The hidden power of implicit collective memory

Abstract

Over the past decades, the field of memory studies has produced a wealth of research on explicit (conscious, commemorative, official) collective memory. But beyond this realm of the visible, there is a largely hidden world of ‘implicit collective memory’. Elements of this invisible world include narrative schemata, stereotypes, patterns of framing, or world models, which are usually not explicitly known or addressed, but get passed on from generation to generation – in order to shape perception and action in new situations. Implicit collective memory is pervasive and powerful. But it is difficult to trace. It is therefore time to join forces for its systematic study: Drawing on approaches from psychology, sociology, communication studies, anthropology, media culture studies, literary studies, and mnemohistory, this article proposes some building blocks for a future transdisciplinary field of research on implicit collective memory.

‘Unusual suspects’: moving beyond commemorative memory

‘Round up the Unusual Suspects!’ Vered Vinitzky-Seroussi's (2011) exhortation is the starting point of this article. Certain forms of collective memory are by now very well-researched. Among the ‘usual suspects’ of interdisciplinary memory studies are forms of explicit, identity-creating, and often official commemoration – acts of memory in any case, which actors are aware of. Such conscious acts of memory are an important, visible and much-discussed part of memory culture. But they do not represent the entire range of possible relations between minds, media communication, and (social) environments.

In fact, as Michael Schudson (1997, 3) remarked, conducting memory studies exclusively as commemoration studies is a bit like the drunk man who only looks for his car keys under the street lamp. In other words: ‘Not all of what societies remember is recalled through or in relation to self-conscious or dedicated memory projects. Instead, the past is often incorporated into the present in ways that do not aim at commemoration’ (Schudson 2014, 85). Schudson thus draws an important distinction between ‘commemorative memory’, on the one hand, and a whole variety of other forms of collective memory, which he subsumes under the term ‘non-commemorative memory’.

The present article builds on such calls to move beyond the predominance of memory studies as commemoration studies. For reasons of terminological precision and interdisciplinary interconnectability (outlined below), it suggests a distinction between explicit and implicit forms of collective memory. It proposes ‘implicit collective memory’ as a cover term for the myriad possibilities of the past affecting the present in ways that most people remain unaware of.

Memory studies have not even begun to systematically address the usually hidden but powerful dynamics of implicit collective memory. In what follows, I will discuss the role of media communication for processes of remembering that remain non-conscious on a collective level. And to show that implicit phenomena are (and have always been) a(n) (equally quite hidden) key concern of memory studies, I will engage with perspectives from a wide range of memory research in various disciplines.

What does ‘implicit memory’ mean? Psychological perspectives

In cognitive psychology, where the term was coined (Graf and Schacter 1985),‘implicit memory’ refers to processes of individual memory that subjects are not conscious of. Cognitive psychology posits different systems of memory. Endel Tulving (1983) famously differentiated between semantic memory (‘knowing that’) and episodic memory (‘remembering’). Both are forms of explicit memory, which enables conscious (declarative, intentional) recall. Implicit memory, on the other hand, is behind all non-conscious acts of memory, which can range from procedural memories (i.e., for binding shoelaces or riding a bike, ‘knowing how’) all the way to effects of perceptual and conceptual priming (see Schacter 1987; on procedural vs. declarative memory, see Squire 1987).¹

Priming is a standard psychological method to make implicit memory empirically observable in the laboratory. One typical example of (perceptual) priming research in cognitive psychology is the word-stem completion task, where participants are first shown a list of words and later asked to quickly generate words from the first three letters (so that ele__ could be completed into elephant, elevator, etc.). The result is that participants tend to resort to the words seen earlier, but without being aware of the connection. Importantly, amnesiacs, who cannot have conscious memory for the previously shown lists of words, perform on this task just as well as healthy people. (For an overview, see Roediger 1990) Put in a nutshell, when seen through the lens of priming implicit memory appears, in the words of cognitive psychologist Henry L. Roediger (1990), as a form of ‘retention without remembering’. Similarly, social psychologist Don Carlston (2010, 4) explains in the Handbook of Implicit Social Cognition: ‘Although measured in a variety of ways, implicit memory has been defined fairly consistently as influences of past experience on later performance, in the absence of conscious memory for the earlier experience’.

In a broader memory studies perspective, there are some particularly relevant effects of priming, among them the ‘mere exposure effect’ (Bornstein and Craver-Lemley 2017), and the ‘validity effect’ (Renner 2017), i.e., the fact that what is seen and heard again and again, often without awareness, seems familiar, is liked, and is even considered true. Behavioural economist Daniel Kahneman (2011, 85) explains: ‘A reliable method to bring people to believe in false statements is frequent repetition, because familiarity cannot easily be distinguished from truth’. Priming can also lead to ‘mental contamination’ (Schacter 2002, 301) by sexist and racist stereotypes as well as to non-conscious forms of plagiarism – cryptomnesia.²

One reason for all these effects is that implicit memory involves source amnesia. The source of a piece of information and its status (is it someone's lived experience or hearsay? Is the source reliable or unreliable, fictional or factual?) do not seem to play a great role whenever items are encoded and activated non-intentionally. What is retained in the framework of implicit memory feels familiar (pleasant, natural, true) and can even assume the subjective, ‘episodic’ quality of personal experience.³

In a psychological perspective, implicit memory is an apparatus of automatic, ‘fast thinking’, as Kahneman calls it in Thinking, Fast and Slow (2011). He posits a psychological ‘double process model’ and distinguishes between controlled (conscious, intentional) and automatic (non-conscious) forms of human information processing. For the social psychologist John Bargh, automatic cognitive processes have four characteristics: ‘lack of awareness, lack of intentionality, lack of controllability, and high efficiency (nonreliance on cognitive resources)’ (Bargh 1994; Carlston 2010, 40).

‘High efficiency’ – this is a central attribute of implicit memory. At the same time, such powerful ‘fast’ forms of memory often remain entirely hidden. They are non-conscious, non-intentional, and uncontrollable. This is why, in his now-classic introduction to the psychology and neuroscience of memory, Searching for Memory (1996, 161), Daniel Schacter writes about ‘the hidden world of implicit memory’. Moving on to the broader field of memory studies, and with a nod to Schacter, I point to ‘the hidden power of implicit collective memory’, thus accentuating not only the invisibility, but also the forward-pushing, future-making capacities of implicit collective memory.

A hidden power: implicit memory as a collective phenomenon

But how can we start thinking about implicit memory as a collective phenomenon? To begin with, it is worth recalling what is meant by ‘remembering’ and ‘forgetting’, when these processes are understood not just as phenomena of individual psychology but seen in their (actual) distribution across assemblages of biological, mental, sociocultural, and material elements. For memory studies, ‘memory’ exists only in this extended form as an ‘ecology of memory’ (Hoskins 2016; see also Sutton et al 2010). ‘Collective memory’ is therefore a tautology. I use the term only to be clear and also with a nod to a terminological tradition going back to Maurice Halbwachs (1925), who claimed that all memory is always already collective memory.⁴

Collective remembering does not mean that all individuals would have identical mental representations in their minds. Instead, it means that certain versions of the past are actualised again and again within social groups (via discourses, media, practices), and that they are well-networked with other topics. Similarly, collective forgetting does not mean, as Guy Beiner (2018) has shown, that all traces, all knowledge of a past event would be lost. It means that within certain social frameworks, there are no acts of remembering traceable. For example, memories of particular past events can be avoided, kept secret, tabooed, or may seem difficult to articulate in public. But often, these events will ‘live on’ as familial or local memories – or within the semantic collective memory of science (as has been the case with the Spanish Flu, see Beiner 2022). The logic of collective memory thus implies, in Barry Schwartz’ (2009, 23) words, that ‘remembering and forgetting are distributed unevenly among different communities, groups, and individuals’.

Implicit collective memory has to be conceptualised according to this logic, too. What the majority of people remain unaware of can be quite obvious to some observers: Cultural stereotypes, tabooed pasts, emotional regimes, non-conscious master narratives – for some people, such as newcomers to a society or critical observers, the daily activations and effects of implicit collective memory immediately catch the eye. Certain actors will moreover work quite intentionally with implicit collective memory: journalists using framing techniques; politicians who play with historical allusions; film industries and advertising agencies who exploit cultural schemata and narrative habits. Last not least, it belongs to the tasks of critical media, art, and literature to make the dynamics of implicit collective memory visible – and thus potentially transformable.⁵

How can we then draw a distinction between forms of memory that are collectively non-conscious and those that belong to explicit knowledge? Here as elsewhere in memory studies functional definitions are key: Within social groups, the explicit and the implicit are particular usages and effects of collective memory, and not separate systems. (Whether forms of explicit and implicit memory are located within different regions of the brain and thus dissociable is still a question under discussion; see Carlston 2010.) This means that certain memorata can be the object of knowledge or official commemoration at one point of time, while at another point of time they unfold their world-making power in an unnoticed way.

A striking example is the presence of colonial monuments or racist forms of speaking in post-colonial societies (for their role in Brexit, see Ward and Rasch 2019). Such ‘colonial remains’ can stay implicit – unnoticed by most – for a long time, until their power of stabilising and transmitting forms of (direct and structural) violence across generations is exposed. In this way, memory activists (such as Black Lives Matter) have made explicit in recent years what had long been a powerful, but largely implicit presence (see Otele et al 2021; Rigney 2022). In our self-reflective memory cultures of what Ulrich Beck (2006) has termed ‘second’ or ‘reflexive modernity’, the transformation of implicit collective memory into explicit knowledge and commemoration has been a key concern. This operation was already at the heart of Victor Klemperer's (2013 [1947]) analysis of the language of the ‘Third Reich’ (LTI, lingua tertii imperii) and its detrimental non-conscious afterlives. Another example of a transforming agent between implicit and explicit collective memory is all feminist work that exposes and seeks to rectify the (let's hopefully say: often) non-conscious gender biases of historiography and science (Reading 2016).

But how is implicit collective memory transmitted from generation to generation when it is never explicitly addressed? Schematisation seems to be a powerful process in the travels of implicit collective memory. Examples of schematised memorata are visual icons, narrative patterns, stereotypes, metaphors, world models, values and norms, or certain ways of acting or ‘doing’ things (on habit memory, see Connerton 1989).⁶ They are often loaded with affect. Think of the discursive formula ‘dying for the fatherland’, the millennia-old iconic image of the Pietà, the binary of light-versus-darkness, the narrative pattern of ‘rise and fall’, the stereotype of the ‘yellow peril’ (actualised again in the Coronavirus pandemic). Such potentially perilous packages of schema-cum-affect are passed on – often (though not exclusively) non-consciously – from human to human, from generation to generation. Media in all their forms and appearances play a decisive role in the process: Implicit collective memory is mediated and remediated via gesture and mimics, via orality and literacy, via analogue and digital media (on remediation as the process of conveying memory contents again and again, across time, cultures, and different media, see Erll and Rigney 2009).

With implicit collective memory, I describe the recurrent use of mostly schematised memorata, which remains – for a majority of the group or society – not intentional, non-conscious, and not visible. As the examples above show, with the collective dynamics of non-awareness come questions of ethics. What responsibilities arise from not-remembering and not-knowing? (On ‘agnotology’, see Proctor and Schiebinger 2008). Such ethical questions can be addressed with Michael Rothberg's (2019) concept of the ‘implicated subject’, which helps describe actors’ differentiated ability and willingness to be aware of the past and its continuous presence. The accent of the present intervention, however, is on the generative power of implicit memory: while remaining unexposed, it is likely to produce more of the same in the future.

How has research in different disciplines addressed the phenomena that I suggest bringing and thinking together under the banner of ‘implicit collective memory’? In what follows, I will discuss narrative templates, media framing and priming, and premediation as concepts for the study of implicit memory – thus staging a dialogue between psychology, anthropology, sociology, communication studies, media culture studies, and mnemohistorical research.

A perspective from anthropology: National narratives

National narratives are a striking example of the hidden power of implicit collective memory.⁷ In How Nations Remember (2021), anthropologist James Wertsch studied the differences in national narrative templates between Russia and the United States. Collaborating with cognitive psychologists on an interdisciplinary survey study (Abel et al 2019), Wertsch and his colleagues could show how Russian narrative templates about the Second World War are fundamentally different from those not only of the US, but also most other nations. Abel et al (2019) asked more than 100 people from each of 11 countries to state what they thought were the ten most important events of the Second World War. The results are striking: Even in China and Japan, subjects produced items that are largely in consensus with the American perspective on the Second World War. Among the most important events range Pearl Harbor, D-Day, the Holocaust, and Hiroshima and Nagasaki. In Russia, however, the results are very different: With an extraordinary level of consensus (and in combination with the greatest general knowledge about the war, which was also tested), Russian subjects usually did not mention Pearl Harbor or the dropping of atomic bombs, but came up with the following core set of events: The Battle of Stalingrad, the Battle of Kursk, the Siege of Leningrad, the Battle of Moscow, the German invasion of the USSR, and the Battle of Berlin.

How can such differences be explained? And what holds this stable set of remembered events together? Analysing Putin's uses of historical memory, Wertsch (2021, 25) argues that these are basic elements of a story that ‘positions Russia as a victim of attacks by alien enemies’. Importantly, this is a narrative pattern that most people in Russia do not (always) seem to be aware of. But it guides not only the way they select remembered events, but also how they imagine the future. According to Wertsch, for Putin the narrative template of Russian victimhood is a ‘fast cognitive tool’: ‘It is a narrative tool that could almost be said to be doing some of his thinking and speaking for him’ (Wertsch 2021, 24). Wertsch's book came out just a couple of months before Russia's invasion of Ukraine. It shows not only the extent to which the war in Ukraine is a war of differing collective memories, but also how memory studies contributes to a better understanding of how implicit memory operates through narrative and can be used and abused as part of aggressive politics.

But how do such largely invisible narrative templates circulate in society? How are they passed on across decades and centuries? In the case of Russia, Wertsch (2021, 14) discerns a dynamics where the template of national victimhood permeates major societal spheres and communication media (from family conversations to textbooks to popular culture and the national press). But what about the striking transnational consensus on the ‘American narrative’? With regard to subjects in Germany, Italy, or China, it remains a conundrum why they should remember ‘Pearl Harbor’ and not, say, the ‘Battle of Kursk’. At least part of the answer must lie in globalising media culture, where American historical narratives have been widely disseminated via Hollywood films such as Pearl Harbor (2001, dir. Michael Bay). The crucial point here is that their enormous influence on viewers’ images of history remains, in the sense of a ‘collective source amnesia’, largely unknown to audiences and under-researched in media studies (on the logic of the ‘memory film’, see Erll 2010, 2012).

What remains equally unknown to most people is the selectivity, narrativity, and perspectivity of their own images of history. Wertsch (2021, 13) emphasises that ‘narrative tools often operate under the radar of conscious reflection, leaving us with the impression that we have a direct, unmediated picture of reality’. He reminds us of the words of narrative psychologist Jerome Bruner: ‘Common sense stoutly holds that the story form is a transparent window on reality, not a cookie cutter imposing a shape on it’ (Bruner 2003, 6–7). Most of the time, we do not even see which cutter we hold in our hands. But the fact that we successfully used it yesterday and the day before makes it more likely that we will use it tomorrow, too. This is the forward-facing power of implicit collective memory.

Perspectives from sociology and communication studies: Framing and priming

As we've seen, anthropology and cultural studies can contribute their insights about narrative patterns to the project of tracing the hidden power of implicit collective memory. Promising perspectives from sociology⁸ and communication studies include research on framing and priming. While communication studies work on priming is based on the psychological research outlined above, the concept of framing has its own tradition in sociology, going back all the way to Erving Goffman, Emile Durkheim, and Georg Simmel. Metaphors of the frame and framing are used in many different ways today: In communication studies (as ‘media frames’), in political philosophy (Judith Butler's ‘frames of war’, 2009), and last but not least in memory studies, where Halbwachs’ cadres sociaux (social frameworks) remains a key concept.

For Erving Goffman, the founding figure of sociological frame analysis, ‘framing’ means ‘the organization of experience’ (Goffman 1986 [1974], 11). ‘Observers actively project their frames of reference into the world immediately around them’ (Goffman 1986 [1974], 39). Such mental frames are acquired in the socialisation process. The production of realities by means of framing is an active process, but not necessarily a conscious one. On the contrary, our functioning in the everyday world is contingent on our automatic usage of frames.

This understanding resonates strongly with Maurice Halbwachs’ theory of collective memory. The concept of the frame was so important to Halbwachs that he devoted an entire monograph to it: Les cadres sociaux de la mémoire (1925, The social frameworks of memory). Halbwachs’ metaphor of the cadres sociaux merges into one term the social and mental properties of the ‘frame’: It is via the people who surround us (French: cadres sociaux) that we acquire our socially shaped mental frames, scaffolds, or schemata of remembering (again, French: cadres sociaux). But these twofold frameworks not only enable and shape acts of remembering. They also – and this is what Goffman will accentuate 50 years later – lead the understanding of new experience along certain paths. For Halbwachs, frames remain largely a hidden power. In La mémoire collective (1997 [1950], 70), he states that a ‘social current of thought’ is ‘ordinarily as invisible as the air that we breathe. In normal life we recognise its existence only when we resist it’ (my translation).

Being a sociologist, Halbwachs directs his attention to the social dimension of memory. But in a hypothetical anecdote about a ‘walk through London’, he imagines how media such as the words of his architect friend or Charles Dickens's novels shape his current experience as he is walking through the city (Halbwachs 1997 [1950], 52–53). Halbwachs thus exemplifies something that media memory studies (e.g., Edy 2006; Erll 2011, 129) have emphasised again and again: Social frames are mediated phenomena. They are medial frames, cadres médiaux. Social relations and meanings are constituted and transmitted through oral speech, letters, or books; and in the age of digital media they have acquired an intrinsically mediated form (Hoskins 2018; van Dijck 2007). Medial frames both shape collective remembering of the past and have a forward-facing power, as we see in Halbwachs’ anecdote, where media preform his perception of London. It can be supposed that in everyday life (i.e., without the sociologist's introspection) they will do so in a mainly non-conscious way.

In the late twentieth century, communication studies adopted the concept of framing in order to understand the logic of the news. For Robert M. Entman (1993, 52), ‘[f]raming essentially involves selection and salience. To frame is to select some aspects of a perceived reality and make them more salient in a communicating text in such a way as to promote a particular problem definition, causal interpretation, moral evaluation, and/or treatment recommendation for the item described’. In this usage, framing becomes a conscious strategy on the production side. But its power nonetheless resides in the non-conscious effects of news framings on the reception side.

Framing and priming have become key concepts of communication studies and media effects research (see Lecheler and De Vreese 2019; Roskos-Ewoldsen et al 2009; Tewksbury and Scheufele 2009). The two terms describe different phenomena involved in the dynamics of implicit collective memory. Both are metaphors. John Sonnett (2019, 227) explains that ‘framing’ is based on a spatial and visual metaphor (the framing or arranging of paintings), while ‘priming’ is based on a temporal and sequential metaphor (only after the presentation of a piece of information can certain effects come to pass). Therefore, research on framing is typically interested in the ‘how’ of communication, while research on priming asks about the ‘what’.

Vincent Price and David Tewksbury (1997) differentiate between ‘applicability’ and ‘accessibility’. ‘Applicability’ is concerned with the semantic uses and usability of a frame (how?), while ‘accessibility’ points to the temporally restricted activation potentials of a prime (what? when?) (see also Scheufele 1999). In communication studies, framing is thus about the ways in which mediations can shape and change perceiving and understanding, while priming is about the automatic activation of non-consciously existent memorata, which may include frames. Media can prime people on certain frames.

The challenge: Implicit collective memory across the longue durée

Psychology, sociology, and communication studies have shown that framing and priming are social and medial phenomena, and that they remain implicit for most of the actors involved. The possibility of linking these approaches with memory studies depends on questions of time: It is only when their power to frame and to prime unfolds long-term effects that mediations can shape collective memory. But the problem is that in most psychological experiments priming effects are counted merely in minutes and hours.⁹

Communication studies have tried to model the temporal stability of frames as ‘chronic accessibility’, which can be heightened by ‘frequent priming’ or ‘repetitive framing’ (Roskos-Ewoldsen et al 2009, 83). Christian Baden and Sophie Lecheler (2012, 359) have made an important foray into the theoretical modelling of the duration of framing effects. They emphasise, much in the sense of memory studies, that ‘the social relevance of framing effects hinges upon their ability to persist’. But empirical research on media framing so far has only provided evidence for effects that last 10 days to a maximum of 3 weeks (Schemer 2013, 161).

From a memory studies perspective, this sounds sobering. Vastly different temporal horizons are at stake in the study of collective memory. Memory studies is concerned with the question of how medial (pre)formations can exert effects over years, decades, centuries, even millennia – and that means, not only across the life spans of individual subjects, but also across multiple generational thresholds: the longue durée of collective memory.¹⁰

The narrative patterns of the Odyssey or of Exodus, the iconic formula of Pietà, as well as antisemitic stereotypes have quite obviously had ‘chronic accessibility’ over long periods of time. But what connections must be made to turn the psychological, sociological, and communications studies concepts discussed here into useful tools for the study of the long-term dynamics of collective memory?

Two aspects need to be taken into account. First, examples ranging from Homeric myths and their narrative templates to Christian iconography all the way to tenacious stereotypes and conspiracy theories show that long-term memorata are always built-up plurimedially (see Erll 2010). They are transmedial phenomena, remediated again and again across the spectrum of available media. This is, second, a social process through and through: Interaction, collaboration, dialogue, negotiation, agonism – the entire spectrum of the dynamics of social memory-making needs to be taken into account here. Psychological research on implicit memory, on the contrary, tends to focus solely on individual memory performance and not on the question of how memorata can be shared and thus become part of collective memory. More generally, in mainstream psychology, social interaction tends to be seen as ‘memory contamination’, rather than (as in the field of memory studies) as a means of ‘memory production’.¹¹

Implicit collective memory is produced and passed on in complex social and plurimedial constellations – often across the longue durée. This insight opens up an entirely new range of questions: How do such constellations come into existence? How is it that certain framings become strong primes, preforming action again and again in social groups? What media technologies and genres tend to have enough authority and power to disseminate frames and make them appear ‘applicable’? What forms of institutionalisation, canonisation, dissemination, educational politics, or marketing make the ‘chronic accessibility’ of certain primes possible? Such questions necessitate the combination of psychological memory studies with the conceptual toolbox, archives, and methods of media culture studies and mnemohistory. An example of such a combination will be presented in the following section.

Perspectives from mnemohistory: Remediating and premediating ‘Mutiny’

European post-colonial memory cultures remain a strong residue of implicit collective memory. European societies are replete with mostly non-conscious and unacknowledged afterlives of its Empires – all across Portugal, Spain, Italy, France, Belgium, the Netherlands, the UK, Denmark, and Germany. This is why research fields such as post-colonial studies and new imperial history have a lot to offer the study of implicit collective memory (Bijl 2016; Craps 2013; Rothermund 2015; Schwarz 2011). Research on ‘post-colonial melancholia’ (Gilroy 2004), ‘colonial durabilities’ (Stoler 2016), or ‘embers of empire’ (Ward and Rasch 2019) address the ways in which implicit legacies of colonialism (affective, archival, discursive) continue to shape mentalities and guide political action long after decolonisation.

Part of these largely invisible and unspoken legacies are ways of framing certain events of colonial history. In Germany, for example, for over a century the Herero and Nama Genocide of 1904–1908 (in what is now Namibia) had been framed as merely a ‘small war’ in a ‘short German colonial history’ (de Wolff 2021). The case I discuss in this section concerns British imperial and post-imperial memory, where an ‘insurgent frame’ has lingered for at least two centuries. This frame has many mnemohistorical sources, but one particular important genealogy points back to the Indian rebellion of 1857/8.

The press coverage of the 1857/8 rebellion in northern India is a case in point for the potential tenacity of initial press framings, and for the quite seamless migration of frames from colonial to post-colonial times. The rebellion brought together Indian soldiers, farmers and princes of different ethnicities and religions against the British, and it was so successful that it almost cost them their Raj. The British called the rebellion a ‘Mutiny’ (it had indeed sprung from colonial soldiers’ uprising) and thus inserted a powerful framing into British imperial mnemohistory that still reverberates today. The term ‘mutiny’ implies an unlawful uprising as well as an event that is restricted to the military. Both interpretations of the events of 1857/8 are debatable.

Anglo-Indian and British press texts that immediately covered the ‘mutiny’ for the imperial metropolis worked with selections, highlighting, word choices, and narrative structures that could still be felt decades, even a century later in the ways in which the years 1857/8 were presented in British historiography and across broader media culture. The first reports printed in The London Times about the rebellion in Northern India were real or feigned eyewitnesses accounts. Their framing is unequivocally one-sided: The rebellion was cast as a perfidious mutiny of ungrateful, religiously fanatical, and cruel subjects against the just colonial rule of the unsuspecting and benevolent British. Atrocity stories became the most powerful genre in the ‘Mutiny’ coverage. The rape and killing of British women and children were specifically highlighted and embroidered with gory details. In 1857, Karl Marx, who offered an alternative framing of the events as the ‘first Indian war of independence’, had already exposed one of the ‘eye-witnesses’ featured in The Times as a liar (Sharpe 1993, 66). However, the early British press framings and their narrative plots migrated untarnished, sometimes word-for-word, first into imperial historiography, then into English novels and theatre of the nineteenth century, early cinema, as well as – after Indian independence 1947 – into post-imperial historiography, fiction, and television.

Those ‘Mutiny’-frames erupted again with a vengeance in 2005 in a debate around a Bollywood movie of the rebellion (Mangal Pandey: The Rising, 2005, dir. Ketan Mehta). In unison with a diverse range of newspapers, and backed by the comments of eminent British historians, the Daily Mail (19 August 2005) criticised the movie as ‘fanatically anti-British’. ‘Fanatical’ – for one and a half century, this had been a standard attribute to frame resistance against British colonial rule. Under the workings of implicit collective memory, a filmic twentieth-century rendition of the ‘Mutiny’ seems to have turned into yet another mutiny.¹²

What I can only sketch here (the whole story is set out in Erll 2007) is the dynamics of remediation in memory culture – the transcription of memorata into ever-changing new media, a process, in which traces (here: framings) of older mediations travel along, often unheeded, across potentially very long stretches of time. Remediations are a vital agent in the dynamic plurimedial constellations which emerge around remembered events and keep them on the agenda of memory culture.

With time, remediations lead to mnemonic premediation: Much-repeated frames, narrative schemata, or visual patterns become ‘household items’ of a media culture. They can become detached (unlocked, unbound) from the stories they were originally used to convey, and reattached to new experience, which they then medially preform (i.e., shape even before the events take place).¹³ Think of Bruner's cookie cutters.

In the case of the ‘Mutiny’, such mnemonically preformed events can include debates about the relative merits of a Bollywood drama. But more deadly, among the events that were quite possibly premediated by ‘Mutiny’-memory is the Amritsar massacre of 1919, a decisive turning point in the Indian independence movement of the early twentieth century. Facing a crowd of largely peaceful protesters, Colonel Dyer of the British Indian Army had his soldiers open fire and kill hundreds of Indians, who could not escape from the enclosed compound of the Jallianwala Bagh, where they had gathered. In a perceptive book about the ‘shadows of the Mutiny’ in the British Empire, The Other Side of the Medal (1925, 53–54), historian Edward John Thompson surmised that the massacre was a knee-jerk reaction, the result of ‘inherited thought concerning the Mutiny’, and evidence of ‘the workings of imperfectly informed minds obsessed with’ the stories about Indian atrocities of 70 years earlier.¹⁴ Seen from hindsight, the framings of mid-nineteenth century media culture were thus still palpable in the early twentieth century, and they are still active today, 150 years later. Surely, they were themselves the result of centuries-old dynamics of premediation. Already in the eighteenth century, Indian peasant uprisings had been framed by British colonisers as outpourings of religious fanaticism (see Guha 1983).

‘Implicitness’ on a collective level does not mean ignorance of all. Perhaps some imperial historians used the gory and one-sided early press framings of the ‘Mutiny’ quite deliberately. Perhaps some journalists today critically revisit the old press archives. And perhaps some are astutely aware of the plurimedial mnemohistory of the rebellion and understand the logic of the ‘Mutiny cookie cutter’. But what is crucial is that the majority of people in British post-imperial memory culture are not aware of the chains (or better, cascades) of imperial remediations. They suffer from what might be termed collective source amnesia.

Thinking and talking about colonialism, actors in post-colonial memory cultures (not just in Britain, but also, say, in Germany or Russia) are often ‘thought and spoken’ by implicit mnemohistories. A word choice like ‘fanatic’ for the description of colonial subjects taking collective action may thus feel natural to some people – and in the logic of implicit memory therefore also ‘true’. These are ‘mere exposure effects’ and ‘validity effects’ on a collective level – effects which are prepared in the longue durée, by the self-reinforcing dynamics of much (re)mediated frames.

For mnemohistorical research, framing and priming are thus very relevant concepts. But they have to be ‘translated’ according to different dynamics on different scales of the complex ecologies of collective memory: for example, into ‘remediation’ (as a form of plurimedial repetitive framing in the longue durée) and into ‘mnemonic premediation’ (as a form of media priming, which is based on longue durée-frames). Mnemonic premediation has effects, on the one hand, on individual minds (fast accessibility and seeming ‘fitness’ of a frame for many people in memory culture). On the other hand, mnemonic premediation does become visible (and thus researchable) in mediations, when certain discernible framings, schemata, or narrative patterns migrate to new topics.

Perspectives for memory studies: Beyond the street lamp

What are the consequences of the hidden power of implicit collective memory for interdisciplinary memory research? Implicit collective memory remains largely an invisible agent. In everyday life, most people will neither notice its power to shape perception and action nor realise that there are sometimes enormous differences between implicit repertoires of different social groups. Think of Russian and ‘Western’ World War II narratives (Abel et al 2019). Research into implicit collective memory means making visible what remains invisible to most people – and thus moving out of the (academically quite safe) light cone of a street lamp that illuminates only explicit (and mostly commemorative) memory.

Implicit collective memory is, contrary to commemoration, not primarily backward-looking, but fundamentally a preforming, a forward-facing dynamics. In cognitive psychology, priming is defined as the influence of past experience on subsequent action. A mnemohistorical perspective shows that new experience and action is often mnemonically premediated, i.e., implicitly preformed by the mediations of (sometimes longue durée) collective memories, in which subjects participate. What is ultimately at stake with the term of implicit collective memory is the social, medial, and for most people non-conscious aspects of a cultural remembering-imagining system (Conway et al 2016). Implicit collective memory is a form of ‘collective future thinking without thinking’.¹⁵

What significance does existent research on memory cultures (sensu A Assmann 2011 and J Assmann 2011) have for the study of implicit collective memory? It is indispensable, because the non-conscious is mostly a long-term effect of active memory culture and its key processes: remediation, canonisation, institutionalisation, commemoration, and the wide dissemination of certain memorata. It is not surprising that some of the most influential long-term frames spring from canonised texts such as the Bible, from frequently remediated national myths, much-used textbooks, or from globally disseminated Hollywood movies. The resources of implicit collective memory emerge from explicit memory culture.

The term ‘implicit collective memory’ is proposed here as an umbrella term for a wide array of phenomena, which all share a certain onto-epistemological mode (they derive from collective memory and most people remain unaware of them) and a function (they preform thought and action). The term aims to bring together research from different quarters and traditions in order to study a transdisciplinary phenomenon. Among these traditions are not only those discussed in this article: psychological, sociological, and communication studies approaches to priming and framing; anthropological approaches to national narratives; mnemohistorical research on remediation and premediation. Implicit collective memory is moreover discernible as one of the key concerns of social movement studies (Lorenzo Zamponi's ‘repertoires’, 2018), research on conspiracy myths (van Prooijen and Douglas 2017), recent discussions in the philosophy of history about the ‘presence’ of the past (Bevernage and Lorenz, 2013), as well as post-colonial and decolonial debates about the continuities of imperial practices and forms of thinking (Stoler 2016). Last not least, the implicit poses a key challenge to theories of new media. It has been addressed as ‘the digital unconscious’ (Monk 1998) and is part of what Andrew Hoskins (Hoskins and Halstead 2021) calls a ‘new grey in digital memory’.

The preoccupation with implicit forms of collective memory can boast a long tradition – perhaps an even longer one than the study of commemorative memory. A key figure is surely by Warburg (2000 [1924]), whose understanding of the afterlives of antiquity gestures far beyond a deliberate recourse to a ‘classical tradition’. Warburg describes the workings of visual ‘pathos formula’ as a non-conscious activation of older forms and affects in new artworks. As we have seen, Halbwachs was also interested in invisible ‘social currents’. And Frederic Bartlettt's (1932) foundational research in experimental psychology is fundamental, too, for an understanding of non-conscious narrative schemata that seem specific to particular (memory) cultures.¹⁶

For many humanities scholars, the ‘collective non-conscious’ may be suggestive of the psychoanalytic ‘unconscious’ and the ways it has been used for the description of social processes. Well-known examples include CG Jung's (2014) quite problematic ‘collective unconscious’, a concept developed in the early 1900s, as well as Sigmund Freud's mass psychology (2004 [1921], 2010 [1939]). In memory studies, psychoanalytical thought has long played a key role in attempts to describe the dynamics of difficult non-conscious collective memory – all the way from Theodor W. Adorno's deliberations on the afterlife of fascism (Adorno 1977 [1959]) to Marianne Hirsch's (2012) ‘postmemory’ as a form of non-intentional transgenerational transmission of traumatic memory (for psychoanalytic approaches to historical consciousness, see Straub and Rüsen (2011 [1998]).

Why ‘implicit collective memory’, and not ‘the collective unconscious’? My use of a technical term coming from cognitive psychology is meant to act as a reminder that there is no easy equation between the Freudian unconscious, on the one hand, and the social effects of past experience and media reception that most people remain unaware of, on the other. It moreover directs attention to a possible dialogue between memory studies and recent cognitive science approaches to the ‘new unconscious’.¹⁷ Moreover, applied in a conversation with post-human studies, the term opens up the possibility of studying the bio-technological co-production of forms of non-conscious memory.¹⁸

But the essential rationale for the term ‘implicit collective memory’ is pragmatic: As Daniel Schacter reminds us, the ‘nonconscious world of implicit memory revealed by cognitive neuroscience differs markedly from the Freudian unconscious’, because it is ‘far more mundane’ than the Freudian drama about the forces of repression. Implicit memories ‘arise as a natural consequence of such everyday activities as perceiving, understanding, and acting’ (Schacter 1996, 190–191). What is at stake is an understanding of the quantitatively most frequent memory processes. Implicit memory – in its cognitive as well as in its social or medial instantiations – is not a pathology, but a basic and ubiquitous dynamics that first of all enables memory ecologies to function. What is therefore needed is a better understanding of the hidden power of everyday, automatic forms of collective remembering across its various dimensions. What is not needed is a playoff between, say, psychoanalysis and the cognitive sciences. Instead, only a joining of forces from different quarters will enable us to make sense of the invisible phenomena of collective remembering – of those ‘unusual suspects’, which are not found under the Freudian, or any other, single street lamp.

Upon closer inspection, implicit collective memory has in fact already emerged as a key concern of present-day interdisciplinary memory studies – but in a characteristically unexamined way. Jeffrey Olick's (2016, 60) studies on the path-dependence of social memory points in this direction, as does Robyn Fivush and Azriel Grysman's (2022) distinction between explicit and implicit gendered narratives, or Barbie Zelizer's (2022) discussion of framing as part of ‘journalism's backstage’. Implicit collective memory is, in the words of Eviatar Zerubavel (2008, 2015), the ‘elephant in the room’ of memory studies, so far remaining ‘hidden in plain sight’.

The greatest challenge of research on implicit collective memory is of a methodological kind: How can we make hidden phenomena graspable, and thus researchable? Perhaps, to begin with, by creating connections between the rich methodological repertoires that already exist in memory studies across its diverse disciplines and that range from experimental and quantitative methods all the way to archival, discourse-analytic and narratological approaches and finally to the possibilities that digital humanities now open up. The conundrums of implicit memory phenomena provide one more reason to turn the multidisciplinary field of memory studies into a site of intensified interdisciplinary collaboration.