David Deutsch and Chiara Marletto have demonstrated a way to formulate the laws of physics without explicitly including time, a longstanding challenge in both quantum mechanics and general relativity. Their research, completed in June 2026, uses ‘constructor theory’, which defines physical laws by the possible transformations of systems using cyclical devices.
“Fundamental laws of physics do not refer to time,” the researchers write, instead focusing on what transformations are possible or impossible. This approach avoids the problematic treatment of time as a non-physical variable, while still accounting for duration and dynamics, a step that is important for a truly timeless physics.
Time’s troublesome role in physics
The ability to define isolated systems without reference to time is central to a new formulation of physical laws, allowing for the specification of possible and impossible transformations with unbounded accuracy. This approach, detailed in work completed in June 2026, moves beyond simply eliminating time as a variable; it actively addresses how duration and dynamics can still be meaningfully described within a framework that doesn’t rely on a temporal parameter.
The research establishes that a substrate is considered isolated if the possibility of any task performed on it depends solely on its own attributes, a definition independent of temporal considerations. Consequently, a composite substrate comprised of multiple isolated substrates also maintains that isolation, a principle mirroring the traditional understanding of time’s role in synchronizing mutually isolated clocks. The concept of isolation necessitates a re-evaluation of how motion and spacetime are incorporated into fundamental laws.
The work infers that the state of motion of a substrate must be considered an attribute, while the spacetime region it occupies must be treated as an additional substrate within constructor-theory-compatible laws of physics. The implications extend to the very definition of dynamics, requiring the identification of a law specifying how variables of an isolated substrate change relative to the standard labels of a class of timers, effectively recreating the sense of temporal evolution without explicitly invoking time itself.
The ability to express the dynamical evolution of a clock, with its discrete or continuous pointer variable, is a key outcome of this approach. This isn’t merely a mathematical trick to remove time from equations. It’s a fundamental restructuring of how physical laws are conceived, focusing on what transformations are achievable rather than how quantities change over time.
The researchers’ work shares some features with other ‘timeless’ approaches to time (Wheeler 1968, Dewitt 1967, Page & Wootters 1983, Rovelli 2004). The traditional conception of physics often treats an isolated system as one whose interactions don’t depend on variables of other systems. But this definition relies on an assumed background of time. In contrast, constructor theory defines this property without referencing time or dynamics, focusing instead on the attributes of the system itself.
This shift in perspective allows for a more fundamental understanding of isolation, establishing it as an intrinsic property rather than a consequence of temporal independence. The definition of accuracy, important to the constructor theory framework, is defined as the distance between the desired output of a task and the actual output of a system performing it, measured during a single run.
Reliability, equally important, is the degree to which a system maintains that accuracy when repeatedly used to perform the same task. These definitions of accuracy and reliability are not merely technical details; they are integral to the entire constructor theory program, providing a rigorous way to assess the feasibility of physical transformations. The framework also requires consideration of ‘superinformation media’, potentially imposing further conditions on any complete theory.
The researchers acknowledge that additional requirements may arise from specific theories. But these must be expressed as constraints on possible and impossible tasks, rather than as relationships between variables in a traditional time-dependent framework. The work doesn’t simply offer an alternative mathematical formulation. It proposes a different way of thinking about the fundamental laws of physics, one that prioritizes what can happen over how things change with time.
Defining laws without defining time
The framework establishes isolation not as a consequence of dynamics, but as a fundamental property definable without reference to how systems change over time. This redefinition moves beyond simply describing how isolated systems behave; it specifies what constitutes isolation itself, based on the feasibility of tasks performed on a substrate. This criterion, unlike traditional definitions, sidesteps the need to define isolation in terms of interactions evolving over time, instead focusing on the inherent properties of the substrate itself.
This approach to isolation has implications for understanding the emergence of clocks, devices traditionally used to measure time’s passage. The research demonstrates how the dynamical evolution of a clock can be expressed within this timeless framework, not as a process unfolding in time, but as a series of possible transformations.
Rather than describing how a clock ticks, the framework defines the conditions under which a clock’s behavior is consistent with the fundamental laws, framing it as a system capable of undergoing specific, repeatable tasks. This is distinct from existing timeless approaches, which often rely on pre-existing assumptions about dynamics and the properties of natural laws, such as the existence of good clocks and the principle of locality.
The work diverges from these existing approaches by offering an explanatory foundation for them, rather than simply adopting their formalisms. The researchers posit that properties like the possibility of good clocks and locality are not axioms to be assumed, but consequences of deeper, timeless laws. These laws are expressed as requirements about possible and impossible tasks, potentially involving the existence of ‘superinformation media’, generalized quantum systems, as a further condition.
The possibility of superinformation media, as defined in earlier work, is a technical detail and a potential constraint on the laws themselves. This emphasis on task-based definitions is central to the entire framework, providing a consistent way to formulate laws without relying on a time parameter. A task is considered possible if a single constructor can effect the transformation on any member of a set of substrates without needing to know which substrate it is operating on.
This requirement is stricter than simply allowing different constructors for each substrate, highlighting the need for a universal capability. The example of reversing the spin of a qubit illustrates this point; the laws of quantum theory forbid such a task because it would violate unitarity, demonstrating how the framework can encode fundamental physical constraints.
By focusing on what transformations are possible, the framework offers a new perspective on the nature of physical laws themselves. The researchers suggest that these laws are not statements about how the universe evolves, but about what the universe allows to happen, framing physics as a theory of possibilities rather than a description of a pre-determined timeline.
Duration and dynamics remain elusive
Defining duration within a framework that intentionally excludes time presents a unique challenge, as the very concept of a time interval requires a standard against which to measure change. The research details how a ‘timer’, a specific type of constructor, can implicitly define duration through its attributes, circumventing the need to explicitly invoke a time parameter. These timers rely on distinguishable states, including a ‘halt flag’ to signal completion, but the choice of attributes defining these states is not unique. A substrate may satisfy the conditions for multiple configurations.
Catalysts and heat engines are familiar approximate constructors. This requirement is important for establishing equivalence classes of timers, each labeled by a specific duration, allowing for the recovery of dynamics from timeless principles.
The implications extend to existing theories attempting to reconcile quantum mechanics with gravity, such as those proposed by Wheeler & DeWitt, Ashtekar, Kiefer and Rovelli, all of which have explored timeless formulations. While these approaches share the characteristic of not explicitly referencing time, the current research provides a foundation for rooting dynamics in general, timeless principles of physics, specifically through the use of timers and clocks.
Theories conforming to constructor theory but not supporting timers are not necessarily invalidated by their conditions, leaving room for further exploration of alternative formulations. The definition of an isolated system, traditionally linked to the absence of external dynamical interactions, is reformulated in constructor theory to remove any reference to time or dynamics. This isolation is a key component in defining the behavior of timers, ensuring that their internal states are not influenced by external factors during operation.
The research establishes conditions for timer-supporting theories that are constructor-theory compliant, providing a pathway to recover dynamics as emerging from timeless principles. This approach differs from traditional formulations where time is often treated as a fundamental parameter governing change. Instead, duration is seen as an emergent property arising from the relationships between physical systems and the tasks they can perform. The authors emphasize that a constructor-theoretic specification of a task must not mention the constructor itself, focusing instead on the transformation achieved.
The concept of equivalence classes is central to defining duration; timers belonging to the same class share a common duration, regardless of their specific attributes. This allows for the establishment of a consistent framework for understanding how time and dynamics can emerge from timeless principles.
The work builds on previous formulations of quantum theory, such as those by Page & Wootters, Smith, Kuypers and Rijavec, which have also explored timeless approaches to physics. The researchers have provided the foundations for timeless approaches to dynamics, rooting them in general, timeless principles of physics, and establishing a framework for future investigations into the nature of time and change. The authors’ work provides an important step towards a more complete and consistent theory of physics, one that does not rely on the problematic concept of time as a fundamental parameter.
Rebuilding physics with constructors
The concept of a ‘null task’, one devoid of constraints on input or output, forms an important element in expressing dynamics within constructor theory, allowing for the description of physical laws without explicit reference to time. This task, represented by an empty set, permits the formulation of evolution for a system, such as a clock with a measurable pointer, in a way that aligns with timeless principles.
The ability to define even the absence of constraints as a task emphasises the theory’s focus on what is possible and impossible, rather than on the passage of time itself. This approach differs from traditional physics, where evolution is often described using parameters that directly measure time’s flow. The composition principle, a foundational tenet of constructor theory, dictates that combining possible tasks must also yield a possible task, extending this logic to isolated systems.
This principle, alongside the locality principle, establishes a framework for understanding how interactions can occur without relying on a pre-defined temporal order. The implications extend beyond simply removing time as a variable. It necessitates a re-evaluation of how the researchers define and understand interactions between physical systems. While constructor theory does not mandate the existence of substrates in complete isolation, it does require the possibility of a constructor operating on them.
This distinction is subtle but significant, emphasizing the focus on the feasibility of transformations rather than the absolute state of isolation. The theory’s compatibility with ‘timeless’ approaches to physics, those that seek to formulate laws without explicitly referencing time, is a key area of exploration, and this work provides conditions for such theories to be consistent with constructor theory.
The possibility of constructing theories that adhere to these principles, yet do not rely on timers, remains open, demonstrating the flexibility of the constructor theory framework. This flexibility is not a limitation, but rather shows the theory’s ability to accommodate a range of potential physical laws. The definition of an isolated substrate, within constructor theory, hinges on the possibility of performing tasks depending only on its own attributes, a departure from traditional physics’ reliance on variables describing the state of the universe.
This approach allows for a description of physical laws that is independent of any external reference frame or temporal parameter. The ability to express laws in terms of possibilities and impossibilities of tasks, rather than equations governing the evolution of systems, represents a fundamental shift in perspective.
The framework’s strength lies in its ability to define constraints on physical systems without invoking the concept of time itself. The researchers emphasize that the conditions they have established do not preclude theories that do not support timers, highlighting the theory’s capacity to accommodate diverse physical models. This suggests that the existence of timers, while consistent with the framework, is not a necessary condition for a constructor-theory compliant theory of time.
The implications of this approach extend beyond theoretical physics, potentially influencing their understanding of causality and the nature of reality. The concept of a ‘constructor 1’, an idealized device capable of performing transformations perfectly and repeatedly, is a useful abstraction, even though real-world devices are inherently approximate.
As in those approaches, it regards time as having no existence independently of physical objects, and as existing only as relationships between their attributes such as clock readings. This abstraction is not merely a mathematical convenience; it reflects a fundamental shift in how the researchers think about the nature of physical laws and their relationship to the physical world.
See today’s quantum computing news on Quantum Zeitgeist for the latest breakthroughs in qubits, hardware, algorithms, and industry deals.
