On this page
A Feynman diagram looks like a doodle: a few straight lines, a wavy line, some arrows. It is actually a precise instruction for a calculation. This post explains where the diagrams came from, how to read one, the most common misreadings, and a worked example showing why a handful of diagrams is enough to predict a measured number to many decimal places. None of this is in the JEE or NEET-UG syllabus; it is the physics you meet at university, and a good example of how a clear picture can carry hard mathematics.
Where did Feynman diagrams come from?
Quantum electrodynamics (QED) is the theory of how electrons and light interact. As the 1965 presentation speech recalls, early versions gave "an infinite and therefore useless result" when physicists tried to calculate the electron's mass. In the 1940s three physicists found ways round this. The Nobel Prize in Physics 1965 went to Sin-Itiro Tomonaga, Julian Schwinger and Richard P. Feynman "for their fundamental work in quantum electrodynamics, with deep-ploughing consequences for the physics of elementary particles".
Feynman's version appeared in two 1949 papers in Physical Review. In Space-Time Approach to Quantum Electrodynamics he showed that the expressions for complex processes could be simplified and "written down directly for any specific problem". Freeman Dyson's paper The Radiation Theories of Tomonaga, Schwinger, and Feynman, published the same year, discussed all three approaches together.
In his Nobel lecture, Feynman described the rest of his work as improving the techniques for calculation, "making diagrams to help analyze perturbation theory quicker."
How do you read a Feynman diagram?
Take the simplest case: two electrons repel each other.
Draw time running up the page and space running across.
Each electron is a straight line with an arrow. Two lines come in from the bottom.
At one point, called a vertex, one electron gives off a photon, drawn as a wavy line. At another vertex, the other electron absorbs it.
Two electron lines leave at the top, moving in new directions.
At every vertex, charge is conserved: one electron line comes in and one goes out, with a photon attached. Each diagram stands for one term in a sum. To get the full answer, you add up the contributions of every diagram that starts and ends with the same particles.
Why does a positron's arrow point backwards?
In The Theory of Positrons, Feynman reinterpreted the Dirac equation's "negative energy states" as waves that can be pictured travelling backwards in time. Experimentally, such a wave corresponds to a positron, the electron's antiparticle, moving forward in time. So in a diagram, a positron is drawn as an electron line with its arrow pointing down the page. One line that turns around can then show an electron and positron meeting and annihilating, or a pair being created. The story of antimatter and Paul Dirac explains where those states came from.
Worked example: why are a few diagrams enough?
Each vertex brings in the electron's charge. A diagram with two more vertices than another is weighted by an extra factor of the fine-structure constant, α. NIST's CODATA 2022 value is α = 7.2973525643 × 10^-3, about 1/137.
Simplest diagram (Dirac's theory, no loops): the electron's magnetic moment in units of the Bohr magneton is g/2 = 1 exactly.
The first correction comes from one loop diagram with one extra photon. It contributes α/(2π) = 0.0072974 ÷ 6.2832 ≈ 0.0011614.
Prediction so far: g/2 ≈ 1.0011614.
Measurement: Fan and colleagues (2023) report g/2 = 1.00115965218059.
One diagram's correction already gets the answer right to about 1 part in 600,000 of the total (the gap is 0.0000017). The remaining gap comes from more complex diagrams, and each further set is suppressed by another power of α.
The 1965 presentation speech said that QED's predictions for the electron's magnetic moment agreed with experiment to within about a part in a million. The newer measurement tests the theory to 1 part in 10^12.
What do students commonly get wrong?
Reading lines as paths. A diagram is a term in a sum, not a photograph of where the particles went.
Treating the exchanged photon as a real photon. The internal wavy line is a virtual particle; you never detect it directly.
Counting only one diagram. Real predictions add many diagrams with the same start and end.
Mixing up time direction. Some books draw time going left to right. Check the axis before you read the arrows.
What can a student take from Feynman's way of working?
Feynman said in his Nobel lecture that much of his method was "first worked out by guessing", then checked against known results. He also built pictures that turned long algebra into something you could see at a glance. Both habits work in Class 11 physics: sketch the situation before writing equations, and test any formula on a case where you already know the answer. Our posts on the Feynman Lectures on Physics and the Feynman technique show more of how he explained ideas.
Field lines, charges and forces are a good place to practise seeing physics. The Fermi Simulation Lab has interactive simulations for ideas that are easier to see than to read, so you can watch how a field changes as you move a charge.
Next step: draw the diagram for two electrons repelling, label every line and vertex, and check that charge is conserved at each vertex.
