Axioms of Quantum Mechanics

The Axiomatic Framework: A Unified View

At its core, quantum mechanics can be distilled into a handful of foundational principles. These postulates connect the abstract mathematical formalism we've developed to the physical reality of measurement and dynamics. While different textbooks may present them with slight variations in number or phrasing, the essential content remains the same.

To begin, let's get a clear, high-level overview of a standard set of five axioms.

Now that we have the complete picture, let's dissect each postulate. We will examine not only what it says, but also the deeper structural reasons for its formulation, connecting it to the powerful theorems that underpin the theory. The following resource provides a compact, formal statement of the postulates which we will refer to as we proceed.

Postulate 1: The State of the System

The state of a physical system is described by a state vector $|\psi ⟩$, which is a normalized element of a complex Hilbert space $\mathcal{H}$. More generally, a state is described by a density operator $\rho$ on $\mathcal{H}$.

The choice of a Hilbert space is the foundational decision of quantum mechanics. The vector space structure allows for the superposition principle: if $|{\psi }_1⟩$ and $|{\psi }_2⟩$ are possible states, then so is any linear combination $\alpha |{\psi }_1⟩+\beta |{\psi }_2⟩$. The complex nature of the space is essential for describing interference phenomena.

But why a Hilbert space specifically, and not just any vector space? The crucial property is completeness. In quantum mechanics, we often encounter infinite superpositions, such as expressing a localized wave packet as a sum of infinitely many plane waves. Completeness ensures that any such convergent infinite sum of state vectors results in another valid state vector within the same space. Without this property, an infinite superposition could "converge" to something outside the space of valid quantum states, leading to nonsensical physics.

The following video offers a brilliant intuitive explanation for why this property is so essential.

As we established in the last lesson, the most general description of a state is the density operator $\rho$. This postulate sets the stage: the entire "arena" for quantum mechanics is a Hilbert space, and the "actors" are the states that live within it.

Postulate 2 & 3: Observables and Measurement Outcomes

Postulate 2: Every measurable physical quantity (observable) is described by a self-adjoint operator acting on $\mathcal{H}$.

Postulate 3: The only possible values which can result from the measurement of an observable are the eigenvalues of the corresponding operator.

These two postulates are deeply intertwined. The reason observables are represented by self-adjoint (Hermitian) operators is precisely because of the properties of their eigenvalues and eigenvectors:

1. Real Eigenvalues: A self-adjoint operator is guaranteed to have real eigenvalues. This is a physical necessity, as the result of any laboratory measurement is a real number (e.g., position in meters, energy in joules).
2. Orthogonal Eigenvectors: Eigenvectors of a self-adjoint operator corresponding to distinct eigenvalues are orthogonal. This allows the set of all eigenvectors to form a basis for the Hilbert space, representing a complete set of mutually exclusive measurement outcomes.

The profound justification for this structure comes from the Spectral Theorem. This theorem guarantees that any self-adjoint operator $\hat{A}$ (and more generally, any normal operator, where $[\hat{A},{\hat{A}}^{†}]=0$) can be decomposed in terms of its eigenvalues and a corresponding set of projection operators. For a discrete spectrum, this looks like:

where $a_n$ are the real eigenvalues and ${\hat{P}}_n$ are projectors onto the eigenspaces. The collection of projectors $\{{\hat{P}}_n\}$ forms what is known as a Projection-Valued Measure (PVM). This PVM is the quantum mechanical equivalent of a classical sample space; it partitions the Hilbert space into orthogonal subspaces, each corresponding to a distinct, real-valued outcome. The spectral theorem thus provides the rigorous mathematical link between an abstract operator and a well-defined experimental measurement.

Postulate 4 & 5: Probabilities and State Collapse

Postulate 4 (The Born Rule): If a system is in a state $|\psi ⟩$, the probability of measuring the eigenvalue $a_n$ of an observable $\hat{A}$ is given by $P(a_n)=|⟨a_n|\psi ⟩{|}^2$, where $|a_n⟩$ is the corresponding normalized eigenvector.

Postulate 5 (The Projection Postulate): Immediately after a measurement of $\hat{A}$ yields the result $a_n$, the state of the system "collapses" to the corresponding eigenstate $|a_n⟩$. (More generally, the state $\rho$ becomes $\frac{P_n\rho P_n}{\text{Tr}(P_n\rho P_n)}$).

The Born rule is perhaps the most quintessentially "quantum" postulate, introducing fundamental indeterminism. But why this specific mathematical form? This question is answered by Gleason's Theorem. It's a deep and powerful result which states that for any Hilbert space of dimension three or greater, the only way to consistently assign probabilities to the outcomes of measurements (represented by projectors) is via the rule $P(P_n)=\text{Tr}(\rho P_n)$. For a pure state $\rho =|\psi ⟩⟨\psi |$ and a simple projector $P_n=|a_n⟩⟨a_n|$, this becomes:

Thus, the Born rule is not an arbitrary choice but a mathematical necessity for any theory that uses the Hilbert space framework to predict probabilities.

The projection postulate describes the effect of a measurement on the system. It is responsible for the reproducibility of measurements: if you measure an observable and get $a_n$, an immediate subsequent measurement of the same observable will yield $a_n$ with certainty. This postulate introduces a non-unitary, irreversible evolution into the theory, which stands in stark contrast to the smooth, deterministic evolution we will see next. This dichotomy is the essence of the famous measurement problem.

Postulate 6: Time Evolution

The time evolution of the state vector $|\psi (t)⟩$ of a closed system is governed by the Schrödinger equation:

$$i\hslash \frac{d}{dt}|\psi (t)⟩=\hat{H}|\psi (t)⟩$$

where $\hat{H}$ is the Hamiltonian operator of the system.

This postulate dictates the dynamics of a quantum system between measurements. The evolution is deterministic, linear, and unitary. Unitarity is crucial, as it ensures that the norm of the state vector is preserved, meaning total probability is conserved over time.

The form of this evolution can be derived from symmetry principles. If we postulate that physical laws are invariant under time translations—a fundamental symmetry of nature—then the transformation that evolves the system in time, $|\psi (t)⟩=\hat{U}(t)|\psi (0)⟩$, must form a continuous one-parameter group of unitary operators. Stone's Theorem on one-parameter unitary groups provides the final piece. It states that any such group can be written as $\hat{U}(t)=e^{-i\hat{G}t}$ for some self-adjoint operator $\hat{G}$. By scaling this generator with $\hslash$ to match units of action, we identify it as the Hamiltonian, $\hat{H}$, the generator of time translations. The Schrödinger equation is simply the differential form of this exponential evolution operator.

Conclusion

In this lesson, we have assembled the mathematical machinery from our first module into a complete axiomatic framework. We saw that the postulates of quantum mechanics are not arbitrary rules but are deeply rooted in the logical and mathematical structure required to build a consistent probabilistic theory of nature.

Key Takeaways:

• The state of a system is a vector in a Hilbert space, whose completeness is essential for handling infinite superpositions.
• Observables are self-adjoint operators because the Spectral Theorem guarantees they have real eigenvalues and a complete, orthogonal set of eigenvectors, corresponding to a well-defined set of measurement outcomes.
• The Born rule is not an ad-hoc assumption but a necessary consequence of imposing a consistent probability structure on the Hilbert space of states, a result formalized by Gleason's Theorem.
• The Schrödinger equation arises directly from the assumption of continuous, unitary time evolution, whose generator is identified as the Hamiltonian via Stone's Theorem.
• The postulates reveal a fundamental duality in quantum dynamics: deterministic, unitary evolution via the Schrödinger equation, and probabilistic, non-unitary collapse during measurement.

With this axiomatic foundation firmly in place, we can now move on to applying it. In the next lesson, we will use these postulates to derive the Born rule for measurement probabilities and analyze the consequences of the projection postulate for sequential measurements.