INTRODUCTION

In this article, drawing on the introductory chapter of Steven Weinberg’s The Quantum Theory of Fields, Volume 1—the book in question, which we worked through at Biricik Bilim for more than four months—I have tried to discuss the theoretical limitations of relativistic quantum mechanics and why quantum field theory is an unavoidable point of departure. Throughout, Weinberg’s fantastically elaborate original notation has been retained, including the factors of $\hbar$ and $c$ in relativistic transformations, among the simplest examples. I prepared the article at undergraduate level, touching on the historical background—my favorite part—proofs, and effective field theory.

WHY THE TRANSITION FROM CLASSICAL FIELDS TO QUANTUM FIELDS IS NECESSARY

The Classical Field and the Paradox of Instantaneous Action at a Distance

When Newton—the Washington of classical mechanics—presented universal gravitation in his 1687 Philosophiæ Naturalis Principia Mathematica, he confronted a conceptual problem that deeply troubled both him and his contemporaries: instantaneous action at a distance. In Newtonian gravity, moving a mass $M_1$ at one end of the universe instantly changes the force on a mass $M_2$ light-years away, with no delay. In a famous letter to his critic Richard Bentley, Newton called it a great absurdity that one body could act upon another across empty space without any mediator. Yet because the mathematics worked, he left the mechanism’s origin to philosophy and future physicists as hypotheses non fingo^*, avoiding judgment.

^* “I frame no hypotheses.”

This conceptual crisis was resolved in the nineteenth century through Faraday’s experimental intuition and Maxwell’s mathematical formulation of it. To describe electromagnetic phenomena, a variable value was assigned to every point $(x,t)$ in spacetime: a field. Maxwell’s equations showed that electric charges do not interact directly and instantaneously. A charge creates local fields $\mathbf{E}$ and $\mathbf{B}$ in the surrounding fabric of spacetime. A change in this field—for example, an accelerating charge—produces electromagnetic waves that propagate through space at a finite speed $(c)$. The field thus ceased to be merely a mathematical convenience and became a real physical entity, independent of particles and capable of carrying energy, momentum, and angular momentum.

Special Relativity, Causality, and Cluster Decomposition

With Einstein’s 1905 announcement of special relativity, and $c=1/\sqrt{\mu_0\epsilon_0}$ in Maxwell’s equations, finite-speed propagation became an absolute universal law. In Minkowski spacetime, no causal information can travel faster than light $(c)$. This is necessary to preserve causality. If an interaction propagated faster than light, Lorentz transformations would allow event A to cause event B for one observer while B occurred before A for another observer in a different Lorentz frame—a logical paradox in which effect precedes cause.

Another axiom Steven Weinberg repeatedly emphasizes in relativistic quantum theories is the cluster-decomposition principle.^* Roughly, the outcomes of two experiments performed very far apart, especially in spacelike-separated regions, must be independent. A particle collision in a laboratory in Ankara cannot instantaneously and nonstatistically affect a particle decay in Andromeda. As Weinberg shows later in the book, the workable way to reconcile quantum mechanics with both special relativity—also called Lorentz invariance—and cluster decomposition is to formulate the interaction Hamiltonian in terms of local field operators.

^* The mathematics behind cluster decomposition is not the ordinary notion of “partitioning a set into nonempty disjoint subsets.” In Weinberg’s sense, scattering amplitudes for widely separated clusters of experiments must factor into products of disconnected pieces. More technically, connected S-matrix contributions linking distant experiments must vanish as their separation tends to infinity. Probabilities for independently prepared experiments then factor like independent probabilities.

The Quantum Breakdown of Single-Particle Wave Mechanics

In nonrelativistic quantum mechanics, particle number is fixed. For example, we write a wave function $\psi(x,t)$ symbolically representing one electron and, according to Max Born, its absolute square $|\psi(x,t)|^2$ gives the probability density of finding the electron at position $x$ and time $t$. At the relativistic limit, however, this single-particle probability interpretation must break down because of Heisenberg’s uncertainty principle.

If we try to determine a particle’s position extremely precisely—to confine it within a very small region $\Delta x$—its momentum uncertainty $(\Delta p)$ increases as follows:

$$\Delta p \geq \frac{\hbar}{2\Delta x}.$$

Special relativity tells us that a particle’s total relativistic energy obeys

$$E=\sqrt{p^2c^2+m^2c^4}$$

Momentum uncertainty therefore causes energy uncertainty $(\Delta E)$. If the confinement scale $(\Delta x)$ approaches the particle’s reduced Compton wavelength,

$$\Delta x \approx \bar{\lambda}_C=\frac{\hbar}{mc}.$$

then the momentum uncertainty reaches the particle’s own mass–momentum scale: $\Delta p\approx mc$. In terms of Einstein’s mass–energy equivalence, its energy scale is approximately

$$\Delta E\approx \Delta p\,c\approx mc^2.$$

This is a statement of relativistic quantum mechanics. The uncertainty $\Delta E\approx mc^2$ has reached the order of the particle’s rest energy. Below this scale, the single-particle picture becomes unreliable. Creating an electron–positron pair in free space requires a threshold energy of roughly $2mc^2$; because of energy–momentum conservation, the process in practice also requires an external field, nucleus, or another particle.

Consequently, when you confine a relativistic electron to a region smaller than $\hbar/mc$ to measure its position exactly, the energy transferred to the system can excite the vacuum and create new particles. You may begin with one electron yet have three, five, or more particles at measurement. Then $|\psi(x,t)|^2$ can no longer be the probability of finding one particle there, because it is impossible to distinguish which particle is the original electron.

This shows that a single-particle Hilbert space^* $(\mathcal{H}_1)$ is inadequate for relativistic physics and must be replaced by a Fock space $(\mathcal{F})$ in which particle number can change. Quantum field theory is the only consistent framework whose fundamental objects are not particles, but fields that undergo quantum fluctuations at every spacetime point, with particles understood as quantized local excitations of those continuous fields.

^* A Hilbert space is a vector space equipped with an inner product and complete in the norm induced by that inner product. It can be viewed as the more general, often infinite-dimensional quantum-mechanical counterpart of finite-dimensional Euclidean space.

Making the walls of a potential well infinite in the Schrödinger equation confines a particle within it. In the relativistic world, however, attempting to confine a particle with a very strong potential can supply enough external-field energy to create particle–antiparticle pairs—the Klein paradox. This is why, when reading QFT, you should place the field rather than the particle at center stage.

Attempting to define a position operator $(x)$ for a free particle in relativistic single-particle quantum mechanics leads to the Newton–Wigner localization problem. A Newton–Wigner-type position operator can be defined, but it does not transform under Lorentz transformations like a simple, covariant local position operator. A wave packet sharply localized in one frame may not appear sharply localized in the same sense in another moving frame. Locality can be preserved only at the level of field operators. Under Wigner’s classification, what we call a particle is classified by labels—mass $m$ and spin $s$—on quantum states carrying unitary irreducible representations of the Poincaré group.^*

Quantum fields are operator-valued tools that connect these abstract Poincaré states to spacetime coordinates and ensure that local observables at spacelike separation cannot causally influence one another. This condition is usually expressed as microcausality: the commutator or anticommutator of appropriate field operators at spacelike-separated points must vanish.

^* The Poincaré group—pronounced “Pwan-ka-RAY”—is the Lie group of all spacetime symmetries that leave the laws of physics invariant in four-dimensional Minkowski spacetime. Its Lie algebra is expressed through the generators of energy, momentum, and angular momentum. It represents the symmetry structure of special relativity and includes both spacetime translations and Lorentz transformations. In other words, it comprises every transformation that guarantees the laws retain the same mathematical form when a physical system is displaced in space or time, the coordinates are rotated, or one passes to an observer moving at another constant velocity. The group is the combination of spacetime translations with the Lorentz group that preserves the Minkowski metric $\eta_{\mu\nu}=\mathrm{diag}(-1,+1,+1,+1)$ (see special relativity). In short, it is the mathematical expression of Einstein’s special-relativistic postulate that “the laws of physics are invariant for inertial observers.” It is foundational to modern theoretical physics, particularly QFT, particle physics, and relativistic quantum mechanics, because elementary particles’ properties, such as mass and spin, are classified through its representations. Conservation laws for energy, momentum, and angular momentum are also directly related to these symmetries and acquire physical meaning through Noether’s theorem.

RELATIVISTIC WAVE MECHANICS AND ITS IMPASSES

Schrödinger’s First Attempt and the Derivation of the Klein–Gordon Equation

Efforts to make quantum mechanics relativistic historically began even before the nonrelativistic Schrödinger equation. When formulating wave mechanics in late 1925, Erwin Schrödinger began from special relativity’s fundamental mass–energy relation $(E^2=p^2c^2+m^2c^4)$. For a free particle of mass $m$ moving in vacuum, the relativistic energy–momentum relation is

$$E^2=p^2c^2+m^2c^4.$$

In an external electromagnetic field, the minimal-coupling principle requires the canonical energy $(E)$ and momentum $(p)$^** to transform in a way that preserves gauge symmetry. For an electron of charge $-e$, in terms of scalar potential $\phi$ and vector potential $\mathbf{A}$, these transformations are

$$E\rightarrow H+e\phi,\qquad \mathbf{p}\rightarrow \mathbf{p}+\frac{e\mathbf{A}}{c}.$$

^** It is the partial derivative with respect to the time derivative of the position coordinate: $p_{\rm kanonik}=\partial L/\partial \dot{x}$.

^*** More precisely, minimal coupling specifies how the electromagnetic potential enters the theory so as to preserve local $U(1)$ gauge invariance. The photon’s emergence as a particle instead concerns quantization of the electromagnetic field.

Substituting these relativistic mappings into the mass–energy relation yields the classical relativistic quadratic relation also presented in Weinberg’s book:

$$0=(H+e\phi)^2-c^2\left(\mathbf{p}+\frac{e\mathbf{A}}{c}\right)^2-m^2c^4.$$

To obtain the standard quantum-mechanical wave equation, differential-operator substitutions are defined from the phase structure of de Broglie plane-wave solutions. The temporal and spatial derivatives become

$$H\rightarrow i\hbar\frac{\partial}{\partial t},\qquad \mathbf{p}\rightarrow -i\hbar\nabla.$$

Substitution of these differential operators into the relativistic quadratic relation produces the second-order partial differential equation now called the Klein–Gordon equation, intended to represent spin-0 particles and, at the time, the electron:

$$\left[\left(i\hbar\frac{\partial}{\partial t}+e\phi\right)^2-c^2\left(-i\hbar\nabla+\frac{e\mathbf{A}}{c}\right)^2-m^2c^4\right]\psi(x,t)=0.$$

The Second-Order Time Problem and Negative Probability Density

The Klein–Gordon equation has a highly symmetric structure fully consistent with special relativity, yet it seriously conflicts with the probabilistic interpretation central to quantum mechanics. In the nonrelativistic Schrödinger equation, the time derivative is first order:

$$i\hbar\frac{\partial\psi}{\partial t}=H\psi.$$

This is ideal for quantum mechanics: knowing the initial wave function $\psi(x,0)$ lets you calculate its subsequent evolution exactly. Moreover, the probability density derived from this equation is always nonnegative: $\rho=|\psi(x,t)|^2\geq 0$. The probability of finding a particle somewhere cannot be negative.

For the free-particle Klein–Gordon equation, however, matters differ:

$$\hbar^2\frac{\partial^2\psi}{\partial t^2}=c^2\hbar^2\nabla^2\psi-m^2c^4\psi.$$

Standard manipulations to derive its continuity, or conservation, equation yield the following density, which one might expect to admit a single-particle interpretation:

$$\rho=\frac{i\hbar}{2mc^2}\left(\psi^*\frac{\partial\psi}{\partial t}-\psi\frac{\partial\psi^*}{\partial t}\right).$$

Here lies the first impasse of relativistic wave mechanics. A second-order-in-time differential equation requires two independent initial conditions: the function $\psi(x,0)$ and its first time derivative $(\partial\psi/\partial t)(x,0)$. Since that initial rate of change can be chosen arbitrarily, $\rho$ can mathematically be negative in some regions. A physical system cannot say that “the probability of finding an electron in this room is -25%,” so single-particle relativistic quantum mechanics is incomplete.

Dirac’s Linear Approach and Matrix Algebra

In 1928, Paul Dirac sought an elegant solution to the negative-probability problem. His idea was simple: like the Schrödinger equation, the relativistic wave equation should be first order in time. But special relativity requires equal treatment of time and space. If the time derivative is first order, the spatial derivatives must also be first order. Dirac therefore proposed the following linear energy equation for a free particle:

$$H=c\boldsymbol{\alpha}\cdot\mathbf{p}+\beta mc^2.$$

Here, $\mathbf{p}=-i\hbar\nabla$ is the familiar momentum operator. The equation nevertheless had to agree with the classical relativistic relation $(H^2=p^2c^2+m^2c^4)$. Squaring it produces products of momentum components, such as $p_1p_2$:

$$H^2=c^2\sum_{i,j=1}^{3}\frac{1}{2}\{\alpha_i,\alpha_j\}p_ip_j+mc^3\sum_{i=1}^{3}\{\alpha_i,\beta\}p_i+\beta^2m^2c^4.$$

For the cross terms to vanish and the equation to reduce to $H^2=p^2c^2+m^2c^4$, the coefficients $\alpha$ and $\beta$ could not be ordinary numbers. With ordinary numbers, $xy+yx=2xy$, whereas here this sum had to vanish. Only matrices satisfying the following conditions could do so:

$$\{\alpha_i,\alpha_j\}=2\delta_{ij}I,\qquad \{\alpha_i,\beta\}=0,\qquad \beta^2=I.$$

Dirac proved that the smallest matrices satisfying this algebra in $3+1$-dimensional spacetime must be $4\times4$. Once the coefficients became $4\times4$ matrices, $\psi(x,t)$ ceased to be a single scalar and became a four-component vector, the Dirac spinor. The probability density from this new first-order equation was always nonnegative:

$$\rho=\psi^\dagger\psi=\sum_{a=1}^{4}|\psi_a|^2\geq0.$$

The probability problem was solved.

The Negative-Energy Spectrum and the Collapse of the Single-Particle Approach

Although Dirac’s matrices resolved the probability crisis, the equation introduced another profound problem by its very nature: negative energies. Its solutions had both positive and negative energy roots for every momentum:

$$E=\pm\sqrt{p^2c^2+m^2c^4}.$$

In nonrelativistic physics, negative energies can be dismissed as meaningless. But particles in nature interact with electromagnetic fields. Quantum theory says a high-energy electron tends to emit a photon and fall to a lower energy. If unbounded levels extend to negative infinity, every electron in the universe should continually radiate and fall into that infinite negative-energy abyss. Matter could not exist.

To resolve this, Dirac proposed hole theory—the Dirac sea—in 1930. Electrons are particles obeying the Pauli exclusion principle; I mean fermions, so more than one electron cannot occupy the same state. Dirac supposed that every negative-energy state in the universe was already completely filled with electrons. Since every seat was occupied, an ordinary positive-energy electron could not fall into one. If an energetic interaction promoted an electron from this filled sea into a positive-energy state, an ordinary electron and a hole in the sea would remain. Creating an electron–positron pair requires an energy scale of at least about $2mc^2$, but a single photon cannot accomplish this in empty space because of energy–momentum conservation. The process usually occurs where an external field, nucleus, or another particle can exchange momentum. To an external observer, the missing electron in the sea appears as a positively charged particle with the electron’s mass. The discovery of the positron in cosmic rays in 1932 confirmed this prediction and became a historic triumph.

Yet this triumph also marked the end of single-particle wave mechanics: the Dirac sea implied that empty space was filled with infinitely many particles. The problem was no longer the dynamics of one particle. Moreover, the solution applied only to fermions, spin-$1/2$ particles obeying Pauli exclusion. It failed for spin-0 bosons such as pions, which do not obey Pauli exclusion and could all accumulate in the same negative-energy state.

In 1934, Wolfgang Pauli and Victor Weisskopf overcame this impasse with a revolutionary interpretation. They showed that $\psi(x,t)$ in the Klein–Gordon equation should not be treated as a one-particle probability wave but as a field variable that, once quantized, becomes a field operator creating and annihilating particles. The possibly negative $\rho$ of Section 2.2 was then understood not as probability but as electric charge density. A charge density may physically and expectedly be positive or negative. Quantum field theory was thus born without any artificial sea hypothesis.

The Birth of Quantum Field Theory

In quantum field theory, the fundamental building blocks are not particles but the fields filling all space. Quantization is applied directly to the classical field, not to a wave function.

Quantization of the Radiation Field and Vacuum Fluctuations

The first field studied was Maxwell’s electromagnetic field. Physicists modeled it as a sum of infinitely many quantum harmonic oscillators, each with a particular frequency $\omega_k$. The Hamiltonian giving the system’s energy became

$$H=\sum_k\hbar\omega_k\left(a_k^\dagger a_k+\frac{1}{2}\right).$$

Here, $a_k^\dagger$ creates a photon in the system, while $a_k$ annihilates one.

The $1/2$ term at the end of this equation has enormous consequences. Even when the vacuum contains no photons, every mode has zero-point energy. This shows that the vacuum state of a quantized field is not classically an utterly motionless void. Standard quantum mechanics with a classical electromagnetic field cannot explain why an excited atom spontaneously radiates and drops to a lower energy. Once the electromagnetic field is quantized, however, the atom interacts with the field’s vacuum modes and spontaneous emission is explained.

The Spin–Statistics Connection

In standard quantum mechanics, particle symmetry—whether particles are bosons or fermions—is imposed as an external rule. Relativistic local quantum field theory instead produces it naturally from the equations.

To explain fermions such as electrons, Pascual Jordan and Eugene Wigner changed the operators’ commutation and exchange rules. Instead of $AB=BA$, they used anticommutation relations based on $AB=-BA$:

$$\{a_k,a_j^\dagger\}\equiv a_ka_j^\dagger+a_j^\dagger a_k=\delta_{jk}.$$

This mathematical structure automatically makes it impossible for two identical fermions to occupy the same quantum state and thus enforces the Pauli principle. Enrico Fermi used this operator logic to explain beta decay. Electrons had once been thought to wait physically inside neutrons. Through field theory, Fermi showed that an electron is created at the instant of interaction, just like a photon. Quantum field theory thereby became the only consistent framework capable of describing relativistic processes in which energy becomes matter and particle number changes.

The Problem of Infinities and Renormalization

When more precise quantum-electrodynamic calculations, especially closed-loop diagrams, began in the 1930s, theorists encountered a disaster: the integrals kept yielding infinity.

The Electron’s Electromagnetic Mass

A pointlike electron also interacts with its own electromagnetic field. The formula for this interaction’s contribution $(m_{\rm em})$ to the electron mass contains a short-distance regulator $a$:

$$m_{\rm em}=\frac{3\alpha}{2\pi}m\ln\left(\frac{\hbar}{mca}\right).$$

If the electron is taken as strictly pointlike and $a$ is sent to zero, the mass contribution diverges logarithmically. The same problem arises in calculating the electron charge because of vacuum polarization by virtual particle pairs. The relation between bare and measured physical charge likewise contains regulator-dependent divergences.

The Lamb Shift and the Solution

Despite these infinities, Willis Lamb announced at the 1947 Shelter Island Conference that he had measured a minute difference between the hydrogen $2s_{1/2}$ and $2p_{1/2}$ energy levels. Dirac theory predicted exact degeneracy, so the difference was laboratory evidence of the very quantum loops theorists were trying to calculate. The theory could not simply be wrong.

Hans Bethe, followed by Richard Feynman and others, established the essential logic: the initial mass $(m_0)$ and charge $(e_0)$ written in our equations are mathematical parameters never measured bare in nature. The laboratory value is the sum of this bare quantity and loop corrections generated by the field. If the divergences are systematically absorbed into counterterms and redefined physical mass and charge parameters—a procedure called renormalization—the remaining physical predictions are finite and extraordinarily accurate.

One of Steven Weinberg’s great contributions to modern physics was showing that absorbing these infinities is not a sleight of hand. We may not yet possess the universe’s most fundamental theory, but when relativity and quantum mechanics are combined at low energies, nature necessarily behaves like an effective field theory. The infinities are simply the natural mathematical bill indicating a boundary to the energy range over which our theory applies.

Conclusion

Historically, once the principles of quantum mechanics are combined with special relativity and spatial causality, QFT is not an arbitrary model invented by physicists but the single unavoidable destination reached by logic and mathematics. I hope you and Furkan enjoy this article. Thank you for taking the time to read it. $<3$