Philosophy of Science: Knowledge and Evidence

• Study notes for the provided text are below:

Why Philosophy of Science?

• Critically evaluates knowledge, data, evidence, and truth
• Critical thinking helps question science's subjectivity
• Unlike art, science aims for reliable, testable, evidence-based knowledge
• Helps distinguish reliable information from misleading claims for entrepreneurs and citizens

Scientific Knowledge

• Belief: acceptance as true
• Justification: good reasons or evidence
• Truth: belief being true and verified
• Needs proper research and recognition of what is unknown
• Requires skepticism and self-revision for true beliefs

Defining Science: Key Characteristics

• Naturalism investigates natural phenomena using empirical methods, avoiding supernatural causes

Evidentialism

• Beliefs must be based on evidence from ongoing, critical investigation
• Evidence must be replicable and reviewed

Falsification

• Scientific claims must be testable and falsifiable, with a willingness to abandon claims proven wrong

Accountability

• Science is a social process involving peer review, collaboration, and competition, where scientists hold each other accountable

Pseudoscience

• Pretends to be science but lacks legitimacy
• Uses cherry-picked or non-transparent data and studies non-natural phenomena
• Claims are non-falsifiable, meaning they cant be proven wrong

Science

• Provides natural explanations for phenomena, based on empirical evidence and involves scientific communities
• Open to revision/falsification, using mathematical tools when appropriate

Importance of Science

• It satisfies practical goals like medicine, innovation, and energy
• Seeks pure knowledge through basic research
• Applied research aims to solve real-world problems, often profit-driven
• Basic and applied research work synergistically
• Science is distinct from literature, religion, or art and aims to solve real-world problems
• Google creates surface-level knowledge, and cause the illusion of understanding

Defining Science in 3 Ways

• History traces science back to the Scientific Revolution (1550–1700) and the Islamic Golden Age, noting blurry origins overlapping with religion/philosophy
• Studies natural phenomena through regular, observable processes that must be detectable with senses or tools
• Natural explanations that don't assume supernatural causes like methodological and ontological naturalism
• Uses empirical investigation, evidentialism, quantitative and qualitative methods
• Must be falsifiable and built on social and institutional collaboration

Climate Change

• Scientists know humans are altering the climate from temperature records, ice cores, and ocean patterns
• Climate models show human impact, expertise requires input from multiple scientific disciplines
• Public concern lags behind research due to misinformation, complexity of evidence, gradual impact, and illusion of understanding

Expertise

• Not all scientists can speak on all topics
• It is domain-specific, and genuine expertise involves credentials, peer recognition, and field relevance

Recognizing Experts

• Look for degrees & affiliations, peer-reviewed publications, consensus within the expert community, and support from other recognized experts

Knowledge

• Defined as justified, true belief.

Basic Research

• Seeks knowledge for its own sake

Applied Research

• Uses science to solve real-world problems

Naturalism

• Focuses on natural observable phenomena

Evidentialism

• Beliefs supported by ongoing evidence

Falsifiability

• A claim must be testable and disprovable

Pseudoscience

• Imitates science by lacks legitimacy/falsifiability

Evidence & Hypothesis Testing

• Scientific knowledge aims for universal claims
• Experience is limited, leading to testing problems

Roles of Evidence

• Verification & Confirmation supports or confirms a hypothesis based on positive results
• Falsification disproves a hypothesis with a single counter-instance

Empiricism

• All scientific beliefs are justified by experience, observation, experimentation, and data
• Justifies confirmation and falsification

Structuring Hypotheses for Falsifiability

• Should be framed to be proven wrong
• Good science attempts to disprove theories

Risky Hypotheses

• Easier to falsify, improving science
• Survival strengthens, not proves, theories

Vienna Circle - Confirmation

• Key Members: Carnap, Hempel, Neurath, Hahn, Gödel

Key Ideas

• Empirical experience is the only valid basis for science
• Knowledge is legitimate if empirically verified Science is descriptive, not evaluative, and philosophy should function like logic/math without speculation

Reliability

• Science is reliable because of its empirical base, which allows testing and observation
• Limits are synonymous with the limits of what can be verified

Counting Evidence

• Evidence needs public, structured, observable, repeatable, and intersubjective elements

Challenges in Verification

• Hypotheses rely on other assumptions (instruments, conditions, prior data)
• Can't isolate a single claim in testing; evidence doesn't confirm in isolation

Paradox of Confirmation

• What counts as real confirmation, logically yes, but is intuitively wrong

Problem of Induction

• We can't observe every case
• Limited experience prevents conclusively verified claims
• Example: the discovery of a black swan in Australia

Popper - Falsification

• Verification is flawed, emphasizing disproving
• Claims it has clear criteria to separate science from pseudoscience
• Defines corroboration as a theory the has repeatedly survived falsification attempts and defines that it isn't correct, just not yet falsified and improves the more risky and tested it gets

Verification vs. Corroboration

• Verification looks for repeated confirmation vs. corroboration seeks disconfirmation testing bold testing
• Verification tries to prove claims right vs. corroboration tries to falsify claims
• Verification sees Positive evidence = confirmation & believes evidence confirms truth vs. corroboration sees Surviving negative evidence = strength & evidence can only refute but not confirm

Auxiliary Hypotheses

• No theory is tested alone, it always includes assumptions
• Disproof may reflect error and may reflect error in any part of the system

Underdetermination

• Empirical evidence might equally support multiple theories
• Shows that evidence alone can’t always pick the "true" theory and questions the objectivity

Intuition

• Confirms intuition is not equal to logic

Hypothesis

• Tested in groups

Theories

• Not always equivalent in terminology

Empiricism

• Cannot always determine the truth

Knowledge

• Comes from experience/observation

Verification

• Supporting hypotheses through positive results

Falsification

• Disproving hypotheses through counterexamples

Risky Hypotheses

• Bold and easily falsified

Auxiliary Assumptions

• Needed for testing

Corroboration

• Hypothesis survives multiple falsifications

Underdetermination

• Multiple theories explain evidence

Positivism

• Knowledge must be verified

Confirmation

• Has a logical mismatch and an intuitive judgement

Goal & Result

• Confirm & verify hypotheses by gathering evidence
• To disprove models by finding exceptions

Empiricism Based

• Grounded in experience and uses objective experiments

Key Problems

• No general claims can be made from a limited experiences and must rely of falsification to find and test disproving claims

Scientific Testing

• It is more useful in industrial and early-stage environments and should challenge assumptions

Popper Pseudoscience

• Avoids verification but can only claim to be falsifiable

Data Design

• Testing experiments and data can confirm falsification

Exam

• Recognize that support claims and theories aren't always equal

Big Question

• How can we trust scientific knowledge that we can't prove

Limited

• Data that uses conclusions does not fully preserve the truth and use probability

Scientific Reasoning

• Is not fully certain and relies on induction

Scientific Claims

• Degrees of testing and not always based on evidence

Belief Strength

• Is due to disagreements during testing

Bayes Theorem

• Supports a solution to the hypothesis

Term Meaning

• Claims that induction is never certain

Subjectivity

• Has benefits but priors can be based on past events

Testing Structure

• Theory is only true if they can be observed

Uniformity

• Past future events is not conclusive