| From: Steve T | 14/06/00
8:48:57 |
| Subject: Scientific Method | post id:
83748 |
| Just reading through the "not the
FAQ's" list, Science and the Scientific Method : Is the scientific method
valid? I felt I was making good progress until I came across the Black crows analogy (I only see black crows therefore all crows must be black etc.) At this point my eyes rolled back and smoke started pouring from my ears but just before I lost consciousness a thought occured to me which I will now attempt to "convert to words". "Science is essentially a method of observation and classification (crude over-simplification).As such, the scientific method, when considered in purely logical terms, is limited by the language it uses (the Queen's(?) English). Thus, a black crow will always be different from a white crow by its very definition. The only way to get around this is to attrribute a seperate and distinct name to every observable object, organism and phenomenon in existence (without automatically relating one with the other). A black crow would then become just a blackcrow and anything that resembled a crow and indeed belonged to the same species (another classification) but was white in color would be a whitecrow (or any name, not necessarily including the words white or crow). This would necessarily lead to the production of very large dictionaries but might also lead to considerably shorter scientific debates." Any thoughts? | |
| From: Kothos | 14/06/00
9:09:10 |
| Subject: re: Scientific Method | post id:
83753 |
Wouldn't work. There would still be a distinction to be made between very black crows and only mostly black crows, and off-white crows and bright white crows and white crows with a freak black streak etc... The logical conclusion to your idea is the re-classification of every object in the universe on its own, ie, every crow would need a seperate name. While this would be a valid way to do things, we have to use classification in order to simplify things. When I eat chicken I don't care which chicken it is. When I look for my lost dog I care very much which dog it is, and not just what colour/species/breed etc... When we conduct scientific observation, I suppose it's up to the judgement of the observer what level of classification is needed. Are there rules for that? | |
| From: recher | 14/06/00
9:09:32 |
| Subject: re: Scientific Method | post id:
83754 |
| to extrapolate that all crows
must be black since all you ever saw were black crows is not proper
scientific deduction. | |
| From: James Richmond (Avatar) | 14/06/00
9:24:30 |
| Subject: re: Scientific Method | post id:
83757 |
| Ah - but there's the rub, recher.
Science is not based on deduction; it relies on induction, which is
exactly the sort of reasoning as with the crows. Granted, once a
hypothesis of some kind is formed, there is an element of deduction
too. JR | |
| From: Schmeardo | 14/06/00
9:49:50 |
| Subject: re: Scientific Method | post id:
83763 |
| I would say: "After a given survey of the colour of crows in X population, we can say with P<0.05 or 95% probability that crows are black in colour." (only of course if I actually get this result after surveying a statistically significant proportion of the population, calculated through a power analysis) This of course doesn't say that ALL crows are black and this methodology (the one I learnt in my science course at least) allows for variations. There are no certainties in science or anything else, just probabilities to work with, that allow you to actually get things done, instead of debating and debating and debating. Matt | |
| From: recher | 14/06/00
9:50:02 |
| Subject: re: Scientific Method | post id:
83764 |
| I cant accept that the crow
analogy is proper use of induction. The correct induction is there are
definitely black crows and accumalating evidence indicates all crows are
black but you can not induce that all crows are always black. It is a good
example of Ockams Razor. The more black crows you see the weight of
evidence increases towards the logical assumption that all crows are black
but finally we locate an albino and can induce that All crows are probably
black except for the occcasional albino. After milliions of crow
observation everywhere and they are all black with are albino we can
pretty confidently induce with certainity but it still aint 100% sure
stuff. Induction is "the process of discovering expalnations for a set of particular facts, by estimating the weight of observational evidence in favour of a proposition which (usu) assers saomething abouth that entire class of facts." Deduction is "the process of drawing a conclusion about something known." | |
| From: Sam O'Reilley | 14/06/00
10:27:33 |
| Subject: re: Scientific Method | post id:
83791 |
| First; classification is based on
"universals" (desciptions which, by observation, apply to different
individual things). We don't know how it is that we are able to perceive
universals to be common elements of the individual things. It is a
question of the use of intellect; and Aristotle called it intuitive
induction. There is much work to be done on conscious awareness before the
bases of mind giving validity to universals are known with any confidence.
That said, the scientific method proceeds to accumulate universal
descriptions by induction from particular examples; thus grouping the
individuals by common classification into something that resembles a
universal proposition for the class; ie. I only see black crows, therefore
I assume (induce/propose) that all crows are black. I have assumed here,
of course, from seeing different colours of different things (ie. cars)
that there might be other coloured crows, but I've never observed them and
so I induce the proposition (from the facts) that they are all black. Next
comes deduction, by which I take my proposition (or theory) and deduced by
sheer logic ( not by assumption from available facts-as in induction) that
if all crows are black then (given the extremely limited nature of the
proposition I can't actually deduce much) no crows are white, and, if I
see a white crow, then I will have to make a thorough examination
(experiment) of the crowness or the whiteness to decide whether to revise
my original proposition or dismiss the white crow as not being a crow, or
not being white(ie. actually being more black than white). And so,
deduction ups the ante of induction, prompts us to test, and may cause us
to revise. Thus secure knowledge advances, but is never totally secure
(because induced propositions are of endless variety and are endlessly
revisable; and testing can make ever greater distinctions). And,
ultimately, what we consider to be of value depends on the individual
scientists. Remember, this is secure method, not secure, ultimate, truth.
And, back to the beginning, we don't have a secure basis on which to
conclude that intuitive induction is anthing more than a loose
intellectual exercise incapable of reflecting the truth of the world
around us (which I don't believe, but I mention
nonetheless). | |
| From: Spazgaz (AIT) ® | 14/06/00
19:20:12 |
| Subject: re: Scientific Method | post id:
84291 |
| you can not
induce that all crows are always black Step away from the light switch James.... Thanks. SPAZGAZ (AIT) | |