while i'm not a scientist, it seems to me that the word that scientists use to describe their ultimate accomplishment is inherintly flawed. they must be as tired of the misinterpretation of "theory" by the uninformed or uneducated as we are. given that they spend innordinate amounts of time correcting the layperson who equates scientific theory to guesswork, why don't they come up with a new word for theory?
in Dawkins' the God Delusion, he added an intro to the paperback version that discussed this. his minor modification was to call it a "theorum", much like the usage in mathematics. i don't feel like that is good enough.
at the same time, i don't have a better word. i'm open to ideas though...
Engineering a change in well-established usage is not an easy thing to do. There is a great deal of history of science in which the word theory is used to mean an intellectual construct uniting relevant facts and ideas. As Poincaré pointed out a collection of facts is no more science than a pile of stones is a house. Facts must be fit together in a meaningful way to form a theory and that is the fundamental structure of scientific thought.
Maybe we should amend it to simply be a "Proven Theory", or maybe "tested theory"?
It just seems odd that all theories, whether they are starting out or approved, carry the same name.
Who ever came up with this, !!!!!!
It goes back a long way— to classical Greek to a word that means "something seen, a vision, a concept."
Hence in the geometry of Euclid the statement of a proposition is labeled theorem and is followed by the proof. The same usage is retained in mathematics papers and textbooks today.
(That much I knew as a mathematician, the rest of this I had to recover from various sources.)
The Latin cognate theoria is not classical, but does occur in the Septuagint in Ezekiel 8:3. as the translation by Jerome of the Hebrew word cemel and is rendered in English translations from KJV onward as image.
In English the word theory has been used in both its present senses for over 400 years according to the OED. The OED even gives a quote "Theory of what is and theory of what ought to be are perpetually compounded" which suggests that the two senses have been a problem for a while.
With such a long history and broad usage in diverse contexts, change by fiat is out of the question, I suppose.
Great response man! I enjoyed very much.
Man, what a debacle. Seems we are trapped into this and will have to repeat ourselves like parrots. Not something I'm looking forward to!
You see the origin of the dual sense of the word theory already in Euclid's use: a theorem was the statement of the idea in a geometric proposition, but it was also something requiring proof.
Just reading the latest information regarding discovery of the Higgs Boson particle. It occurred to me there might be a tie in for this thread.
Perhaps what you're looking for already exists. I'm going to go out on a limb here and show my complete ignorance of all things scientific so be nice in your responses.
What you're seeking to do is use the scientific term "Law" for a proven theory. This puts us in a box because wrinkles are continuously discovered that require rewrites to what we know. We currently consider those modifications to existing theories or replace them with completely new theories.
The only way you're going to get around the complete misuse of the word "Law", is to quantify the confidence level in any given theory. The only thing I know that would do that would be along the lines of the Sigma system. I see it being used to describe the confidence level of the Higgs finding but I'm sure it would be over my head as far as how they come up with each sigma level.
It would seem odd to me to see a theory of evolution described as a 5.9 Sigma theory of evolution. No doubt, anything less than a 6 would leave just enough room for a theist to say "so it's only 50/50 that this theory is true. Even your own scientist aren't sure".
I imagine there are diminishing returns to application of a system such as this to most theories.
I think the whole attempt to reclassify the word we use to describe theories is a fools errand. We're just kowtowing to those who have their own mythical explanations. No logic or degree of proof is going to move them from that position.
Yeah, I agree, we can't use anything that would leave a door open for argument, when there isn't any in i.e. evolution. No matter what details change, there is ZERO chance of evolution it's self been thrown out.
Which brings us back to the original point.
A proven theory's core cannot change, only details of it. Someone already mentioned that a theory holds all the information including the original hypothesis, the laws and changes for ever.
The confusion part is that folks don't know that. So we are still in the same boat, we don't have a way to differentiate between an original unproven or tested theory against one that is proven like evolution.
Until we have, they will continue to throw at us the old "that's a theory" and we'll just continue to repeat ourselves over and over.
I guess that's life.
Sigma is just the name for the standard deviation in a normal probability distribution. A normal distribution is completely determined by its mean, usually denoted by the Greek letter μ, and its standard deviation σ. The mean shows that height in the middle and the standard deviation shows how "fat" it is. The typical plot is shown below:
If you measure something and the measurement errors are random, they follow a normal distribution. 68.2% will fall within one standard deviation of the average of all, 95.45% will fall within two standard deviations, and 99.73% will fall within 3 standard deviations. (This is the usual 6σ standard talked about.)
The term is also used for the confidence level attached to the results of experiments to express the probability that the results are correct. In this case of the Higgs boson, it seems likely to me they are speaking of a confidence level within 5σ of the mean, which in the popular terminology used in industry would be a 10σ standard.
Wow...I hope you didn't go through much trouble to draw that chart ;-)
Thank you for the explanation of that. I'm sure several of us will find it useful.
If I remember correctly, they were talking about several decimal places (i.e. 99.9999%) probability of confidence with respect to the Higgs. I guess they only see one every trillion collisions so that must play into why they're using sigma.
Thrilling to think it is true but I won't be spending any sleepless nights if the lights instantaneously go out on the universe several billion years from now. At least it will keep those physicists off the streets and away from a life of crime.
What the physicists want to estimate is the probability that their events are not just random errors.
Correction: the sentence
The mean shows that height in the middle and the standard deviation shows how "fat" it is.
The mean shows where the height in the middle occurs and the standard deviation shows how "fat" it is.
I shouldn't type while I am on the phone.
A couple of additions. First of all, scientists usually refer to statements that are not yet established by good experimental results as conjectures. A conjecture is an educated guess.
Second, all science, including well-established theories, are held tentatively. It is inductive inference that is involved and since you cannot test exhaustively under all conditions, times, and places, the generalization encapsulated in a law is never quite absolute. However, some things are so well tested that assuming they hold for future conditions is a very good bet.
An excellent example of this is given by Newtonian mechanics, which were superseded by relativistic mechanics. Newton had no way of testing his mechanics at very high speeds and no reason not to think that what he observed and used would extend to them. Only when relativity was developed did it become clear that a modification needed to be made. Newton's mechanics are still an excellent approximation, so close that at all ordinary speeds they suffice.
In a sense then, all scientific conclusions are tentative, but most are so well tested that it's foolish to bet against them except under extreme circumstances. In other words, the conclusions should be accepted as factual.