1800s / Two routes to one screen
When adding light
makes darkness.
Imagine expecting two openings to make a screen simply brighter. A wave model offers a different prediction: contributions reinforce at some positions and cancel at others. The intellectual shift is not the striped picture alone, but a rule that predicts how the stripes move.
Use the wavefront illustration above, then change the separation in the Young teaching model. It is a modern idealisation, not a reconstruction from Young’s measurements.
Try the interference model →Checked primary text: Young, Experiments and Calculations Relative to Physical Optics (1804), facsimile. Read to the Royal Society on 24 November 1803; its card-edge interference experiment is not this modern two-slit schematic. Scan inspected 30 October 2026. Index date caveat →
Dialectic: does a picture prove a theory?
Fictional teaching dialogue; not historical quotations.
Observer: The bright and dark bands look persuasive. Is that enough?
Sceptic: Could a different model draw the same picture?
Investigator: Change one parameter. Ask both models to predict the next pattern before you see it.
Your challenge: Which measurement would separate the explanations rather than merely illustrate one?
1910s / The rare event
Let the outlier
change the model.
Most particles passing through a thin foil seem unsurprising. The difficult question is what a large deflection says about the distribution of charge. A scattering model makes the angle dependence inspectable; a dramatic image alone does not tell you whether the assumptions hold.
The index distinguishes the 1909 scattering observation from Rutherford’s 1911 interpretation. The laboratory uses ideal single Coulomb scattering, omitting screening, multiple scattering and nuclear size.
Inspect the scattering model →Checked primary text: Rutherford’s 1911 paper (institution-hosted transcription), which cites Geiger and Marsden’s 1909 observations. Checked 30 October 2026. Index record →
Dialectic: when is an outlier evidence?
Fictional teaching dialogue; not historical quotations.
Observer: A rare deflection contradicts my expectation.
Sceptic: Or it exposes your apparatus, contamination or selection rule.
Investigator: Compare the rate, angular distribution and controls—not just one memorable event.
Your challenge: What control would you ask for before changing the model of the atom?
1980s / What correlations can decide
Turn a philosophy
into a bound.
Instead of asking only what reality “really is,” ask what combinations of correlations an explicitly defined class of explanations permits. A Bell-type inequality makes assumptions answerable to experiment. The model here samples a quantum prediction; it cannot independently prove that prediction.
Change visibility and compare the analytical CHSH value with the local bound of 2. Synthetic outcomes are drawn from an assumed distribution, not from a detector.
Explore correlations →Checked primary abstract: Aspect, Dalibard & Roger (1982), reporting time-varying analyser measurements consistent with quantum predictions and violating a Bell inequality. Checked 30 October 2026. Index record →
Dialectic: what did the test rule out?
Fictional teaching dialogue; not historical quotations.
Observer: The inequality is violated. Have we explained everything?
Sceptic: Only if we forget the assumptions and experimental loopholes.
Investigator: State the class of models being tested. Separate an experimental violation from a simulation designed to produce one.
Your challenge: Why is the synthetic sampler on this page not a Bell experiment?
1990s / A changing identity
What if “missing”
means transformed?
A shortfall in one detected particle flavour can invite more than one explanation. Oscillation models turn the question into a dependence on distance and energy. The useful story is not “the missing particles were found,” but how an alternative explanation makes testable predictions.
The two-flavour vacuum laboratory compares numerical evolution with an analytical appearance probability. It leaves out matter effects, three-flavour mixing and detector response.
Run the oscillation model →Checked primary abstract: Super-Kamiokande: Evidence for Oscillation of Atmospheric Neutrinos (1998). The observed directional dependence was consistent with flavour oscillation, not explained by the studied experimental and flux uncertainties. Checked 30 October 2026. Index record →
Dialectic: absence or transformation?
Fictional teaching dialogue; not historical quotations.
Observer: Fewer events arrived than expected.
Sceptic: Is the source wrong, the detector inefficient, or the propagation model incomplete?
Investigator: Look for a structured pattern, not only an overall deficit.
Your challenge: Which dependence would help distinguish those possibilities?
2010s / From a signal to a claim
A discovery is
a chain of evidence.
On 4 July 2012, ATLAS and CMS announced a new particle consistent with the Higgs boson. Identifying its properties took further study. That distinction—initial observation versus later characterisation—is part of the story, not a footnote.
GW150914 was detected on 14 September 2015 and announced on 11 February 2016. Detection date and publication or announcement date answer different questions. Preserve both when retelling a frontier.
Checked primary-institution anchors: CERN: discovering the Higgs; ATLAS: Higgs physics; LIGO discovery paper; LIGO: GW150914. Checked 30 October 2026. No measured-event animation is supplied here.
Dialectic: when may we call it a discovery?
Fictional teaching dialogue; not historical quotations.
Observer: The signal matches a prediction.
Sceptic: How were backgrounds, calibration and alternative explanations tested?
Investigator: Read the claim at its actual strength. Then follow the independent checks and later measurements.
Your challenge: What would be lost if the headline collapsed detection, announcement and confirmation into one date?
Editorial roadmap / Coverage, not a promise
Keep the gaps
in the open.
The index ranges from 1704 to 2024. Some decades have several entries; others, including the 1940s, have none in this selection. The selector shows only decades with records. A missing entry does not mean no physics happened.
Future episodes can connect a frontier question, an original source, an animated explanation and a debate over assumptions. They remain proposed until their evidence and teaching assets are ready. There is no complete paid course behind this page.
Inspect the available index →
Free to explore / Proposed collaborations
Bring a question worth teaching.
Interested in an educator workshop, a commissioned physics story or an institutional learning resource? Tell MAI Physics about your audience, topic and intended use. Scope, deliverables, permissions and any fees would be agreed separately.
Discuss a learning project →This opens your email app; nothing is sent automatically. Memberships are a proposed direction, not available for purchase. No accounts, payment collection, credits or progress tracking are enabled.
