How Do the No-Deleting and No-Cloning Theorems Relate?

This article explores the foundational relationship between the no-deleting theorem and the no-cloning theorem in quantum information theory. Together, these two principles establish the conservation of quantum information by proving that an arbitrary unknown quantum state can neither be duplicated nor destroyed in isolation. Understanding how these dual theorems complement one another reveals why quantum data behaves like a conserved physical quantity rather than classical digital bits, and how this symmetry governs quantum computing, memory management, and cryptography.

In classical computing, copying and deleting data are routine operations performed billions of times per second. A classical bit exists in a definite state of either zero or one, allowing a processor to read its value without disturbance, replicate it across multiple memory registers, or overwrite it with zeros. Quantum mechanics operates under fundamentally different rules because a qubit can exist in a linear superposition of basis states. Measuring an unknown quantum state collapses its superposition, destroying the original coherent information before it can be duplicated or erased through classical observation.

Formulated in 1982 by Wootters, Zurek, and Dieks, the no-cloning theorem proves that no unitary quantum operation can take an arbitrary unknown quantum state and a blank target qubit and transform them into two identical, independent copies of the original state. This restriction follows directly from the linearity of quantum mechanics. If a hypothetical cloning operator worked for basis states, applying it to a superposition of those states would produce an entangled state rather than a separable product of two identical copies. Only orthogonal states that are known to belong to a specific basis can be copied reliably.

Nearly two decades later, in 2000, Arun Kumar Pati and Samuel L. Braunstein introduced the no-deleting theorem, which addresses the opposite process. Suppose a system already possesses two identical copies of an arbitrary unknown quantum state. The no-deleting theorem proves that there is no linear, isometric quantum operation capable of resetting one of those copies to a standard blank state while leaving the remaining copy intact and keeping the ancillary environment independent of the deleted state. While a known state can easily be rotated back to a reference state, an unknown state cannot simply be wiped out.

The no-deleting theorem complements the no-cloning theorem as its time-reversed dual. In a closed quantum system, all state evolution is governed by unitary operators, which are mathematically reversible and preserve inner products between states. Cloning attempts to transform one copy and a blank state into two copies, whereas deleting attempts to transform two copies into one copy and a blank state. If a universal quantum deleter existed that did not leak information into an ancillary system, reversing its unitary dynamics would yield a universal quantum cloner. Even when allowing the ancillary state to change, linearity forces any valid deletion process to simply move the quantum state intact into the ancilla rather than destroying it.

Together, these two theorems establish a profound conservation law for quantum information. Just as the first law of thermodynamics states that energy can neither be created nor destroyed in an isolated system, the combination of no-cloning and no-deleting demonstrates that unknown quantum information cannot be created from nothing or annihilated into nothing under unitary evolution. When a quantum state appears to be erased through environmental decoherence or resetting, the information is not destroyed; instead, as further formalized by the no-hiding theorem, it is transferred into the surrounding environment or into correlations across a larger composite system.

This symmetry has direct practical consequences for quantum algorithm design and hardware engineering. Because quantum programmers cannot simply delete temporary working qubits at the end of a subroutine without leaving the main register entangled with the discarded qubits, they must rely on uncomputation. Uncomputation runs reversible gate sequences backward to disentangle ancillary qubits and return them cleanly to their initial blank state. In quantum cryptography and distributed quantum networks, the inability to either clone or silently delete unknown states guarantees that an eavesdropper cannot copy transmitted qubits or erase evidence of interception without introducing detectable errors.