Philosophy and Nature of Science

8/14/2019 Philosophy and Nature of Science

1/58

Philosophy and Nature of

SciencePart 1. Philosophy

Part 2. Philosophers


2/58

Part 1.

Observation,

Hypothesis,

induction,Falsification,

Theory

Explanation


3/58

Science divides into Methods and

Applications Method - Procedures for acquiring

knowledge Application - Use and purpose of

discoveries


4/58

Questions asked in the

philosophy of science Is science based on faith? What is the scientific method? How are new discoveries treated? Is everything reducible to physics and

mathematics? Is everything reducible to a few rules?


5/58


6/58

Ideal Scientific Method

Observation Repetition

Induction(1)Hypothesis

Deduction or generalizationConsequence or prediction

Testing Induction(2) Induction (1) not successful


7/58

Critique of the Ideal scientific

Method Whats observed and studied depends on the currently acceptedexplanation

Explanation selects the observationExplanation Influenced by:

Brain hardwareGestalt formationOptical illusions

Brain Software

Education


8/58

Induction

inductionObservation -----------------> Hypothesis


9/58

Induction Induction goes from effect to cause. Effect can possibly have many causes.

A cause may have a single effect. Hypothesis is a kind of cause

cause effect


10/58

Critique of Induction

There is no logical way of going fromobservation to hypothesis

Hypothesis is a simple guess

Frequently hypothesis precedes observation


11/58

Maybe Hypotheses should be considered only

as Statements of Probability The universe is a series of stochastic events with

ill-defined boundaries

An hypothesis is neither true nor false. It is astatement of probability for success or failure.

Replace All swans are white

withWhat is the probability of finding a green orblack or blue swan?


12/58

Genealogy of Certainty

Hypothesis --> Theory --> Fact

INCREASING CERTAINTY


13/58

Are These:

Explanations, Hypotheses,Theories, or Facts? 2nd Law of Thermodynamics Newtons Law of Universal Gravitation Gas Law Ohms Law Electromagnetic Theory Kinetic Gas Theory Atomic Theory Theory of Relativity String Theory


14/58

Hypothesis, Theory, Fact

Hypothesis are Guesses not logicallyderivable from deduction or Induction

Theories are statement of Probability

Facts do not exist- nothing is 100% certain


15/58

Verification & Falsification

What is meant by explanation?

What is a fact?

When is a

Fact verified?

How many observations needed?


16/58

Form Hypothesis

I put a balloon in my refrigerator and funny thingshappened . I really need an explanation.

First the balloon shriveled up.

Next, the balloon changed color from red to blue.

Finally the balloon said, Get me out of here. Its too cold!


17/58

Deduction and Induction

induction Observation ------------> Hypothesis

deduction Hypothesis ------------> Observation


18/58

Deduction

If there is no cogent way of going fromobservation to hypothesis

Then there is no cogent way of deducingfrom hypothesis to observation


19/58

Critique of deduction

Is Deduction outside of mathematics possible?If this X, then that Y

Are parameters really known? Is X truly an X Is the equipment properly controlled? Can one recognize an X or a Y ?


20/58

Critique of Deduction

Modern Science does not seek causes butseeks relationship among variables

Independent variables are not causes anddependent variables are not effects

If one knows Y =g(x), can one predict(deduce) the future?


21/58

Critique of Hypothetico-

Deductive Method Y = g(X) implies X is selectable and Y is

determinable

We know X and Y only within someprecision range: x1, x2, x3,y1,y2,y3,..

Within the area defined by the range of Xand the range of Y there is no Y=g(X)

All equations are statement of probability


22/58

Verification and Falsification

Replace Verification with Falsification Verification and falsification are asymmetrical

Multiple verification does not establish a theorymore than a single verification

A single falsification overturns a theory

It takes only one green swan overturns the theory that all swansare white. Observing one million white swans does no more to prove allswans are white than witnessing ten white


23/58

Falsification

It is nearly impossible to falsify an hypothesis. Since a test depends on many factors it is difficult

to determine whether the hypothesis failed are oneof the other factors failed.

Some failures of dependent factors:

precision and accuracy of instrumentation, correctinterpretation of data, flawless recording of data,improper experimental conditions


24/58

What is an Explanation?

Hypothesis is not always an explanation Explanations reference non-observables

Science explains objective reality in termsof a non-objective, non-observable reality

Among non-observable objects are

electrons, quarks, photons, gluons,gravitons, positrons, black holes, dark matter, dark energy


25/58

Explanation

Explanation is to ask a quantifiable question

Explanation is not to ask a why question

Explanation is a sequence of events.


26/58

Limits of Explanation

The ultimate building blocks of the universe donot interact with our instruments at all or elseinteract too strongly so that physical nature alters.

It is possible we are at the end of traditionalempiricism where observations suggest for or elsetest for hypothesis.

String theory perhaps is an harbinger of Post-scientism wherein hypothesis andobservation formally separate


27/58

Theory Have Limits of

Application Ptolemaic Astronomy replaced by Copernican Astronomy replaced by Newtonian Astronomy replaced by Einsteinium Astronomy replaced by

Quantum Gravity Theory


28/58

Gloss

A Law is an late 18th and early19th century was of saying theory:LAW = THEORY

Theory gives a mathematical relationship between observable

dependent and observable independent variables .The distinction between independent and dependent variable is arbitrary.

Hypothesis gives a mathematical relationship between non-observableand observable variables


29/58

Overall Criteria for Theories

Consistent Parsimonious

Correlative Empirically Testable ( verifiable& falsifiable) Useful

Progressive Retrogressive


30/58

Criteria for Theory

Logical Empirical Sociological Historical


31/58

Logical Criterion

Clear and explicit boundary conditions

Falsifiable ( imagine conditions theory is invalid)

Every theory had limits of applicability( falsifiable means not disproving a theory but ratherdetermining limits of use)


32/58

Empirical Criterion

Provides guidelines to interpret data

Verifiable prediction and retrodiction ( pastevents )


33/58

Sociological Criterion

Resolves anomalies of current theories

Defines new concepts of operations to solveproblems

Propose new paradigm or problems(a paradigm is a model, a world in miniature

mathematically)


34/58

Historical Criterion

Consistence or coherent with pre-existingtheories of established validity


35/58

Part 2

David Hume Karl Popper Thomas Kuhn Imre Lakatos


36/58

David Hume(1711-1776)

Scottish Empiricist- All we know is throughexperience

Hume argued we are not justified in makinggeneralizations from observations because ourobservation are finite in number

To know something we need to experience it.Obviously we cannot have experience of thefuture so we cannot know it.


37/58

Causation

Hume attacked reasoning from cause to effect Two colliding billiard balls are not predictable. No

logical connection between first ball, causing second ballto roll. We know this only by experience.

We cannot use reason to make predictions (future).Experience cannot tell us about the world (future) eitherbecause we experience only the past.

We believe Sun will rise tomorrow based on experiencenot reason.


38/58

Newtons Laws

Does Newtons Three laws of dynamics predictthe future of colliding billiard balls?

Newtons laws cannot predict trajectory of realbilliard balls due to unknown boundaryconditions, and to first order nature of the Laws.

If Newtons Laws were exact and boundary

conditions known, then past and present arecontemporaneous


39/58

Thomas Kuhn (1922-1996)

All research presupposes a world-view,a collection of fundamental objects, natural laws, definitions, andabove all a definition of what research is.

Kuhn called a world-view, paradigms

Mature science have established paradigms Example of mature sciences are chemistry, physics,

geology; whereas, economics and psychology are

immature sciences.


40/58

Model or Paradigm

Intellectual excision (setting boundary conditions)from the universe or sub-set of universe, entitiesarbitrary in design but relevant to problem understudy

Assembly of entities into a simplified representationof the universe or subset of universe

Determining probability of events with set of variables spanning selected ranges


41/58

Paradigm Thomas Kuhn popularized the term in his book The Structure

of Scientific Revolutions (1996) by using it to describe howscience works. According to Kuhn, scientific explanations of the world are controlled by a paradigm, some model of howthe world is expected to work and into which actualobservations are fitted, even if the fit is not very exact. Asinexact fits accumulate, it becomes more apparent that thedominant paradigm is inadequate as a model of reality. Whenenough contradictions exist, a paradigm revolution occurs anda new paradigm is adopted.

The word paradigm comes from the Greek paradeiknunai andmeans "to compare." In science and philosophy it has thesame basic meaning as in common usage: a model orinstance used as a basis or example for further work.


42/58

Need for Paradigms

Research requires paradigms Paradigms are models of the way the world works

Without paradigms research is a random collectionof observations lacking unification of structureinto a whole.

Without paradigms, it is not possible to decidewhich are and which are not importantobservations


43/58

Dominant Paradigms

As a field matures, one paradigm becomesthe dominant one. Once paradigms isestablished research progresses quickly


44/58

Paradigm guides direction of Research

It becomes clear with aid of paradigmwhich research areas are fruitful. These

areas are ones not totally explained

Paradigms give concepts and laws to build

on.


45/58

Paradigm Shift

Paradigm shift occurs when old paradigmshown inadequate

What is defined as research is reevaluatedConcepts turn upside down

Earlier research is reinterpreted


46/58

Real research

Real research occurs during a paradigmshift

Once a paradigm dominates, researchbecomes puzzle solving

Puzzle solving is not research since answers

known beforehand


47/58

Example of Puzzle Solving

After Newton explained solar system, laterscientists using Newtons theory predicted

The presence of the then unknown planetsNeptune, Uranus, and Pluto


48/58

What New Paradigms Do

Discovery of new paradigm results in newquestions being asked and old questions

abandoned Newton saw gravitation as a property of

matter. Earlier theories tried to find a

mechanical explanation as whirlpools inspace or angels.


49/58

Paradigms are Incommensurable

Paradigms have different world view. It isdifficult to compare them

Consequently, science defines truth relativeto a paradigm and not absolutely. Truth is a

story


50/58

Science

Science is a conformist society which present onlythe currently accepted theoryConsequently science defines reality relative to theaccepted paradigm

Students are educated into the accepted paradigmand to ignore alternative paradigms

The society of science determines what scientistsobserve


51/58

Paradigm Diagram

old paradigm unexplained observations competing newparadigms

one dominant paradigmpuzzle solving

i n c o m m e n s u r a t e

unsolved puzzles ignored

Mopping up operation

unexplained observations andalternative interpretation ignoreduntil enough accumulates to overturncurrent paradigm

unexplained observations


52/58

Sir Karl Popper (1902-1994)

Falsification is the idea that science advances by unjustified,exaggerated guesses followed by unstinting criticism.

Any "positive support" for theories is both unobtainable andsuperfluous; all we can and need do is create theories andeliminate error

Scientists never actually use induction. It is impossible toverify propositions by reference to experience


53/58

Karl Popper

All observations are selective and theory laden A demarcation between science and pseudo-

science is established by falsification. A theory isscientific only if it is refutable by a conceivableevent

Every genuine test of a scientific theory is basedon an asymmetry between verification andfalsification


54/58

Karl Popper

Popper replaces induction with falsification Science is not distinguished from non-

science on basis of methodology. Nounique methodology specific to science

Science consists mostly of problem solving.


55/58

Imre Lakatos (1922 - 1974) `

All scientific theories are equally unprovable

Falsification doesnt work due to rescuehypotheses

the "basic unit" of scientific development is not

the scientific theory, such that scienceprogresses when one theory proves to bemore successful than another .


56/58

Imre Lakatos

the "basic unit" is actually the researchprogram . Science progresses when oneresearch program becomes more productive

and more useful than other and, hence,receives a greater share of social resourcesthrough funding and younger scientists lookingto join. A research program is characterized bya particular set of "hard core" fundamental

ideas and is deemed successful so long as itcontents continue to increase .


57/58

Imre Lakatos

In reality scientists do not abandon theories. They inventrescue hypotheses or ignore anomalies or refutations

Popperian crucial experiments and Kuhnian revolutionsturn out to be myths. What happens is progressive

research replaces degenerating ones .

Progressive scientific programs predict and producedramatic, unexpected observations and results


58/58

Philosophy and Nature of Science

Documents