Natural Born Historians

Hamish Linehan (University of Stirling) and Matthew Sims (Macquarie University)

Philosophers often maintain that human memory is distinctive. This purported uniqueness is typically grounded in two claims: first, that human memory possesses a normative epistemic character -- i.e. there are conditions under which memory can be true or false and second, that it can be encoded, shared, and communicated using external objects and systems -- so-called “memory technologies.”

One explanatory move linking these two claims is that memory technologies are used to address the natural limitations of “bio-memory” – understood here as our capacity for remembering without any external aids. On this account, such technologies compensate for the constraints of our cognitive system. We propose that an alternative account is possible. Rather than treating the normative epistemic character of memory as primary, it may instead be partially the result of an evolved disposition toward the externalisation of memory.

Our memory systems have co-evolved with the development of external objects used to record the past, and this co-evolution may be responsible for the normative, epistemic character of memory.

Some philosophers take the constructive, error-prone nature of memory to indicate that its primary function is not to accurately record the past. While memory can, under certain conditions, represent the past accurately, specifying these conditions differs from providing a descriptive account of the cognitive systems responsible for memory. Accordingly, “memory” can be understood as equivocating between two concepts: memory in an epistemic sense, as a source of knowledge about the past; and memory in an empirical sense, as an explanation of what causes one to have a memory at all.

In its epistemic sense, memory is normative: it can yield knowledge about the past, and that knowledge can be correct or incorrect. We regard memories as accurate insofar as their contents are akin to the past events they represent. To establish this, we judge whether a memory can be corroborated by other records. For example, if the contents of a remembered event align with contemporaneous photographs or written accounts, it will be judged more accurate than one that is misaligned.

Memory technologies, then, can be used to establish standards for the accuracy of memory. We judge whether our memories are correct by deferring to external records. Like other technologies, memory technologies address recognised limitations of our cognitive systems. Consider the use of a calendar: appointments are recorded externally, and this record serves as a check against our tendency to forget important events. The underlying rationale is straightforward: we recognise that unaided memory is unreliable for certain tasks, and therefore external supports are needed to mitigate these shortcomings. In this sense “our memory systems are characterised by an openness to the world, in that we incorporate artifacts and technologies into our memory systems”.

Memory technologies can be closely linked to the normative epistemic dimension of memory. At the individual level, they support tasks such as remembering intentions or preserving information that would otherwise be lost. At the collective level, they function as channels of cultural inheritance, enabling the transmission and refinement of knowledge across generations (e.g., in tool use). Artifacts -- written language, maps, information stored in smartphones -- are often treated as paradigmatic memory technologies, serving as external representations that sustain memory both individually and diachronically. In this way, they preserve information about the past in forms intelligible to others, thereby mitigating the limitations of fallible memory.

These examples are exclusively human. This reflects a broader tendency in the literature. Most philosophers assume that the normative epistemic character of memory is unique to humans. Nonhuman animals, it is argued, are not “knowers” in the relevant sense: they lack the capacity for self-conscious reflection on their representations and cannot evaluate them as true or false. Consequently, their memory is taken to lack a normative epistemic component and therefore to have no need for memory technologies of comparable scale or complexity.

Nevertheless, certain cases suggest that other animals exhibit a similar -- albeit more limited -- openness to external artifacts. We can illustrate this with the case of the bowerbird. These Australian birds construct elaborate courtship structures (“bowers”), decorated with collected objects. The intricacy and composition of these displays plausibly signal past foraging success, thereby indicating fitness as a mate. Importantly, bowerbirds preferentially select materials -- often choosing non-organic objects -- that decay slowly, allowing the display to accumulate over time.

Similarly, consider the acellular slime mould Physarum polycephalum – a protist. As this large unicellular organism moves across surfaces while foraging for microbes, it deposits a trail of extracellular slime. Upon re-encountering this trail, Physarum avoids the area. In doing so, it effectively tracks previously foraged locations, conserving metabolic energy by not revisiting depleted regions. The production and layout of these slime trails function as part of the organism’s navigational memory system, providing information about past foraging events that its internal structure alone could not store.

This can plausibly be interpreted as a case in which the cognitive system has evolved in dependence on environmental scaffolding, possibly through evolutionary repurposing (exaptation) of extracellular slime’s original lubricant function to guide future navigation based on past activity.

While we may hesitate to equate human memory technologies with the use of such materials by these organisms, their complex behaviours — mating and navigation -- rely on the presence of, and response to, external materials. Their involvement is not self-reflective but suggests that these cognitive systems are organised in ways that depend on external structures for specific tasks. A similar possibility arises in the human case. Although memory technologies are often developed to address identifiable problems, their prevalence across cultures and historical periods suggests that reliance on external memory aids may be a constitutive feature of our cognitive systems. Our memory systems’ evolution may have resulted in a reliance on external materials to veridically capture the past.

As such, human memory can be framed as fundamentally open to, and reliant upon, the use of memory technologies.

Offloading information about the past onto external materials can be understood as an ongoing, pre-reflexive feature of cognition. This openness, rather than being uniquely human, may be continuous with the forms of environmental reliance observed in other organisms. Just as bowerbirds are natural-born foragers, and Physarum natural-born microbe-eaters, humans might be framed as natural-born historians.

Our contention is that this reliance on memory technologies may have contributed to memory’s normative epistemic character. Human memory systems may have developed an openness to such technologies, in much the same way that the behaviours of Physarum and bowerbirds presuppose particular environmental conditions. Our memory systems presuppose the availability of memory technologies and the possibility of veridically recording the past, just as the bowerbird’s cognition presupposes an environment rich in materials for constructing bowers.

On this view, unaided memory should not be regarded as inherently fallible or error-prone; rather, remembering without external supports is an impoverished version of a system that normally incorporates them. Similarly, without access to slime trails, Physarum can still navigate, but in a way more prone to error. This is not because Physarum is a poor navigator, but because its navigational system has developed an ability to offload information onto the environment.

Why, we might argue, would our memory systems prioritise accuracy when we could offload this task to external resources? If our memory systems include external aids, and if these aids support accurate representation of the past, then the normative epistemic character of memory becomes less puzzling. By offloading aspects of accurate record-keeping onto the environment, memory systems may have been free to evolve in other directions – supporting further sociality, cooperation, and reflection.

Hamish Linehan is a philosophy PhD researcher currently enrolled in the St Andrews and Stirling Graduate Programme. He works in the University of Stirling's Centre for the Sciences of Place and Memory, where he explores the ways in which our autobiographical memories and sense of self are enacted and distributed across people, places, and objects.

Matthew Sims is an affiliate at the Department of History and Philosophy of Science at the University of Cambridge and a postdoctoral researcher at Macquarie University where his research focuses on biological agency, memory in non-neuronal organisms, and evolutionary theory. He is the author of the book Slime Mould and Philosophy, Cambridge University Press.