Capstone 3: The Cross-Domain Research Project — Find a New Parallel

Overview

This is the most ambitious capstone in the textbook. It asks you to do what Aiko Tanaka does in Chapter 39: find a new parallel between physics and music that was not in the textbook. Not a metaphor. Not a resemblance. A structural parallel — meaning the same mathematics appears in both domains, and that mathematical identity illuminates something about each domain that looking at it alone would not reveal.

The bar is high deliberately. There are many ways to draw connections between physics and music that sound interesting but are not structural in the relevant sense. "Both music and physics deal with waves" is true but trivial — we knew that from Chapter 1, and it does not illuminate anything new about either domain. "The rhythm of a jazz improvisation resembles the fluctuations of a quantum system" is evocative but vague — without a precise mathematical mapping, it is poetry, not physics. The parallels in this textbook have been characterized by precision: the same equation, applied to different objects, producing new understanding of both. That is the standard.

It is also an achievable standard, because the space of unexplored parallels is genuinely large. Physics and music each contain hundreds of deep structures, and the overlap between those structures has been explored only at the fringes. The examples in this textbook — harmonic series and particle resonances, symmetry breaking and key changes, the Fourier uncertainty principle and the time-frequency trade-off in music perception, Bose-Einstein statistics and the physics of musical consonance — are real parallels that took real effort to develop. There are more.

Your parallel must meet three criteria. It must be structural: the same mathematics must appear in both domains, not just similar words or similar shapes. It must be non-trivial: it must not already appear in this textbook (or in the most common physics-and-music survey literature). And it must be illuminating: it must show something new about at least one of the two domains — a result, a prediction, or a question that you could not have formulated without the cross-domain view.

Aiko's Standard

In Chapter 39, Aiko defends her dissertation parallel — she argues that symmetry breaking in particle physics and tonal modulation in music are not merely analogous but structurally identical in a mathematically precise sense. Her committee challenges her on three fronts: Is the mathematics genuinely the same, or approximately similar? Is the mapping between the domains precise enough to be falsifiable? And is the parallel productive — does it generate new questions, or is it a closed observation?

These are the same three questions your committee (your instructor and peers) will ask. Keep them in mind throughout the project. A well-developed parallel that fails one of these tests is still excellent work; a superficially impressive claim that fails all three is not a successful project.

Learning Objectives

By the end of this project, students will be able to:

1. Identify structural parallels between physics and music by examining mathematical frameworks for shared formal properties rather than surface-level resemblance.
2. Articulate the distinction between mathematical identity, structural analogy, and metaphorical similarity, and accurately classify a proposed parallel at the appropriate level.
3. Develop a precise mathematical argument for a cross-domain parallel, including explicit identification of the corresponding objects, relations, and equations in each domain.
4. Design and execute a falsifiability test for a proposed parallel — identifying evidence that would confirm or refute the structural claim.
5. Conduct a scholarly literature review spanning both physics and musicology, synthesizing sources across disciplinary boundaries.
6. Write a research paper in standard academic format that presents a novel argument, acknowledges its limitations honestly, and situates it within existing scholarship in both fields.
7. Assess the "illumination value" of a parallel — the degree to which the cross-domain view generates new questions, predictions, or understandings that neither domain would reach alone.
8. Present a complex, interdisciplinary argument clearly to an audience with mixed expertise, adjusting the level of mathematical detail appropriately.

Background Reading

All students should review the following before beginning:

• Chapter 1 (Wave Basics) and Chapter 7 (Fourier Analysis): Foundational mathematical tools that appear in many parallels.
• Chapter 16 (Consonance and Dissonance): The physics and mathematics of consonance — a domain where many interesting unexplored parallels exist.
• Chapter 18 (Symmetry in Physics): The role of symmetry groups in physics, with examples from particle physics and condensed matter.
• Chapter 21 (Quantum Analogs in Music): The most explicitly cross-domain chapter in the textbook — study it as a model of how parallels are developed and qualified.
• Chapter 22 (Uncertainty Principles): A worked example of a precise parallel and its physical and musical consequences.
• Chapter 24 (Statistical Patterns in Music): Zipf's law, power-law distributions, and statistical regularity — an area where many unexplored parallels exist.
• Chapter 39 (Bridging Domains): Aiko's defense chapter — read this as a model of how to present a cross-domain argument.

Students pursuing parallels in specific areas should also read the chapters most relevant to their topic area (see Sample Approved Topic Areas at the end of this document).

Phase 1: Identify a Candidate Parallel

Estimated time: 3–6 hours

Brainstorming Protocol

The most productive way to find a parallel is not to look for one directly, but to build two independent lists and look for mathematical overlaps.

Step 1: List 10 physics concepts or frameworks you find interesting or have studied in depth. For each, identify the core mathematical structure (not the physical application, but the mathematics itself): a differential equation, a symmetry group, a probability distribution, an optimization principle, a network structure, a geometrical object.

Step 2: List 10 music concepts or phenomena you find interesting or have studied. For each, identify the mathematical structure that underlies it (not the musical description, but the mathematics): an interval ratio, a symmetry, a statistical distribution, a graph structure, a topological property.

Step 3: Draw lines between your two lists wherever the same mathematical structure appears on both sides. These are your candidate parallels.

Step 4: For each candidate, ask: is this already in the textbook? Is this well-known in the existing physics-and-music literature (see Further Reading in Chapter 39 for a bibliography)? If yes, cross it off. What remains is your set of genuine candidates.

Guided Starting Points

If your brainstorm does not produce strong candidates, consider these starting areas. Each is described at the level of a hint, not a full parallel — you will need to develop the mathematics.

Thermodynamics and Musical Dynamics Thermodynamic systems evolve toward maximum entropy (disorder). Musical performances move through states of more and less order — tight rhythmic coordination, dense harmonic texture, melodic predictability. Is there a meaningful formal relationship between thermodynamic entropy and musical information content? Shannon entropy of pitch distributions has been studied; thermal equilibrium and musical "settling" less so.

Network Theory and Orchestration An orchestra is a network: each instrument is a node, and edges represent acoustic coupling (instruments that often double each other, blend perceptually, or interact harmonically). Network theory provides tools for analyzing graph properties: degree distribution, clustering coefficient, path length, and centrality. Are these tools applicable to orchestration? Do orchestral networks have small-world or scale-free properties?

Topology and Voice-Leading Geometry Music theorist Dmitri Tymoczko has shown that voice-leading spaces have specific topological structures (orbifolds). Physics uses topology extensively — topological insulators, knot invariants in quantum field theory, the topology of phase spaces. Are there physical systems whose configuration spaces have the same topological structure as voice-leading spaces? What would this imply?

Statistical Mechanics and Musical Genre Evolution Statistical mechanics describes how macroscopic behavior emerges from the statistics of many microscopic components. Musical genres are population-level phenomena: they emerge from the decisions of many individual composers and listeners, and they evolve over time in ways that look, statistically, like physical processes. Does a spin-glass model, a Potts model, or a Boltzmann distribution capture aspects of genre evolution in the Spotify Spectral Dataset?

Quantum Entanglement and Musical Co-Improvisation In quantum entanglement, measurements on spatially separated particles are correlated in ways that cannot be explained by local hidden variables (Bell's theorem). In co-improvisation, musicians who cannot hear each other well still produce correlated outputs (they anticipate each other, converge on shared tempos, respond to perceived but not heard intentions). Is there a formal sense in which co-improvisation violates a "musical Bell inequality" — that is, produces correlations that cannot be explained by locally available information?

Gauge Theory and Harmonic Function Gauge invariance is the physics principle that certain physical laws are unchanged under local symmetry transformations. Harmonic function in tonal music has a similar property: a chord's "function" (tonic, dominant, subdominant) is not an absolute property but a relational one that changes depending on context. Is there a formal sense in which harmonic function is "gauge-like" — invariant under certain transformations but not others?

Writing the Proposal

Once you have a candidate parallel, write a 1-page proposal (approximately 400 words) that answers:

1. Statement: In one sentence, what does your parallel claim? (Example: "The mathematical structure of Boltzmann statistics, which describes the distribution of energy among particles in thermal equilibrium, also describes the distribution of harmonic interval usage in large corpora of tonal music.")

2. The physics: What is the physics concept? What is its mathematical statement? (Give the equation, theorem, or framework.)

3. The music: What is the music concept? What is its mathematical statement?

4. The connection: What mathematical object appears in both? How precisely do the two statements resemble each other?

5. The test: What evidence would support your parallel? What evidence would refute it?

Submit your proposal to your instructor before proceeding. Your instructor will confirm that the parallel is non-trivial and suggest resources for Phase 2.

Phase 2: Develop the Mathematical Argument

Estimated time: 4–7 hours

The Structure of a Mathematical Argument for a Cross-Domain Parallel

A well-developed cross-domain parallel has a specific structure. You must provide all of the following:

The physics statement: A precise mathematical statement from physics, including the equation or theorem, its standard interpretation, and the objects it applies to.

The music statement: A precise mathematical statement from music theory or musicology, including the equation or theorem (or, if the mathematics has not previously been made explicit, your formulation of it), and the objects it applies to.

The mapping: An explicit dictionary of correspondences:

Every term in the physics equation must have a corresponding term in the music equation. If any term lacks a clear correspondent, that is a limitation of the parallel that must be acknowledged.

The level of the parallel: Classify your parallel honestly at one of three levels:

• Mathematically identical: The same equation, with a precise term-by-term mapping, applies to both domains. This is the strongest claim. Example: the wave equation applies to both acoustic waves and electromagnetic waves — not analogously, but identically.

• Structurally analogous: Different equations with the same formal structure (same type of equation, same qualitative behavior, but different constants or functional forms). Example: the harmonic oscillator equation appears in both mechanical vibration and electrical circuits — structurally identical, but the physical objects are different.

• Metaphorically similar: The same qualitative behavior or conceptual pattern in both domains, without shared mathematical structure. This is the weakest level — not a structural parallel in the sense required by this project, though it may be a productive starting point.

Your project requires at least a structural analogy. Mathematical identity is preferable but much harder to achieve.

Developing the Argument

For most parallels, the development requires these steps:

1. Source the physics: Find the primary literature (textbooks or research papers) for the physics side. Do not paraphrase; reproduce the exact mathematical statement.

2. Source or formulate the music: Find or formulate the mathematical statement of the music side. In some cases, this already exists in the music theory literature (Tymoczko, Noll, or the Journal of Mathematics and Music are good starting points). In other cases, you will need to formulate it yourself, which is a significant intellectual contribution.

3. Write the dictionary: Construct your correspondence table. Be precise about units and interpretation.

4. Check the mapping: Does the physics equation remain true (or become a meaningful statement) when you substitute the music objects for the physics objects? This is the heart of the argument.

5. Identify where the parallel breaks down: Every finite parallel has limits. Identify the conditions under which the mapping fails — either because the mathematics diverges or because the musical interpretation becomes absurd. This is not a weakness of your project; it is part of the intellectual honesty that makes it credible.

Phase 3: Find Evidence

Estimated time: 4–6 hours

What Counts as Evidence

A cross-domain parallel is more convincing when it is supported by evidence beyond the mathematical argument itself. There are several kinds of evidence available to you:

Quantitative data analysis: If your parallel makes a quantitative prediction about music (e.g., "the distribution of interval usage should follow a Boltzmann distribution with an effective temperature corresponding to the genre's stylistic 'temperature'"), test that prediction against real data. The Spotify Spectral Dataset, the MIDI archives at the Center for Computer Assisted Research in the Humanities, or the music21 corpus (a Python library providing access to a large symbolic music corpus) can provide the data.

Historical evidence: If your parallel implies that composers have intuitively exploited the structural relationship you identify (even without knowing its physics analog), look for historical evidence. Did composers write about the musical phenomenon in terms that resonate with the physics concept? Did they discover the musical analog of a physics result independently?

Theoretical predictions: If your parallel is strong enough, it should make predictions — things that must be true about music if the parallel holds, which are not already known. Even if you cannot test these predictions yourself, formulating them is a contribution. Write them as explicit, falsifiable claims.

The Falsifiability Test

Write a formal falsifiability test for your parallel:

Claim: My parallel predicts that [specific, measurable property of music] will have [specific, quantitative value or qualitative property].

Test: To check this prediction, one would [describe the measurement procedure].

Refutation condition: The parallel would be refuted if the measured value of [property] differs from the prediction by more than [threshold], or if [alternative result].

This exercise forces precision. A parallel that cannot be falsified — even in principle — is not a scientific claim. It may still be interesting as a mathematical observation or a compositional inspiration, but it is not a physics claim.

Phase 3 Exercise

Write your falsifiability test (above). Then, if your parallel is testable with available data, execute the test. Report the results honestly — a negative result (the data do not support the parallel) is not a failure of the project. A negative result, reported with integrity, is a significant contribution: it tells us where the parallel fails, which constrains future theories.

Phase 4: Assess the Illumination

Estimated time: 2–3 hours

The Illumination Standard

Aiko's standard, from Chapter 39, is that a good cross-domain parallel offers "a new way to feel the physics." We extend this: a good cross-domain parallel should offer a new way to understand at least one of the two domains — a question you can now ask that you could not have asked without the parallel, or a result you can now see that was invisible from within a single discipline.

Assess your parallel against the following questions:

What does this parallel teach you about physics? Is there a physical result that becomes newly intuitive when translated into musical terms? Is there a physical question (about the behavior of the system, about its limits, about its generalizations) that the musical analog suggests?

What does this parallel teach you about music? Is there a musical phenomenon that becomes newly understandable when described in physics terms? Does the physics side of the parallel suggest musical experiments, compositional strategies, or analytical tools that did not previously exist?

Is the parallel productive or decorative? A productive parallel generates new questions and research directions. A decorative parallel is interesting to contemplate but does not change what you do or think in either domain. Be honest about which yours is — or where it falls on the spectrum.

Comparison to Aiko's work Aiko argues in Chapter 39 that her parallel is productive because it predicts specific properties of musical transitions that can be tested in large corpora, and because it suggests new questions about symmetry breaking in physical systems that the music analogy makes intuitive. Apply the same standard to your parallel. What would you predict? What would you do next?

Phase 5: Write the Research Paper

Estimated time: 5–10 hours

Format

Your research paper should be 3,000–5,000 words (excluding references and figures) and follow standard academic structure:

Abstract (~150 words): State the parallel, its mathematical basis, the evidence you gathered, and your main conclusion.

Introduction (~500 words): Motivate the parallel. Why is it worth investigating? What would it mean if it is correct? Briefly situate it in the existing literature (what has been said about the relationship between physics and music, and where does your parallel fit or extend that conversation?).

Background: The Physics (~500 words): Present the physics side of the parallel clearly. Assume a reader with undergraduate physics knowledge. State the mathematical framework, give the key equation or theorem, and explain what it describes and why it is important in physics.

Background: The Music (~500 words): Present the music side of the parallel with comparable clarity. Assume the same reader has taken an introductory music theory course. State the mathematical framework (or your formulation of it), explain the musical phenomenon it describes, and explain why it is important in musicology or composition.

The Parallel (~700 words): The heart of the paper. Present the correspondence table. State the mathematical claim precisely. Classify the parallel (mathematically identical / structurally analogous / your assessment). Show the key step where the same mathematics appears in both domains.

Evidence (~400 words): Present your evidence. If quantitative: show the data, the analysis, and the result. If theoretical: state the predictions. If historical: present the relevant sources and their interpretation. State honestly whether the evidence supports, partially supports, or fails to support the parallel.

Limitations (~300 words): What does the parallel not capture? Where does the mapping break down? What would need to be true for the parallel to be upgraded (from structural analogy to mathematical identity, or from metaphor to structural analogy)?

Implications (~400 words): What new questions does the parallel generate? What would a research program pursuing this parallel look like? What would it need?

Conclusion (~250 words): Restate the main claim, summarize the evidence and its interpretation, and place the work in context.

References: Minimum 10 scholarly sources, spanning both physics and musicology literature. Use a standard citation format (APA, Chicago, or MLA).

Citation Standards

A minimum of 10 scholarly sources is required. These should include: - At least 3 physics sources (textbooks or peer-reviewed papers) - At least 3 musicology or music theory sources (textbooks or peer-reviewed papers) - At least 2 cross-disciplinary sources (papers or books addressing both physics and music, or mathematical music theory) - Additional sources as appropriate

Avoid relying primarily on general-audience books or Wikipedia. The Journal of Mathematics and Music, Music Perception, Physical Review, and the Journal of the Acoustical Society of America are all appropriate venues.

Deliverables and Grading Rubric

1. Proposal (10 points) The 1-page proposal from Phase 1, submitted before beginning Phases 2–5. Graded on clarity of the claim, non-triviality, and viability of the falsifiability test.

2. Mathematical Development (25 points) The correspondence table, the level classification, and the written argument from Phase 2.

3. Evidence and Analysis (25 points) The falsifiability test and evidence report from Phase 3, plus the illumination assessment from Phase 4.

4. Research Paper (30 points) The full 3,000–5,000 word paper.

5. Presentation (10 points) A 10–15 minute presentation of your parallel to the class. You must explain the physics, explain the music, present the correspondence, show your evidence, and state your conclusion — accessibly, for an audience that includes both physics students and music students.

Sample Approved Topic Areas

The following topic areas have been reviewed by the textbook authors and confirmed as non-trivial and likely to support genuine structural parallels. Brief guidance is provided for each.

Thermodynamics and Musical Dynamics Start with the Boltzmann distribution and Shannon entropy. The connection to music information theory (Markov models of pitch sequences, entropy of musical corpora) is well-established; the connection to thermodynamic equilibrium and musical settling is less explored. Key sources: Cover & Thomas, Elements of Information Theory; Temperley, Music and Probability.

Topology and Voice-Leading Geometry Tymoczko's work (A Geometry of Music, Oxford University Press) establishes the topological structure of voice-leading spaces. Physics uses topological methods in condensed matter (topological insulators, Chern numbers). The parallel between Tymoczko's orbifolds and physical phase spaces is under-explored. Key warning: this topic requires genuine mathematical topology. Audit your preparation before committing.

Network Theory and Orchestration Start with Barabási, Network Science for the physics side. For the music side: who has studied the network structure of orchestral scores? Explore whether orchestral voice distributions show scale-free or small-world properties. The Choir and the Particle Accelerator running example is directly relevant here.

Statistical Mechanics and Musical Genre Evolution The Spotify Spectral Dataset provides historical audio feature data across genres. Ising model, Potts model, or agent-based models from statistical mechanics may describe genre evolution dynamics. Start with Sethna, Statistical Mechanics: Entropy, Order Parameters, and Complexity for the physics.

Quantum Entanglement and Musical Co-Improvisation This is a high-risk, high-reward topic. The Bell inequality has a mathematical structure that could potentially be applied to co-improvisation correlations. The challenge: designing an experimental setup for co-improvisation that tests a musical Bell inequality rigorously. Without a credible experiment, this parallel remains metaphorical. Start with Bell's original 1964 paper and with literature on improvisation and musical interaction (Sawyer, Improvised Dialogues).

Gauge Theory and Harmonic Function Gauge invariance and harmonic function both involve context-dependence of "local" properties. The mathematics of gauge theories (fiber bundles, connections) may have analogs in music theory's treatment of Roman numeral analysis. Start with Baez & Muniain, Gauge Fields, Knots and Gravity for an accessible physics treatment; Riemann and Riemannian function theory for the music side.