PhilSci - Exam Questions

All 49 exam-style questions for Philosophy of Science, one set per note. Each answer opens with the key points a grader looks for, then a full model answer. Write or say your answer before revealing. The short drilling cards are on the flashcards page.

Back to the Philosophy of Science home.

L01 Introduction and Logical Empiricism

Exam questions

Mock exam Q1: "Explain briefly the verifiability criterion of meaning, and its significance for the problem of demarcation." (10 points)

Key points: Meaningful (scientific) iff verifiable in principle by observation or experiment; meaning of a statement is its method of verification; unverifiable statements are pseudo-statements; theories made of them are pseudo-science; pseudo-science (like metaphysics) seems to say something but is literally meaningless.

Follow the official key:

1. The criterion. A statement is meaningful (scientific) if it is verifiable: its truth can in principle be determined through observation or experiment. Slogan: “the meaning of a statement is its method of verification”.
2. The other kind of meaningful statement. Analytic statements (logic and mathematics), true in virtue of form, saying nothing about the world.
3. Demarcation. Theories whose statements are not verifiable are made of pseudo-statements, so they are pseudo-science: they seem to say something (like metaphysics), but they are literally meaningless.
4. An example. Not in the key, but have one ready: the arthropod reduction, or Heidegger’s “The Nothing itself noths” against $\neg \exists xP(x)$.

Contrast worth one sentence: Popper’s falsifiability separates science from non-science without calling non-science meaningless.

Explain Carnap's diagnosis of Heidegger's "Das Nichts selbst nichtet". What exactly is wrong with it, and what does Carnap conclude?

Key points: “Nothing” treated as a name and a verb; logically “nothing is $P$” is $\neg \exists xP(x)$; type confusion not translatable into logical form; meaningless pseudo-statement rather than false; philosophy exposes pseudo-statements.

• Source: Heidegger’s 1929 Freiburg inaugural lecture What Is Metaphysics?; Carnap’s attack is in “The Elimination of Metaphysics Through Logical Analysis of Language” (1932).
• The sentence treats “nothing” as a name of a thing and then as a verb. In modern logic it is neither. “Nothing is $P$” is $\neg \exists xP(x)$: it is not the case that some $x$ is $P$. There is no object called Nothing that could do anything.
• This is a type confusion: grammar allows the sentence, but it cannot be translated into a logically correct form.
• Conclusion: the sentence is not false but meaningless, a pseudo-statement. Philosophy’s job is to expose such pseudo-statements, not to treat them as deep discoveries.

Describe the syntactic conception of scientific theories (the received view) and illustrate it with kinetic theory.

Key points: Theory as sentences in a formal logical language; axioms, derived sentences, correspondence rules; theoretical vs observational terms; kinetic theory: rules for $P$ and $T$ turn the derived equation into $PV=NkT$; normative ideal.

• A theory is a collection of sentences in a formal logical language: (1) axioms, read as basic laws; (2) derived sentences, obtained by formal deduction; (3) correspondence rules linking theoretical terms to observational terms.
• Essential: the split between theoretical terms (atom, electron: not directly observable) and observational terms. Theoretical terms get their meaning only through correspondence rules.
• Kinetic theory. Axioms: gases are molecules in motion; energy and momentum are conserved; molecules are elastic and Newtonian. Correspondence rules: pressure is the mean force of molecules on the walls; $kT=\frac{2}{3}⟨\frac{1}{2}m{v}^2⟩$. Derivation: $PV=\frac{2}{3}N⟨\frac{1}{2}m{v}^2⟩$, and substituting gives the ideal gas law $PV=NkT$.
• It is a normative ideal: a rigorous standard for evaluating scientific knowledge, not a description of what labs actually do.

Compare the syntactic and semantic conceptions of theories. Why was the semantic view proposed, and does the distinction survive?

Key points: Syntactic: sentences plus correspondence rules; semantic: class of mathematical models; motivations: no axiomatization, multiple formulations, models in practice; Halvorson, Lutz, Frigg: views collapse or are complementary.

• Syntactic: theory = set of sentences (axioms, derivations, correspondence rules), tied to observation by a vocabulary distinction.
• Semantic: theory = a class of mathematical models representing possible systems, tied to targets by conditions of application.
• Motivations for the semantic view: (1) many theories are not axiomatized; (2) one theory has several equivalent formulations, so identifying it with one is wrong; (3) scientists actually work with models.
• Does it survive? Halvorson (2012, 2013) argues the semantic view, if plausible, is syntactic; replies from Glymour (2013) and van Fraassen (2014). Lutz (2017): the debate does not capture any significant differences. Frigg (2022): syntactic is too strict (science needs models and natural language), semantic also needs a language to specify structures. So they collapse into each other or are complementary.

Link to original

L01b Popper and Lakatos

Exam questions

Mock exam Q2 (10 points): Imre Lakatos analyses the development of science in terms of "scientific research programmes". Describe the components of research programmes, and explain what, according to Lakatos, scientific progress consists in.

Key points: Hard core; protective belt of auxiliary hypotheses; positive heuristic; negative heuristic; progressive leads to new discoveries, degenerating only accommodates known facts.

Follow the official key: “(1) Hard core; (2) protective belt with auxiliary hypotheses; (3) positive heuristic; (4) negative heuristic. Progressive programme: leads to new discoveries. Degenerating programme: only accommodates known facts.”

1. Unit of appraisal: a research programme, a series of theories over time sharing a common structure, rather than an isolated hypothesis.
2. Hard core: the fundamental assumptions that give the programme its identity (Newton: three laws of mechanics plus the law of gravitation), held irrefutable by methodological decision.
3. Protective belt: auxiliary hypotheses, initial conditions, theories of the instruments. It meets the evidence and is what gets modified when an anomaly appears.
4. Positive heuristic: the problem-solving machinery, a partially articulated plan for developing the belt (which models to build, which anomalies to digest). Example: the Newtonian checks atmospheric refraction, light in magnetic storms, may invent an unknown planet and calculate its position, mass and velocity.
5. Negative heuristic: never direct modus tollens at the hard core; absorb anomalies in the belt.
6. Progress: cannot be “unrefuted”, since all programmes grow in an ocean of anomalies. Progressive = predicts novel facts (Halley’s comet returning after seventy-two years; Einstein’s eclipse star separations). Degenerating = theories fabricated only to accommodate known facts, lagging behind them (Marxism, theory of imperialism cooked up after the event).
7. Revolutions: scientists desert a degenerating programme for a progressive rival.

Losing marks: calling the hard core unfalsifiable in principle (it is protected by decision), or treating every belt modification as ad hoc.

, for the theory is correct." Would, and-or could, this be acceptable for Popper? Explain your answer.

Key points: No, if the measurement was correct: theory falsified, Einstein dogmatic; perhaps the measurement was wrong; Duhem-Quine: blame can always go elsewhere; acceptable only with independent reasons for rejecting the measurement (no ad hoc hypothesis).

Follow the official key:

1. No, assuming the measurement was correct: the theory is falsified and has to be rejected. Einstein shows a dogmatic attitude. Eddington’s test is exactly the risky prediction that makes relativity scientific: it forbids the null result everyone before Einstein expected.
2. Einstein’s move is the Marxists’ move: re-interpreting to escape refutation, which rescues the theory only at the price of destroying or lowering its scientific status.
3. However, perhaps the measurement was not correct. The prediction comes from the theory plus auxiliary assumptions (telescope optics, photographic plates, data reduction, star identification). Given the Duhem-Quine thesis, scientists can always put the blame elsewhere and retain the theory.
4. For Popper this is acceptable only if there are independent reasons for rejecting the measurement result, reasons other than that the theory would otherwise be refuted. Without them it is an ad hoc hypothesis (a conventionalist stratagem).
5. Einstein as quoted gives no independent reason, only confidence in the theory, so the reply is not acceptable.

Losing marks: stopping at “no”; letting Duhem-Quine excuse Einstein without the independent-reasons condition; saying Popper thinks a falsified theory is proved false and gone forever.

State Popper's criterion of demarcation and use it to classify Einstein's relativity, astrology, Marx's theory of history and psychoanalysis. Why do the last two fail in different ways?

Key points: Falsifiability as the criterion; Einstein scientific (risky prediction); astrology fails (vague prophecies); Marxism testable, falsified, then rescued by a conventionalist twist; psychoanalysis never testable.

• Criterion: “the criterion of the scientific status of a theory is its falsifiability, or refutability, or testability.” A theory must be capable of conflicting with possible, conceivable observations.
• Einstein: scientific. The 1919 eclipse prediction was risky: had the shift been absent the theory would simply have been refuted.
• Astrology: fails. Astrologers were impressed by confirmations and unimpressed by unfavourable evidence, and made prophecies vague enough to explain anything: “in order to escape falsification they destroyed the testability of their theory”.
• Marxism: fails, but not originally. Early versions (the coming social revolution) were testable and in fact falsified; followers re-interpreted theory and evidence to agree, a conventionalist twist that made it irrefutable.
• Psychoanalysis (Freud, Adler): simply non-testable from the start. No conceivable human behaviour could contradict it (the two men and the child).
• So Marxism was once science and was destroyed by its defenders; psychoanalysis never made a risky claim.

What is Lakatos's main objection to Popper's falsificationism, and how does the methodology of scientific research programmes answer it without following Kuhn?

Key points: Scientists have thick skins (tenacity); all theories born refuted; Kuhn’s irrationalism rejected; unit becomes the research programme; novel prediction replaces refutation.

• Objection: Popper’s criterion “ignores the remarkable tenacity of scientific theories. Scientists have thick skins.” Faced with contrary facts they invent a rescue hypothesis for a “mere anomaly”, or ignore it and work on other problems.
• Crucial-experiment stories are “fabricated long after the theory had been abandoned”. A Newtonian asked what would make him abandon Newton “would have been exactly as nonplussed as are some Marxists”, so Popper’s test fails to separate Newton from Marx.
• All theories are born refuted (lunar motion and Newton; Kaufmann and relativity), so “unrefuted” cannot mark science.
• Not Kuhn: Kuhn concluded theory change is irrational; then there is no demarcation, no distinction between progress and decay, no objective standard of honesty.
• Answer: shift the unit from the theory to the research programme (hard core, belt, heuristics) and the test from refutation to novel prediction: progressive programmes predict novel facts, degenerating ones accommodate known facts.
• Revolutions are rational: scientists join the progressive programme, with no instant rationality.

Popper and Lakatos both discuss modifying auxiliary hypotheses to save a theory from contrary evidence. Compare how each judges such a modification, using an example.

Key points: Popper judges motive and form, ad hoc lowers status; independent reasons required to blame the measurement; Lakatos: belt modification normal and necessary; judged by later novel facts; same manoeuvre, different outcomes (invented planet, theory of imperialism).

• Popper: the modification is judged by its motive and form. If introduced only to absorb the refutation (ad hoc: for this purpose only, adding no new testable consequences) it is a conventionalist stratagem that lowers scientific status. Blaming the measurement is allowed only with independent reasons.
• Lakatos: belt modification is normal, constant and necessary, since that is what the positive heuristic does. It is judged only by its consequences: did the modified programme go on to predict a novel fact that was found?
• Example: an invented planet to explain an orbital anomaly and the Marxists’ theory of imperialism are the same manoeuvre in Popper’s terms. In Lakatos’s terms one can lead to a novel discovery and the other produced nothing. (Background example, which Lakatos does not name: the planet postulated to explain Uranus’s orbit was found as Neptune, while the same move for Mercury gave Vulcan, never found.)

Losing marks: treating every protective-belt modification as automatically ad hoc.

Contrast Carnap's verifiability criterion with Popper's falsifiability criterion: what each demarcates, its verdict on metaphysics, and the direction of its logic.

Key points: Verifiability demarcates meaningful from meaningless; falsifiability demarcates science from non-science; metaphysics nonsense for Carnap, possibly valuable for Popper; universal laws cannot be verified but can be falsified by modus tollens.

• Verifiability (Carnap, Vienna Circle): a statement is meaningful if its truth can in principle be determined by observation or experiment; “the meaning of a statement is its method of verification”. Demarcates meaningful from meaningless. Metaphysics is nonsense, pseudo-statements to be eliminated.
• Falsifiability (Popper): a statement is scientific if capable of conflicting with possible observations. Demarcates science from non-science. Non-scientific statements are meaningful, sometimes important, sometimes the ancestors of science (Empedocles, Parmenides).
• Logic: verification wants observations to establish universal laws, which is impossible; falsification wants observations able to destroy them, which modus tollens permits.
• Popper: his problem “was neither a problem of meaningfulness or significance, nor a problem of truth or acceptability” but of separating the empirical sciences from all other statements.

Compare Popper, Kuhn and Lakatos on the unit of appraisal, the demarcation criterion, and how theory change happens.

Key points: Unit: theory, paradigm, programme; criterion: falsifiability, none explicit, novel prediction; contrary evidence: refutes, anomaly, absorbed by belt; theory change: crucial experiment, Gestalt switch, desertion of a degenerating programme.

• Unit: Popper, the individual theory; Kuhn, the paradigm and its community; Lakatos, the research programme (a series of theories over time).
• Demarcation: Popper, falsifiability (a potential falsifier specified in advance); Kuhn, no explicit criterion (puzzle-solving tradition, stable community); Lakatos, progressive versus degenerating (novel predictions).
• Contrary evidence: Popper, refutes the theory, which must be rejected; Kuhn, an anomaly tolerated until anomalies pile up into crisis; Lakatos, hits the belt and is digested by the positive heuristic.
• Tenacity: Popper, a vice; Kuhn, functional in normal science; Lakatos, rational and often correct.
• Theory change: Popper, a crucial experiment kills the theory (rational by logic); Kuhn, a revolution or Gestalt switch, no algorithm (charged with irrationalism); Lakatos, desertion of a degenerating programme for a progressive rival, rational but no instant rationality.
• Crucial experiments: Popper, the heart of method; Kuhn, rare and reconstructed after the fact; Lakatos, a myth.

Link to original

L02 Kuhn on Scientific Practice

Exam questions

Mock exam Q4: Thomas Kuhn compares scientific revolutions, which consist in a change of paradigm, to visual "Gestalt switches", in which "what were ducks in the scientists' world before the revolution are rabbits afterwards". Explain why Kuhn makes this comparison.

Key points: revolution is a sudden, holistic shift, like a conversion; observation is theory-laden (Gestalt-inspired); after a revolution the same data form a different whole; support with priming or playing-cards experiment; no theory-neutral observation language (incommensurability).

Follow the model answer (two reasons, both needed):

1. A revolution is a radical, sudden shift. Like a conversion experience, it makes one see the world in a different light all at once, without piecemeal accumulation of new facts. The duck becomes a rabbit as a whole.
2. It expresses his view of observation, inspired by the Gestalt psychologists: observation is theory-laden. What scientists see, and what they take their observations to show, depends on the paradigm they bring to it, so after a revolution the same data are organised into a different whole.

Support with the young woman / old woman priming experiment or the Bruner and Postman playing-cards experiment, and name the consequence: there is no fully theory-neutral observation language to adjudicate between paradigms (incommensurability).

Losing marks: giving only one of the two reasons.

Mock exam Q5: What does Helen Longino mean by the statement that objective scientific knowledge has a "social character"? How does something become "scientific knowledge"? Give some examples of social mechanisms that secure objectivity.

Key points: two or more individuals; critical emendation and modification, each step transformative; knowledge is what survives this, freed from subjective preferences; replication and peer review; treatment after publication and absorption into knowledge.

Follow the model answer:

1. Social character. Knowledge becomes scientific through the participation of two or more individuals, in a process of critical emendation and modification of individual contributions. Each such step is transformative.
2. How something becomes scientific knowledge. By surviving this process: a claim criticised from several points of view and modified in response is freed from the subjective preferences of any single investigator. Objectivity is a property of the community’s practice.
3. Mechanisms: replication of experiments with variations; peer review, which brings another point of view on the phenomena so the interpretation is freed from subjective preferences; treatment after publication (refining new ideas and techniques); absorption into knowledge (subsequent citation, use and modification by others).

For depth: tie it to contextual empiricism (data support a hypothesis only against background assumptions, which may be value-laden, so an individual cannot guarantee objectivity alone) and list the four conditions for transformative criticism: recognised avenues for criticism, shared standards, responsiveness to criticism, equality of intellectual authority.

Describe Kuhn's model of scientific development, illustrate it with one historical case, and explain why it challenges the idea of cumulative progress.

Key points: paradigm, normal science, anomaly, crisis, extraordinary science, revolution; one case (phlogiston to oxygen, or Ptolemy to Copernicus); revolutions are non-cumulative; theory-ladenness and incommensurability block neutral comparison; Kuhn’s five shared values as reply.

• Cycle: a shared paradigm organises normal science (puzzle solving); a persistent anomaly leads to crisis; in extraordinary science rival approaches are explored and standards are contested; a revolution installs a new paradigm, which defines new puzzles, and normal science resumes.
• Case: chemistry went from a preparadigmatic stage (alchemy) to the phlogiston paradigm (Priestley: combustion releases phlogiston) to the oxygen paradigm (Lavoisier: combustion is combination with oxygen). The substance, the explanatory vocabulary and the interpretation of weight changes all changed. Alternatively Ptolemy (geocentric) to Copernicus (heliocentric), where the use of “planet” changed.
• Against cumulative progress: a revolution is non-cumulative. Observation is theory-laden, and paradigms differ in meanings (semantic incommensurability) and standards (methodological incommensurability), so no neutral algorithm compares them and it becomes unclear in what sense the new paradigm is objectively better.
• Close with Kuhn’s reply: five shared epistemic values keep theory choice rational.

Kuhn is often accused of relativism and irrationalism. State the criticisms and explain how Kuhn's account of theory choice answers them.

Key points: irrational social victory, relativism, unclear “paradigm”, no demarcation criterion; five shared values: accuracy, consistency, broad scope, simplicity, fruitfulness; values, not algorithmic rules: imprecise and can conflict; rational disagreement is legitimate and needed for progress.

• Criticisms: paradigm victory is an irrational social victory; no clear demarcation criterion; “paradigm” is unclear (Masterman counted 21 uses in SSR); no objective criteria for progress, hence relativism; Lakatos: Kuhn vindicates “mob psychology”.
• Reply (“Objectivity, Value Judgment, and Theory Choice”, 1977): five shared criteria, accuracy, consistency, broad scope, simplicity, fruitfulness, are “the shared basis for theory choice”. They are objective in the sense that the scientific community shares them.
• They are values, not algorithmic rules: each is imprecise, and they can conflict (more accurate but less simple), so scientists weigh them differently. Every choice mixes shared and individual criteria, and previous experience counts.
• Hence rational disagreement is possible and legitimate, and it is needed for progress: if everyone switched at once, competing possibilities would not be developed or tested.

Can science be value-free? Compare the value-free ideal with the positions of Kuhn, Douglas and Longino.

Key points: value-free ideal: values only in agenda and application; weak ideal: epistemic values allowed in research; Kuhn: epistemic values, theory choice not mechanical; Douglas: indirect role via inductive risk; Longino: values in background assumptions, objectivity through transformative criticism.

• Value-free ideal (Weber, logical positivists): values may enter agenda setting and application (panels 1 and 3 of the triptych), never scientific research itself (panel 2).
• Weak value-free ideal: non-epistemic values allowed in panels 1 and 3, epistemic values allowed in panel 2.
• Kuhn: his five values are epistemic, and they show that internal theory choice is not mechanical.
• Douglas: values have an indirect internal role through inductive risk: under uncertainty, the different costs of false positives and false negatives set how much evidence is required. Values do not become evidence.
• Longino: values can enter the background assumptions that connect data to hypotheses (what is measured, which models, which comparison groups, what counts as explanation). Objectivity is secured socially, by transformative criticism.
• Douglas and Longino are internality positions: non-epistemic values can legitimately play a role inside scientific reasoning.

Link to original

L03 Under-determination

Exam questions

Mock exam Q6: "The under-determination thesis was first formulated by Duhem and later strengthened by Quine. State this thesis in Duhem's own version, and explain it with an example." (half a page)

Key points: premises entail $O$, experiment gives not-$O$; only the conjunction is false, no particular premise; no crucial experiment; hypotheses tested with auxiliary assumptions incl. apparatus; example (telescope or Poincare’s sphere).

Must hit:

1. The formal version. If premises $\{P_1,\dots ,P_n\}$ deductively entail an observation statement $O$, and experiment establishes $O^{\prime }$ where $O^{\prime }$ entails not-$O$, then the conjunction of the premises is false. It does not follow that any particular premise is false.
2. The consequence. There is no crucial experiment that irrefutably decides between two hypotheses, because a hypothesis is only ever tested together with auxiliary assumptions (including the theory of the apparatus).
3. An example. An astronomical theory conflicts with telescope observations: the conflict does not say whether the astronomy or the optics of the telescope is at fault. (Poincare’s sphere is the impressive alternative.)

Losing marks: stating the slogan (“theory is under-determined by data”) instead of the premises-and-conjunction version, or giving no example.

Mock exam Q3: Einstein (1953) said that if the eclipse measurement had not shown the shift, "Then I am very sorry for the dear Lord, for the theory is correct." Would, and-or could, this be acceptable for Popper? (10 points)

Key points: no, if the measurement is correct: falsified, dogmatic attitude; Duhem-Quine: blame can be put elsewhere (measurement, auxiliaries); acceptable only with independent reasons to reject the measurement; otherwise ad hoc, lowering scientific status.

Follow the official key:

1. No, assuming the measurement was correct: the theory is falsified and has to be rejected. Einstein shows a dogmatic attitude.
2. However, perhaps the measurement was not correct. Given the Duhem-Quine thesis, scientists can always put the blame elsewhere (an auxiliary assumption, the measurement) and retain the theory.
3. But for Popper this is acceptable only if there are independent reasons for rejecting the measurement result. Otherwise it is an ad hoc hypothesis, which lowers the theory’s scientific status.

Losing marks: stopping at “no”, or saying Duhem-Quine lets Einstein off without the independent-reasons condition.

Does under-determination show that theory choice is arbitrary? Answer using Duhem, Quine and Laudan.

Key points: Duhem: good sense decides, not formal is not arbitrary; Quine: any statement held true come what may; Laudan: logically possible is not rational, evaluative vacuum; only the egalitarian thesis would make choice arbitrary.

• Duhem: no. Logic alone does not settle which premise to give up, but good sense does: some opinions are “perfectly unreasonable” even though no contradiction forces you to drop them. Choice is not formal, which is different from arbitrary.
• Quine: pushes hardest. Any statement can be held true “come what may” if you make drastic enough adjustments elsewhere (pleading hallucination, even revising logic).
• Laudan: logically possible is not rational. Quine argues in an “evaluative vacuum” and never examines the ampliative rules of theory choice to show they are too weak to decide. Only the egalitarian thesis (every theory is as well supported as every rival) would make choice arbitrary, and nothing in Quine establishes it.
• Verdict: holism shows a theory can be saved, not that it is reasonable to save it.

Explain under-determination using Poincare's sphere, and say what conclusion Poincare draws.

Key points: Euclidean disk with $T=R^2-r^2$, bodies shrink towards the edge; inhabitants find infinite non-Euclidean space; no experiment separates the two; geometry is conventional, no fact of the matter; geometry is an auxiliary assumption (Duhem).

• A world inside a sphere of radius $R$. Description 1: a Euclidean disk with temperature $T=R^2-r^2$ (hottest at the centre, zero at the edge), and every body expands in proportion to $T$, so everything shrinks towards the edge. Description 2: inhabitants, whose rulers shrink with them, never reach the edge and find space infinite and non-Euclidean.
• No experiment separates the two, because every experimental set-up can itself be described in either picture.
• Conclusion: geometry is conventional, neither a priori (against Kant) nor a posteriori. We pick the description that is simplest for us, like choosing metres over yards. There is no fact of the matter.
• Link to Duhem: the geometry is one of the auxiliary assumptions that rides along in every test.

Link to original

L03b GenAI, Writing and Philosophical Learning

Exam questions

De Haro says the central question about GenAI in study is "whether AI strengthens or replaces your own learning". Explain the distinction with examples, and say why the permitted uses are permitted.

Key points: rejects both enthusiasm and prohibition; supporting uses (clarify, examples, test, feedback on own text); replacing uses (outsource reading, submit AI answers, unjudged text, no acknowledgement); permitted uses presuppose prior effort; test understanding, own voice first, transparency.

Must hit:

1. The refusal of both slogans. De Haro rejects both enthusiasm and prohibition: whether AI use is allowed matters less than what a given use does to your learning, and he thinks this has a determinate answer in most concrete cases.
2. Supports learning: clarifying a difficult definition, requesting alternative examples, testing whether you can explain an argument, getting feedback on text you have written.
3. Replaces learning: outsourcing the reading, submitting AI-generated answers as your own, using generated text before you can judge its quality and accuracy, failing to acknowledge AI use.
4. Why: every permitted use presupposes prior effort (you read and got stuck, you can judge whether an example fits, you have an argument to be tested on, you wrote the text). The AI is downstream of your own work. The forbidden list is the same list with the effort removed.
5. De Haro’s conclusion: use AI to test your understanding; develop your own philosophical voice before relying on AI to draft text; be transparent about AI use and responsible for every claim you submit.

Losing marks: listing permitted and forbidden uses without the principle that connects them (prior effort, AI downstream of your work).

What does Gartenberg et al. (2026) show about GenAI and academic publishing, what explains it, and what does it not show?

Key points: more rather than better writing; submissions up, readability down; costs shifted to editors and reviewers; institutional incentive chain; systems-level, one journal, not every AI use.

• Study: submissions and peer reviews at Organization Science, five years from January 2021, manuscripts and reviews both analysed. Central claim: AI tools combined with incentives push research towards more, rather than better, writing.
• Results: submissions +42% since late 2022; Flesch Reading Ease −1.28 SD in January 2026 relative to January 2021. The claim is about the pair of numbers together.
• Peer review: over 30% AI use, desk-rejection rate is 30% higher and revise-and-resubmit decisions are rare; more than 30% of reviews use some AI, and these are harder to read and narrower. Easy text generation shifts costs to editors and reviewers (an externality: the author saves, evaluators pay).
• Explanation (institutional): universities reward publications in top journals, so researchers face pressure to increase output; GenAI reduces the cost of producing text; journals get more weak submissions. Nobody in the chain behaves badly, so the place to attack the diagnosis is the incentive structure (the first link).
• Limits: systems-level rather than individual-level, one journal, and it does not establish that every use of AI makes text worse.

Losing marks: claiming the study shows that any individual author writes worse with AI.

According to De Haro, under what conditions can someone responsibly use AI to help write philosophy, and what follows for a student?

Key points: already an expert; can already write philosophy; both conditions are about detection; 500-page unfinished review chapter; unassisted practice first.

• Two conditions, both necessary: (A) you are already an expert: you can check claims, detect omissions and revise to improve the argument; (B) you can already write philosophy: you recognise unnatural text and weak or incomplete arguments.
• Both conditions are about detecting errors; neither is about generating text. So AI-assisted writing is safe only for someone who could have caught the errors anyway, and the tool is least useful to the person who most wants it.
• Evidence: De Haro’s own review chapter with Hans Halvorson. He supplied outline, argumentation, style and literature on a paper requiring no original ideas, and after 500 pages of conversation it was still unfinished. Expert guidance was needed throughout.
• Ordering claim (the bootstrapping problem): the competence needed to use the tool is acquired by writing without it, so unassisted practice comes first and assistance after. Before that stage, the priority is to develop your own philosophical voice and judgement.
• Philosophy of science specifically: GenAI is good at surface summaries and poor at argumentative nuance, which is what the discipline’s aims (reconstruct arguments, evaluate them, write and criticise by the same standards) require.

Link to original

L04 Scientific Explanation and Understanding

Exam questions

Mock exam Q7: "Carl Hempel's "deductive-nomological model" for a long time was the "received view" of scientific explanation. Briefly summarize Hempel's model, and give at least one argument against the model's being a satisfactory account of scientific explanation." (half a page)

Key points: explanandum deduced from general laws plus particular conditions; law used essentially, empirical content, explanans true; explanation and prediction symmetric; asymmetry objection (flagpole); barometer: correlation is not explanation.

Follow the official key:

1. The model. An explanation is a deductive argument: the explanandum (the phenomenon to be explained) is deduced from an explanans of general laws plus particular facts and initial or boundary conditions.
2. Conditions of adequacy. The explanandum is a logical consequence of the explanans; the explanans contains a general law used essentially; it has empirical content; it is true.
3. Explanation and prediction are symmetric: same logical structure. To explain is to show the phenomenon was to be expected given the laws.
4. Objections (the key lists all of these; one argued well is enough):
   • Asymmetry: flagpole height, sun’s position and optics yield the shadow’s length (explains); shadow length and the same laws yield the pole’s height (predicts, does not explain). Both satisfy every D-N condition, so the model is not sufficient.
   • Irrelevance: salt hexed by a magician dissolves in water. Valid, lawlike, true, and the hex explains nothing.
   • Apparently missing causation (correlation or expectation $\ne$ explanation): a falling barometer predicts a storm but does not explain it, since both have a common cause (the drop in atmospheric pressure).
   • Explanations do not always involve laws: citing a causally relevant fact can be enough (“the window broke because the ball hit it”).

Lead with the asymmetry (the strongest), and have the barometer ready as the second, since it is named in the official answer.

Causation or unification: which gives the better account of scientific explanation? Use the flagpole, Salmon, Kitcher and Godfrey-Smith.

Key points: Salmon: explanation situates the explanandum in the causal nexus, solves the flagpole; causal costs: Hume’s problem, too narrow; Friedman and Kitcher: unification reduces brute facts, covers non-causal cases; Godfrey-Smith: not competitors, contextualism about standards.

• Causal (Salmon 1984): explanation situates the explanandum in the causal nexus, showing the mechanisms that produce it. Solves the flagpole at once: the pole causes the shadow. Costs: it inherits Hume’s problem (telling causation from correlation), and it may be too narrow, since explanations in quantum theory or of a law by a more general law are hard to phrase causally.
• Unification (Friedman 1974, Kitcher 1981, 1989): explanation fits phenomena into a broader pattern, using the same patterns of derivation again and again and reducing the number of brute facts. Handles non-causal explanation. On the flagpole, Kitcher says causal talk summarises deeper asymmetries in unification: deriving shadows from poles belongs to a far wider pattern.
• Godfrey-Smith: treating them as competitors is the mistake. Salmon came to accept unification, Kitcher came to accept causation. He goes past this pluralism to contextualism: there is no single special explanatory relation (or fixed short list), and standards of good explanation partly depend on the scientific field and period (following Kuhn). Not “anything goes”: a standard can embed a factual error.

Is explanation internal to science or external to it? Contrast van Fraassen with Godfrey-Smith.

Key points: both treat explanation as context-dependent; van Fraassen: three-term relation, external, not an epistemic aim; Godfrey-Smith: internal, standards vary by field and period; stake: inference to the best explanation and unobservables.

• Shared ground: both say explanation varies with context.
• Van Fraassen (The Scientific Image, 1980): explanation is pragmatic and external. It is a three-term relation between theory, fact and context; an explanation is an answer to a why-question. It counts among reasons to accept a theory for some purpose but adds nothing to our beliefs about how theory relates to world. So it is not an epistemic aim of science: “scientific explanation is not (pure) science but an application of science”, “a use of science to satisfy certain of our desires”.
• Godfrey-Smith: explanation is thoroughly internal. Assessing explanatory power is an important part of scientific reasoning, but each field (and period) sets its own standards.
• Why it matters: van Fraassen’s constructive empiricism needs explanation to be non-epistemic, or inference to the best explanation would push him to believe in unobservables (Musgrave’s reply targets exactly this).

Is scientific understanding anything more than having an explanation? Answer with Hempel, the reductivists, and De Regt and Dieks.

Key points: Hempel: understanding epistemically irrelevant; reductivists: Khalifa, Lipton, Trout; De Regt and Dieks: understanding as skill (useability); two senses of pragmatic, meso-level standards, CUP and CIT; intelligibility standards vary historically.

• Hempel (eliminativist): understanding belongs to the psychological, “pragmatic” side of explanation, with pragmatic meaning subjective. It is epistemically irrelevant, a topic for psychology.
• Reductivists: explanation is understanding enough. Khalifa: understanding is a form of knowledge reducible to explanation (still philosophically interesting). Lipton: “simply more knowledge: knowledge of causes”. Trout: the feeling of understanding is a product of cognitive biases.
• De Regt and Dieks: understanding adds a third requirement, useability: the skill to use a theory to build models ($T\sim M\sim P$) that explain phenomena. This is an ability, not knowledge.
• Against the subjectivity charge: separate “pragmatic” as subjective from “pragmatic” as useable for aims; the second is objective (inter-subjective). Standards of intelligibility are set at the meso-level of scientific communities. CUP and CIT make it testable: can scientists recognise qualitatively characteristic consequences of $T$ without exact calculation?
• Evidence: standards of intelligibility vary historically (Huygens on Newton’s action at a distance) and between contemporaries (Schrödinger against Heisenberg and Pauli, 1926).

Losing marks: using “pragmatic” without saying which sense.

Link to original

L05 Scientific Realism and its Critiques

Exam questions

Mock exam Q8: "One of the arguments discussed by van Fraassen and Musgrave, in their debate on scientific realism, is about scientific explanation. Van Fraassen attacks the realist demand for explanation, and argues that explanation is not one of the aims of science. Musgrave then responds that van Fraassen tacitly conflates 'realism with essentialism, ... the demand for explanation with the demand for ultimate explanation'. What does Musgrave mean by this, and how is it a reply to van Fraassen?" (10 points)

Key points: ultimate explanations remove puzzlement; non-ultimate explanations relocate and enhance puzzlement yet are real explanations; better or deeper explanations always possible; science does aim to explain; no unlimited demand for explanation.

Follow the official key (Musgrave, pp. 1102-1103):

1. Ultimate scientific explanations serve to remove puzzlement.
2. Non-ultimate scientific explanations do not serve this pragmatic function: they relocate and enhance puzzlement. Yet they are real explanations.
3. There can always be better or deeper explanations, and scientific theories certainly do aim to give explanations, but there is no unlimited demand for explanations.

How it answers van Fraassen (a reconstruction from the question and the key; Musgrave’s paper has not been checked, so this part is not settled): van Fraassen argues the realist’s demand for explanation cannot be unlimited and concludes explanation is no aim of science. That only follows if “explanation” means ultimate explanation, the essentialist ideal of explanation bottoming out in essences. Since non-ultimate explanations are genuine, science can aim at better and deeper explanations without an unlimited demand for ultimate ones. That removes van Fraassen’s reason for denying explanation is an aim of science, and with it his reason for dismissing IBE.

What does scientific realism claim, and what are the two main arguments for it? Present them and say where an anti-realist pushes back.

Key points: well-confirmed theories approximately true, including about unobservables; conjunction of semantic, epistemic, metaphysical realism; no-miracles argument with Higgs and top quark; NMA as IBE, approximate truth best explains success; pushback: explanation not an aim, van Fraassen’s rival hypothesis, PMI.

• Claim: empirically well-confirmed theories are true or approximately true, including what they say about unobservables. This goes beyond empirical adequacy.
• It is the conjunction of three theses: semantic (read theories literally), epistemic (we are justified in accepting them as true), metaphysical (the world is as the theory says; its entities exist). Rejecting any one makes you an anti-realist.
• No-miracles argument (Putnam): “Realism is the only philosophy that does not make the success of science a miracle.” Novel predictions such as the Higgs boson and the top quark, made on theoretical grounds before any experiment, would be incredible coincidences if the theories were false.
• IBE: the NMA is an inference to the best explanation. Explanandum: theories successfully predict experiments never performed. Best explanation: their approximate truth. Conclusion: sufficiently successful theories are (probably) approximately true.
• Pushback: anti-realists often deny explanation is an aim of science, so “best explanation” gives no reason to believe. Van Fraassen offers a rival hypothesis (we believe the best explanation is empirically adequate). Laudan’s PMI shows success has gone with falsity before.

State van Fraassen's constructive empiricism and explain how it survives Maxwell's objection that the theory-observation distinction cannot be drawn sharply.

Key points: aim is empirical adequacy, acceptance is belief only in adequacy; semantic realist, epistemic anti-realist, agnostic about unobservables; Maxwell’s glasses-microscope continuum; category mistake: terms versus entities; “observable” vague but usable, only belief hangs on it.

• Constructive empiricism: “Science aims to give us theories which are empirically adequate; and acceptance of a theory involves as belief only that it is empirically adequate.” Contrast realism: the aim is “a literally true story”, and acceptance involves belief that the theory is true.
• Position: semantic realist (claims about unobservables are literally true or false), epistemic and metaphysical anti-realist. Belief about unobservables is “up to us”. He is agnostic about electrons; he does not say they do not exist.
• Maxwell’s threat: glasses, window, microscope lie on a continuum with no principled cut-off. Fatal to the positivist, who tied meaning (all-or-nothing) to the line.
• Three moves: (1) take theories literally in all respects, dropping verifiability; (2) the distinction is a category mistake: theoretical/non-theoretical terms (language) versus observable/unobservable entities (objects); (3) “observable” is vague but usable, like “bald”.
• Why it works: only epistemic commitment hangs on the line, so a vague boundary is harmless. Clear cases on each side (a table, an electron) are all the argument needs.

Losing marks: calling van Fraassen a positivist, or saying he denies unobservables exist.

Present Laudan's pessimistic meta-induction against convergent realism, and explain Saatsi's two readings of it.

Key points: convergent realism: approximate truth, preserved reference, explained success; PMI: successful theories with non-referring terms (aether, phlogiston), induction over theories; three problems rooted in approximate truth; Saatsi reading 1: current theories likely false; Saatsi reading 2: success not a reliable indicator of truth.

• Target, convergent realism: mature theories are approximately true, refer and preserve reference across theory change, and new theories explain the success of old ones, so science converges on the truth.
• PMI: many once empirically successful theories are now considered false; they took terms like “aether” and “phlogiston” to refer, and we now take those terms not to refer. By induction over theories, present successful theories may well turn out false too, so we are not justified in believing theories true or their posits real.
• Three problems (Laudan 1981): (1) no good definition of approximate truth; (2) no justification for the realist’s link success → approximate truth → reference; (3) central terms of once-successful theories do not refer, so those theories were not approximately true, yet were successful. All three stem from approximate truth.
• Saatsi (2005): reading 1, an induction on the past (current theories are likely false); reading 2, a timeless argument that undermines the NMA by showing success is not a reliable indicator of truth. Reading 2 is more modest and harder to resist: even if our situation differs from the 1800s, past successful-but-false theories make the NMA’s “miracles” less miraculous.

Losing marks: treating the PMI as an induction over observations, or as just “science has been wrong before” without the target link success → approximate truth → reference.

Link to original

L06 Scientific Realism and the Pessimistic Meta-Induction

Exam questions

Practice question (not from the mock): Psillos responds to Laudan's pessimistic meta-induction with a divide et impera strategy. Explain it with Maxwell's theory and the aether, and explain why Stanford thinks it gives realism only a "Pyrrhic victory".

Key points: Causal core versus idle constituents, core retained; aether idle in Maxwell’s theory; anachronism: Maxwell held the aether essential; convergence guaranteed, no prospective criterion; dilemma: past beliefs or repeated misidentification.

• PMI: once-successful theories posited entities (the aether) now thought not to exist, so success does not warrant approximate truth.
• Divide et impera: theories have parts. The causal core explains the phenomena and generates success, and is typically retained; idle constituents play no role in success and typically disappear. Success confirms only the core.
• Example: in Maxwell’s theory the equations and mechanisms of explanation are the core and do not require the aether; the aether is idle, so its loss leaves the successful part intact.
• Stanford: the split is anachronistic (Maxwell regarded the aether as essential); convergence on the core is virtually guaranteed, since whatever current theories retain counts as core by definition; and without prospectively applicable criteria we cannot apply it to our own theories, which future scientists will judge the same way.
• Pyrrhic victory: the realist must either claim knowledge of past scientists’ beliefs and intentions, or say they repeatedly misidentified the parts that made their theories successful.
• Evaluation point: De Haro replies that convergence is not guaranteed, since none of the original terms need survive.

Explain De Haro's extensional scientific realism and how it answers the pessimistic meta-induction, using the aether as the example.

Key points: PMI assumes a single kind of meaning; intension versus extension (Venus); extension fixed by intension plus domain circumstances; extensional equivalence via predictive, material, conceptual correspondence; aether is extensionally the field plus fixed frame, so it refers.

• Diagnosis: the PMI, and the literature from Putnam to Stanford, assumes a single kind of meaning; that is why “aether” seems both to refer and not to refer.
• Two kinds of meaning (Frege, Carnap): intension/sense (linguistic meaning) and extension/reference (the object). “Morning star” and “evening star” differ in sense and both refer to Venus.
• Extensions are determined by the intension plus the circumstances in which a phenomenon is studied (model, parameter values, extra-theoretical facts, approximations), i.e. relative to a domain of application.
• Step 1: successive theories are extensionally equivalent on the domain, shown by predictive, material and conceptual correspondence. Step 2: approximate truth defined via extensions.
• Slogan: we are justified in being realists about extensions, not necessarily about intensions.
• Aether: Maxwell’s aether is extensionally equivalent to the electromagnetic field plus a fixed frame of reference, $v\ll c$; his energy-conservation argument for it was very good. So “aether” refers, while being intensionally distinct from “field”.
• Versus Stanford: no theory parts are distinguished, so no Pyrrhic victory; Stanford’s “repeated, profound” changes miss restricted domains where extensions are continuous.

Present Kitcher's reply to Laudan about the reference of terms like "aether", and Stanford's objection to it.

Key points: Terms do not automatically refer, reference fixed by context; aether as field, empty space or nothing; Stanford: the theories were radically misguided; realism needs approximate truth, reference is only a precondition.

• Kitcher (1993): terms do not automatically refer; look at the specific context. Reference is fixed differently in different cases: “aether” could refer to the electromagnetic field in some cases, to empty space in others, and to nothing in others. So Laudan is wrong that such terms are simply non-referring.
• The mechanism: reference is assigned to individual uses (tokens) of a term, guided by the speaker’s intentions, so some of Priestley’s uses of “dephlogisticated air” can refer and others not.
• Stanford (2006): grants the distinction may make sense, but it misses the point of the meta-induction. Even if central terms refer, they sit in theories that repeatedly turn out radically misguided, which cannot be approximately true.
• Stanford’s words: little comfort if the accounts of those entities “were mistaken about virtually everything except the fact that the entities in question played some causal role in producing observable phenomena”.
• Point: realism needs approximate truth; reference is only a precondition.

How does extensional scientific realism avoid Stanford's "Pyrrhic victory" objection, and how is it parallel to and different from van Fraassen's constructive empiricism?

Key points: Dilemma comes from sorting theory parts; extensional realism sorts no parts, all claimed terms refer; parallel: literal reading, restricted belief; difference: extensions include unobservables.

• Stanford’s objection targets selective confirmation: sorting theories into parts requires knowing past scientists’ intentions or saying they misidentified what made their theories succeed.
• Extensional realism distinguishes no theory parts. It is a linguistic/philosophical theory of how to interpret scientific statements: assuming empirically adequate theories and sound arguments, all terms scientists claim refer, do refer.
• So it never overrules the scientists: it can agree with Maxwell that the aether was essential and real.
• Parallel: van Fraassen reads theories literally but believes only what they say about the observable; De Haro takes literally what the scientist says but believes it only in its extensional, not intensional, meaning.
• Difference: extensions include unobservables (fields, electrons), so extensional realism is a realism; the observable/unobservable line plays no role.

Link to original

Tutorials

Exam questions

Popper: what makes a theory scientific, how does falsifiability relate to verificationism, and which criterion is more accurate?

Key points: Falsifiability is Popper’s criterion; verificationism is about meaning; falsifiability is about demarcation; unfalsifiable claims can be meaningful; the two answer different questions.

• Popper’s criterion: a theory is scientific if it is falsifiable. The mark of pseudo-science is that it explains everything and forbids nothing.
• Different criteria doing different jobs. Verificationism is a theory of meaning: unverifiable statements are meaningless. Falsifiability is a criterion of demarcation: it separates science from non-science.
• Popper is explicit that unfalsifiable statements can be perfectly meaningful, just not scientific. Metaphysics, for Popper, is meaningful and sometimes fruitful.
• Before judging which is more accurate, say that the two answer different questions (meaning versus demarcation).

Losing marks: treating them as two versions of the same idea. That is the intuitive answer, and it is wrong.

Why does Kuhn argue that "cumulative acquisition of novelty is not only rare in fact but improbable in principle" (p. 89), and why does that make scientific revolutions necessary?

Key points: Normal science is puzzle-solving inside a paradigm; novelty appears only as anomaly; novelty arrives by rupture; normal science cannot correct its own foundations; revolution restores puzzle-solving.

1. Normal science is puzzle-solving inside a paradigm, and the paradigm fixes what counts as a legitimate problem and a legitimate solution.
2. Genuine novelty is by definition what the paradigm does not anticipate, so it can only appear as an anomaly, which counts as a failure rather than a contribution.
3. So novelty arrives by rupture rather than by accumulation.
4. Necessity of revolutions: normal science cannot correct its own foundations. Anomalies accumulate until the paradigm’s puzzle-solving capacity is exhausted, and only a change of paradigm can restore it. Revolution is the mechanism by which science changes its mind at all.

Why are arguments in paradigm debates necessarily circular for Kuhn, why are successive paradigms incommensurable, and do you agree?

Key points: Paradigms supply the standards for judging arguments; each side uses criteria the other rejects; paradigms differ in problems, standards and meanings; no neutral vocabulary; pushback: scientists communicate across paradigms.

• Circularity: each side argues from within its own paradigm, and paradigms supply the standards by which arguments are judged. So each side can only defend its position using criteria the other does not accept.
• Kuhn’s comparison (p. 88): “Like the choice between competing political institutions, that between competing paradigms proves to be a choice between incompatible modes of community life.”
• Incommensurability: paradigms differ in problems, standards and the meanings of their terms, so there is no neutral vocabulary in which to state a fair comparison.
• Standard pushback: incommensurability is often overstated. Scientists in practice do communicate across paradigm boundaries and do cite empirical successes both sides recognise.
• Same territory as the mock exam question asking why Kuhn compares revolutions to Gestalt switches.

Lakatos says neither Popper nor Kuhn solves the demarcation problem. Why, what makes a research programme scientific, and how does Lakatos nuance Popper and modify Kuhn?

Key points: Popper naive: scientists keep theories despite contrary facts; Kuhn irrational: no rational criterion for paradigm change; research programme as unit of appraisal; progressive means corroborated novel predictions; series of theories over time plus rational criterion.

• Popper is naive: scientists do not in fact abandon a theory the moment it meets a contrary fact.
• Kuhn makes science irrational: paradigm change is a change in commitment with no rational criterion, so no demarcation is possible at all.
• Lakatos’s move: shift the unit of appraisal from the single theory to the research programme.
• Scientific means progressive: the programme predicts novel facts, some of which are corroborated. A degenerating programme only accommodates facts already known.
• Nuances Popper: appraises a series of theories over time rather than a single theory against a single test, which makes room for the tenacity scientists actually show.
• Modifies Kuhn: keeps the historical unit while restoring a rational criterion for choosing between programmes.

Is scientific value an "objective" property of a theory on Lakatos's account? Give the case for and the case against.

Key points: For: public record of predictions and corroborations; against: no principled cut-off for degeneration; verdict only with hindsight.

• For: progressiveness is assessed against the public record of predictions and corroborations, which is independent of anyone’s opinion.
• Against: judging a programme degenerating requires deciding how long to wait before calling it, and Lakatos gives no principled cut-off. So the verdict is only available with hindsight.

Longino: what produces scientific knowledge, and how can it be objective?

Key points: Community criticism produces knowledge; knowledge survives transformative interrogation; objectivity from science as a social enterprise; four conditions and examples of social mechanisms; not an ethical theory.

• What produces knowledge: the community subjecting individual work to criticism, rather than the individual. Knowledge is what survives transformative interrogation.
• Objectivity is a consequence of science being a social enterprise. It is a property of a community’s practice, and depends on the depth and scope of transformative interrogation.
• It does not differ in kind for non-empirical sciences, against the logical empiricists.
• A full answer also states Longino’s four necessary conditions for objectivity, and the mock exam version of this question (the social character of objective knowledge, how something becomes scientific knowledge) also asks for examples of social mechanisms.
• Contextual empiricism is a theory of scientific objectivity and epistemic methodology. It is not an ethical theory and does not say whether a piece of research is permissible.

Losing marks: conflating objectivity with ethical permissibility. De Haro wrote an unprompted course announcement to correct exactly this confusion.

What is the problem of underdetermination according to Quine, and according to Laudan? Which version would make theory choice arbitrary?

Key points: Quine: empirically equivalent, theoretically inequivalent formulations; underdetermined by all possible observations; Laudan: non-uniqueness versus egalitarian thesis; only egalitarian makes choice arbitrary; Quine does not establish it.

• Quine: our theory of the world is underdetermined by all possible observations. There can be theory formulations that are empirically equivalent (they imply the same observation sentences) but theoretically inequivalent, so no possible evidence decides between them.
• Laudan: the problem is a family of theses that get run together, and he separates them. Non-uniqueness: for any theory there is at least one rival as well supported by the evidence. Egalitarian: every theory is as well supported as any rival.
• Only the egalitarian thesis would make theory choice arbitrary, and Laudan argues nothing in Quine establishes it.

Why does Quine argue that holism (the Duhem-Quine thesis) lends credence to underdetermination, and why does Laudan argue that it does not?

Key points: Quine: many ways to repair the web, so many fitting systems; holism shows logical possibility; the threat needs rational warrant (egalitarian); ampliative rules unexamined, an evaluative vacuum; only harmless non-uniqueness follows.

Quine: if hypotheses face experience only as a whole, a recalcitrant observation can be accommodated by revising any of many sentences in the web. There is always more than one way to repair the system, so it seems there must always be more than one system that fits all the data.

Laudan, reconstructed in four steps:

1. Holism shows that any hypothesis can be retained in the face of any evidence by adjusting something else. That is a claim about logical possibility.
2. Underdetermination in the form that threatens methodology is a claim about rational warrant: that the evidence supports rivals equally (the egalitarian thesis).
3. Getting from 1 to 2 requires showing that the ampliative rules of theory choice are too weak to prefer one repair over another. Quine never examines those rules: he argues in an “evaluative vacuum”.
4. So holism supports at most non-uniqueness (some rival exists), and that is harmless, because choosing among rivals is exactly what methodology is for.

This is the standard reply to anyone who uses underdetermination to argue that theory choice is arbitrary.

What is a "theory" according to Quine? Reconstruct his principle of individuation for theories and say why underdetermination depends on it.

Key points: Formulation (conjunction of axioms) versus theory expressed; same theory: empirically equivalent plus reconstrual of predicates; molecule/electron swap; no reconstrual means different theories, which underdetermination needs.

• Theory formulation: a sentence, in practice the conjunction of a theory’s axioms. Theory: what a formulation expresses.
• Same theory when two formulations are empirically equivalent and one can be converted into the other by reconstrual of predicates: systematically swapping predicates for other predicates or open sentences, as when two physical theories differ only by exchanging the words “molecule” and “electron”.
• Empirically equivalent formulations that no reconstrual can bring into agreement are different theories, and those are the cases underdetermination needs.
• Link to Quine’s “common core” point: if a theory is nothing more than the conjunction of its observation conditionals, rivals share it rather than compete with it.

Which account of explanation is most convincing: causation, unification, or Godfrey-Smith's pluralist and contextualist view?

Key points: Causation (Salmon) and its challenges; unification (Friedman, Kitcher) and the flagpole asymmetry; Godfrey-Smith: no fixed list, field-relative standards; objection: what makes them all explanation; understanding as common aim.

• Causation (Salmon): to explain is to describe what caused the phenomenon. It fixes the flagpole at once, since sunlight and pole cause the shadow. Challenge: it needs an account of causation, which empiricists since Hume regard as suspect, and it seems too narrow for explanations that are hard to state causally.
• Unification (Friedman, Kitcher): to explain is to fit phenomena into a broader pattern, reducing the number of facts accepted as brute. Challenge: recovering the flagpole asymmetry without appealing to causes. Kitcher tries by arguing causal talk is a loose summary of deeper unification asymmetries.
• For Godfrey-Smith: each account handles cases the other struggles with, and their champions conceded ground to each other (Salmon accepted unification, Kitcher accepted causation). He goes further than pluralism: there is no fixed list of explanatory relations, because each field sets its own standards (as Kuhn argued with Newton’s gravity). It is not “anything goes”, since a standard can embed a factual error.
• Against: if “explanation” means different things in different fields, it is unclear what makes them all explanation, and contextualism risks describing practice without saying what makes an explanation good.
• De Regt and Dieks’s answer: understanding is the common aim that unifies the plural forms of explanation.

Is explanation crucial to the inner workings of science, or external to it? Compare van Fraassen's pragmatic account with Godfrey-Smith's view.

Key points: Van Fraassen: external application, three-term relation; Godfrey-Smith: internal to scientific reasoning; both context-dependent; real disagreement: whether explanation is epistemic; stakes: IBE and unobservables.

• Van Fraassen, external: explanation is a three-term relation between theory, fact and context, an application of science to answer our questions. It adds nothing to what we believe about the world.
• Godfrey-Smith, internal: assessing what explains what is an important part of scientific reasoning. Fields just differ in the standards they use.
• The catch: both say explanation is context-dependent. That agreement hides the real disagreement, over whether explanation is epistemic.
• Stakes: if explanation is internal and epistemic, inference to the best explanation can support belief in unobservables, which van Fraassen’s constructive empiricism must deny.
• Mock exam question 8 turns on this difference.

"Contra Hempel, van Fraassen, and Trout, we hold that the pragmatic nature of understanding is not inconsistent with it being epistemically relevant" (De Regt and Dieks, p. 141). What is the epistemic relevance of understanding, and what are the conditions for understanding a theory?

Key points: Opponents: Hempel, van Fraassen, Trout; pragmatic in the sense of useable for aims; understanding a macro-level aim; needed to produce explanations and models; condition: intelligibility via CIT.

• The opponents: Hempel treats “pragmatic” as “subjective” and so as epistemically irrelevant. Van Fraassen makes explanation a non-epistemic application of science. Trout reduces the sense of understanding to cognitive bias.
• Two senses of “pragmatic”: (1) subjective; (2) useable for certain aims. Understanding is pragmatic in the second sense.
• Epistemic relevance: understanding is an aim of science in its own right (a macro-level aim), and it is needed to produce the explanations and models through which science gets knowledge at all.
• Conditions for understanding a theory: its intelligibility. A theory is understood by scientists who can use it, as measured by the criterion for intelligibility of theories (CIT).
• Do you agree? It turns on whether “able to use” is an epistemic achievement or merely a practical one.

De Regt and Dieks's CUP and CIT make understanding relative to scientists in a context. State both criteria. Does this reduce understanding to a subjective affair?

Key points: State CUP; state CIT; community standards at the meso-level, a public testable skill; contextual rather than subjective; Schrödinger-Heisenberg pressure point, settled by the community.

• CUP: a phenomenon is understood scientifically iff there is an explanation of it based on an intelligible theory that conforms to empirical adequacy and internal consistency.
• CIT: a theory $T$ is intelligible for a scientist in a context if they can recognise qualitatively characteristic consequences of $T$ without performing exact calculations.
• No: the context is a scientific community with shared, trained standards of intelligibility (the meso-level). Whether a physicist can recognise qualitative consequences without calculating is a public, testable fact about her skills rather than a feeling.
• What it does make understanding is contextual: relative to a community and its level of progress.
• Pressure point: Schrödinger against Heisenberg in 1926, two groups in the same discipline disagreeing about what was intelligible. There, contextual starts to look a lot like subjective.
• Defence: the disagreement was settled by the community over time, as Pauli predicted.

De Regt and Dieks argue that understanding is an "additional epistemic aim of science" (p. 153). Is their account convincing?

Key points: For: explains historical variation in intelligibility; answers Trout: a skill is more than a feeling; against: skill may be a means to truth or empirical adequacy; Khalifa: knowledge plus competence.

• For: it explains the historical variation (Newton, Huygens, Maxwell; Schrödinger versus Heisenberg) that a single fixed standard of explanation cannot. It also answers Trout, since a skill that produces empirically adequate models is more than a feeling.
• Against: a skill may be a means to the real epistemic aims (truth, empirical adequacy) rather than an epistemic aim itself. If understanding just is the ability to build good models, a reductivist like Khalifa can call it knowledge plus competence, and nothing needs adding to the list of aims.

Link to original