Hps100 Midterm Review
Theory: A set of propositions that attempt to describe something. Scientific Mosaic: A set of all accepted theories. Note that scientific mosaics have changed over time.
Scientific Change: Any change in the scientific mosaic.
Question: Is there anything unchangeable in the mosaic? Is there such a thing as absolute knowledge?
Case 1: Mathematics
1 + 2 = 3. What makes this statement true?
We've defined 2 as: 1 + 1 = 2
We've defined 3 as: 1 + 1 + 1 = 1 + (1 + 1) = 1 + 2
Mathematics: General Template
- We start with some basic definitions
- Some basic theorems follow immediately from the definitions. - Other theorems follow from definitions and basic theorems. - More theorems can be deduced from those theorems and so on.
Don't forget about the age old question of How do you calculate molecular mass and formula mass?
Case 2: Swans
How can you justify the statement: All swans are white.
The theory that all swans are white is based on our experience (observation) of individual swans.
Case 3: Gravity
Newton's law of universal gravitation → F = (G * m1 * m2) / r^2
How do we know this is true?
→ It must be somehow based on experience.
Two types of Propositions:
- deducible from definitions.
- cannot contradict the results of experiments or observations - must necessarily hold true in all possible worlds. i.e. the opposite is inconceivable
- all propositions of formal sciences (e.g. mathematics, logic) are analytic.
If you want to learn more check out Why do firms innovate?
- not deducible from definitions
- can contradict the results of experiments or observations - most (not all) propositions of empirical sciences (e.g. physics, biology, economics) are synthetic because the opposite is conceivable - E.g. We can conceive a world where floating mountains are possible.
Key point to note: Synthetic propositions can be confirmed by experience and observation but cannot be proved. There is no such thing as “proof” when it comes to synthetic propositions.
1. Can analytic propositions be absolutely certain? → Yes.
2. Can synthetic propositions be absolutely certain? → No.
Why synthetic propositions can't be absolutely true: 3 Problems: If you want to learn more check out Why use bacteria and bacteriophage model systems?
1. Problem of Sensations:
Our sensations/perceptions are questionable. E.g. Optical illusions. There is a disconnection (or possibility of a disconnection) between reality and perception. We also discuss several other topics like Which famous italian family were huge patrons of the arts during the renaissance?
2. Problem of Induction: We also discuss several other topics like What is residence time?
How can we arrive at a general conclusion if experience provides us only singular propositions. (e.g. We see a swan that's white. How do we know that there isn't a black swan somewhere that no one has seen? How could we know if we've observed all the swans out there? How could we know that there won't be a black swan in the future?
The point is, our experiences are limited.
3. Problem of Theory-Ladenness:
Let's say you “observe' mountains on the Moon through a telescope. What's the problem here? → You've placed your trust in the instrument (telescope) and theories of optics. Even trusting your vision requires trusting things like having good lighting, you're not under the influence of any drugs, you're not hallucinating, etc.
Our observations rely on or are laden with theories. Theories can change. Theories are shaping our perspectives.
Can synthetic propositions be absolutely certain? There are two stances you can take:
1. Infallibilism: Synthetic propositions can be infallible, i.e. incapable of making mistakes. Empirical knowledge can be absolutely certain. 2. Fallibilism: No synthetic propositions can be infallible. Empirical knowledge cannot be absolutely certain. This is the position accepted nowadays.
Theory Assessment - Question: Don't forget about the age old question of What do answers do?
While all theories in empirical science are fallible, we still believe that some theories are better than others. How do we decide which theories should become accepted into the scientific mosaic?
A theory is said to be accepted if it is taken as the best available description of its object. The object could be something physical (e.g. falling apple, revolving planet), something social (e.g. a group of people, social institutions), etc.
A theory is said to be in use if it is taken as an adequate tool for practical application, regardless of whether it is accepted or not.
A theory is said to be pursued if it is considered worthy of further development, regardless of whether it is accepted or whether there is any use for it at that moment.
Accepted: Quantum Physics, General Relativity
Used: Classical Newtonian Physics
Pursued: String Theory
Two incompatible theories can be simultaneously used, pursued but not accepted. Only one of the alternatives can be accepted at a time.
How do we decide which theory is the best available description of its object? We need a list of rules, criteria.
We need a method of appraisal to determine which theory is better in light of the available evidence.
Method: A set of requirements (criteria, rules, standards, etc.) for employment in theory assessment (evaluation, appraisal, comparison, etc.) Example of methods (not necessarily good methods): Accept theories that are simpler. Accept theories with confirmed novel predictions. Accept theories that solve more problems. Accept more precise and accurate theories.
Methods should not be confused with methodologies.
Methodology: A set of explicitly formulated rules of theory assessment.
Methodologies are the rules openly prescribed by the community as the correct way of doing science.
Methods are the actual (implicit) expectations of the scientific community.
Our openly formulated requirements are often very different from our actual expectations.
You may or may not have an openly stated methodology, but you do have a method. You may or may not be aware of what is it that guides you in your choices (that would be your methodology) but you do have a method (your implicit expectations which allow you to choose between competing movies for example).
Methodologies are what scientists say they should be doing; methods are what scientists actually do.
Focus on the method, not on the methodologies.
Methods of theory assessment are not to be confused with techniques used by scientists to construct theories.
Research Technique: A set of procedures for theory construction (generation, invention). E.g. brainstorming.
If there were a fixed set of rules employed by the scientific community in theory assessment then we would be in a position to say that our current theories are better than the theories of the past.
The whole process of scientific change would be governed by this fixed scientific method.
Are there such unchangeable rules of theory assessment?
What are the requirements of the scientific method? How can they be explicated? To answer this question we must study how theories actually become accepted into the mosaic.
Why is it that, in some cases, theories become accepted simply because they are more precise and accurate than their predecessors, while in other cases, theories are required to provide confirmed novel predictions in order to become accepted?
Accepted Ontology: The accepted views on the types of entities and interactions that populate the world.
Our attitude seems to depend on whether the theory attempts to modify the accepted ontology.
Does a new theory try to modify the accepted ontology?
No: In order to become accepted, the new theory must fit the known data with more precision and accuracy than the accepted theory. Yes: In order to become accepted, the new theory must also provide confirmed predictions of hitherto unobserved phenomena.
Precision and Accuracy: Coulomb’s Law
This law didn’t introduce any new ontological elements. It merely quantified a known relation.
Coulomb’s Law became accepted because it fit the data. It had no confirmed novel predictions.
Novel Predictions: Wave theory of Light
Light is a wave that spreads in a universally present medium, ether. Light waves can diffract and interfere when they meet obstacles similar to water waves in a way. The theory’s prediction was confirmed and the theory was accepted after that.
Novel Predictions: Superstring theory
All known particles and fundamental forces are vibrations of tiny supersymmetric strings – superstrings. This theory introduces new ontological element -> superstrings.
Hypothetico-deductive method: A hypothesis is allowed to introduce unobservable entities (e.g. particles, forces, superstrings, etc.) provided that it predicts something novel, hitherto unobserved, and some of these novel predictions are confirmed.
Among other things, nowadays, new theories must satisfy these requirements in order to become accepted.
Question: Is this scientific method fixed (unchangeable, transhistorical)? -> No
Laws of Scientific Change:
General Theory of Scientific Change
Both scientific theories and methods of theory evaluation are changeable.
Is the choice of methods arbitrary, or is there a logic that governs the process of method change?
Is scientific change a rational (law-governed) process?
Can there be a general theory of scientific change?
Yes: Generalism: There can be a general theory of scientific change. A theory that explains all the transitions from one theory to the next and from one method to the next.
No: Particularism: There cannot possibly be such a thing as a general theory of scientific change. Two reasons:
o There is no universal and fixed method of science. This says the method is changeable.
o There is nothing universal in science; each historical episode is unique. This says there is nothing permanent is science.
Everything is changeable. (This is a stronger argument than the first one).
The details of each scientific change in the mosaic are so specific that no two changes will be explained in the same way. There is nothing universal guiding those changes.
Particularism is what’s accepted today.
Why would we believe that there can be a general theory of scientific change?
Drug Testing 1: Experimental Confirmation -> Problem: Unaccounted affects Drug Testing 2: Controlled Trial -> Problem: Placebo effect
Drug Testing 3: Blind Trial -> Experimenter’s bias
Drug Testing 4: Double-Blind Trial
Hypothetico Deductive Method:
Even this method is based on some assumptions.
1. Complexity: The world as it appears in observations is a product of some more fundamental inner mechanism. We believe there are things that are not directly observable. This is an assumption implicit in all of contemporary science. That’s the reason why we posit the existence of such unobservables such as particles (e.g. atoms) or forces. The idea is that there is more to the world than meets the eye.
∙ We wouldn’t tolerate unobservables if we thought the world is composed only of what is immediately observable.
2. Post hoc explanations: Any phenomenon can be given many different contrasting post hoc explanations which are equally precise. Post hoc means after that fact.
∙ Novel predictions wouldn’t be necessarily if there were no risks of cooked up post hoc explanations.
Method: A hypothesis is allowed to introduce unobservable entities (e.g. microparticles, forces, supersrings, etc.) provided that it predicts something novel, hitherto unobserved (not yet been observed), and some of these novel predictions are confirmed.
Since assumptions are changeable, methods are changeable.
Aristotelian-Medieval Method: Assumptions
1. Nature of Things: A thing has its nature, an indispensable (substantial) quality that makes a thing what it is.
2. Intuition Grasps Nature: The nature of a thing can be grasped intuitively by an experienced person.
Aristotelian-Medieval Method: A proposition if acceptable if it grasps the nature of a thing through intuition schooled by experience or it is deduced from the general intuitive propositions.
Scientific Change: The Laws (These are NOT accepted and NOT used but are pursued)
The Third Law: Method Employment
A method becomes employed only when it is deducible from other employed methods and accepted theories of the time. This is true for all cases of method employment. (This is to do with how methods change; nothing to do with changes in theories)
The Zeroth Law: Compatibility
At any moment of time, the elements of the scientific mosaic are compatible with each other.
The First Law: Scientific Inertia
An element of the mosaic maintains its state in the mosaic unless replaced by some other elements. The theory or method will stay in the scientific mosaic until it is replaced.
The Second Law: Law of Theory Acceptance:
In order to become accepted into the mosaic, a theory is assessed by the method actually employed at the time. (How theories change. Nothing to do with changes in methods)
1st Law for Theories:
An accepted theory remains accepted unless replaced by other theories.
A theory is assessed only if it attempts to become accepted, not when already accepted.
Contextual Appraisal Theorem:
Theory assessment is an appraisal of a proposed modification of the mosaic by the method of the time.
A theory becomes rejected only when other theories that are incompatible with the theory become accepted.
A new theory is often incompatible with some parts of the mosaic.
How do we decide if two theories are or aren’t compatible?
Method: Three Components
Method: A set of criteria for employment in theory evaluation. A method can consist of different criteria.
1. Acceptance Criteria: Criteria for determining whether a theory is acceptable or unacceptable.
2. Demarcation Criteria: Criteria for determining whether a theory is scientific or unscientific.
3. Compatibility Criteria: Criteria for determining whether two theories are compatible or incompatible.
Since methods are changeable, compatibility criteria are also changeable.
Formal Science: Inconsistency-Intolerant
In formal sciences, we are inconsistency intolerant.
Empirical Science: Inconsistency-Tolerant
In empirical sciences, we are inconsistency tolerant.
Two analytic propositions are compatible only when they are mutually consistent.
Two synthetic propositions can be compatible even if they contradict each other. Contradictory propositions cannot both be true at the same time, but they can both be quasi-true (truthlike). Synthetic propositions can be only quasi-true. Only analytic propositions can be strictly true.
What happens when two incompatible theories satisfy the requirements of the method of the time?
Mosaic Split Theorem:
When two incompatible theories meet the requirements of the method at the time, the mosaic splits in two.
There is no universal fixed method of science <- This is true but it doesn’t mean you cannot have a theory on how the methods of science change (general theory of scientific change).
Particularism is wrong in insisting that there is nothing permanent in science. All cases of scientific change seem to obey certain logic; there is a certain pattern. Scientific change is a rational (law-governed) process.
Given that even time-honoured theories become rejected, can we still claim that our current theories succeed in describing the world?
Do our best scientific theories correctly describe the nature of the external (mind-independent) world?
Do our theories describe things as they actually are? Or do they only describe things as they appear to us?
Yes -> Scientific Realism: Our best scientific theories correctly describe the nature of the mind-independent world. Thus, we can both use theories in practical applications and accept them as best available descriptions of the external world.
No -> Scientific Anti-Realism (Instrumentalism): We do not know whether our correctly describe the nature of the mind-independent world. Note this does not say our theories fail to describe the nature of the mind independent world. Thus, we shouldn’t accept theories, but should only use them in practical applications, for theories are nothing more than useful tools. It doesn’t say we can’t know (at some time), we don’t know currently.
Scientific Realism, nowadays, comes in two basic flavours: Structural Realism: Although our theories about the natures of the unobservable entities can be false, our knowledge of the relations between them is true.
Entity Realism: Our knowledge about unobservable entities is true, although our theories concerning relations between those entities can be false.
Structural Realism: Although our theories about the natures of the unobservable entities can be false, our knowledge of the relations between them is true. Although in the future, our theories might be rejected, most of the equations will survive in one way or another.
Entity Realism: Our knowledge about unobservable entities is true, although our theories concerning relations between those entities can be false. Although in the future, our knowledge about relations may change, the accepted entities will remain accepted. We may learn new things about these entities but the entities themselves remain.
Both of those conceptions face serious problems.
Structural Realism is flawed, for our knowledge about relations is not immune to change. It is a historical fact that our knowledge about relations between entities is changeable.
Entity Realism is flawed, because for the list of entities that presumably populate the world has changed through time.
Both Structural Realism and Entity Realism adopt a selective approach.
Scientific Selective Approach (Selective Optimism): Although strictly speaking all (even our best) theories are false, some aspects of our best theories are true. This is based on Infallibilism.
The selective approach is in conflict with the position of Fallibilism -> No synthetic proposition is infallible. Empirical knowledge cannot be absolute certain.
The selective approach is incorrect, because every part of our empirical theories may turn out to be false.
Since all theory assessment is comparative, the only thing we can hope to know is whether our current theories are closer to the truth than our past theories.
Does science actually progress towards truth?
In many fields of human endeavor, the notion of progress is problematic. E.g. music, art, sports, etc.
Is it true that our theories simply provide us with different pictures of reality, none of which is better or worse? Or can we say that our theories gradually improve our understanding of reality?
If scientific theories are merely different forms of approaching the world then there is no scientific progress, and the external world remains unknown. However, if our theories gradually improve our understanding of reality, then there is scientific progress and the external world Is knowable, albeit always imperfectly.
Does science actually progress towards truth?
Yes: Progress Thesis: Science progresses towards truth. i.e. scientific theories provide increasingly correct descriptions of the external truth. No: No-Progress Thesis: We can’t know whether science progresses towards truth, i.e. whether some descriptions are closer to truth than others.
Scientific Community: Progress
The scientific community always holds that compared to previously accepted theories, the currently accepted theory provides a better description of its object.
The question is not what scientists themselves think about progress but whether they actually succeed in approximating the truth. We need another approach.
Empirical Success of Science: Science has been empirically successful; predictions of our theories become increasingly precise and accurate.
Reality affects Phenomena: The world of phenomena (the world of experiments and observations) is somehow affected by the external world.
Progress Thesis: Science has been progressing towards truth; the explanations provided by scientific theories are increasingly correct. The idea in simple; had our theories been far from the truth, their empirical success would have been a miracle.
No-Miracles Argument: Since our theories become increasingly accurate and precise in their predictions, this can only indicate that we gradually uncover the inner structure of the world. We have these successes in science. We believe these successes are not due to complete chance, but we are actually getting closer to the truth.
But how about constant ontological mistakes? Accepted Ontology is the accepted views on the types of entities and interactions that populate the world.
Pessimistic Induction Argument (Pessimistic Meta-Induction Argument):
Ontological “Mistakes”: The ontologies of past theories are usually considered mistaken from the perspective of later theories. Progress Definition: A process of acquiring increasingly correct descriptions of the world (including its ontology) -> Increasingly correct ontologies.
Because we keep making mistakes, we’re likely to keep making mistakes in the future. So we can’t ever know if we’re progressing towards the truth. Our current theories/ontologies will probably be proven to be incorrect as well.
No-Progress Thesis: We can’t know whether science progresses towards truth, i.e. whether some descriptions are closer to truth than others. Note
that, this doesn’t deny that the predictions get more precise and accurate. But this doesn’t mean we’re getting closer to the truth.
What is wrong with this argument?
We do not think that our current ontology is absolutely correct, merely that it is a better approximation than the ontologies of the past. Thus, we can’t say that past ontologies were absolutely fast; they were merely not as correct as our current ontology.
We believe in a succession of ontologies, where future ontologies try to improve on past ontologies. Therefore, the premise of Ontological “Mistakes” is wrong. The whole Pessimistic Induction Argument is wrong.
The major argument for Progress Thesis is No-Miracles Argument. The major argument for No-Progress Thesis is the Pessimistic Induction Argument (which is flawed).
Does science progress towards truth? -> Professor’s view is Yes. You can’t say that this is the accepted view. It is open question for debate.