Outline:
Overview of issues (plus occasional references to the earlier history we have examined)
Ch. 1: on the (spurious) distinction between "theory" and "hard facts," and the issue of "fit" in evaluating competing theories (leading to the class discussion of the fundamental assumption - apparent since Thales - that the universe forms a kosmos, a unity, so that our theories about that unitary/coherent universe must be consistent [i.e., they should fit with one another]).
Ch. 2, "the positivist model of scientific theories" -
while positivism has roots in the early 1800's (namely, the French
philosopher August Comte), the discovery/development of formal
logic at the end of the 19th ct. issues in a new optimism regarding
the power of formal (axiomatic) systems to define and demarcate
knowledge, leading to the explicitly epistemological effort
to define natural science as the only legitimate form of knowledge,
primarily by (apparently) succeeding in defining "natural
law" in formal terms - i.e., the correspondance rules ("c-rules")
which ground the meaning of theoretical terms in
observational terms (as "theoretical terms observational
terms" - i.e., as each is the necessary and sufficient condition
of the other).
The historically astute recognize this as the positivist version
of Hume's empiricist criterion of meaning.
But it also means that everything - including the ability to
demarcate "real" science (knowledge) in contrast
with pseudoscience (including philosophy and religion as dealing
with non-observational entities [rights, Divinity, values, etc.])
- turns on this "centerpiece" notion of c-rules - and
the distinction they rest upon between theoretical
and observational terms.
Ch. 3, "trouble in positivist paradise"
In the "Introduction," he refers to the discovery of cosmic background radiation, which we have explored via the video segment from Stephen Hawking's Universe - as another example of the specific issues he will deal with in this text:
theory - and the point that science also involves "a practical engagement with the natural world." (3)
the distinction between observational and theoretical terms in science
the nature of scientific explanation - coupled with the problem of the underdetermination of theory
(we have seen this problem several times before:
how to pare down the (potentially infinite) theories/explanations which "fit the data" (e.g., the Ptolemaic vs. Copernican)
- and the implications of this problem for the realism / instrumentalism debate)
his defense of realism - "that scientific explanations not only seek the truth about [inquiry-independent] phenomena but actually achieve it in a number of cases...." (5) vs.
antirealism, including the various species of social constructivism - especially as social constructivism leads to relativism.
[the historically aware will recognize here precisely the problem we examined in:
a) the Sophists / Religious Dogmatists - relativism / absolutism conflict in ancient Greece - one ostensibly resolved by the Socratic/Platonic focus on the mathematicals and forms as knowable by mind [issuing in physics and the sciences - as focused on "things and their shadows" as the objects of sense-data - as epistemologically defined as "a likely story," i.e., one which cannot achieve the same level of truth/certainty as our knowledge of mathematicals and forms]; and
b) the questions raised by Kant, Einstein, and Quantum Mechanics as we
(i) recognize the role of the observer in determining his/her "experience" of the observed (most dramatically, by forcing "reality" to "make a decision" about its location, etc., through our acts of attempting to measure that reality), and
(ii) especially our having to rely on mathematical accounts of sub-atomic phenomena (the particle/wave duality, the uncertainty principle, etc.) over against "common-sense"/macro-atomic concepts - epistemologically "relativizes" our knowledge of nature (heading us towards instrumentalism once again) - but does not support an ethical or moral relativism.]
reductionism - the effort to reduce all branches of science to a single form (probably physics) - vs. antireductionism (or, less oppostionally/dualistically - pluralism) which argues that different modes of knowledge cannot be reduced to a single paradigm without loosing essential elements of those diverse modes.
"Reductionism is where science turns blatantly metaphysical in the sense of turning ontological." (6) That is, at this juncture science begins to make claims about what is most real - and thereby, what is not real.
(Whether or not science, without the aid of philosophy, can successfully undertake this classic philosophical task, is an important issue of debate as well.)
Includes the important point:
A frequent contrast some critics of science often use is that between theories and so-called "hard data." What those critics fail to realize is that there is no such thing as hard data minus some theory under which they count as hard data. It is a theory that tells the observer that what is being observed is relevant in the right way to what the observer is interested in finding out. (12)
(there is also good discussion on this page of "maps" and "proof" - raising the question: is he as much of a realist as he seems to want to be?)
As well, consider the epistmological point:
A scientific theory does not live isolated in epistemological space. The universe is a cosmos, not a chaos, and its various levels and parts fit together , albeit roughly, to form an overall fabric. (22)
This means that one aspect of considering a new theory is how well it "fits" with previously existing theory as a larger explanation for what we take to be an ordered, coherent, and consistent universe.
Ch. 2, "The Positivist Model of Scientific Theories"
2.1. Theories as Formal Languages
(Logical positivism and its connection with the rise of mathematical logic - but also an explicitly epistemological position - 29ff.
An issue here: what do we mean when we speak of a scientific "law"? (30f.)
The positivist solution: "material conditionals of universal scope" (31f.)
What is important here - "the central assumption behind the positivist model of science":
The positivists assumed that the meaning of logically defined symbols could mirror or capture the lawful relationship between an effect and its cause. Or, to put it in more general terms, the positivists assumed that logical relations between formal symbols could accurately represent external relations between the objects, properties, or events to which the formal symbols referred. (32)
[The historically aware will see here a
specific example of the general issue we have pursued with regard
to mathematics: why should we assume that our formal notation
and systems have anything to do with external "reality"?]
This leads us to the positivist version of the verifiability criterion of meaning (can you say "David Hume?" I knew you could...)
...every theoretical term in a scientific theory must be provided with an explicit definition composed entirely of observational terms. (33)
This requirement is fulfilled by "correspondence-rules" (c-rules) in which the observational side of the definition is the "material equivalent" (each side of the definition - the observational and the theoretical - is the necessary and sufficient condition of the other).
These rules, most centrally, are what positivists
point to as distinguishing science from "pseudoscience"
- ostensibly, claims to knowledge which cannot be grounded (through
c-rules) in observation (35f.)
Ch. 3, "Trouble in Positivist Paradise"
Given the central role of c-rules - and, thereby, the distinction between theoretical and observational terms - in the positivist paradigm, especially as these demarcate science from pseudoscience (much less from religion, philosophy, etc.) -
if this distinction becomes problematic, the positivist paradigm threatens to tumble like a house of cards.
In fancier terms:
....the observational/theoretical distinction is intended to carry methodological, epistemological, and ontological significance. That is, something observable is something that we can come through systematic investigation, to know, exists. (42)
But while the observational/theoretical distinction thus bears the weight of positivist methodology, epistemology, and ontology - the distinction itself was more or less taken for granted by most positivists.
And, it turns out, various efforts to refine and ground this distinction more fully seem to run aground. Klee identifies the following:
Carnap: a term T is observational if a practitioner using the Theory T occurs can determine T's correct application with relative ease,... otherwise, T is a theoretical term.
Relative to what? Ease of application relative to the greater difficulty of application for a theoretical term. But leads to the further, fatal problem: the criterion is dependent on the experience and expertise of the practitioner. (Example of the allergist: an expert can "observe" eosinophilia without instruments, lab tests, etc. - but the novice cannot - meaning that the term is observational or theoretical depending on the practitioner.)
3.1.2 Instrumentation (Amy):
What's observational, what's theoretical? Something is theoretical if it requires some sort of instrument to observe, document, etc.
BUT: so do eyeglasses count (as in the example of looking at a
patient's skin to examine reactions to a skin allergy test)?
If not - maybe expensive instruments? (Counterexample: expensive
eyeglasses)
If not - perhaps complex instruments? But this varies from person
to person.
SO: the criterion is too vague to be of use. [parallels the problem with "ease of application," above.]
3.1.3 Putnamian Cynicism (Tomo)
No problem with the observational/theoretical distinction - it was invented to solve a problem that does not exist.
"White blood cell" - white is observational, "white blood cell" is theoretical
--> there is no observational/theoretical distinction.
This solves the ontological problem - i.e., the problematic distinction between theoretical and observational entities, but...
not the epistemological problem: can we really apply theoretical terms to observation, vice-versa - are there "c-rules" for theoretical/observational terms?
-->
3.1.4 Encapsulated Information Processing (Kelley)
(Example of having to abstain from antihistamines, etc., in order to ensure good results from allergy tests.)
Fodor: the brain processes information in encapsulated units. (so a theoretical/observationl distinction might be workable).
Churchland: no - everything goes into the thalamus which then
forwards it to the prefrontal cortex which "processes"
as the theory area. This is a move from a strong distinction
between observations and theory to a tight (Kantian) interaction
between observation and theory...
indeed, for Kelley, a nice Platonic analogy of the line
which correlates shadow/thing//mathematicals/ideas :: cerebrum
(consciousness) <-- thalamus: coordinates all the sense data
3.1.5 Sensory Surface Distance (Kristen)
Example of rash on the arm (which we can see) ... the reactions occuring at the cellular and chemical levels. The latter require lots of fancy equipment (cf. instrument criterion) - but this introduces more risk of error, plus additional theoretical assumptions. There is no procedure for generating a genuinely faithful (i.e., non-distorted) "picture" of these domains of reality/observation.
In addition to the collective failure of each of these criteria to fully anchor the observational/theoretical distinction - there are additional problems with the correspondance rules themselves - specifically, the demand that each term on each side of the material equivalence refer to exactly the same entities. While this seems on the face of things to be a reasonable and easy-to-satisfy demand...
3.2.1. Dispositional Properties
Operationalism - introduced by Bridgman in 1927. While it does not last long in the physical sciences - oddly, it remains influential in the social and behavioral sciences (53).
Two problems:
b) General problem for any positivist model of the sciences:
Surely it is true that a system under scientific study, be it a human allergy patient or the planet Mars, can possess a theoretical property even when no observable operations are being performed on that system - indeed, even if no observable operation is ever performed on that system. If this is so, then the observational side of at least some c-rules must be written in the subjunctive mood. (56)
But if this is the case, two problems emerge:
ii) bigger problem: the subjunctive mood "...lets in everything." (57)
Now what, Murky? HOLISM! (Carl Hempel) -- but this leads to a diffusion of the observational/theoretical distinction - such that it is no longer clear that the observational/theoretical distinction can "carry" the methological / epistemological / ontological weight the positivists claimed for it. (59)
But this in turn opens the door to the point in Hanson: all scientific terms are theory-laden (59)
-->Fodor: you can see anything, given a given theory (--> social construction)
Ch. 4:
Baconian view :: Quine-Duhem Thesis (pragmatism)
Popperian Falsificationism
Is Underdetermination a Myth?
Pragmatism and Realism