Postulates and Equivalent Presentations

A postulate set is a compact specification of a theory’s state space, measurable events, probability rule, dynamics, composition rule, and update conventions. Different textbooks can package these ingredients differently without changing the physical predictions, provided the translations are explicit and the domain of validity is the same.

This chapter states one standard nonrelativistic package, then shows how matrix mechanics, wave mechanics, density operators, and alternative pictures reorganize it. The aim is neither to present the postulates as arbitrary declarations nor to claim that a compact list settles interpretation, detector physics, model building, or the limits of nonrelativistic quantum theory.

Why a Postulate Package Matters

Postulates identify which mathematical objects carry physical meaning and which rules connect them to observations. They prevent derivations from quietly assuming what they claim to prove. They also expose where extensions are needed: open-system channels extend closed unitary dynamics, POVMs and instruments extend ideal projective measurements, and quantum field theory changes the framework required for relativistic locality and particle creation.

Three kinds of choice should be distinguished:

• Empirical structure: interference, probabilistic outcomes, quantum composition, and experimentally tested dynamical predictions constrain viable postulates.
• Mathematical packaging: rays or rank-one density operators, Schrödinger or Heisenberg pictures, and basis-free or coordinate forms can encode the same predictions.
• Conventions: basis order, Fourier normalization, phase convention, units, and sign choices affect formulas but not consistently calculated observables.

Changing the package is legitimate only when a translation preserves normalized states, outcome probabilities, composition, and dynamics over the class of systems being discussed. Merely using similar notation does not establish equivalence.

One Minimal Package

The chapter uses the following five-part organization.

States

An isolated system is associated with a complex Hilbert space $\mathcal H$. Pure states are rays. General states are density operators satisfying

$$\rho\ge0, \qquad \operatorname{Tr}\rho=1.$$

The distinction between an abstract state and a basis-dependent wavefunction or column vector remains essential.

Measurements and probabilities

A sharp discrete measurement is represented by orthogonal projectors $\{P_a\}$ with $\sum_aP_a=I$. More generally, a POVM uses positive effects $\{E_a\}$ satisfying $\sum_aE_a=I$. The Born rule is

$$p(a) = \operatorname{Tr}(\rho E_a),$$

with $E_a=P_a$ for a projective outcome. A complete generalized measurement also requires an instrument when post-measurement states matter.

Closed-system dynamics

Between interventions, a closed system evolves unitarily:

$$\rho(t) = U(t,t_0) \rho(t_0) U(t,t_0)^\dagger.$$

For state vectors, the Hamiltonian generates the Schrödinger equation

$$i\hbar \frac{d}{dt} \lvert\psi(t)\rangle = H(t)\lvert\psi(t)\rangle.$$

Open-system evolution requires a broader channel or master-equation description and additional assumptions.

Composition

For distinguishable subsystems,

$$\mathcal H_{AB} = \mathcal H_A\otimes\mathcal H_B.$$

The tensor product permits entangled states and local observables. Identical particles require symmetrization or antisymmetrization and are not exhausted by simply attaching subsystem labels.

Conditional state update

For an ideal projective outcome $a$ with nonzero probability, the Lüders update is

$$\rho \longmapsto \rho_a = \frac{P_a\rho P_a} {\operatorname{Tr}(\rho P_a)}.$$

Some presentations include this as part of the measurement postulate rather than as a separate item. That is a packaging choice. What matters is that the probability rule and conditioned transformation are both specified without being conflated.

Finite-Dimensional Presentation

For qubits, spin systems, and finite-level models, one may begin with $\mathcal H\cong\mathbb C^d$. States are normalized vectors or positive trace-one matrices, sharp observables are Hermitian matrices, closed dynamics is unitary, and composite dimensions multiply under tensor products.

This presentation is complete for its stated class of models and allows readers to learn superposition, incompatibility, entanglement, and generalized measurements without first introducing differential equations. It also hides issues that return in infinite dimensions:

• unbounded operators and their domains;
• continuous spectra and generalized eigenvectors;
• boundary conditions and self-adjoint extensions;
• inequivalent representations possible in systems with infinitely many degrees of freedom.

A finite truncation can approximate an infinite-dimensional model, but it does not make all operator identities exact. Convergence and the treatment of discarded states must be checked.

Wave-Mechanics Presentation

In a position representation, a pure state is described by

$$\psi(\mathbf x,t) = \langle\mathbf x\rvert\psi(t)\rangle,$$

with normalization

$$\int d^n x\, \lvert\psi(\mathbf x,t)\rvert^2 = 1.$$

Position acts by multiplication, momentum acts on a suitable domain as

$$\mathbf P = -i\hbar\boldsymbol\nabla,$$

and the abstract Schrödinger equation becomes a partial differential equation. The wavefunction is not an additional kind of physical state; it is a representation of the Hilbert-space state. Boundary conditions and the integration measure are part of the problem, not decorative details.

Wave mechanics is especially natural for particles moving in space. It should not be treated as conceptually prior to finite-dimensional spin or qubit systems, which may have no useful position-space wavefunction in the model under study.

Density-Operator Presentation

The density-operator package uses $\rho$ as the state from the beginning. Probabilities and expectations are trace pairings, closed evolution is unitary conjugation, projective updates use the Lüders rule, and generalized operations are completely positive maps.

This presentation contains the pure-state formalism as the special case

$$\rho = \lvert\psi\rangle\langle\psi\rvert.$$

It is the natural language for mixed preparations, reduced states, open systems, quantum information, statistical mechanics, and unread measurements. It does not eliminate state vectors: vectors remain efficient representatives of pure states and appear naturally in purifications, amplitudes, and spectral decompositions.

What Equivalent Means

Two presentations are equivalent over a stated domain when there is a translation between their states, observables, and dynamics that preserves every operational prediction. Examples include

• changing from abstract kets to basis components;
• changing between position and momentum representations;
• moving time dependence between states and observables in Schrödinger and Heisenberg pictures;
• replacing a pure-state vector by its rank-one density operator.

The invariant check is a probability or expectation value, such as

$$\operatorname{Tr}(\rho E) = \operatorname{Tr}(\rho' E').$$

Path integrals, canonical quantization, and algebraic formulations can agree with operator quantum mechanics when their mathematical definitions, boundary data, regularization, and observable dictionaries are controlled. The word “equivalent” should not conceal anomalies, inequivalent representations, measure problems, or a mismatch in scope. Detailed translations belong to Quantum Dynamics.

Assumptions and Scope

Unless a page says otherwise, the compact package assumes

• a nonrelativistic quantum model;
• an externally specified time parameter;
• fixed particle content or a finite set of modeled degrees of freedom;
• closed-system unitary dynamics between specified interventions;
• idealized measurement rules unless an apparatus or instrument is modeled;
• distinguishable-subsystem tensor products before identical-particle constraints are imposed;
• Hamiltonians and observables with appropriate domains and boundary conditions.

The postulates do not choose the Hamiltonian. Constructing $H$ requires physical modeling, symmetry, effective-theory judgment, experimental parameters, and approximation choices. A formally valid evolution rule cannot rescue a physically inappropriate model.

What the Postulates Do Not Settle

The standard package does not by itself

• select a unique interpretation or ontology;
• explain why one particular outcome is experienced or recorded;
• model detector amplification, noise, calibration, or readout;
• derive the Born rule merely by restating it;
• decide which degrees of freedom should enter an effective Hamiltonian;
• specify how nonrelativistic quantum mechanics is embedded in relativistic quantum field theory;
• prove that every formal expression is mathematically well defined.

These are not defects to hide. They mark distinct questions for foundations, experiment, rigorous mathematics, effective theory, open systems, and the QFT bridge.

Page Map

Question	Canonical page	Main distinction
Why state assumptions explicitly?	Why Postulates Matter	theory specification versus arbitrary convention
What is the compact standard package?	Minimal Postulates	core rules versus detailed implementation
How does the theory look for finite systems?	Finite-Dimensional Postulates	matrix-first clarity versus hidden domain issues
How does the theory look in position space?	Wave-Mechanics Postulates	representation versus abstract state
How does the theory look for general states?	Density-Matrix Formulation	pure-state special case versus general state language
When are formulations equivalent?	Equivalent Formulations	shared predictions versus similar notation
Where does the compact package apply?	Assumptions and Scope	declared model domain versus universal claim
Which questions remain outside it?	What the Postulates Do Not Say	formal rule versus interpretation and mechanism

These eight articles form the planned chapter.

Suggested Routes

First systematic pass

Read why postulates matter, the minimal package, assumptions and scope, and what the postulates do not say. This gives the theory and its limits before choosing a representation.

Matrix-first route

Read finite-dimensional postulates and the density-matrix formulation, then apply them to spin, qubits, entanglement, and generalized measurements.

Wave-mechanics route

Read wave-mechanics postulates after states and representations, then continue to the foundations and solvable models of the canonical-systems volume.

Formulations route

Read equivalent formulations and then use the Translation Table of Formulations and Which Formulation Should I Use?.

Consistency Checks

• State the system class and Hilbert space before choosing coordinates.
• Identify which rule supplies each probability, evolution, composition, and update.
• Verify translations by comparing operational predictions, not notation alone.
• Transform states and observables consistently under representation or picture changes.
• Distinguish a theorem derived within a postulate set from an empirical or interpretive claim about the postulates.
• Make domains, boundary conditions, regularization, and approximation assumptions visible when they matter.
• Do not apply a finite-dimensional identity blindly to unbounded operators.
• Check whether open-system, relativistic, identical-particle, or field-theoretic structure requires an extended framework.

Common Mistakes

• Treating one textbook list as uniquely canonical. Equivalent packages can combine or separate the same ingredients.
• Calling postulates arbitrary because they are stated rather than derived internally. Their empirical adequacy and mutual consistency constrain them.
• Confusing a representation with a new physical theory. Wavefunctions and columns can represent the same abstract state.
• Assuming every formalism is automatically equivalent. Domains, anomalies, regularization, and scope can obstruct a claimed translation.
• Using finite-dimensional intuition for unbounded operators without qualification. Domains and continuous spectra matter.
• Treating state update as a settled ontology. The formal conditional rule does not select one interpretation.
• Expecting postulates to determine a realistic Hamiltonian. Model construction requires additional physics.
• Applying closed-system unitarity directly to a reduced subsystem. Open dynamics generally needs channels or master equations.
• Assuming the compact package is already quantum field theory. Relativistic fields require additional structure.

Cross-Links

• Formalism Overview
• Measurement and State Update
• Time Evolution
• Composite Systems and Entanglement Basics
• Density Operators and Mixed States
• Quantum Dynamics
• Objects and Meaning
• Representation Translation Table

References

• P. A. M. Dirac, The Principles of Quantum Mechanics, 4th ed., Oxford University Press, 1958.
• J. von Neumann, Mathematical Foundations of Quantum Mechanics, Princeton University Press, 1955.
• G. W. Mackey, Mathematical Foundations of Quantum Mechanics, Dover, 2004.
• R. Shankar, Principles of Quantum Mechanics, 2nd ed., Springer, 1994.
• A. Peres, Quantum Theory: Concepts and Methods, Kluwer, 1995.
• L. E. Ballentine, Quantum Mechanics: A Modern Development, 2nd ed., World Scientific, 2014.