Mnemonic agency?
Marina Trakas
You forget your mother’s birthday. You remember telling your partner to pick up your child from school, but he insists you did not (and in fact, you didn’t). You are considering promoting an employee, and all the mistakes made by the default male candidate fade away, along with the accomplishments of women, Black employees, and other minorities. These are all common, everyday cases of memory errors and distortions, arising from forgetting, memory biases, or interference. None of them is intentional. Despite this, it is common practice, at least in Western societies, to blame those who misremember or forget, as if they could have acted differently. We often say things like: How could you forget your mother’s birthday? You should have told me to pick up Marty. You didn’t remember everything Maria did for the company? Through such speech acts, we routinely hold people responsible for their memory errors and distortions, guided by the intuitive belief that they exercise some form of control over their remembering, even if most of us would struggle to specify what that control consists in.
This intuitive belief is reinforced by the fact that memory errors and distortions are often not innocent and can have negative consequences for others. If your mother attaches great importance to birthdays, forgetting hers will almost certainly cause her affective harm. Similarly, a child who is not picked up by their parents may experience significant stress and anxiety. In another register, if a careless male employee is promoted over a woman who worked harder because her achievements were forgotten due to a stereotype, this not only harms the unrecognized employee in multiple ways but also constitutes an injustice (more specifically a mnemonic injustice: Puddifoot, forthcoming; Trakas & Puddifoot, 2025).
But it is not only moral and normative considerations that lead us to treat people as responsible for their memories, but also legal ones. Cases in which parents have been criminally charged with negligence for hot-car deaths involving forgotten babies (Breitfeld, 2020) suggest that society does not view memory as something that merely happens or fails to happen to us. Rather, they reflect the assumption that remembering is not a purely passive process and that some form of mnemonic agency may be involved.
Yet, given that control over our memories is far from obvious, a key question remains: what kind of mnemonic agency, if any, can actually be exercised? Answering this question is crucial, because although there are moral, normative, and legal reasons for holding people responsible for their memories, we may simply be mistaken in thinking that there is some form of control we can exert over them.
This is one possibility. One might argue that there is in fact very little we can do. We can intentionally direct our attention to the event or information we want to remember, and we can rehearse it repeatedly in order to increase the likelihood of successful encoding. But this may be all that is available to us. These are diachronic strategies (Kennett & Smith, 1996; Burdman, 2023), that is, strategies adopted in advance to increase the likelihood of remembering. Studying for an exam is a good illustration of this point. You can study carefully, with focus and repetition, and nonetheless fail to recall the relevant information at the moment of the exam. The information may have been lost, not properly encoded, or may simply fail to be retrieved, and there is nothing that can be done to avoid this outcome. Let us call this view the beyond-our-control hypothesis. According to this hypothesis, whatever mnemonic agency we can exercise operates only prior to the core memory processes of encoding, consolidation, and remembering. In this sense, the agency at stake has little claim to being genuinely mnemonic.
Is the beyond-our-control hypothesis plausible? Are we simply mistaken whenever we hold people responsible for misremembering and forgetting? This hypothesis assumes that we are not genuinely agents when we remember, and that memories simply pop up in our minds (or fail to do so), in a way not fundamentally different from compulsory actions. But is this really the case?
In principle, this description fits one kind of memory, namely what is known in cognitive psychology as “involuntary memory.” Conceptualized at least since Aristotle’s distinction between mnēmē, understood as a passive affection resulting from perception, and anamnēsis, conceived as an active and organized search of the mind through associative processes (Aristotle, ca. 350 B.C.E./1966), involuntary memories have been largely neglected in cognitive science, except in trauma-related research. In no small part due to the work of cognitive psychologist Dorthe Berntsen over the last decades (see Berntsen, 2009), the view that involuntary memories are not an uncommon phenomenon restricted to trauma, but rather a universal phenomenon occurring as frequently as voluntary memories, has become increasingly widespread.
It is not obvious that the voluntary-involuntary distinction, originally modeled on an analogy with physical actions, can be straightforwardly applied to mental states such as memories. Even if it can, it is unclear whether it captures a strict, binary distinction rather than a continuum. Further theoretical work is clearly needed on this front. Still, if involuntary memories were taken to support the beyond-our-control hypothesis, their counterpart, voluntary memories, would seem to provide evidence against it. Voluntary memories are typically conceived as intentionally initiated, involving controlled and goal-directed associative processes, and thus as requiring cognitive effort and a form of agency that is properly mnemonic.
Given that some degree of control appears to be exercisable in voluntary remembering, the hypothesis could be reformulated so as to accommodate this possibility. On this reformulation, it may be called the voluntariness-based control hypothesis: the kind of memory retrieval determines whether mnemonic agency is possible. Nonetheless, equating a specific kind of retrieval with mnemonic agency does not seem to be a good move, as it may obscure the fact that memories that initially come to mind as if we were passive receptacles can also become objects of voluntary control. In fact, retrieval processes are ill-conceived when they are reduced to a single mental act that terminates in a memory, whether a memory that simply pops up or one that is eventually found. Many retrieval processes are complex and extended in time, and this temporal extension is often filled with intentional and goal-directed mental actions aimed at monitoring and controlling the information retrieved. Metacognition does not need to be automatic and implicit; it can also be high-level, explicit, and rational (Arango-Muñoz, 2011; Trakas, 2019).
Accordingly, if we care about not harming others through forgetfulness or misremembering, we have at our disposal synchronic strategies (Kennett & Smith, 1996) that allow us to further scrutinize our memories independently of how they are retrieved. We can compare them with other memories and beliefs, assess whether something relevant is being forgotten, and examine our interests and prejudices that may lead to distortions. Ultimately, nothing forces us to endorse a memory we distrust, and being cautious and refraining from blindly trusting the first memories that come to mind can sometimes be the most appropriate response (Trakas, 2021; 2024). Furthermore, not only mental actions but also behavior can be part of these explicit, conscious, and temporally extended metacognitive processes. Interaction with other people and with the external world can function as scaffolding processes for memory, positively shaping and refining our individual ways of remembering the past (Sutton et al., 2010).
Therefore, both intrapsychic and externally scaffolded strategies (Burdman, 2023) are available to subjects to exert control over their memories, independently of the way in which those memories are retrieved. Let us call this view the general mnemonic agency hypothesis. Unlike the previous hypothesis, it holds that mnemonic agency is, in principle, possible with respect to all memories. This does not imply that mnemonic agency can in fact be exercised over every memory, nor that all memories allow for the same degree or kind of such agency. Different brain dysfunctions and dysregulations, as well as specific circumstances and situations, may significantly limit or even impede the exercise of mnemonic agency. Nor does the hypothesis require that mnemonic agency be exercised every time we remember. Doing so would be cognitively costly, often unnecessary, and in many contexts simply unviable. What it does imply is that we possess the cognitive capacity to monitor and control our memories, and that this capacity is part of our cognitive architecture. Importantly, this also means that moral, normative, and legal considerations that lead us to treat people as responsible for their memory errors are not groundless. Rather, they are anchored in genuine cognitive possibilities. The general mnemonic agency hypothesis therefore provides the most plausible framework for understanding mnemonic agency. It grounds our practices of responsibility while leaving open the normative question of their appropriate scope.
References:
Arango-Muñoz, S. (2011). Two levels of metacognition. Philosophia, 39(1), 71-82.
Aristotle. (1966). On memory and recollection (W. S. Hett, Trans.). Harvard University Press. (Original work published ca. 350 B.C.E.)
Berntsen, D. (2009). Involuntary autobiographical memories. An introduction to the unbidden past. New York: Cambridge University Press.
Breitfeld, E. (2020). Hot-car deaths and forgotten-baby syndrome: A case against prosecution. Berkeley Journal of International Law, 25, 72-106.
Burdman, F. (2023). Diachronic and externally-scaffolded self-control in addiction. Manuscrito, 46, 77-116.
Kennett, J., & Smith, M. (1996). Frog and Toad lose control. Analysis, 56(2), 63–73.
Puddifoot, K. (forthcoming). ‘Mnemonic Injustice.’ In Wright, S. and Goldberg, S. (Eds), Memory and Testimony: New Essays in Epistemology. OUP.
Sutton, J., Harris, C. B., Keil, P. G., & Barnier, A. J. (2010). The psychology of memory, extended cognition, and socially distributed remembering. Phenomenology and the cognitive sciences, 9(4), 521-560.
Trakas, M. (2019). On epistemic responsibility while remembering the past: the case of individual and historical memories. Les ateliers de l’éthique, 14(2), 240-273.
Trakas, M. (2021). The sense of mineness in personal memory: problems for the endorsement model. Estudios de Filosofía, 64, 155-172.
Trakas, M. (2024, June 19). “I forgot that you existed”: Making people responsible for their memories. Imperfect Cognitions. https://imperfectcognitions.blogspot.com/2024/06/i-forgot-that-you-existed-making-people.html
Trakas M, Puddifoot K. (2025). Fear Generalization and Mnemonic Injustice. Episteme, 22(2), 379-405.