really helpful.

[Grumpy]

spidergoat

Science is based on two assumptions and two assumptions only.

1. That the Universe is as it appears to be.

2. That the human mind is capable of perceiving and understanding the Universe.

It is also put in these terms

1. The Universe exists.

2. Man is capable of understanding that which exists.

Or

1. The Universe is real(IE we do not live in a Matrix).

2. That man can understand reality.

Number one is the basic assumption(precept, if you like), Number 2 is not a given, it requires hard work and disciplined and objective testing at all times(the Scientific Method is the best system we have yet found for determining what is objectively true). Plato's navel gazing need not apply.

Grumpy

[kwhilborn]

One could argue that,
"All science must be built upon the back of existing science". Even though there is a periodic table, it could not be considered science without the discovery and acceptance of the atom.

This may seem logical and not worth noting, but I think it has validity as an assumption. There have been scientific events (like fire) that mankind has not been able to explain, and until the (fire) event can be described fully scientifically it is not valid science. Was fire accepted as science before it was described? This requires some deep thinking for the philosopher types here.

[Grumpy]

kwhilborn

There have been scientific events (like fire) that mankind has not been able to explain, and until the (fire) event can be described fully scientifically it is not valid science

Fire is simply any sufficiently exothermic chemical reaction. Fire was probably one of the first things understood in a scientific way over a million years ago(and maybe earlier if some recently found evidence proves to be valid). Capturing fire and finding out what burned and what didn't was one of the first scientific studies. Determining the best methods of keeping fire going 24/7 was probably the second. Finding ways of creating fire when you have none was probably third. Using fire to improve food and harden spear tips was probably fourth and so on. Study of fire may have been the birth of true science.

Grumpy

[Rav]

I think the fundamental assumptions are that there is indeed an external world, and that we can determine at least some of it's properties through rational principles of empirical investigation.

[kwhilborn]

Yes. I agree fire was near the beginning of science, but the point I was trying to make was different. I probably explained it badly then.
I am trying to say all new science must now be explainable by old science or it cannot be accepted. I am sure there is room to create a new scientific assumption based on this.

I was trying to avoid bringing pseudoscience into this thread, but I will purely as an example, as I cannot think of many current observations about our surroundings that may apply. We even have theories about the furthest reaches of our universe now so I am pressed to come up with an example.

Pretend that something like telepathy was real for the sake of discussion (I am not advocating such in this thread). Let us pretend that everyone could easily send thoughts from one to another in a way that certainly is not possible now. Pretend we can talk without words. Now we have a (pretend) phenomenon that cannot be explained by science yet everyone can do it. Would the telepathy be scientific now, or would it need to wait until an accepted theory could explain it. I hope this makes my question and assumption clearer.

There are probably better examples (I'm also trying to avoid LENR here), but I am not versed in any phenomenon that science cannot yet explain.

[SciWriter]

An external world exists, for we have senses to take it in, the senses being spy outposts that directly contact external reality. That we have an understanding of reality is shown since we can manipulate it out there using stuff out there to make things that work out there.

[Dinosaur]

Logic itself is based on unproven axioms accepted as true. It also includes some undefined primitive terms, required to avoid meanless circular definitions. In plane geometry, line & point are (I think) undefined primitives.

Einstein in describing his fundamental definition of time used before & after as undefined primitives.

BTW: There have been many Threads relating to time. Einstein's very simple basic definition did the job very well, while almost all the posts were basicaly verbal diarehia.

[Yazata]

Are there unproven or unprovable assumptions on which all of science is based? Specifically the principles of causality and uniformity (of natural laws)?

There's the problem of induction -- How does a finite set of observations of particular cases justify universal conclusions about ALL X's (even unobserved ones)?

Deduction isn't without its problems either. How can we demonstrate its validity without circularity, without assuming the very principles that we are questioning?

What are natural laws? Science typically assumes that general principles exist in reality, but it isn't exactly clear how to recognize a natural law when we encounter it, let alone how to best characterize and explain it.

And what's up with probabilities? Do probabilities exist in reality itself, or do they only apply to our imperfect knowledge of absolutely determined events?

There are problems regarding knowledge of unobservables -- scientists and philosophers are divided on this issue into realist and instrumentalist camps. Examples of unobservables might be physical fields of various sorts, or sub-atomic particles. Lots of discussion about the foundations of quantum mechanics revolves around this issue.

Causality has been a major topic of discussion since David Hume criticized the concept. If all we can observe is constant conjunction between cause and effect, even if we shrug off the problem of induction, we still have problems accounting for what the connection between cause and effect really is, ontologically speaking. In some cases it isn't even clear how cause and effect should be distinguished one from the other. If all that connects cause and effect are observed constant conjunctions, then what is it that distinguishes scientific laws from accidental correlations?

There's scientific reductionism -- We typically assume that sciences like biology, chemistry, geology or astronomy all reduce to the principles of physics without any remainder. It isn't entirely clear how we know that or whether it's always true. Arguments about emergentism arise here.

And obviously our scientific knowledge depends on the broader principles of epistemology in general. How do we know not only scientific truths, but anything at all? The ancient problems of skepticism arise here -- how can we be sure that we aren't dreaming? How can we be sure that what we perceive isn't just our own private perceptions and there isn't any external world outside our imagination? How can we banish the possibility of solipsism?

And who are "we" in all this? What account should we give of minds and knowing subjects in science?

[SciWriter]

And what's up with probabilities? Do probabilities exist in reality itself, or do they only apply to our imperfect knowledge of absolutely determined events?

Since we cannot predict from 'chaos', which doesn't imply that the happenings are random or even that 'random is possible, the probability things is a 'prediction' of the possible happenings overall, during some duration.

Causality has been a major topic of discussion since David Hume criticized the concept. If all we can observe is constant conjunction between cause and effect, even if we shrug off the problem of induction, we still have problems accounting for what the connection between cause and effect really is, ontologically speaking. In some cases it isn't even clear how cause and effect should be distinguished one from the other. If all that connects cause and effect are observed constant conjunctions, then what is it that distinguishes scientific laws from accidental correlations?

We make what seem to be useful boundaries on the flow of things to note 'cause and effect', but there is only the one initial, circumstantial, cause ever continuing to unfold as the one big event.[/QUOTE]

[Yazata]

One could argue that,
"All science must be built upon the back of existing science". Even though there is a periodic table, it could not be considered science without the discovery and acceptance of the atom.

That's very true.

Individual scientists don't all start at the beginning and reinvent all of science from the ground up. Scientists exist and do their work in the context of a broader scientific culture, in the historical context of a huge set of more-or-less established results, problems and procedures. Scientists don't necessarily accept everything they are taught as indubitable and inerrant religious-style Truth, but they do assume that a critical mass of existing science is true enough that it can form a basis for their own work.

[C C]

Are there unproven or unprovable assumptions on which all of science is based? Specifically the principles of causality and uniformity (of natural laws)?

Judea Pearl and his associates launched a contemporary defense of causality, apparently inspired by either need for causal principles arising in robotics and AI or their falling out of such work: "I now wish to summarize briefly the central message of this lecture. It is true that testing for cause and effect is difficult. Discovering causes of effects is even more difficult. But causality is not MYSTICAL OR METAPHYSICAL. It can be understood in terms of simple processes, and it can be expressed in a friendly mathematical language, ready for computer analysis."

He gives an overview of the problems which causality has suffered in philosophy and statistics, here, which leads into introduction of the new approach:

http://singapore.cs.ucla.edu/LECTURE/lecture_sec1b.htm

http://singapore.cs.ucla.edu/LECTURE/lecture_sec2.htm

http://singapore.cs.ucla.edu/LECTURE/lecture_sec3.htm


The quandaries of natural laws are examined in this article:

LAWS OF NATURE, SOURCE UNKNOWN
http://www.nytimes.com/2007/12/18/science/18law.html

"Are they merely fancy bookkeeping, a way of organizing facts about the world? Do they govern nature or just describe it? And does it matter that we don’t know and that most scientists don’t seem to know or care where they come from?"

[Dinosaur]

Yazata: Quantum Theory indicates that there are many truly random processes.

And what's up with probabilities? Do probabilities exist in reality itself, or do they only apply to our imperfect knowledge of absolutely determined events?

The data resulting from most (all?) quantum level processes indicates that probabilistic rather than classical causal laws apply. This & the Uncertainty Principle strongly negate causality.

The classical physics POV was that probability was used due to lack of knowledge of deterministic laws or due to it being impractical to apply deterministic laws. Modern physicistis do not hold this POV.

BTW: Many people think that the Uncertainty Principle relates to practical limitations of measurment technology. It really is a claim that quantum level entities cannot have both a precise momentum & a precise location at the same time. Bose-Einstein condensates provides experimental confirmation of this claim.