In recent years, the idea of collective memory has been discussed to an increasing degree in relation to the pre-modern cultures of the North, illustrated, for instance, by the recent edited by Jürg Glauser, Pernille Hermann, and Stephen Mitchell.For other, more narrowly focused studies one could point out the long-time engagement of members around the scholarly network Memory and the Pre-Modern North and their individual publications. I am grateful to the network for sponsoring the session Memory in Runes and Community, held Tuesday 3 July 2018 at the 25th International Medieval Congress in Leeds, at which occasion an earlier version of this article was presented. In particular, I would like to express my thanks to the session organizers and editors of the present issue, Simon Nygaard and Yoav Tirosh for their continuous patience, support, and advice. I also have to thank Katie Beard and Zachary Melton for proof-reading and improving my English. I also owe my gratitude to the two anonymous reviewers and the editors of the journal for helpful comments and aid during the publication process. This orientation towards past societies and historical cultures has indeed been a crucial moment in the formation of modern collective memory studies. One may mention the scholarly work of Jan Assmann in particular, who outlined some of the significant foundations of collective (or cultural) memory as results from his research on the early-historical cultures of the Mediterranean and Mesopotamia. These modern collective memory studies grew especially out of the belated reception of the writings by Maurice Halbwachs (1877–1945), who coined the term collective memory (la mémoire collective) alongside developing a theory about the social existence and determination of memory. Indeed, this new field of study developed from investigations into different historical modes of communication and their manifestations in various media.

It should be mentioned right from the start that ever since the formation of collective memory studies, different conceptions have been used in the field, albeit quite often interchangeably. A number of terms like public memory, communicative memory, social memory, cultural memory, or collective memory are often used synonymously and on a broader basis. Despite this, attempts have been made by both Jan Assmann 2008 and Aleida Assmann 2006 to define collective memory more specifically as a paramount quality of both short-lived individual and socially communicated memory as well as long-living political and cultural memory. For the purposes of this article, it is Jan and Aleida Assmanns’ notion of cultural memory that is implied when speaking of collective memory. I use the term collective memory more inclusively, encompassing both communal and public functions of the discussed rune-stones, thus refiguring and modifying their approach slightly to suit the sources.The line between different kinds of collective memory is thin and fragile, threatened by oblivion—as a random and wilful act of forgetting. However, collective memory is also cultural, that is, defined by conventions and institutions central to a culture that guard, maintain, and control collective memory J. Assmann 2008, involving the concept of actualization or reconstruction stressed by Jan Assmann 1995, 130; cf. A. Assmann 1999, 134 as the process of renewed activation, recontextualization, and re-interpretation of memories by communities. It is conceptualizations like these and the relating nuanced vocabulary that makes the Assmanns’ theories still one of the most helpful and underestimated frameworks for memory studies. Using collective memory and its cultural dimensions as analytical concepts may provide new insights into an otherwise quite inaccessible source material. Such is attempted in the present article with its socio-cultural focus on inscriptions with runes from the earlier periods. As I demonstrate and argue in this article, this theoretical framework provides a helpful addition to runological studies, particularly in developing more holistic oriented approaches to the runic source material.

The early rune-stones that are treated in this article have regularly been associated with commemoration and other cultural praxes related with burial habits. The reasoning behind these hypotheses or interpretations, I argue, have been insufficient. This study proposes a new approach, not only by directing effort at considering several of the serious problems of both modern scholarship and the sources themselves, but by evaluating the benefits and constraints of collective memory theory and how to use it. The use of collective memory theory supports the study of these early runic inscriptions in particular, given the fundamental assumption that rune-stones, also in the earliest periods, were a socio-cultural phenomenon involved in the memory-work of communities. Thus, this article aims to investigate the vacuum that has arisen in the rather one-sided debate around the early rune-stones, which touches on (collective) memory solely within the constraints of a socially and communicatively limited culture of individual commemoration and remembrance, or public memorialization paralleled by other cultures. Instead, this contribution studies issues and features characteristic for the older rune-stone culture specifically, thus contributing to the study of collective memory more generally.

This article focuses only on the early rune-stones, c.150–750 CE, as noted above. In Scandinavian archaeology this ranges from the Late Roman Iron Age, to the Migration Period, and the Merovingian/Vendel Period. In Nordic linguistics, the period covers the pre- or Ancient Nordic stage (also Northwest-Germanic or Proto-Nordic) to Old Nordic (also Old Norse), which relates to runological periodization.For helpful introductions to runological periodization and historical linguistics, see Knirk 2002. The present article follows Knirk’s terminology of historical language stages. To combine these different periodizations, reflecting crucial historical, socio-cultural developments, Early Scandinavian may be used as a term to refer to the formation of a distinctive Iron Age culture across Scandinavia. The earliest inscriptions in runes appear with the shift from the Early to the Late Roman Iron Age and mark the beginning of the earliest transmitted Ancient NordicProbably already during the first half of the first century CE or the Early Roman Iron Age, people in the (subarctic) North began for the first time to incise runes, which they had developed likely in close contact with the Roman world Düwel 2008; cf. Williams 2004.—and thus lay the foundation for important developments in the cultural history of collective memory in Early Scandinavia, as argued in this study. These early inscriptions were carved in runes, which in their oldest forms are traditionally called the older rune-row (or elder fuþark),The Early Scandinavian runic inscriptions are written in runes that are called the older (or elder) fuþark, after the first six signs of several preserved complete rune-rows. This writing system of 24 characters was gradually replaced some 500–750 years later by the younger fuþąrk. For helpful introductions to such runological principles see the English overviews by Knirk 2002 or Barnes 2012. What is interesting about these runographic developments is that they also represent drastic linguistic changes from (Late) Ancient/Pre-Nordic to (Early) Old Nordic. A series of systematic syllable shortenings, or syncopes, and sound changes known as Umlaut and breaking, are reflected in the changing appearance and grapho-phonological relations among the runes Nielsen; Torp in Kristoffersen and Torp; Schulte in Schulte and Williams, 81–90. Within this period of change, large-scale socio-cultural transitions occurred during the fifth to eighth century that led up to the Viking Age. For a brief evaluation of different explanatory models of these complex matters, see the recent contribution by Michael Schulte (in Schulte and Williams 58–61). There are good reasons to assume that both the language and the runic writing system changed alongside major social developments. However, these events are far from well understood, and many questions remain open beyond the scope of this work. marking the first phase of runic writing until a transition to the younger rune-row (or younger fuþąrk), which occurred at the onset of the Viking Age of Scandinavia.

While the earliest known runic inscriptions are found on a broad variety of loose objects—especially in the cultural contact area between the Romanized and Germanic culturesIn the present context, contact area refers to the hinterlands of the Roman limes, its northern parts reaching further into southern Scandinavia. From this area stem potentially proto-runic inscriptions found between the Rhein and Elbe. In the present context this implies that while the beginning of runic writing, as it is known to the present day, may show strong interlinking, Common Germanic features, a high degree of interculturality can be expected for the inscriptions.—runes were soon incised on large boulders and other surfaces of stone too. Although the dating of such stones is problematic, this practice may have arisen early on during this first phase of runic writing Imer 2011, 170–71; 2015. As this practice to carve rune-stones appears to employ several inner regularities or patterns that remain difficult to explain, it might be useful to refer to these early rune-stones as an epigraphic habit, thus following Bernard Mees 2017; cf. McMullan. Such a perspective stresses the cultural rootedness of the phenomenon. Of course, one may argue that on the whole the inscriptions of the early rune-stones form a rather heterogenous material, as the examples discussed below may demonstrate. Still, significant changes of the early rune-stones only seem to take place during their late or transitional phase, characterized by some longer inscriptions and diversifications of the texts—as far as the material allows one to judge. Nonetheless, the early rune-stones appear to draw on some of the same epigraphic strategies until these were seriously affected by a series of significant social changes in or by the Early Viking Age leading to new epigraphic habits.

Rune-stones were, of course, continuously erected through the Viking Age, throughout which even more significant changes occurred. Impressively, the Late-Viking-Age stones, from the late tenth to twelfth century, amount to a total of nearly 3,000 stones Barnes 2012, 66,The total count is made up of the finds from Scandinavia (modern Sweden, Denmark, and Norway), though expanding throughout the Norse diaspora (primarily the Isle of Man, smaller numbers from Orkney, Shetland, Ireland, England, Latvia, Ukraine), underlining the universality of the phenomenon. and thus, they have, without a doubt, attracted the most attention. These late rune-stones form a very different source material from the current focus, as they are prominently Christian memorials. There is, however, also another diverse category of Early-Viking-Age stones, which has recently been studied and for the first time inventoried by Hanna Åkerström. With these stones a new distinctive rune-tradition can be defined for the period c.700–950 CE with new centres in Denmark and Sweden.

The early rune-stones, however, are exclusively found from what is today the west coast of Norway to south-eastern parts of Sweden. Compared to the abundance of Viking-Age stones, early rune-stones are surprisingly small in number, counting some 65–70 stone-inscriptions.Numbers depend heavily on the actual state of carvings as actual runic inscriptions or merely scratches, which affects a handful of inscriptions, e.g. Opparaun dRäF 55Anm2; NIæR Ikke medtagne Indskrifter 11, or the dating and thus placement in runographic chronology of several inscriptions, most eminently those of the so-called transitional inscriptions Barnes 1998. This transition however is a problematic issue in runology, both from graphematical, linguistic, and also analytical perspectives. Different opinions and frameworks have therefore been aired by runologists like Barnes 2010, Schulte 2010b, or Stoklund. Nonetheless, there is also a cultural-historical perspective and actual holes in the transmission, marking potentially drastic changes in the transmission and habits of runic writing. Despite their relative rarity, the spread of early rune-stones in space and time is particularly significant. Following Lisbeth Imer’s recent re-evaluations of early datings 2011; 2015, the early rune-stones cover a period of about 500 years. Thus it should be noted that the early rune-stones convey a long-lasting and widely distributed phenomenon under the given circumstances. Therefore, it would be rather surprising if no considerable diversification and developments would have occurred within the corpus. Such aspects have, however, been underexplored so far when it comes to the early rune-stones. This article hopes to add on new perspectives to these stones by proposing an alternative, more holistic approach to rune-stones (and other inscriptions) that might indeed be needed. The established lines between the corpora, especially those from within Scandinavia, may have their legitimation in terms of grapho-phonological developments and in the context of Nordic language history. This, however, ignores the possible role of different developments on a larger scale within the history of the Nordic peoples during the Iron Ages, raising also the question of the integrity of runic writing culture beyond purely epigraphic and linguistic issues. Therefore, one could argue that sociological dimensions are indeed an area of investigation where much can still be gained in the context of runology. Rune-stones pose a particularly meaningful object for investigation due to their long history, showing different facets and phases. Taking a socio-cultural approach is a matter of high complexity, and many unknown factors and variables are missing or only partially understood for the present context. Nonetheless, questions of a socio-cultural nature may help us look further beyond classical runological issues like those mentioned above, and contextualize epigraphic and linguistic matters by means of socio-cultural religious, climatic-environmental changes.

As discussed in greater detail below, this article suggests that rune-stones as a phenomenon were highly social, both in their own (runological) development alongside societal changes, but also in terms of their role or place within their surrounding societies. Given the complexity of this issue, my current study can hardly cover all individual problems equally thoroughly. Nonetheless, it attempts to open a discussion on how judgements about functions, purposes, or uses of runes as a writing system are made on a general basis, although the particular scope of the present study is limited to inscriptions on rune-stones alone. Since many collective ideas about early rune-inscriptions in stone (and elsewhere) have dominated their study thus far, a brief discussion of runological methodology is helpful to identify a number of key issues within research history. While a more thorough understanding of the history and current state of scholarship about rune-stones, especially the early rune-stone culture, is essential (see below), this discussion is likely to create more questions than it is able to answer.

Before moving on to the question of how collective memory studies can aid the discussions of early runes and runic inscriptions, this section first seeks to gain a better understanding of the nature and main characteristics of the early rune-stones as potential bearers of (collective) memory. Here, however, one also must address some key issues when working with these types of sources. Runic inscriptions in general, especially the early material, form a problematic source group. On the one hand, the representability of these stones as sources is diminished because they only exist in rather limited number; on the other, in many cases there are serious issues regarding the preservation and documentation of the material I am discussing.

An illustrative case for the problems of preservation and documentation is the little-known stone N KJ82 Saude NIæR.10 from Telemark,Runic inscriptions are referred to according to common runological praxis by signum and finding place. The present article largely adheres to the standards set by ; thus I give their signum first together with the designation by place-name. In certain cases, different signa or editions are added after this in parentheses, usually from the earliest standard edition or original publication, which were also consulted for their find reports. which is now lost. This stone is only mentioned once, by Danish Renaissance scholar Olaus Wormianus, or Ole Worm (1588–1654), who refers to it in one of the earliest runological works, his 1636, 68; 1651, 66. The publication renders the inscription differently from other runic inscriptions presented by Worm, which in general are reproduced by several sets of runic types. Saude, however, is printed in a fashion (Figure 1) that can only be a direct reproduction of a foreign rendering. The original source to the Saude-stone appears to be from a letter that Worm received.

While for some runic inscriptions, such as the Saude-stone, the only source that remains are vague memories, other inscriptions like the stone N KJ64 Barmen Olsen from Sogn of Fjordane, have survived and are still found in situ. This situation is somewhat unique among the early stone inscriptions. The inscription is given in the following based on the initial publication by Olsen and the edition in dRäF.64:

Here, on the Barmen-stone, the runes are legible to a certain degree. However, the runes are also quite worn, as noted by Oliver Grimm (for the ). Nonetheless, a reading provided by Magnus Olsen does exist and is still followed by scholars today. In Olsen’s further interpretation, the first two runes most likely render a sequence of runes not uncommon in older inscriptions, designating the personal pronoun 1st-person singular ek [I]. What follows in runes 3 to 9 is less certain, yet by referring to analogous inscriptions, scholars have commonly assumed a personal name or epithet in this runic sequence. An adjective þjarfr [unleavened; common; flat] can indeed be found in Old Icelandic, and a Viking-Age stone from U 90 Säby in Järfälla, Uppland, Sweden, has the cognate þerfʀ* (þerf). These are believed to be developments from the Germanic root *þerƀ- to which the Proto-Nordic þirƀijaʀ would form a nomen agentis derivation the one who makes strong Antonsen 1975, 48. The last two runes, however, are believed to be a unique case of intertextual reference through abbreviation. Thus, they would give only the beginning of a longer phrase such as rūnō/-ōʀ rune/-s; runic text + faihido/faihiu paint, or semantically similar verbs, which can be compared to other inscriptions, for instance the sequences runofaihido on N KJ63 Einang NIæR.5, runofahi on Vg 63 Noleby dRäF.67, or simply fahido on Bo KJ73 Rö von Friesen.

Based on these two exemplary cases it may seem that rune-stones can provide insights into a somewhat well-developed writing culture albeit often containing little more than single (rather laconic) expressions. However, precisely in the reoccurrence of typological-pragmatic patterns, formulae, and intertextual references lies the basis to assume an independent runic writing culture. Nonetheless, many conclusions that go beyond this assumption are often only achieved by conjectural interpretations and circular arguments, as will become clear later in this article. Thus, when speculating on the exact nature or functions and roles of this early written culture with runes, one is easily leaving the grounds of certainty that the inscriptions provide on their own. This has primarily to do with the challenges and limitations of the early inscriptions. When engaging with these inscriptions of the earliest phase, the modern observer is inevitably confronted with several obstacles concerning the very reading of rune-inscriptions—that is, the epigraphical investigation of a carving before its (linguistic-philological) interpretation.The process of reading runic inscriptions, one may note, may refer to a mere interpretation of texts as such, and could therefore be separated from the preceding process in which the carvings of an inscription are identified purely graphemically as signs (before texts as such). Nonetheless, one can object that this is complicated by the intrinsic issue of determining which signs are thought to (re)present proper runes, rune-like signs, other symbols (of possible conveyed or conventionalized meaning), and further iconographical elements, for instance. Importantly, reading is a process of describing what one sees, determining whether it is deliberate carvings or scratches that one sees, and decoding what one sees, that is rendering a carving in normalized transcription and/or romanized transliteration. See Barnes 2012, 178–79 for helpful thoughts. Also at issue is their often sparse documentation and poor preservation, which includes their contexts and thus makes interpretation a difficult venture. Yet it is fair to say that the issue of linguistic interpretation—besides all its inherent constraints—has hitherto preoccupied most of the scholarship too one-sidedly, as criticized in recent articles, for instance, by Bernard Mees 2015; 2017; 2019.

While other scholars such as Michael P. Barnes criticize the same issue, only to provoke a more runographic-oriented runology 2013, 10; Mees points out that since runology has had a predominant interest in etymological interpretations 2015, 516, it is rare to encounter runological interpretations that are influenced by modern socio-cultural theory 2019, 1. Similarly, Lisbeth Imer has argued in favour of various contextual approaches, thus criticizing that rune-inscriptions have often been studied singularly, which makes her suggest that en sådan fremgangsmåde er upraktisk, når man vil undersøge et samfunds skriftkultur, fordi man med en enkelt indskrift i fokus ikke tager højde for paralleller eller den samfundsmæssige baggrund such an approach is unpractical if one wants to study a society’s writing culture, because with a single inscription in focus one does not pay enough attention to parallels or the societal background 2015, 9.Unless otherwise indicated, all translations are my own. While Imer’s contextual approach to runology is primarily archaeologically oriented, she also underlines the importance of a general focus towards rune-inscriptions within cultural history and the history of writing. With collective memory studies as a framework, this article aims to touch on similar aspects, investigating the social dimensions of runic writing, yet from a different angle.In connection with the socio-historical dimensions of rune-inscriptions, see the helpful discussion of some basic criteria by Williams (2013), comparing it to other studies, e.g. Jesch (1994), Sawyer. Before getting there, however, the common grounds of runological theory and method need to be considered.

The field of runological research has long suffered from both methodological and theoretical issues,This is for instance expressed in the 2013/2014 project by the Center for Advanced Study at the Norwegian Academy of Science and Letters (CAS Oslo) under the lead of James Knirk, and with the aim to produce a . This has yet to be published but see some helpful thoughts ahead of its publication that were formulated by Michael P. Barnes (2013). some of which I discuss here. Certain problems are inevitably linked to the rather long and complicated history of runology. The study of runic inscriptions traces back to antiquarian scholars of the seventeenth century like Worm and Johannes Bureus (1568–1652), though a narrower investigation of runology’s early history must be left out here.Helpful overviews can be found in recent introductions to runology, see e.g. Barnes 2012 or Düwel 2008. Accordingly, modern runology as it developed from their work is a relatively new discipline. In particular, the branch of runology studying inscriptions in the older runes does not go back much further than a hundred years, as runes of the older rune-row were not adequately decoded before the late nineteenth century. In scholarship since then, the first modern editions were compiled starting around the turn of the nineteenth to the twentieth century, while new editions, still in use, were issued half a century ago or more. Here, one needs to mention especially dRäF by Wolfgang Krause and Herbert Jankuhn from 1966 despite the considerable ideological bias it carries under the surface Barnes 2012, 201. Under certain circumstances it may still be helpful to consult some of the early editions, for instance to arrive at the first and at times best description of (epi)graphical and contextual features of an inscription. Nonetheless, it can only be hoped that currently ongoing projects publishing online databases will solve most editorial issues,It is only now with the ongoing project RuneS and their database, as well as the project and their database that new digital standards are underway. See also the and the . while also producing new tools for research.

In the absence of better methods for runologists, critics such as Kurt Braunmüller (1991; 1998) have not even found satisfactory heuristics for the study of runes. One of the greatest hazards here is inflicting personal opinions or common sense on scientific measures, as R. I. Page (10, 12) and Michael P. Barnes (1994) famously pointed out. Additionally, it is important to acknowledge that scholarly opinions of the early twentieth century were basically nothing but a highly ideological loaded conflict of artificial opinions on the nature of runes, as Bernard Mees (2015; 2017; 2019) has stressed. Most prominently, for instance, this can be seen in the works of Wolfgang Krause. Based on the belief that runes themselves possess magical power, early rune-stones were made artificially extraordinary as apotropaic grave-protection besides commemorating death Krause 1937; 1970, 46–63. Some scholars, for instance Klaus Düwel (1978; 2008, 37–42), continued on several of Krause’s ideas. Other scholars like Elmer H. Antonsen (1975; 2002), Terje Spurkland 2001, and already Anders Bæksted, however, developed their own sceptical opinions that runes were a writing system like any other, used for all possible purposes.

Turning back to the two examples Saude and Barmen: the interpretations mentioned for both inscriptions can prove reasonable on linguistical grounds, regarding the grammatical form or morphological and syntactic structure, as far as the runographical readings allow. Regarding the second stage of linguistical interpretation, the final decisions on semantics and pragmatics has to consider contextual information. For Saude, such data simply does not exist. Still, it is commonly counted to the type that Krause labelled einseitige Gedenkinschrift one-sided commemorative inscription dRäF.78–94, a type largely made up by nomina propria in the genitive. For the Barmen-stone, however, there is a context, although no certain burial. Instead, the stone stands close to the shore possibly surrounded by traces of a circular stone-setting around it and with an undecorated bauta-type standing-stone in close vicinity. The Barmen-inscription is commonly counted to the inscriptions of the rune-carver type (Der Runenmeister the rune-master in Krause’s terminology, comprising dRäF.63–70). While this could be a neutral terminology, for Krause—see similar Looijenga (340)—the Barmen-stone was supposed to perform a magical act of protection not further specified by Krause but simply listed together with other rune-master inscriptions that he judges to be apotropaically guarding burials.

However, Krause himself states that the lines between his categories are rather blurred, and he rightfully remarks that not all stones are found in or at graves. Nonetheless, the openness of his categories leads him to place almost all early rune-stones in burial contexts by default, either directly in or at graves or cenotaphs, commemorating the dead, protecting the dead, if not referring directly to cult or magic. This tendency has been rightfully criticized by later scholars, one of the earliest, Elmer Antonsen (1975, vii)—yet, in his case, only to develop views to the other extreme, as later pointed out by Barnes (1994). One can see that interpretations like in these two cases often have circular arguments. On the one hand, an inscription like Saude that lacks any context becomes a commemorative inscription for a burial because other topologically similar inscriptions are assumed to be dedicated to the dead. An inscription like Barmen, on the other hand, becomes associated with a burial because this is said to be the common function of rune-stones at graves, which would be further confirmed by the stone’s atypical context. Such speculative runology—as can easily be argued—does not withstand new approaches.

In recent years, runology has taken up several new academic trends, particularly from archaeology. This trend can be seen, amongst other things, in detailed spatial archaeological (or archaeogeographical) investigations, like Grimm and Stylegar on the early rune-stones in general, Stylegar on the N KJ72 Tune, or Carstens and Grimm on the so-called Blekinge-stones Bl 3 Stentoften, Bl 4 Istaby, and Bl 6 Björketorp. Besides this, one can especially note sociological approaches in which rune-stones have played a role. Thus, Bernard Mees has studied N Viking2011;28 Hogganvik see Knirk 2011 but also the Tune-stone and a number of other inscriptions from a new, pronounced socio-cultural angle, led by principles of intertextuality and discourse theory.As a similar trend one may observe that also linguistically trained runologists, for instance Michael Schulte (2007; 2015b), Kurt Braunmüller (2012), or Christiane Zimmermann, have tried different methods or models such as text-pragmatics or oral speech markers to tie the potentially epigraphic mode(s) of writing closer to social realities. One can also observe an increasing awareness of the special characteristics—or perhaps the very nature—of these inscribed stones as monumental media, which typically have concentrated on Late-Viking-Age rune-stones.A particular emphasis on the multimodal visuality can be found in Marco Bianchi’s dissertation, for which Terje Spurkland (2012) provides a useful discussion. Kristel Zilmer (2010) goes beyond the mere material dimension and elaborates on communication in all its modalities between oral monumentality and commemorative literacy, covering quite different perspectives. This newly emerging academic tendency views rune-stones as commemorative media, analyzing them for their multimodal, medial, and material features, causing the monumentality of rune-stones to be refined and reformulated.

So far, as it concerns such major trends in recent runological scholarship as outlined above, collective memory studies have been largely, yet not completely, overlooked. In fact, (collective) memory has gained some attention in runology, especially in a broader scholarship that also includes runological sources, and particularly among scholars working on stones from the Late Viking Age. Mats Malm has recently provided a helpful overview of such studies that rarely touch the surface of collective memory theory explicitly, but more often focus on studying the lexicon or certain linguistic expressions of commemoration as well as modes of commemoration across media. To these works one should make a few additions that provide more fruitful accounts for the context of this article—even though they focus almost entirely on the (Late) Viking Age.

Clear references to collective memory are presented by Ing-Marie Back Danielsson (2015) in her study of Viking-Age rune-stones placed along pathways, combining collective memory (with mention of Halbwachs) with a phenomenological approach. Her work stresses the bodily experience of rune-stones and how rune-stones themselves structure their surrounding landscape, human perception, and collective memory.The dimension of phenomenology in Back Danielsson’s article is influenced by Merleau-Ponty, as can be seen when Danielsson states that the stones together structured the landscape in a certain way, and also regulated how the body was to enter, encounter and experience this index, nexus, or gate to other worlds 2015, 81. A slightly different tact is taken by Back Danielsson in another article 2016 on Late-Viking-Age rune-stones, examining their social qualia, a framework borrowed from Piercian semiotics to denote qualitative experiences. Thus, Back Danielsson can state that rune-stones were not only memorial stones to deceased people …, but also a type of artefact that encouraged, renegotiated and/or reinforced different embodied, and therefore, mnemonic practices among the living 2016, 75. She maintains that not only places but also families were tied together in the landscape through the rune-stones. They resulted in shared experiences, life and death … In this way, individuals, collective memory and rune-stones were seamlessly interwoven 2015, 81. In another article, Back Danielsson (2016) articulates some further aspects of the embodied processes involved in human-object interactions. Here, she arrives at a noteworthy point when literally referring directly back to the above quote and adding to it that rune-stones were individual monuments, but they were also monuments that shaped and influenced social memory, which provide a foundation and context for them 2016, 80. These are indeed vital points and can only support the observations made later in this article, although based on different material.

Adopting similar lines of argument, Elisabeth Arwill-Nordbladh has discussed the famous stone Ög 136 Rök from the Early Viking Age in several contributions from an explicitly collective memory-oriented angle. Mainly guided by more recent archeologists and their focus on the bonds between memory and spatial, material manifestations, she also refers the idea of inscribed and embodied memory 2007, 56, both originally introduced by Paul Connerton as incorporated and inscribed bodily practices of collective (or social) memory 72–3. Later, she also cites Pierre Nora and his lieux de mémoires places of memory and milieu de mémoire milieu of memory Arwill-Nordbladh 2008, 170. Thus, she can conclude that people in Late Iron Age Östergötland were elaborating an active attitude towards memory, where apprehensions about time and the past were interacting in different ways 2007, 59. She also proposes that the Rök-stone was inspired by Old English or continental manuscript cultures in her attempt to find a suitable model of the seemingly new, unprecedented way the Rök-stone creates collective memory. Nonetheless, her arguments remain vague or ad hoc in the absence of any direct evidence for the use of manuscripts or wax-tablets as source for inspiration by Rök-stone’s carver or commissioner. Thus, her interpretation of Rök is in the end less convincing compared to more holistic approaches but nonetheless employ memory theories in a useful way.

When it comes to the Rök-stone, this has indeed been one of the most popular rune-stones,The Rök-stone has created a vast body of research literature and almost as many different interpretations of its role and function. For a helpful discussion of memory-related discussions of the Rök-stone, see Malm 222–24. For the present article, the Rök-stone forms too large a topic to be adequately covered, though it will be a worthwhile matter of discussion for a future study. Indeed, it should be worth re-evaluating the common opinion which sees the Rök-stone standing for a distinctively new development in the history of erecting rune-stones (see e.g. Arwill-Nordbladh 2007, 59). One may wonder whether this perspective ultimately comes from the fact that the stone’s main inscription is carved with runes of the younger rune-row, making it apparently automatically part of the corpus of the later (or younger) inscriptions. Interestingly, however, the Rök-stone also shows some older runes, thus taking in a strange intermediary position. which is especially true in regard to the number of works discussing aspects of memory in the inscription. These have in particular dealt with the runologically and linguistically difficult sakummukmini of the Rök-stone. Others, such as Per Holmberg for instance, have provided more holistic socio-cultural contributions about the inscription, with Holmberg stressing a locally bound, collective, and performative dimension of memory and knowledge. In the most recent article on Rök, Holmberg, Gräslund, Sundqvist, and Williams describe a remarkable concept of memory beyond ordinary recollection. Instead, they see the memories expressed in the inscription as ritual acts of social and religious significance relating to the past, present, and future, that together contribute to the maintenance and renewal of the world 18.This most recent paper on the Rök-stone was published just at the time the first draft of this article was finished, hence too late to be systematically considered here. Interestingly, Holmberg, Gräslund, Sundqvist, and Williams interpret Rök in the light of myths known from medieval Old Icelandic poetic sources, and possibly from some iconographical representations from the Viking Age. While I do not share all new interpretations brought forward by the authors, I highly regard their interdisciplinary approach and the attempt to interpret Rök as both relating to actual climatic events predating the stone some two hundred years as well as relating to much later myths—all involving collective memory on different levels. Stephen Mitchell mentions Rök as well in an article on memory in the medieval North. He makes the interesting claim that it is anything other than happenstance that the Rök stone’s opening focuses on memory, mediality, and performance: these functions were at the heart of such monuments, their production, and their performance 284. However, as Mitchell uses Viking-Age rune-stones only to dive into other aspects of memory and performance culture in the North, he remains outside the focus of my study.

The only article addressing memory specifically in the older runic inscriptions, focusing on rune-stones in their development until the Late Viking Age is by Elena Melnikova. Melnikova speaks of historical memory in rune-stones, or memorative function that rune-stones acquired 312, from the Early Iron Age to Late Viking Age. However, her contribution could be criticized for referring to problematic views on the elder inscriptions as magical and for going too far in her own speculations. Further below, I come back to recent studies dedicated particularly to the early rune-stones that introduce helpful new concepts—yet hardly coming closer to collective memory than referring to concepts like memorialization Mees 2017.

While most of the contributions discussed in the previous section are of great value in their own right, it is surprising that collective memory theory has hardly entered runological studies explicitly as a paradigm. It appears that collective memory is commonly brought into discussion because of the observation that monuments perform memory-work in a memoria-culture, which consequently makes rune-stones a logical object of research from varied angles involving their (collective) memory in different ways when studying past cultures of Scandinavia. Thus, studies of rune-stones and how they perform memory work, often in terms of mediality, have regularly been based on rather simplistic assumptions regarding the nature of collective memory. What has been lacking are investigations into the specific kinds and different roles of collective memory in rune-stones. It is for this reason that this deficit will receive more attention in the following sections of this study, especially because the early stones relate to the collective sphere in a number of ways and thus also to collective memory.

A key problem which appears when rune-stones have been addressed as bearers of memory is that largely modern conceptions and understandings are projected back onto early or proto-historical societies of Scandinavia, for instance what a memorial ought to be and stand for, or how memory is expressed. Of course, this study also utilizes modern theories about collective memory to apply it on very different material. To avoid major pitfalls and circular arguments, however, this article is asking what memory could imply outside the private sphere of grief and display modes for commemoration—as through elegy for instance. As the early stones hardly confirm such aspects of collective memory,An exception might be N KJ101 Eggja NIæR.55, which however does not adhere to any of the patterns among the remaining body of currently known early rune-stones. and thus are distinguished from the later Viking-Age rune-stones, there appears to be a gap in the debate worth investigating. An attempt at this is made in the following, utilizing collective memory theory for the discussion of a number of exemplary early rune-stones.

When applying the concept of collective memory, it should be kept in mind that its study involves neither a single theory nor does it have a concrete method for investigation. Instead it is represented by a multitude of theories or perspectives that form a framework or paradigm for investigation, instrumentalizing methods from other disciplines rather than having built up its own (cf. Kansteiner). The essence of collective memory is built on the observation that memory is socially determined. This idea, as pointed out already, goes essentially back to French academic circles around Émil Durkheim (1858–1917) and Henri-Louis Bergson (1859–1941), where it was first introduced by their pupil Maurice Halbwachs who thus is seen as the father of collective memory studies (cf. Gensburger).

As collective memory studies developed, the concept has not remained unchallenged. Important critiques have maintained that memory cannot be collective in the sense that a whole society remembers. So, as Susan Sontag stressed, societies can only stipulate memories 86. Collectives are selective in their maintenance of memory as Halbwachs already noticed 1925, 376; 1992, 172. Sontag speaks of collective memories as what a society chooses to think about, so only collective instruction (not collective memory) exists 85. Collective memories therefore are a fiction, according to Sontag, detached from their original meaning. This point can be brought in line with Reinhardt Koselleck’s view that any collective memories are impossible since memory requires experience and (eye)witness accounts as a necessity. However, as Koselleck states, there are kollektive Bedingungen collective conditions that enable memory 6.

While proposed as critiques against collective memory by Koselleck and Sontag, on the contrary, selectivity and preconditioning are important aspects of collective memory as already noted by Halbwachs.Illustrative in this context is Halbwach’s view on society, memory, and communication, which claims that les hommes vivant en société usent de mots don’t ils comprennent le sense : c’est la condition de la pensée collective. Or chaque mot (compris), s’accompagne de souvenirs people living in society use words that they find illegible: this is the precondition for collective thought. But each word is accompanied by recollections 1925, 377; 1992, 173. While the focus here is entirely on language, Halbwachs makes an important addition a few lines later, stating that c’est langage, et c’est tout le système des conventions socìales qui en sont solidaìres, qui nous permet à chaque instant de reconstruire notre passé it is language, and the whole system of social conventions attached to it, that allows us at every moment to reconstruct our past 1925, 377; 1992, 173. While Halbwachs first underlines the general quality and capacity of language (carrying out discourse in order to create memory through communication), he makes an important addition with the système des conventions socìales qui en sont solidaìres. More recent scholarship is essentially built on this observation when studying the media and material manifestations (in space and time) that create and shape collective memory. Later scholars have made useful additions, like Pierre Nora who has presented the concept of lieux des memoire places of memory which can be compared to Aleida Assmann’s variant of it, Erinnerungsräume spaces of memory 1999. Already Halbwachs highlighted the necessity of shared interests in order to make possible successful communication, as discussed above. These theories and ideas have also been related to Brian Stock’s concept of “textual communities” (12) among memory researchers and others (see below), describing a use of textual media that goes further than general communication, even beyond texts, while also being interested in the roles and functions which texts can play in collectives in creating identities and memories.

It is aspects like these stressing the social dimension of (collective) memory that are highly relevant to the present study. Can one find collectives existing around the early rune-stones that somehow resemble textual communities? Can evidence be found for selective, preconditioned acts of collective memory as already defined by Halbwachs? And how could these be brought in line with the Assmanns’ studies on the different mechanisms, kinds, or functions of memory? It is not enough to postulate that a rune-stone is an expression of collective memory simply because it is a monumental structure, as previous scholars using the concept have maintained. Herein lies a fundamental, ahistorical dimension shaped by modern conceptions of monuments and memory as well as the backwards projection of apparently similar structures from later times to the early rune-stones.

If we thus start by looking at the socialness of rune-stones, as previously stated, the collective and cultural dimension of rune-stones may appear obvious on several levels as they form messages, media, and contexts that were publicly displayed and prominently placed in landscapes and social spaces. These stones were loaded with textual and non-textual deixis, which has been discussed largely only for Late-Viking-Age stone inscriptions Jesch 1998, 464; Zilmer, and certain types of early formula. Additionally, the understanding of these predominantly short reference-specific messages—with their many nomina propria, titles, and epithets—presupposes collective knowledge and traditions. Hence, the rune-stones’ interpretability links them strongly to memory and its transmission.

It is important to consider how rune-stones were used in particular and how the different elements—stone, runes, context—were intertwined. The likely existence of different levels of meaning has been discussed by Michael Schulte (2013). In addition, socio-cultural dimensions have been pointed out on several layers, as, for instance, the relation to the slightly older and contemporary custom of erecting undecorated standing-stones Spurkland 2010, 81, so-called bauta(r)steinar, or to the engraving of cliffs, which go back for centuries before the introduction of runes. In opposition to other comparable types of human-made structures, the social aspects of these customs must be underlined, for the making of rune-stones must necessarily have required larger social groups, families, or collectives. Additionally, one may reckon with performative acts of ritual or religious cults conducted in connection with such stones, for which, again, collectives would be engaged, yet not just in performances or ritual acts but also in the transmission of knowledge, meanings, and understandings of world. Especially the latter aspect must be assumed crucial for stones with runic inscriptions, indeed for the general transmission of runes at all.

When it, however, comes to the commemorativeness of the early rune-stones under discussion, the phenomenon rune-stone is far from well understood. This is particularly the case when it comes to the oldest material, considering the fact that source-critical limitations affect the early rune-stones in many ways (see above), making any wide-reaching assumptions and theories difficult. Nonetheless, claiming that the stones should be interpreted as some sort of burial- or grave-markers rooted in a supposed commemorative culture appears even more problematic when noting that the traditional explanation of rune-stones and their functional context as commemorative grave-monuments is largely based on three key-elements.

Firstly, archaeological contexts are used as guidelines wherever there exist associations of certain stones with graves or burial sites; on the one hand by implementation (or re-use?) of rune-stones inside of graves, and, on the other, by association with graves or grave-fields (yet always on a one-per-cemetery basis). This point, in particular, has received some attention and re-evaluation, classically by Anders Bæksted and more recently by Lisbeth Imer (2011; 2015), showing how little evidence there is for direct associations of early rune-stones with burials as their primary function. Secondly, scholars commonly base their argument on a handful of pre-Viking-Age inscriptions that only possibly mention a burial. This interpretation can be rated as commonly overestimated as the inscriptions in question hardly establish any unequivocal links to burials. However, this issue would require more sensitive linguistic discussion than is possible here. Thirdly, rune-stones are generally compared in functional terms with younger monuments from the Late Viking Age, carrying commemorative inscriptions that link to a Christian context. It is this last point that should be the most worrisome and therefore is emphasized below.

Just recently, in a 2017 article, Oliver Grimm and Frans-Arne Stylegar suggested that almost all known Norwegian rune-stones in the older and transitional runes belong to certain or likely burial-contexts. This claim is framed by a sociological approach that should be highly regarded, taking spatial and power-political structures into consideration. Reasonable contexts have been developed for important monuments such as the Southeast-Norwegian rune-stone N KJ72 Tune NIæR.1, discussed in the following. The stone was examined in person by the author on November 12, 2015, and reads:

This stone is a good case to illustrate some crucial issues. While this stone was found implanted in the wall of a churchyard, it has been associated with a large grave-mound close by from where it must have been removed and used as building material already in medieval times. Thus, the Tune-stone once stood at a prominent burial site, yet likely alone over a larger cemetery that possibly was founded around the time of the stone’s erection. In this it is comparable to N KJ63 Einang. This pattern is one of the major divergences between the early rune-stones and Late-Viking-Age rune-stones, the former seemingly more sparsely distributed in time and space, while the latter are commonly found in closer vicinity of another, even on the same site, and within the limits of more private functions, that is, erected for individuals by their relatives.

Understandably, the inscription on the Tune-stone has been used as an important argument to determine the character of this rune-stone. While the second side of the stone likely mentions three daughters that either divided up the inheritance or carried out a heritage-feast for Wōđurīđaʀ*, the first side is said to state that a Wīwaʀ erected the stone in the memory of this Wōđurīđaʀ*. This interpretation depends on the preposition after, which can be paralleled only by one other early rune-stone, Bl 4 Istaby DR.359; RäF.98. The stone was examined by the author on November 15, 2015, and reads:

The interpretations of both Tune and Istaby as commemorative, honouring a deceased person, depend on the preposition after Tune or ᴀfatʀ Istaby, which is a cognate to modern English after. When considering the historical semantics of this lexeme it must be noted that they are far from only referring to acts of commemoration, which is why one should be wary when translating it with in memory of. It could simply mean for or in the name of Fritzner s.v. eptir. Considering that no direct burial-contexts exist for either stone, one might wonder whether the actual purpose of this stone, or Istaby likewise, was indeed that of commemoration. This would also be supported by the fact that primarily legal or property-demarcating connotations cannot be excluded for the Tune-stone as has been stated by a number of scholars (for instance, Spurkland 2001, 52–53). All this could indicate that ideas about rune-stones based on Late-Viking-Age examples usually had the major lead. There, Old Norwegian, Old Danish, and Old Swedish cognates to after are prominently found and are more likely to have formed a rune-stone culture influenced by Christian memoria Düwel 2013; Zilmer. But can such interpretations of eleventh-century stones be applied back in time on cases being possibly up to five hundred years older if not more?

If we therefore recognize that a concept of commemorativeness in its classical sense is compromised in many ways, it could help to return to the question posed above. Asking what specific kinds of (collective) memories we might be dealing with in the present material, it should indeed be worthwhile pondering whether different memories than purely commemorative ones can fit as suitable explanations for the early rune-stones. Before a different tradition takes over towards the (Late) Viking Age, arguably with direct links to Christian culture, earlier rune-stones follow different epigraphic habits. An illustrative case is the Reistad-stone KJ N74; NIæR.14, which has been discussed by numerous scholars since its first publication. The stone was examined by the author on November 12, 2015, and may be read as:

The second rune in the first line (top-line) is commonly read u, but may also be a damaged d, or even h. The reading d was preferred by Elmer H. Antonsen, who also believed the third rune to be an r. Whichever way this line is read, it is commonly held that it contains a nomen proprium. The runes unnam are commonly interpreted as a prefixed verb-form *und-/unþ-nām took. However, Antonsen 1975, 52‒53; 2002, 6–7 believed an ʀ to follow at the end of the second line and therefore postulated unnamʀ as object (to the root nomen *nēm- with negative particle *un-) to wraita as verb-form third-person singular preterit wrote to Germanic *wrītanaⁿ [write]. This explanation changed little regarding the semantical content of the inscription but imposed more linguistic problems than it solves Bammesberger 119‒122. His reading was subsequently rejected. Earlier interpretations saw wraita as the object and attempted to derive it equally from *wrītanaⁿ, which lacks equivalents in the Germanic languages. Following the originally suggested reading, Thórhallur Eythórsson proposed a very different interpretation, where he maintains the previous view of unnam, but links wraita to Old Norse reitr field 199. It is well possible that even the place-name Reistad reflects that form. While Thórhallur’s interpretation has much going for it—it can be considered the best solution on etymological, phonetical, morphological grounds—, it is commonly rejected on typological (semantic and pragmatic) grounds Barnes 2013, 28.

That said, intertextual comparison can be seen as a useful tool to support arguments for interpretation of early rune-inscription, while the value of typology has been underlined in many linguistically difficult cases cf. Braunmüller 2012, 72. Nonetheless, it is worth considering whether it would be more productive not to dismiss otherwise (etymologically, morphologically) rather unproblematic interpretations on solely typological grounds but to require stronger arguments instead. When moving on from an epigraphic investigation to the next step of interpretation, there has been a tendency in runology—as discussed above—to pay little attention to the great spread of the source-material over both time and place, which leaves societal changes and cultural (or collective) variations largely uncovered. Additionally, there has been a tendency to base interpretations on artificially extraordinary ideas about the nature of the runes. With this in mind, it should be reasonable to reconsider interpretations that appear at first sight typologically marked when there is any reason to doubt more conservative, unmarked interpretations, especially when doubts are backed up by other inscriptions.

Interestingly, recent contributions by Oliver Grimm, Lydia Carstens, and Frans Arne Stylegar have brought new views on early rune-stones Stylegar; Carstens and Grimm; Grimm and Stylegar by looking beyond inscriptions in order to take landscape and socio-archeological features into consideration. Michael Schulte’s interpretation of the recently found Hogganvik-stone as an emblem of power 2013; cf. 2015a is especially exemplary of this trend. Another helpful framework could be that of textual communities, as defined by Brian Stock 12, where a text-bearing stone would carry meaning beyond primarily private (even though publicly displayed) purposes like grief and commemoration yet also beyond the literate as Spurkland 2010 has shown.It may seem inappropriate to replace the concept with memorial communities as Mats Malm has suggested, as the notion memorial can imply too many serious semantics and certain functions. Additionally, it is important not to uncritically overuse Stock’s productive concept as it was tailored for medieval and partly literate collectives. Nonetheless, one could consider a further specification of Stock’s concept as an analytical framework to provide reference and reasonable interpretations that support explanations of political and power-related properties in the public beyond the texts itself. Thus, focusing on the medial aspect like Stock does, rune-stone communities could similarly give way to a helpful distinction of the phenomenon rune-stone for the given period and areas by taking into consideration the whole range of possible conventions attached to material and performative aspects of rune-stones, which provided people with means to reference and (re)construct the past in the present and the present in the future. In lieu of contemporary sources that could help us understand the use of rune-stones in their communities, one may arrive at meaningful conclusions using phenomenological approaches to perception and embodiment linked to the active engagement with rune-stones on their spot, which otherwise remain largely out of reach for the modern researcher.For a helpful discussion of such matters, see Back Danielsson 2016. It can also be meaningful to address rune-stones as (epigraphic) habit like Mees 2015; 2017 has done, moving the focus to a pronounced cultural approach. This would stress that rune-stones exist in a socially conventionalized system of collective memory outside language and beyond the inscriptions themselves, within their runic communities.

However, one should be wary not to simplify and reduce collective memory to a custom in which stones act on their own as memory-bearers since memory cannot exist on its own. Instead, memory needs to be lived, requiring a phenomenological process of experience, remembrance, or recollection. Furthermore, memory needs to be shared, as a process of transmission, communication, and thus it is interactive as Andy Clark has stressed. Therefore, while studies into the historical Nordic languages and their graphic systems are of crucial importance to gain insights in what is communicated, it should be acknowledged that language does not operate in a vacuum. Similar remarks have been made already by Anders Andrén 1997, 151 or Terje Spurkland 1987, 53.Indeed, there has been a rather long process within runological scholarship to engage with the objects and contexts of runic inscriptions, especially in terms of the greater literacy debate, going back at least to the 1980s. Some more recent reflections of this discourse can be found in methods of contextual analysis that essentially bring text and object into a dialogue Düwel 2008, 16–17; Imer 2015, 16–20.

Thus, we may have a framework suitable to stress a shift in studies of early rune-stones away from focusing purely on their (historical) functions by acknowledging that rune-stones served certain communities, for instance for the purpose of emblematic display of power and identity (see above), amongst other epigraphic habits that were slowly developing, changing and shifting along the way. In order to understand collective memory, however, one needs to look beyond the cultural tools used for collective memory and pay special attention to the particular use made of them on particular occasions Wertsch and Roediger, 324; my emphases. This also includes early rune-stones—hence the following section discusses exemplary cases, looking into the particularities of how early rune-stones possibly shared collective memories in Clark’s terms.

Of course, there is a danger of over-generalizing memory Kantsteiner; cf. Roediger and Wertsch; Brown, Gutman, Freeman, Sodaro, and Coman. It is not possible to establish a link between every socio-cultural phenomenon or historical media and memory, and there are a number of objections that have been both connected to memory as well as opposed to it e.g. Klein; Confino; Olick; Wertsch and Roediger. Such examples include the categories of history, myth, and tradition, discussed in a helpful way by Astrid Erll 2017, 36–39. Maybe one can say that collective memory relates to the past, but it does not necessarily relate to factual history (or historiography). In a similar way, one would not say that collective memory is myth but rather that myth can be a form of collective memory. Therefore, not every rune-stone relates per se to collective memory just because common opinion has it that a stone-monument relates first and foremost to commemoration.

A definition of collective memory, which Jan Assmann offers, is worth considering in the present context. Assmann claims that memory or remembrance happens in two ways. One he calls the mode of biographic memory, which is based on personal experience in the recent past 1992, 52. This could be a burial-stone commissioned by the relatives of a deceased. The other—the mode of foundational memory—is more relevant here for: der Modus der fundierenden Erinnerung arbeitet stets … mit festen Objektivationen sprachlicher und nichtsprachlicher Art: in Gestalt von Ritualen, Tänzen, Mythen, Mustern, Kleidung, Schmuck, Tätowierung, Wegen, Malen, Landschaften usw. The foundational mode always functions … through fixed objectifications both linguistic and nonlinguistic, such as rituals, dances, myths, patterns, dress, jewelry, tattoos, paintings, landscapes, and so on J. Assmann 1992, 52; J. Assmann 2011, 37. He refers to them as Zeichensysteme sign systems with a mnemotechnical function, that is, to aid the identity of groups through their memory of their history and past 1992, 53. Assmann later specifies the sign systems used in this process as normative and formative forces 1992, 76. Thus, the orientation of memories of the foundational type is that of potential entities that can be activated or actualized to serve the creation of collective culture or identity, not only in the now, but, worth noting, also towards the future. For collective memory is more than the past, as has been underlined by collective memory studies. When Jan Assmann thus investigates further how memory works by building up storage from which memories can be drawn for later re-actualization, he implies an orientation towards the future as an inherent potential.This also becomes explicit in the division of Speichergedächtnis storage memory and Funktionsgedächtnis functional memory provided by Aleida Assmann 1999, now commonly distinguished as canon and archive in English. Interestingly, Pernille Hermann 2010; cf. 2019, 107 has studied precisely such founding memories in the medieval North, especially in the context of early history after the settlement of Iceland.

The next question would be whether the early rune-stones could serve in a similar way as foundational or even founding memory media, thus being more than just storage devices for concrete memories but active mediators in the transmission of collective memory whose mediality rightly must be highlighted Zierold; Rigney. For cultural memory is based on communication through media, as Astrid Erll 2010, 389 puts it. Indeed, it appears that Assmann’s above-cited list of sign systems for (founding) memories calls for the addition of rune-stones. Some noteworthy points about rune-stones and collective memory have been brought forward by Lydia Klos. In her attempt to understand the rune-stones of Sweden from the Late Viking Period as media, Klos refers to the central ideas about collective memory detailed by Jan Assmann Klos 323–25; cf. Malm 188. She highlights especially the selection of subjective information for the future (325–36, cf. 324 mentioning inscriptions like U 114 Runby, which ends with an urge to continue the memory of the commemorated men while humans live). As such information she identifies elements from Early Scandinavian culture(s) like name-giving traditions or the practice of tying memories to place-names 326–28. Similarly, Judith Jesch had already concluded that a stone like U 729 Ågersta, referring to its location in the inscription and stating that it will remain there, is extending the power of the utterance into the future 1998, 468. Of course, here we are dealing with rune-stones from the Late Viking Age, but both scholars attempt to address aspects outside the framework of private commemoration. This altogether implies that collective memories are not simply stored away but born of people’s minds and carried on in the form of a dialogical exchange between humans and other medial manifestations or representations of collective memory. Thus, it appears fruitful to discuss the early stone-inscriptions further as an emerging memory-culture in which rune-stones function as memory-makers, that is both founders but also bearers of collective memory. Such a process can only be understood as developing hand in hand with the Early Scandinavian societies in constant exchange between both cf. Braunmüller 2012, 74.

It can be helpful in the understanding of collective memory in the early period to look into the rather widely acknowledged principles in the history of the runes and their transmission. Especially interesting in the context of this study is the relative uniformity of the runes over time. Of course, there are prominent changes, variations, and assumed local traditions found throughout the material. Yet, the rather high degree of similarity across such a vast area and timespan is still surprising. It could be sufficient to assume some way of transmission of the runes from generation to generation, yet not only in time but over space as well. While there is hardly any agreement on the precise practical circumstances around this transmission of the runes (e.g. Barnes 2012, 23), there is good reason to state that, at least as a writing system, the older rune-row was part of the collective memory. The social distribution of runic writing is also interesting within this model, since runes generally appear restricted to the upper social strata (see Düwel 2015). This social elite might well be the same social frame with which the collective memory conveyed in rune-stones should be identified. Nonetheless, one has to think of this as several smaller Scandinavian collectives of locally or regionally leading families that are culturally closely connected and likely to have exchanged people, goods, but also knowledge.

To what extent oral and performative acts were happening in connection with or as an extension of the rune-stones is difficult if not impossible to say, yet such strategies must be reckoned with and they are likely to have helped the collective memory survive. Indeed, formulaic typology has been studied for the Viking-Age stones in some greater depth, and scholars like Joseph Harris 2010a; 2010b and Judith Jesch 1998; 2001; 2005 have underlined the later rune-stones’ memorial language, focusing on public fame as well as the monument and its materiality, which in turn could be compared to certain features of skaldic poetry (cf. Schulte 2010a, Marold 2012). When considering the pre-Viking-Age stones, we are dealing with less representative and far more complex material—and new finds of inscriptions might change the picture entirely.The recent discovery of the sensational Øverby-stone, nonetheless, only seems to support the argument(s) of the present article. Unfortunately, the publication of this new early Scandinavian rune-stone came too late to be meaningfully integrated in the analyses and discussions of this study, which is why a brief comment on the Øverby-stone is included at the end of this article in form of a postscript. As far as the current state allows any careful conclusions, the occurrence of names (or nomina propria that can serve to designate a person) and other commonly short utterances are far more prominent features of the early stone-inscriptions than possible formulaic language. For scholars like Wolfgang Krause, this alone, then, has led to the assumption of commemorative functions—yet on what grounds?

Looking through the examples discussed above, one can observe that the postulated name on the Saude-stone could be there for a variety of reasons judging from the inscription alone; the memorial for a deceased being just one of many purely speculative assumptions. The Barmen-stone both from the inscription and the context hardly points to a burial memorial, but rather could contain a commissioner’s or carver’s signature, if one follows the comparative evidence. A somewhat more complex case is the Tune-stone, which could point to a burial from its context (at least through conjecture), and also its inscription would at first confirm this idea with little doubt. One might admit, carefully however, that the situation is unclear. Interestingly, the inscription topicalizes wiwaʀ as seemingly connected to the execution of the stone (as commissioner or carver). It further emphasizes the association of the stone with woduride, twice by direct deixis, yet without explicitly stating his death or burial. Instead, attention is drawn to the three daughters, who performed some act in connection with the heritage of their father. What is surprising is that the daughters’ names are apparently of little importance as they remain unmentioned. Obviously, their genealogical relation with Wōđurīđaʀ* was sufficient enough to mention. Maybe the economics of carving a rune-stone played into it, but there are other arguments speaking in favour of another explanation: the inscription closely resembles Nordic patronymic naming-traditions—as long back in time as they can be safely attested—and thus could symbolize the family-founding forefather in genealogical lines. It is this aspect of founding stones that could indeed provide a meaningful framework for these exemplary inscriptions.

As an attempt to make better sense of the previous observations, one should not forget a number of rune-inscriptions that are characteristic for the beginning of this age of memory in runes, namely rune-inscriptions on cliffs or other outcrops. These inscriptions belong to the northernmost rune-stones before the Viking Age. An interesting case is the now perished inscription N KJ56 Veblungsnes NIæR.25 e͡kịrilaʀwiwila|⁄n I, the irilaʀ of W. in Møre og Romsdal, which belongs to a number of other runic inscriptions with ek I and the enigmatic word irilaʀ or erilaʀ, likely denoting a functional title.The word irilaʀ is also found on the newly discovered Øverby-stone. The stone’s primary publication by Iversen, Kjesrud, Bjorvand, Kimball, and Mannsåker also provides a very helpful overview of all irilaʀ/erilaʀ inscriptions 81. Most relevant is the detailed contextualization of Øverby’s irilaʀ within a wider historical framework Iversen, Kjesrud, Bjorvand, Kimball, and Gundersen 79–93, which situates the Øverby-stone at the centre of a system of control, perhaps of military function, despite its fringe rather than centralized location. It appears that the stone may have been an important medium in founding and/or maintaining these structures of control. Thus, it can be argued, the stone took part in the making of collective memory (cf. Postscript). There have been some early ideas linking the remaining runes of the inscription—the sequence wiwilaṇ or wiwila| (the last sign possibly a damaged n-rune or a separation marker)—to the place-name Veblungsnes, which were discussed and swiftly rejected by Sophus Bugge in his edition NiæR, 1:323. There is a possible connection between a new establishment or at least some restructuring of power in the area that supports this link, as it is not completely impossible that the inscription falls in the time of this process. Thus, we potentially get a place(-name) that has lived on in time as a collective memory for which the actual founding is preserved.

The inscription N KJ55 Valsfjord NIæR.28 ekhagustaldạʀþewaʀgodagas | ẹ[ … ⁵⁻⁶ … ]ʀ I, (the) hagustaldaʀ, G.’s servant … , executed less monumentally, but thoughtfully situated at Oksvoldvågen at Valseidet in Sør-Trøndelag, is another case of such a cliff-inscription. The place of the rune-inscription was an important isthmus on the north-way shipping route. Thus, it represents an exemplary case, linking a cliff-inscription with a local power demonstrating its presence in the area it controlled during the Iron Age. Again, the inscription may contain titles or functionary designations in hagustaldạʀ young warrior; retainer and þewaʀ servant, while godagas most likely would be a name Antonsen 1975, 46. Another inscription, N KJ53 Kårstad ekaljamark[ị]ʀ | ḅại[j̣̣]ṣʀ I, (from?) the other (border)land; … , is today preserved in several boulders reassembled outside the University-museum of Bergen. Its original location was in a rock-art field of the Bronze Age, where the inscription was probably added at a later stage. Following the suggestions by archaeologists Liv Helga Dommasnes 68 and Gro Mandt (in Lødøen and Mandt), the inscription was probably carved around the same time when new burial customs were introduced, which demonstrated individual power more clearly than before, at a grave-field a few kilometers away. In their eyes, the inscription might well indicate a change of power structures in the fjord-area.

None of these can be seen to refer to burials on the spot, which is ruled out by the cliff locations. Instead, they show a clear emphasis on a speaking or carving ek, I, stressing its deictic relation with or an actual presence or power in the ek, which then is further enhanced by adding a functionary’s titles or by highlighting possible settlement properties and persons of power in society. Thus, these cliff-inscriptions demonstrate another crucial side of founding memories linked to heavily place-centred inscriptions, for what is remarkable with these inscriptions from Kårstad, Veblungsnes, and Valsfjord is their location at cliffs or rocks in the fjords and waterways of Norway’s western coast.There are actually more rune-stones that have clear links to waterways as has been noted by some scholars. Terje Spurkland (2001) mentioned that the earliest rune-stones appear to represent a coastal phenomenon, even though Lisbeth Imer 2011; 2015 recently has noted that this might simply resemble settlement clusters of the Iron Age. If we give Spurkland’s theory credit, however, detailed studies of rune-stones, such as the Barmen-stone (see above), can reveal close connections to the sea and main water-routes from their placing in the landscape. Hence, it is almost impossible not to stress that other epigraphic habits than classical commemoration are at stake here. The question that remains would be how significant these inscriptions are for the remaining material, and how far they can serve as a model for other rune-stones.

Only a few times, as in the above case of the Reistad-stone, have alternate concepts been proposed to explain an inscription in its context in other terms than commemoration—yet, they happened to be rejected rather soon thereafter. In that particular case, the twofold coverage by collective memory through the place-name Reistad, in addition to the stone itself, is remarkable, however, not that unique since it is possibly paralleled by similar cases like the mentioned Veblungsnes-inscription. For the Reistad-stone, Thórhallur Eythórsson stressed the similar patterns found in Old Icelandic sources, like the famous , in which, as a rule, attention is drawn to the importance of the founding-father of a settlement to whom later descendants trace their claims and rights back. As a kind of epicentre around which land gets claimed, inhabited, used, structured, and maintained, names survive through the age of times and form a symbolic form of collective memory on their own.Thus, one may understand links between a rune-stone and a burial-ground on a more symbolic level of meaning. This possibly relates to similar phenomena of symbolic meaning in certain types of stone settings, like the tree-shaped stone-settings from the Scandinavian Iron Age and even possible real trees—known in modern folk-memory as tuntre farm-tree and by other names. A number of promising examples of such structures have been discussed by Anders Andrén in his recent book on Old Norse cosmology 2014. Looking for intercultural parallels or patterns could be fruitful, and the idea that the pre-Christian structures, which one is facing here, do not necessarily have to remain conformed to modern categories based merely on common sense. This would speak for the point that the very idea of commemoration may instead go beyond its modern conception by including founding memories and other possible aspects, becoming a much larger and multi-faceted concept of memory. Thus, it would not come as a surprise if place-names and rune-stones were interrelated more often—for both worked apparently within the same areas of collective memory.

The present discussions have shown that for the early stone-inscriptions (c.150–750 CE), collective memory, as a socio-cultural phenomenon and analytical concept, is of crucial importance: collective memory conditions the perception and conception of these monuments. Of course, this article is not the first attempt at reconsidering the early runic inscriptions. Research has taken place into their socio-cultural dimensions of semiotic and pragmatic expressions of communication, for instance, or the whole body of socio-cultural anchorage as media. It should also be mentioned that several of the peculiarities dealt with in the previous sections were already touched upon within the discussions of the 1980s to the early 2000s (in particular) about the matter of orality and (early) literacy in Scandinavia.See, for instance, Jesch 1998, Harris [1996] 2010a; 2010b; Spurkland 2010. Though not discussed here, this previous research would be useful in a framework together with collective memory theory since it would add a great deal to some of the fundamental thoughts and concepts.

This article has, however, confirmed assumptions about the socio-cultural qualities of early rune-stones, for instance by demonstrating that the transmission of the older rune-row, or the establishing and maintenance of runographic habits, would not be possible if not part of a collective memory. Thus, the present article has sought to illustrate how collective memory studies provide helpful theoretical frameworks and analytical concepts to support runology. The discussions have also shown that collective memory studies can shed new light on insufficiently answered questions regarding the nature of the oldest runic inscriptions, as well as little questioned models of interpretation within the study of runes.

The outlined difficulties of the scholarly debate make it clearly visible that runes have taken specific connotations in our contemporary collective memory, which must be considered one of the chief principles or even prejudgments colouring scholarly work. As an attempt to navigate around some issues at the heart of this problem in scholarly (and public) reception, this article focuses only on the early rune-stones, a deliberately materially confined source-basis. The primary aim of these medial limitations was to approach the use of runes not universally but to gain insights into their particularities, while still keeping the approach open for comparisons within the corpus. This presupposes a perspective on rune-stones as cultural phenomena, which is to say that the early rune-stones are a culturally specific tool used in certain contexts. Throughout the article, this perspective is tested and adjusted by examining exemplary case-studies. As is illustrated by the case studies, past realities are necessarily clouded in uncertainty, but each revisitation equipped with alternative theories can reveal new plausible explanations. The previously accepted interpretations of early rune-stones as burial-markers rarely correlate completely with the circumstances surrounding the stones, if a grave is present at all the dating, gender, or spatial relation rarely fits. This observation, however, has room for an interpretation of early rune-stones as fundamental to the creation and preservation of collective memories around them.

As a novelty within runic studies—although building on a number of helpful studies by other scholars—the present article makes an attempt to develop a more holistic and interdisciplinary approach based on collective memory theories. In contrast to previous studies and well-established common opinions within runology, this article argues that the early rune-stones performed memory work on a scale beyond being burial-stones, grave-markers, or memorials for the deceased. Rather, they were involved in the foundation of collective memories, founding important focal points that make it possible to relate to the past, present, and the future. It has been demonstrated here that common opinions on the nature of early rune-stones prior to the Late Viking Age or the medieval tradition hold little plausibility, when claiming that these early inscriptions relate primarily to burials, predominantly serving commemorative, at times apotropaic, functions. For the purpose of reassessing the question of which monumental purposes and habits these stones represent, the present article uses collective memory as a tool to glean meaning hidden beneath or beyond text, revisiting contexts and opening up previously overlooked interpretations.

Just upon finishing the first draft of this article, the newly discovered early Scandinavian rune-stone from Øverby in Rakkestad, Vestfold, received its initial publication by Frode Iversen, Karoline Kjesrud, Harald Bjorvand, Justin J. L. Kimball, and Sigrid Mannsåker Gundersen. As is quite common with freshly discovered rune-stones, its first publication seems to have raised more questions than it could answer regarding certain issues. The primary issue, as with most early rune-stones, is the very reading of the inscription. Although new digital methods were applied for the primary publication, producing largely reliable and more objective data right from the start, Michael Schulte 2020 has in the meantime proposed a critique of the first reading as well as supplying a partially different interpretation of the inscription, which may indeed be favoured. Aside from this slightly different reading/interpretation of the inscription, Schulte’s contribution remains however without consequences for the contextualization of the new irilaʀ in its wider historical-geographical horizon down to the local socio-political environment Iversen, Kjesrud, Bjorvand, Kimball, and Gundersen 79–93. Accordingly, irilaʀ/erilaʀ most likely designated a high-ranking local leader with primarily administrative and military functions. Thus, the Øverby-stone essentially supports the arguments developed in the present article and would serve well as a demonstration of the conclusions of sections V and VI above, particularly due to the noticeable association of irilaʀs’ stones with marked borders and later legal districts. One may thus even suggest that the irilaʀ appears to be involved in founding and maintaining the collective memory of these institutions, whose vague memories can be traced through a broad range of contemporary but foreign and later Medieval Scandinavian sources.It should be noted that the original context of the rune-stone is today largely unknown and lost. An association with a grave-field remains uncertain, yet Iversen, Kjesrud, Bjorvand, Kimball, and Gundersen stress that the irilaʀ in general may have been involved in burial-rituals, based on the observation that four of five inscriptions on stone mentioning irilaʀ have tentative associations with graves 82; 85; cf. 75. The fortunate circumstance that at least twelve irilaʀ/erilaʀ inscriptions exist today, however, shows that burials cannot have been the primary functions of such inscriptions.