Science and Non-Science

OHear: Chap. 4

The Demarcation Criterion

Two previous points {prerolled theses: for those having difficulty rolling their own)

There has been a type of growth in knowledge in science that has not been found in elsewhere in human thought. (54)
[I]n science proper there [is] an attention to criticism and the negative instance which [is] lacking in certain other areas of intellectual activity, rather to their detriment. (54)

Contra Bacon:

Bacon's methods are unworkable and his description of formulation of scientific theories a caricature. (54)
[T]he scientific spirit consists not in the way we formulate our theories, so much as in our treatment of them once we have got them. (54)
Science . . . gain[s] its distinctive character not from the elimination of presupposition and intuition, but in the control an impartial nature . . . exercise[s] over them. (54-55)

Context of Discovery vs. Context of Justification

Context of Discovery: stage of "inspiration" (Edison)

misleadingly so called: might be more accurately deemed "context of hypothesis formation"
here -- at the creative stage -- anything goes

Kepler's mysticism
Newton's alchemy
Kekule's dream
Sagan's pot smoking

Context of Justification: stage of "perspiration" (Edison)

also misleadingly so-called if you're Popperian

there is no justification (positive confirmation)
only theories that continue to enjoy a presumption of innocence
have been severely tried and not (yet) falisified

only at this stage that we get to what might properly be called discovery wherein

precisely formulated hypotheses
undergo trial by nature

by experiment
and observation

Preemptive Strike Contra Relativism (prefiguring Sokal)

grounded on a failure to appreciate the Discovery vs. Justification distinction
of course the historical and social context have a lot to do with the creative stage of discovery

Even if, as is sometimes claimed, the spirit of capitalism created a climate in which men would naturally seek to quantify, analyse, and exploit nature,

it does not follow that all the theories produced in this context are not true. (56)

but this doesn't impugn the objectivity of the scientific justification

Whether they are found wanting or not will be determined

by their ability to predict the course of nature,
and not by the desires or beliefs of capitalists, entrepreneurs, mystics, or social historians. (56)

Popper's Falsifiability Criterion

Popper's Proposal

Empirical falsifiability [is] the distinguishing mark of a scientific theory. (56)

As applied to psychoanalysis

[Psychoanalytic] theorise are compatible with any type of behavior in individuals. (56)

Whether you are brave or cowardly, you will still be manifesting your Adlerian will-to-power.
A dutiful son and a rebel may both be suffering from an Oedipus complex.
therefore,

[P]sychoanalysis does not provide explanations of a scientific sort,

in which predictions are derived from theories,
and then tested against the evidence . . . (57)
instead

[Psychoanalysis gives] persuasive redescriptions and interpretations of human activity from specific standpoints, looking at human activity as an instance of will-to-power or of repressed sexuality, for example.

The succcess of such stories will presumably be judged by thier adequacy in describing and making sense of the behavior of the individuals in question
rather than in terms of any predictions they might lead to.

The skill of the analyst is more like that of a novelist than a scientist. (57)

Applied to Marxism

Predictions have been made

In industrialized, capitalist societies there will be increasing polarization between workers and capitalists.
With the workers becoming increasingly impoverished.

Resulting in a revolution by the workers against the capitalists.

The revolution will be followed by a period of proletarian dictatorship.
Leading, finally, to the withering away of the state.

Manifestly, nothing of the sort has happened

But Marxists do not conclude: Marxism is refuted.

they respond as if their "theory" were religious dogma
devising ever more arcane and complex explanations for the failed predictions
e.g., Leninism?!

advanced capitalist countries buy off the workers at home

with "bread": "social welfare programs" (mentioned by O'Hear)
and "circuses": football, television, etc.

with the profits got by exploiting the third world.

Nonscience vs. Pseudo-science

Many valuable human intellectual pursuits are unscientific, e.g.,

philosophy
religions
the arts

What's discreditable . . . is pretending to practice science when you are really enacting a moral vision or a therapy or a religion.

Problems for Popper's demarcation criterion

Popper says,

"Statements or systems of statements, in order to be ranked as scientific, must be capable of conflicting with possible or conceivable observations." (58)

Immediate Problem: on this criterion] many empirically provable statements, which we would intuitively think of as scientific, become unscientific.

Existential generalizations

"There is at least one planet."
"There are electrons."
"Bacteria exist."

Explanation: To refute an existential you have to confirm the corresponding universals (that for everywhere and always):

There are no planets.
There are no electrons.
There are no bacteria.

[Probability statements] cannot be falsified if no limit is put on the possible number of [instances].

[P]robabilistic theories used in science usually refer to open ended runs of events
and so it looks as if they cannot be falisified.
Only looks that way

no conclusive refutation
but "we can . . . stipulate that certain outcomes highly unlikely on our theory are to be treated as refutations on that theory": a "rule of detachment"

my ad hominem (v. Popper, not O'Hear)

probablistic reasoning is inductive
there goes the neighborhood again: the Popperian hope in conclusive refutation

Illustration: prediction 70% chance of rain tommorow.

if it doesn't rain tomorrow, that's perfectly consistent with the hypothesis that there was a 70% chance
and this was one of the 30% days: the weathertron wasn't mistaken
but if this happens a lot -- if it never rains (or only rains 30% of the time) -- when the meteorological theory being applied says 70% then it's increasingly unlikely that this is just a run of bad luck
and increasingly likely that the theory is false.

Yet more fundamental difficulty

scientific laws are not capable of conflicting with possible or conceivable observations
without additional assumptions
the upshot being

no theory is ever conclusively refuted by experiment or observation
the problem could always be with the additional assumptions

and the more scientifically fundamental the law -- e.g., Newton's 1st law

the more reluctant we are to give it up
the more inclined we are to blame the failure on the auxilliary assumptions

Are Theories Ever Falsified?

Example: Newton's 1st Law: Law of Inertia

Covering Law Explanation/Prediction

L1: An object at rest tends to stay at rest and an object in motion tends to stay in motion with the same speed and in the same direction unless acted upon by an unbalanced force.
C1: This mojo is in motion.
C2: This mojo is not being acted upon by an unbalanced force.
:. E: This mojo will stay in motion in the same direction.

Observation: The mojo changed direction!
Conclusion:

Conclude not L1, NOT!
There must be an unbalanced force . . . we hypothesize!

"torsion forces" they say are present
unseen, convenienty ;-)

Worry about C2 -- a negative existential again -- An implicit assumption of every would be falsification

C2 not so much an observation (I didn't see any opposing forces at work, did you?)
as an escape clause generalization -- an all purpose escape clause -- here are no unseen, unbalancing forces present.
Compare Lenin's escape:

there were counterbalancing forces of imperialism present
so Marx's theory is not counterinstances by the predictive failures alleged against it

Other assumptions

reliability of

your senses:

assumption: your eyes don't deceive
corollary: it's no an hallucination. nobody put LSD in the Alma College water supply.
corollary: it's not a holographic projection.

the test apparatus: similar doubts (illustrated by O'Hear's example)

more likely the balance is broken
than that Newtons 3rd Law is Refuted

consequent stress on repeatability:

to control for idiosyncratic variations of test situations
and experimenter bias (including outright fakery)
so there are no conclusive one-shot refutations

Popper's Concession . . . segue to Kuhn

Concession (quoted by O'Hear, p. 63)

[M]y criterion of demarcation

cannot be applied immediately to a system of statments . . .
Only with reference to the methods applied to a theoretical system is it possible to decide whether we are dealing with

a conventionalist [i.e., dogmatically held]
or an empirical [i.e., provisionally held] theory.

Segue to Kuhn

So what's scientific or unscientific does not so much hinge on

the matter or content that's believed
as the manner in which the belief is held

dogmatically:
experimentally

What's scientific is not so much the theories themselves as the disciplines

the accumulated scientific traditions of inquiry
and the research communities who embody these traditions.

Kuhnian Relativism

Received Logical Empiricist View: approach centered on theories (the product)

Theory of X

set of statements about X
empirical scientific theory

observationally or experimentally well confirmed
observationally or experimentally falsifiable

Difficulties in specifying, defining, or articulating

what makes for confirmation

not only skeptical -- Goodmanian & Humean -- difficulties but
(related?) technical difficulties about inductive reasoning

nonmonotonicity & the total evidence requirement
ultimate reliance (in determining relevant evidence)

social "custom" (Hume) whence "projectability (Goodman)
and individual habit

or "custom" (social practice of identifying and projecting some properties & not others) &

what makes for falsification

preliminary worry about negative existentials: falsifiability too strong a requirement?!
difficulties in specifying, definining, or articulating

what makes for falsification
given that one can always save the theory in the face of recalcitrant evidence

by doubting the reports, or instruments, or your senses
by hypothesizing unknown interfering factors

Newton

anomalies in the orbit of Uranus lead to Neptune hypothesis
anomalies in the orbit of Mercury lead to the Vulcan hypothesis

Mojo: unknown (torsion) force hypothesis
Lenin: forces of imperialism? Or is this . . .
modifying the hypothesis
for good: elaboration & refinement of theory
for ill: ad hoc [to this] or post hoc [after this]

you see there's these epicycles . . . tale of two theories
violation of simplicity or Occam's razor

simplicity: other things being equal, prefer the simplest theory or explanation
Occam's Razor: don't multiply entities beyond necessity

troubles: the subjectivity of simplicity

why should simpler be more likely to be true?

easy to see why simpler should be easier for us to understand or easier to compute
not easy to see why simpler should be more likely to be true.

"life's not fair"
and maybe nature ain't simple.

almost an aesthetic judgment

the Ptolemaic sees as a refinement
only seems ad hoc to the Copernican

epicycle has become synonymous with ad hoc
because the Copernicans won and the winners write the history
or at least because hindsight is 20x20.

Kuhn: from Theories to Paradigms: from Product to Process (Kuhnian revolution itself an instance of paradigm shift in the philosophy of science).

Paradigm is an ethos or form of life

a whole way of doing science
centered around some definitive theoretical-experimental result
including not just overall theoretical agreement

but intellectual values & standards
common practices
organizational and professional ties

"A paradigm characterizes ruling sets of scientific techniques and ways of looking at data."
Examples of major shifts & paradigms

Aristotelean to Newtonian and Newtonian to Relativisitic Physics
Ptolemaic to Copernican to Keplerian to Newtonian Astronomy
from Linnean(?) to Darwian biology

Paradigms are not up for falsification at all, normally

it determines the way scientists see the world and approach their data
it is the unquestioned background of their research: normal science

research concerns things yet to be worked out or still up for grabs within the paradigm
the paradigm itself sets the rules for adjudicating these disputes

setting experimental practices
and standards

since the paradigm itself sets the standards (supplies the model of good science)

if it's not immune from criticism by its own standards
it would seem to have a big leg up on the competition

normal science is like puzzle solving within the paradigm

which sets the puzzles
and says what counts as a solution

more radically yet: paradigms (at least) color perception

the readiness with which people say previously "invisible" defects in the heavens once belief in the immutability of the heavens was given up when the Copernican view waxed decisively.

sunspots had been systematically recorded in China (where they had no such belief in heavenly immutability) for centuries
before Galilleo first "discovered" them in the West (where immutability was accepted)
"The very ease and rapidity with wich astronomers saw new things when looking at old objects with old instruments may make us wish to say that, after Copernicus, astronomers lived in a different world." (Kuhn: 66)

in periods of normal science

paradigms themselves (or their central theoretical tenets) are protected from falsification
would-be refutations normally treated as anomalies

come in various stripes

unexpected (not as predicted) results
inexplicable difficulties in extending the theory to cover certain cases

treated as outstanding problems

for further investigation
within the paradigm

Scientific Revolutions

Title of Kuhn's book was The Structure of Scientific Revolutions
Sometimes

perhaps anomalies become so many or pressing
or due to input from other sciences, or general culture, or whatever (context of discovery)

e.g., the advent of the telescope helped revolutionize astronomy
similary, with the microscope, biology.

example: behaviorism to cognitivism.

outstanding problems on behaviorist principles centering on explaining

"higher order" behavior
especially human language use or "linguistic behavior"

defining moment: Chomsky's review of Skinner's Verbal Behavior
outside influences

computers as high order behavers after their own fashion
concepts & tools imported from computer science

to explain and invesigate
human "higher order" capacities

Incommensurability of Paradigms and the Extrarationality of Paradigm Shifts: Kuhnian Relativism

Since paradigms set the standards of rational acceptance
There is no rational basis for comparing paradigms: they will be rationally incommensurable

each -- judged by its own internal standards -- is best by those standards
and there is no neutral court of appeal -- no extraparadigmatic standards

The evidence itself is not neutral but paradigm-dependent or theory laden (c.f. the "different worlds" line)

scientists operating in different paradigms (may) even see different things under the same circumstances

since perception itself is theory laden
"in science the basic similarity-dissimilarity relationships, on which observation and classification depend, are themselves dependent on the paradigm within which we work" (73)

Dramatizing the Upshot:

Two scientists -- one a proponent of theory A, the other of B.
They could agree on all rationally relevant considerations

on the observed facts (or at least on their verbal reports thereof ): Keplerians & Ptolemaics agreed
of the standards of rational acceptability (or at least their verbal formulations thereof)

simplicity
accuracy
fruitfulness
scope

yet still disagree on the fundamental question of which paradigm is best

e.g., the Ptolemaic & the Keplerian may agree that simpler is better
yet disagree over whose theory is simpler
Keplerian thinks ellipses are simpler because you need fewer.

Ptolemaics: thinks circles are simpler (and more perfect) figures than ellipses.

quality of components a more important desideratum
than number

Failure of mutual intelligibility

changes of classification -- if alchemists & modern chemists mean something different

by "element"
even "water"

proponents of different paradigms -- in a sense -- are not even speaking the same language

by "water" the alchemist means "an element such that . . ."
and the modern chemist means "a compound such that . . ."

Relativism: Overdramatizing the Upshot?

proponents of different paradigms speak different languages
see the world differently
and even "live in different worlds"

History vs. Philosophy of Science

Kuhnian appeal to history of science to against received empiricist conceptions -- e.g., Popper's, Inductivists

Logical empiricists are going so: more proscriptive

given that science advances
in ways quite unlike the ways other explantory narratives don't

mythologizing
philosophizing

What logical and epistemic characteristics must it have to be like that?

How must its concepts and statements -- especially its laws -- be grounded in observation
Hence LE's outstanding troubles (anomalies)

about induction and confirmation
and finally, even about, falsification

Kuhn: look how science is -- and it's history: more descriptive

and you'll see is not the steady growth of knowledge

wherein everthing proceeds all nicey-nice & logically as LE suggests
as in periods of normal science

but overall it's by fits and starts

by scientific revolutions: radical changes of world-views
where there is no neutral rational basis for decision
even no clear neutral or extraparadigmatic sense in which such change is cognitive advancement

often cast off paradigms embody insights or valuable approaches

that their successors conspicuously lose
which get reinstated later (perhaps in yet another paradigm shift)

history reveals

a series of incommensurable changes in world view
rather than a "steady advance of knowledge"

O'Hear's Criticism of Kuhn's appeal to history

Back at ya?: Invalidity of Kuhn's appeal to history

The theory ladenness of Kuhn's reconstruction
Possible to reconstruct the same historical events -- e.g., the French revolution -- as

a revolutionary change: things really changed
a continuation of the old regime: things really didn't change, basically

According to one's theoretical (and other) persuasions two historians may interpret the same events quite differently
In the same way there's no neutral basis of comparison

where Kuhn sees a "progress" of revolutionary conversions
inductivists will see a gradual accumulation of better and better confirmed hypotheses
Popperians will see the corroboration

of ever more daring (broad & precise)
as yet unrefuted conjectures

Worry about O'Hear's Criticism: It concedes too much!

Just so says the Kuhnian
the way you describe the interpretive impasse

advocates of different appoaches
see the historical evidence differently
according to their theoretical or paradigmatic persuasions

is just the way you say it is here, between us.
so

where you think you're sawing off the branch I'm sitting on
you're actually buying into my view

O'Hear's rejoinder?

remains to be seen whether there is no neutral observational or logical grounds for comparison

between these different interpretations of the history of science
or between competing "paradigms" more generally

may be yet be a paradigm-neutral basis for a case

against Kuhn's interpretation
for the history of science being unlike Kuhn's reconstruction of it

more continuous and
less fragmented

Further Rejoinder: the historical evidence Kuhn appeals to really cuts very little epistemological ice

even granting that the history of science is a discontinuous and fragmented as Kuhn sees it as being
no relativistic conclusions follow

might still be reason to think reasoned judgments could be made about the superiority and inferiority of competing paradigms
suppose the change from behaviorism to cognitivism really were much as Kuhn says

a sharp break (not a gradual progression)
in which nonrational elements played a major role

nothing follows from this concerning the cognitive superiority, or inferiority, or incommensurability of either view

perhaps cognitivism is flatly scienficially or epistemologically better
despite being born of revolution
just as the fact that this country was born of revolution doesn't entail anything about whether

that the new independent form of governance was better than the old colonial form
or worse
or incommensurable

relativism doesn't follow from the history but rather seems based on the following line of argument

There is no cognition of the world independent of and set apart from the the concepts we use to approach the world.
Different inquirers -- especially those starting from widely different paradigms -- approach the world via different concepts.
So, every inquirer (or group likeminded group thereof) is "imprisoned within their own conceptual scheme"

Herein Kuhn makes the following fundamentally philosophical claim:

that something in the nature of our conceptualization of the world
makes it impossible for observors from different theoretical backgrounds
ever to correctly recognize that they are talking about the same things
and hence to compare observations.

Which O'Hear denies: He will maintain

nothing of this [relativistic] sort follows from the mere fact that
all our knowledge of the world is mediated though our conceptualization and categorizations