A Graduate Study Plan for John Pardon's Symplectic Geometry

John Pardon’s 2026 Fields Medal matters for a graduate student in symplectic geometry not because it supplies a new celebrity to admire, but because it points toward a specific reading problem: if the work everyone is discussing is built around virtual fundamental cycles, Fukaya categories, and the MNOP conjecture, what has to be learned before the papers become readable? Quanta’s July 23, 2026 profile describes Pardon as a mathematician whose relatively short publication list consists of unusually consequential papers, with Kai Cieliebak saying that “every paper he wrote is some kind of breakthrough paper, all in very top high-level journals.” [1]

That is an encouraging fact only if it is interpreted correctly. It does not mean there are fewer pages to study. It means the pages are dense with infrastructure. A student who opens the most famous paper first may recognize the nouns and still have no stable way to parse the argument.

The right entry point is the foundations problem that made Pardon's work so valuable. In 2017, Quanta described a dispute over the foundations of symplectic geometry: the Kuranishi-structure approach introduced by Fukaya and Ono in 1996 had drawn criticism from McDuff and Wehrheim around 2012, and the field lacked a universally accepted rigorous framework for counting pseudo-holomorphic curves. Pardon's 2015 Stanford thesis and 2016 Geometry & Topology paper offered an algebraic approach to virtual fundamental cycles, and Mohammed Abouzaid said it “made it possible for many people to stop worrying about things.” [2]

That sentence is easy to underestimate. In research mathematics, being able to stop worrying about a foundational construction is not a small convenience. It changes what people are willing to build on top of it. The study plan below is therefore not a tour of attractive topics around symplectic geometry. It is a prerequisite chain aimed at one narrow goal: becoming prepared enough to make a serious first pass through Pardon's papers and book manuscript.

StudyMethod usually writes for exam preparation rather than research reading, but the planning logic is the same: start from the target performance, work backward through dependencies, and give each phase a job. The same habit of backward planning appears in The Study Habit That Won Hong Wang the Fields Medal, and the timeline discipline is closer to using a midterm study schedule template than to browsing a reading list. The difference is that here the exam is not a test date; it is the moment when the notation in a research paper stops being decorative.

The 12-18 Month Roadmap, With the Warning Attached

A motivated graduate student with prior exposure to analysis, topology, and differential geometry can use 12-18 months as a rough pacing model. It is not a promise of mastery, and it is not a substitute for seminars, an advisor, or working through exercises carefully. The point of the timeline is to prevent a common mistake: treating Pardon's papers as the beginning of the study plan rather than the destination.

This sequence is not invented for the sake of a neat article. Ana Cannas da Silva’s Lectures on Symplectic Geometry is a standard entry point, with chapters moving from symplectic linear algebra through core constructions in symplectic geometry; the text has been cited 1,406 times. [3] Recent graduate syllabi also support the same progression: Michael Hutchings’s Fall 2024 UC Berkeley Math 242 course and Rui Loja Fernandes’s Fall 2019 UIUC Math 520 course both treat symplectic geometry through canonical material rather than shortcut summaries. [4][5]

Phase 1: Differential Topology Is Not a Warm-Up

The first phase should be brisk only if the material is already stable. A reader needs smooth manifolds, tangent and cotangent bundles, submersions and immersions, transversality, orientations, degree, vector bundles, tubular neighborhoods, and enough comfort with Sard's theorem to recognize why genericity arguments appear everywhere later.

Lee's Introduction to Smooth Manifolds, or a comparable graduate course, is appropriate here. The test is not whether the definitions sound familiar. The test is whether you can follow a proof that replaces a geometric object by a transverse perturbation and know what has been preserved. If that step feels like magic, later discussions of moduli spaces will become unreadable at exactly the wrong moments.

• You should be able to compute tangent spaces and pullbacks without rechecking the definition each time.
• You should know what an orientation convention is doing, even when you do not yet enjoy it.
• You should be able to state transversality in a way that distinguishes the theorem from the wish.
• You should recognize when a proof depends on compactness, boundary behavior, or a choice of perturbation.

Phase 2: Algebraic Topology Supplies the Accounting System

Symplectic geometry constantly turns geometric behavior into algebraic invariants. That means homology, cohomology, cup products, fundamental groups, Poincare duality, exact sequences, and basic characteristic classes cannot remain passive background. They are the accounting system used to record what curves, intersections, and moduli spaces are meant to count.

This phase does not require pausing for every branch of modern homotopy theory. It does require enough algebraic topology that, when a paper assigns a virtual class or compares invariants, the algebraic target is not the mysterious part. If the chain complex itself consumes all available attention, the symplectic content has no room to enter.

Phase 3: Learn Symplectic Geometry Before Curve Counting

Cannas da Silva is a sensible first spine for this phase because it begins with the linear algebra and local geometry that later constructions use constantly: symplectic vector spaces, Darboux's theorem, Moser's trick, Hamiltonian vector fields, moment maps, reduction, and basic examples. [3] McDuff-Salamon's Introduction to Symplectic Topology can then deepen the geometric and topological side.

The important shift is from treating a symplectic form as another differential form to treating it as a rigid structure with flexible-looking local models and surprisingly global consequences. Darboux's theorem says there are no local invariants of symplectic manifolds in the same way Riemannian geometry has curvature; Moser's argument teaches how families of forms can be compared; Hamiltonian flows make the geometry dynamic. These are not ornamental chapters before the real subject. They define the instincts needed to understand why pseudo-holomorphic curves became such a powerful tool.

A useful stopping point is the ability to work through the standard examples without relying on analogy: cotangent bundles, projective spaces, toric examples, and Hamiltonian group actions. You do not need to be fast yet. You do need to know where the symplectic form, almost complex structure, and topological invariant enter separately, because later they will be braided together.

Phase 4: J-Holomorphic Curves Are Where the Difficulty Changes Shape

The middle of the plan deserves the most time. J-holomorphic curves are the bridge from basic symplectic geometry to the foundation work associated with Pardon. At this stage, the reader is no longer merely learning definitions; she is learning why reasonable-looking counts are hard to make rigorous.

The central objects are maps from Riemann surfaces into symplectic manifolds satisfying a Cauchy-Riemann-type equation determined by an almost complex structure. The words are compact; the consequences are not. One has to track Fredholm theory, regularity, transversality, compactness, bubbling, gluing, orientations, and quotienting by automorphisms. Each of these can be understood locally for a while, but curve-counting forces them to interact.

McDuff-Salamon's J-holomorphic Curves and Symplectic Topology is the natural main text for this phase. It is not a book to skim on the way to Pardon's paper. The Fredholm setup explains why moduli spaces have expected dimensions. Compactness and bubbling explain why limits of curves are not as tidy as a first count would like. Transversality explains why the moduli space one wants is not automatically the smooth object one needs.

A hypothetical warning example helps here. Suppose a moduli space is expected to be zero-dimensional, so one hopes to count its points and obtain an invariant. If the space is singular, has excess dimension in some strata, or changes under perturbation in a way that is not coherently controlled, the count is not yet a theorem. The problem is not that geometers forgot to be careful. The problem is that the objects being counted naturally resist the clean manifold behavior that elementary transversality would like to provide.

This is where the 2017 foundations story becomes mathematically useful rather than merely dramatic. The field needed a way to assign fundamental classes to moduli spaces that were not honest smooth manifolds of the expected dimension, while still supporting the curve counts used in symplectic topology and enumerative geometry. [2]

What You Should Be Able to Do Before Moving On

• Explain what the moduli space parameterizes in a basic curve-counting problem.
• Distinguish expected dimension from actual geometric behavior.
• Recognize why compactification introduces broken or bubbled objects.
• Follow the role of perturbations without pretending they are harmless.
• Say why orientations and signs are part of the invariant, not bookkeeping after the fact.

Phase 5: Pardon's Virtual Fundamental Cycles

Only after the previous phase does Pardon's 2016 paper become the right object to attempt. The paper, An algebraic approach to virtual fundamental cycles on moduli spaces of pseudo-holomorphic curves, appeared in Geometry & Topology and is the published version of the work tied to his doctoral research. [6] It has been cited 204 times, but the more important point for a reader is the kind of paper it is: it builds machinery so that other mathematical constructions have firmer ground under them.

The paper should be read with the foundations crisis in mind. If the Kuranishi approach was contested and curve-counting needed a rigorous basis, then an algebraic construction of virtual fundamental cycles is not an optional technical alternative. It is a way of making the counts usable. That is why Abouzaid's phrase about being able to “stop worrying about things” is so revealing: the emotional relief comes from moving a burden out of every subsequent argument. [2]

At this phase, the goal is not to reproduce the paper from memory. The goal is to identify the input data, the construction, the compatibility requirements, and the output class. You should know which earlier anxieties the formalism is designed to control: singularity, nontransversality, compactification, orientations, and functorial behavior. If those words are still vague, go back rather than forward.

Phase 6: Fukaya Categories and Contact Homology as Stress Tests

Pardon's later work on Fukaya categories and contact homology is a natural next checkpoint because it shows foundational technology operating inside richer algebraic structures. His Stony Brook page lists his papers and the book manuscript that now serve as the most direct public route into this part of the work. [7]

This is also the point where a reader should become stricter about algebra. Fukaya categories are not merely a new vocabulary for the same curves. They require comfort with chain-level constructions, A-infinity structures, signs, gradings, and operations defined by counts of pseudo-holomorphic curves. Contact homology brings its own analytic and algebraic demands. A student who has treated virtual fundamental cycles as a black box may still get some orientation here, but the arguments will remain mostly out of reach.

The reading order should therefore stay conservative: first understand what problem the construction solves, then locate the moduli spaces being counted, then ask what algebraic structure the counts are supposed to satisfy. Jumping straight to categorical language can create the comforting illusion of sophistication while hiding the analytic dependency chain.

Phase 7: MNOP Is the Capstone, Not the Shortcut

The MNOP conjecture relates Gromov-Witten and Donaldson-Thomas invariants for Calabi-Yau threefolds. It had been open for 20 years before Pardon's proof, and his paper Universally counting curves in Calabi-Yau threefolds appeared on arXiv in 2023 with a revision in 2025. [8] This is not where a student should go for a first explanation of Gromov-Witten theory, Donaldson-Thomas theory, or Calabi-Yau geometry.

It is, however, where the earlier phases show their purpose. The proof sits at the intersection of curve-counting, moduli theory, and algebraic formalism. Quanta reports that the framework includes a new structure now called the “Pardon algebra,” and Jim Bryan described the result as “the biggest result in enumerative algebraic geometry for the last 20 years.” [1] The phrase should be handled carefully: in July 2026 it is a newly prominent name around very recent work, not a settled historical category with decades of independent use.

There is already an important sign that the framework is being tested beyond the original paper. Kai Behrend's April 2026 arXiv paper A Pardon Algebra for Zero-cycles extends Pardon's algebra to zero-cycles. [9] That is early independent confirmation of significance, not a license to write as if the entire long-term place of the theory has already been decided.

The Book Manuscript Belongs After the Papers Start to Make Sense

Pardon's Stony Brook page hosts the book manuscript Derived moduli spaces of pseudo-holomorphic curves, with an April 2025 PDF updated on the page by July 2026. [7] The manuscript should be treated as a capstone resource, not as an introductory textbook. Its value is precisely that it organizes the formalism at a level where the earlier analytic and topological problems have already been taken seriously.

Quanta reports a line from the preface that “most of the real work has been in finding the ‘right’ formalism, after which the proofs fall into place.” [1] That is a useful guide to the whole study plan. It does not mean formalism replaces examples. It means examples, analytic estimates, moduli spaces, and algebraic structures have to be arranged so that the proof no longer fights its own language.

A Practical Pacing Model

For a student already past first-year graduate coursework, a plausible 12-18 month schedule might give the early prerequisites less time and the J-holomorphic/moduli-space transition more. The calendar should remain adjustable. If Phase 4 slips, the later phases should move, not compress.

A planner helps only if it records dependencies honestly. Use something like an exam countdown planner if you need a visible schedule, but do not let the boxes create a false sense of completion. A week spent repairing transversality is not a delay if it prevents three months of fake progress later.

How to Know You Are Ready to Open Pardon Seriously

The honest readiness test is not whether you can explain the Fields Medal citation to a friend. It is whether you can read the first pages of a Pardon paper and identify the mathematical role of the objects being introduced. Some terms will still need checking. That is normal. What should no longer be happening is total dependency failure: every sentence demanding a separate course.

• You can say what kind of moduli space is under discussion and why it is not automatically smooth.
• You can distinguish geometric compactification issues from algebraic packaging issues.
• You can track what a virtual fundamental class is meant to make possible.
• You can recognize when a later invariant depends on earlier foundational choices.
• You can read slowly without mistaking slowness for evidence that you do not belong in the subject.

That last point matters. Many strong students first encounter celebrated papers at exactly the wrong time: late enough to know the vocabulary, too early to know the machinery. Pardon's work rewards depth over speed. A plan for reading it should do the same.

References

1. John Pardon Wins the 2026 Fields Medal for Work in Symplectic Geometry — Quanta Magazine, 2026-07-23
2. A Fight to Fix Geometry's Foundations — Quanta Magazine, 2017-02-09
3. Lectures on Symplectic Geometry — Cannas da Silva
4. Math 242: Symplectic Geometry — UC Berkeley, Fall 2024
5. Math 520: Symplectic Geometry — UIUC, Fall 2019
6. An algebraic approach to virtual fundamental cycles — Geometry & Topology, 2016
7. John Pardon — Stony Brook Math personal webpage
8. Universally counting curves in Calabi–Yau threefolds — arXiv, 2023
9. A Pardon Algebra for Zero-cycles — arXiv, April 2026