Interpretative cooperation with biographical texts: A semiotic approach to analyzing collective memory

Abstract

This article reports on research into the little-known process of how documents of memory, in this case biographical narratives of life under the communist regime, are received by members of the community of memory. The research is based on an experiment in which three generations of family members (two generations of eyewitnesses of communism and one postcommunist generation) were asked to infer the political identity of narrators from very short fragments of biographical texts concerning the communist era. Using a semiotic approach, the authors perform a detailed analysis of how readers, in their formulation of an acceptable interpretation of the narratives, cooperate with these excerpts of biographical texts. The results show that behind generally weak intergenerational differences in cooperation with the texts, there are certain specific differences in such cooperation between the young postcommunist generation and the generations of their parents and grandparents. The authors suggest that these distinctions indicate the role semiotic competencies play in determining what memory will be accomplished in a given situation.

Keywords

Biography collective memory communism Czech Republic generations narratives

From a functional perspective, the social sciences define collective memory (in the broad sense of the word (Olick and Robbins, 1998)) as knowledge about the past that shapes and maintains a group’s collective identity. In the words of Jan Assmann (1995), “a group bases its consciousness of unity and specificity upon this knowledge and derives formative and normative impulses from it, which allows the group to reproduce its identity” (p. 128). Collective memory enables people—as members of a particular community of memory (Bellah et al., 1985: 333; Irwin-Zarecka, 2007)—to direct their identity work toward obtaining, maintaining, or changing their own status or the status of the social group to which they feel they belong. A community of memory may be understood as any group or society that relates representations of the past to its current identity and ipso facto transforms them into representations of its own past. As such, collective memory exists in society through various media that constitute “documents of memory”; such media-based documents tend to incorporate a combination of discursive, visual, plastic, or acoustic elements (Kansteiner, 2002: 190). Perhaps the most important contribution to the study of expressively complex documents of memory is that of the French historian Pierre Nora (1986–1993) and his colleagues on “sites of memory” (lieux de mémoire).

We can approach collective memory, conceived of as a specific knowledge base, either from the perspective of the sources of its production or through the sources of its reproduction. By the former, we mean social actors that produce publicly accessible “documents of memory” (e.g. memoirs, narratives, films, books, exhibitions, historical documents), while the latter represent social actors that consume them. Most studies of collective memory have focused on the former and the way in which collective memory is intentionally and strategically produced: in other words, on the formation of a group’s collective memory in relation to their past, their current identity, and their political legitimacy (Conway, 2010; Kansteiner, 2002: 180). Studies of the collective memory of the Czech communist past are similar in this respect; examples include F. Mayer’s (2003) study of the construction of the image of the Communist regime, O. Šmídová’s (2006) work on the memory of Czech Germans, or J. Přibáň’s (2005) work on the relationship of memory to the codification of the concept of the nation. Moreover, the production of documents of memory constitutes the main research interest of the Oral History Center at the Institute of Contemporary History, Academy of Sciences of the Czech Republic.

As noted above, relatively little attention has been devoted to how documents of memory are received by members of a community of memory—a pity in the sense that “the crucial issue [of collective memory research] is not what is represented [as the past] but how this representation has been interpreted and perceived” (Confino, 1997: 1390). For instance, we can learn how the period of socialism has been represented in postcommunist countries in the last 20 years by analyzing textbooks published during this time (Beneš and Nigrin, 2010), but nonetheless remain ignorant of how these textbooks have been used in classrooms or at home. In our experience, pupils use Wikipedia rather than textbooks in preparing their schoolwork, so the formative influence of textbooks on collective memory is not obvious. In this respect, it is worth noting W. Kansteiner’s (2002) concern about the growing significance of mass media, which are gradually becoming the dominant (co-)producers of the documents of memory consumed by contemporary communities. In an effort to capture the public’s attention, contemporary mass media re-contextualize documents of memory as fragmented communications (individual documents, pieces of information, testimony) which together make up the substratum for the formation of a “low-intensity” collective memory. In Kansteiner’s (2002) view, communities of memory are not built on shared interpretations of particular past events, but on the shared experience of their medialized representations. As pertains to the postcommunist context, C. Tileagă (2009, 2011) showed how medialized textual representations of biographies (e.g. secret police files), memories (e.g. testimonies), and history (e.g. governmentally sponsored reports) are used together to form a kind of hypertext of memory through which the representation of the national past is collectively achieved.

Despite its necessity, research on the reception of documents of memory is still uncommon, with notable exceptions. Harald Welzer et al. (2002) studied the transmission of memory, via stories and conversations, regarding the period of National Socialism and the Holocaust in forty German families. The authors convincingly demonstrate that while all generations were well informed about the past, the younger generations transformed their parents’ and grandparents’ eyewitness reports into their own stories having different accents and focal points. These perceptual shifts resulted in considerably differing images of the past between generations, where morally ambivalent and often incoherent documents of memory (eyewitnesses histories) were received by younger generations in a manner that integrated them into a morally positive, sometimes even heroic image of their parents and grandparents. The fact that the memories were transmitted in a communicatively cooperative manner allowed all generations to remain unaware of the generational discrepancies in perception and to achieve the impression that there was only one definitive shared understanding of the past (see also Tschuggnall and Welzer, 2002; Welzer, 2010). Another example is a recent article on the reception of the film Hotel Rwanda (Gudehus and Anderson, 2010). The authors argue that “reception research at the microlevel, specifically at the point of actual reception, is indispensable for theorizing [historical] macro-phenomena and has not yet been addressed satisfactorily” (Gudehus and Anderson, 2010: 359). Drawing on analysis of interviews with people who have seen the film, they constructed two idealized types of modes of reception of the narrative document: emotional and factual. The former comprises a moral and evaluative stance toward the events represented, the latter is rather more “reflexive in the face of the possibility of narrating history” (Gudehus and Anderson, 2010: 358). However, because of their small and nonsystematic sample, they could not relate these two reception modes to sociodemographic variables.

Our study presented here is in line with such work in that we share the same assumption: the production of representations of the past is only a half of the collective memory story. We focused on the reception side, at the microlevel, and on how, at the microlevel, the content of a particular textual document is accepted by readers as being such a document. In other words, we set out to determine how, when presented with a text document, members of a community of memory interact with it and are then able to incorporate it into an intersubjectively shared representation of the past. An answer to this question could help us better understand the process of the construction and modification of collective memory. We would argue that collective memory is not just an intersubjectively shared collection of representations of the past, but also a specific competence: the ability of a member of the community to interact with documents of memory. In our view, what makes a person a member of a particular community of memory is not just his or her knowledge of the community’s shared past, but—and perhaps above all—the ability to adequately cooperate with past-related documents circulating in a society.

The experiment

The objective of the study was to gain greater understanding of how people receive documents of memory by analyzing their interaction with them from a semiotic perspective. In order to do so, we needed to examine the ordinary reading process, that is, a routine and self-evident activity, and thus the idea to draw on the ethnomethodological tradition of highlighting a taken-for-granted practice by “breaching” it emerged (Garfinkel, 1967: 35; McHoul, 1982). Breaching experiments, as Garfinkel (1967) described them, “start with familiar scenes and ask what can be done to make trouble” (p. 37). While we did not carry out a regular breaching experiment as Garfinkel defined it (it did not disrupt background expectations nor evoke feelings of unease, confusion, or hostility), our experiment allowed us to easily talk to participants about their understanding of, and conclusions about the texts, and the reasoning behind them—which would be rather difficult if not impossible in naturally occurring settings. Therefore, in order to make the interpretive cooperation with a text observable and accountable (McHoul, 1982, chs 2 and 3), we presented the participants in the experiment with unusual documents in the form of short excerpts from autobiographical narratives of life under communism and provided a structured set of tasks, including to identify the narrator. The brevity of the texts ensured that a minimum of context for the narrative statements was provided and that forming an understanding of them as meaningful representations of the socialist past would require considerable cooperative work from the reader.

A total of 14 respondents participated in the experiment. They represented four students and their families, and comprised three generations in each family—grandparents, parents, and children (the students). Members of the grandparents’ generation spent the majority of their lives under communist rule, the middle generation remembered the period from their youth or middle age, and the youngest generation—people in their 20s—possessed only indirect knowledge of communism. Our choice of participants was guided by the assumption that each of the three generations’ different degree of experience with life in socialism may influence their cooperation with biographical documents about this period, and we wanted to explore such possible differences.

We gave each participant a sheet of paper with 21 short excerpts of narratives selected from transcriptions of biographical interviews. Each of the selected narrative excerpts was between 35 and 45 words long and included the word “we” or “us”: the deliberate inclusion of these pronouns served to ensure that the texts could be associated with the collective identity one of three possible kinds of narrator—the communist functionary, the dissident or “the ordinary person.”¹ The idea was for the excerpted narratives to contain some discursive construction particular to the group to which the narrator belonged, in order to allow the respondents to more easily attempt to infer the narrator’s political identity. The excerpts were randomly selected from the corpus of interviews using a random number set.

Participants in the experiment were presented all 21 excerpts on sheets of paper, and in addition to (1) identifying to which of the three political categories (communist functionary, dissident, or “ordinary person”) the narrator of the given excerpt belonged, they were asked to (2) underline the words or parts of the text that led them to make their decision. Respondents had no idea that each of the three types was equally represented among the excerpts (in order to keep them from making guesses based on such knowledge and from comparing the excerpts to each other too much), nor were they in any way directed to lean toward attributing one particular identity to the texts. Quite on the contrary, if they felt unable to identify the narrator, respondents could choose to identify more than one or none. Upon completion of the task, the respondents were interviewed by us and asked to explain the reasons underlying their selection of identity and of words to underline for each excerpt. The interviews were recorded and transcribed for later analysis.

Method of analysis

The analysis was based primarily on Ricoeur’s hypothesis of threefold mimesis. According to Ricoeur (1984), a narrative encompasses within itself a threefold figuration: first, the world as a course of events understood in a specific way by the narrator (mimesis 1 or pre-figuration); the actual story, in which the narrator encodes the course of events into a specific genre of symbolic communication, that is, narrative emplotment (mimesis 2 or con-figuration); and finally, the reader’s or listener’s interpretation of the narrative (mimesis 3 or re-figuration). Conceptualizing narrative this way is useful from a sociological perspective in that unlike pure formalist–structuralist methods, it integrates both the narrator’s understanding of the world, which precedes the narration, and the audience’s subsequent interpretation into the analysis of the narrative.

We analyzed the respondents’ action of reading—or what Ricoeur would call mimesis 3—using the semiotic approach elaborated by Umberto Eco (1984) in his book on reader cooperation with narrative texts. Eco (1984) was inspired by Pierce’s pragmatic realism, where “to discover an object means to discover the way by which we operate upon the world producing objects or producing practical uses of them” (p. 194). This constituted the basis of his own theory of cooperation with narrative texts, in which the reader’s interpretation involves the purposeful, practical task of activating their content. As Eco (1984) laconically puts it, “the texts are lazy machineries that ask someone [i.e. the reader] to do part of their job” (p. 214). In our research, we accordingly examined the ways in which readers cooperated with the biographical narrative excerpts when interpreting them. Since one of the key strategies employed by readers in interpreting a narrative is to construct their own hypothesis of the narrator’s identity (Bortolussi and Dixon, 2003: 60–66; Kindt and Müller, 2006), we stimulated interpretive cooperation by instructing the respondents to attempt to infer the identity of narrators.

The analysis of textual cooperation with the narrative was observed on several levels, as they are described by Eco (1984: 11–27). The first level of cooperation involves the lexicon, which is the sequence of verbal expressions or the lexemic surface of the text (Eco, 1984: 15). Readers cooperate with a text by applying a system of rules (codes) to the text’s words, which help them convert the text to the first level of content. When readers read “so we argued a bit” in the text, they are able to recognize that in this text, the word “bit” means “a small amount” and “argued” means “verbally communicate mutual disagreements.” Conversely, the phrases “Chinese boudoir” or “sophisticated systems” may not necessarily clearly mean something to all readers on this basic level, so some will try to guess at their meaning from the context of the text. In our research, we were mainly interested in the lexicon in the sense of the words and phrases in the texts identified (underlined) by readers as indication of the narrator’s identify, even if they obviously represent only a fragment of the entire lexicon on which they based their interpretation of the texts. The underlined words tended to relate in some way to something in the past potentially significant in identifying the narrator. To illustrate, here are several verbal expressions that the readers underlined: prison, we secretaries, what we are studying, string-pulling, and so on. We expected that the lexicon would help us answer the question of what role the element of particular words and phrases in the narrative played in leading readers to accept the document as a document of memory and to what extent that element is shared across generations and respondents.

The second level of cooperation relates to the reader’s encyclopedic knowledge. Eco (1984) describes the encyclopedia or encyclopedic competence as the reader’s ability to understand the words in the text in relation to the cultural tradition to which the text’s basic vocabulary refers (p. 23). For example, if the phrase “the opening up of borders” appears in a text, the knowledge required to construct what the text is about and continue reading is drawn from the reader’s encyclopedia, relating it to the sudden fall of communist regimes in Central and Eastern Europe. In sociological thought, the concept of the encyclopedia resembles Schütz’s (1970) concept of “the stock of knowledge at hand,” but Schütz conceives his concept in more general terms, as an interpretative frame for current and past experience of everyday life (p. 74). Eco’s encyclopedia, however, is more narrowly conceived as the ability to relate language to culture the way an encyclopedia book relates each entry to its cultural meaning. In our research, we examined the encyclopedic level because it refers to what readers tend intuitively to link to collective memory, that is, a particular stock of knowledge about the past. We examined the specific connotations readers drew from their underlined words in the basic lexicon on the first level of cooperation with the text and, from a comparative perspective, were interested in the similarities and differences in the respective encyclopedias of the three generations.

We also attempted to analyze the third—and the most complex—level of reader cooperation, namely, the construction of “possible worlds,” which refers to the reader’s expectation of what is going to happen next in the text, to propositional attitudes toward the unfolding of events, and predictions about the state of phenomena and things (Eco, 1984: 217–224). The concept is much discussed in semiotic and literary theory (see Doležel, 1998) as a means of describing and analyzing the complex relation between factual and fictional reality in literary works. We have adopted and used this concept in a manner that differs in one important respect from its common usage. In our research, a possible world is defined as that which the reader construes through his or her knowledge (encyclopedia)—actualized by the text—as to what may or may not have happened in the past. We modified our definition for the reason that biographical narrative is different from fiction: biographical narrators are not entirely at liberty to use their imagination in constructing the world of their life story as they could find themselves in conflict with the reader (listener) were the latter to protest “but that’s not how things were!” By analogy, readers of biographies likewise cannot receive the text uncritically and freely construct hypothetical structures of possible worlds to which it could refer, because they know that it refers to a real past and not fiction (unlike historical novels or films). We believe our modified concept of possible worlds to be useful for analyzing interpretative cooperation with the biographical texts in our study, as it helps us see how their readers (rather than narrators) arrived at the meaning that they derive from the given excerpts. Readers of an autobiographical narrative develop propositions about what is going to happen next and what to expect, based both on the text and on their own encyclopedic knowledge; those propositions are either confirmed or not once the reader reads further on (Eco, 1984: 217). Our readers were confronted with a text that provided them with a very limited amount of material from which to formulate their hypotheses as to what the text is about, and could not verify any part of their hypothesis against other material. Thus, drawing on their encyclopedias, the respondents constructed the situational context or contexts in which, for them, the presented text made sense. From a semiotic perspective, we understand these imagined contexts as “possible worlds” framing the texts, that is, situations that the respondents regard as having been realistically possible. Unsurprisingly, in some cases, the respondents related the situations or events in the excerpts directly to their own or another person’s experience, representing, in our analysis, the most extreme case of a “possible world.” And, in our observation, it was this interpretative movement between a “possible” and the “real” world that constituted the reader’s main cooperative activity with the text.

Findings

We formulated and interpreted our findings to answer the question of what kind of interpretative cooperation respondents engaged in when reading the short excerpts of biographical narratives about the communist period in former Czechoslovakia. We should emphasize that although our respondents were Czech and that we focused on the country’s communist past, the aim of the study is not to explain how Czechs relate to their past, but to use this particular case to analyze a generic process.²

The section devoted to findings is divided into the different dimensions of reader cooperation as outlined above, with special attention devoted to identifying any generational differences among respondents—that is, differences based on respondent’s greater or lesser degree of experience with the given period to which the narratives relate.

Basic dictionary

On the lexical level, we were surprised to find that the influence of generation in respondents’ cooperation with the text was rather weak, and the differences insignificant. The only notable difference was found in the youngest generation, which tended to focus more on the historically relevant meanings of individual words (e.g. prison, iron curtain) than on any phraseological characteristic of a particular group or on potential explanatory contexts. So although the respondents’ personal experience of communism was different based on their generation, this fact was not reflected on the level of the individual words and phrases used by each in interpreting the biographical texts of this period. In most cases, respondents underlined words they believed to be linked directly to the narrator’s identity, as the following examples show:³

R[espondent]:

Well, I think here that they were—I don’t get the sense at all.

I[nterviewer]:

[laughter]

[…] at all, but if they’re talking about fitters, electricians, switchmen, then they were probably, probably, normal, ordinary people. (Excerpt 7, student A; the words quoted from the excerpts here and below are presented in roman font—authors’ note)

Well here I put communist because the word barrack [barák in Czech—authors note] is typical for them, this [“RC”] will definitely be the Regional Committee. (Excerpt 9, parent Cb)

This suggests another finding already indicated in the last quote: the reliance on something that we could call a group-style of speech. Readers, except for the youngest generation, occasionally constructed the image of the narrator from certain words that respondents deemed to be characteristic of the speech of a particular social group in the past. Identification of the narrator as a communist functionary or dissident on the basis of such words is demonstrated in the following interview excerpts:

Well here, a very sophisticated system, that’s a Bolshevik style of speech, right.

Hmm.

There’s various things Gorbachev evidently doesn’t want, that also has a Bolshevik ring to it, doesn’t it.

Hmm.

So that’s why I picked communist. (Excerpt 6, parent Cb)

Well, definitely not a communist. Because he wouldn’t have been in prison at that time. Well … actually just because the reason there … it isn’t given. But I’d say, maybe then, when I think about it, that it was probably a dissident. The reason being that there’s this refined speech … in that sentence. And an ordinary prisoner actually … young … he’d presumably be … a laborer. So he’d speak in a different way. (Excerpt 10, parent B)

The encyclopedic level

The second level of reader cooperation with the text involves the reader’s encyclopedic competence. In our research, we interpreted encyclopedic competence primarily as the past-related knowledge activated in the course of attempting to understand a historical text. The initial hypothesis was that this competence would differ between generations and that the phrases in the text would activate different knowledge or different degrees of knowledge, depending on which generation the reader-respondent belonged to. Thus, the assumption was that encyclopedic competence would be conditioned by personal experience, in this case, by whether the reader had lived part of his or her life under communism or possessed only indirect knowledge of it (through the media, family, literature, friends, etc.).

Our assumption, however, was not confirmed—at least not in the manner we expected. Instead of revealing generation-dependent differences, the encyclopedic knowledge drawn on by respondents in explaining their choices tended to refer to only widely shared stereotypes about (1) dissidents, “ordinary people,” or communist functionaries; and (2) life under communism. As we attempt to show in the interpretation, this does not result from all the respondents knowing “the same things” or “equally amounts” about communism, but rather from the lack of a need to activate an extensive encyclopedia in the given task (receiving the document as a document of memory and identifying the narrator)—respondents usually make do with just a standard range of common knowledge. On the other hand, despite the fact that the perceived stereotypical characteristics of the three main political groups were shared across the generations, the contexts the respondents created for them in their interpretation were diverse, which ultimately led them to identify different narrators. An illustrative example of this is the interpretation of the following biographical excerpt:

Excerpt 21: Today it amuses me that it was like that. There was this train of events, and we had no desire for things to go wrong. With the dissent, when I did come into contact with it, I more or less took it as a marginal affair. (The narrator is a communist functionary)

Each of the three possible identities were assigned to this narrator by different respondents, and in each case, the respondents were able to support their choice with reference to their encyclopedic knowledge of the given identity or period.

In the following interview, the narrator is identified as a “dissident”:

And here there’s train of events, for me a key word, and then I also might underline had no desire for things to go wrong. And now I’m in a quandary here, because … Train of events is used by someone who’s already, who’s already thinking about these things.

Hmm.

And who maybe then saw in these things, as they occurred, a train of events rather than just one thing. That means, it somewhat indicates someone more informed.

Hmm.

So, but … There’s this had no desire for things to go wrong, none of us had a desire for things to go wrong. So I’d say some dissident and he may have been speaking about something when it wasn’t yet ready and when then the timing, […], it was simply supposed to be different, so say they didn’t want that some, that some provocation would take place. (Extract 21, parent C)

Another respondent identifies the narrator as an “ordinary person”:

Well, that’s what I think too, that it was a normal person who wasn’t interested in some change where anything could happen and didn’t give much thought to dissent. It wasn’t that common for people to be part of the dissent and for someone to think something up and for someone to want to change something and so on. (Excerpt 21, grandparent C)

Yet, a third interviewee chooses “communist functionary”:

Oh some bigwig, communist, said that. He considered dissent to be a marginal affair. Well. They underestimated it. For them it was like …

Hmm… So they underestimated it.

I think that they underestimated it a bit. Yes. (Excerpt 21, grandparent B)

If we take the respondents’ comments and summarize the explicit encyclopedic knowledge they used in explaining their choices, we get the following pieces of information: dissidents were more informed about the political situation than ordinary people were, and they threw themselves into confrontation with the regime; “ordinary people” did not think about politics too much and did not get politically engaged, whether for or against the regime; communist functionaries (erroneously) regarded dissent as an insignificant political force. It must be said that the encyclopedic competences of the respondents are not at variance, even though they led to identifying the narrator as belonging to a different political group. It could even be said that all of the knowledge presented above could form part of one and the same encyclopedia and that the divergent classification of the narrators is due to the various indices activated differentially by the text.

From a generational perspective, it is worth mentioning one difference observed between the encyclopedic competence of students and the older generations having direct experience of communism. While the knowledge of the actors in the students’ encyclopedias resembled typological images of the figures that inhabited the previous regime (dissident, communist), the older generation referred primarily to a specific or generalized experience from the past. This is how one student explains why she chose the narrator she did:

And in the last…piece the speaker mentions dissent, but in the end I was leaning more toward it being an ordinary person rather than an active dissident as a person who expresses resistance to the regime. (Excerpt 21, student A)

A dissident is defined here by his or her expected role or attitude of “a person who expresses resistance to the regime.” The members of older generations (see the citations above) based their choices on how people, in their opinion, were likely to behave: “So I’d say some dissident and he may have been speaking about …” The students seem to describe the typical world of historical actors rather than describing what took place. But their typological approach does not imply that they erred more in their guesses at the identity of the narrator and at what the excerpt is about. Since eyewitnesses can use their encyclopedic knowledge to retrieve and actualize multiple plausible contexts for interpreting the excerpt, an element of uncertainty is introduced in identifying the narrator; the students’ encyclopedic knowledge, on the other hand, is based on types of actors and situations, enabling them to outline simple and seemingly superficial proposals, which are nonetheless often correct. Perhaps this is owing to the experiential character of the encyclopedic competence of eyewitnesses, that is, they are able to retrieve possible contexts and explanations from the perspective of the period described, while students tend rather to apply explanations that circulate in contemporary (artistic, documentary, expert) discourse about the communist past.

Possible worlds

The third and most complex level of reader interpretation is the construction of “possible worlds.” As described earlier, for the purposes of our research, we defined a “possible world” as that which the reader, based upon his or her knowledge (encyclopedia) as actualized by the text, construes regarding what may or may not have happened in the past to which the narrator is referring. So a possible world is not a world reconstructed on the basis of a person’s direct experience or individual memories and recollection of the past, it is a proposal; a hypothesized plausible course of events that is arrived at through reference to experience from the past as well as additional knowledge.

From the very first analyses, it was apparent that the character of the possible worlds the respondents constructed in the interviews was in a fundamental respect determined by the task assigned in the experiment: to read the excerpts and search them for indicators pointing to the identity of the narrator. This fact had both a stimulating and limiting effect on the formation of possible worlds—structurally, they most often referred to features ascribed to individual categories of speakers and the situations they could be found in. An example is the proposed possible world of communism in which the status of cook and that of communist functionary are deemed to be mutually exclusive, as demonstrated in the following comments:

Well, given that it was said that if it’s a communist then it would be … uhm … a person of high rank in the Communist Party, then, it’s likely that a cook was not a person of high rank and…uhm…the impression from this is that of an ordinary person under the previous regime. When she says about herself that they were just two cooks and talks about the manager. (Excerpt 1, student A)

To compare, here is a comment from a member of the generation of parents, who identifies the narrator solely on the basis of encyclopedic knowledge and has no need to propose a possible world:

Well, I identified her as an ordinary Party member, and I did so because the words used there included string-pulling and that she received money for an hour more, because the practice was that if a person was a Party member then they had some advantages, even an ordinary member, that a non-member didn’t have. (Excerpt 1, parent Ca)

The comparison reveals one of the differences observed between generations in terms of how they received documents, and to which we already drew attention in reference to encyclopedic knowledge, namely, that students logically deduced the identity of the speakers from a familiar typology of figures and situations they had indirect knowledge of (dissidents were against the regime, lived in difficult circumstances, and were imprisoned; communist functionaries supported the regime, had it good, and were powerful). This means they tended rather to construct possible worlds, while both generations of eyewitnesses relied less on proposals and more on direct knowledge (and lack of knowledge) drawn from their own personal experience.

Rarely were any pure possible worlds observed in respondents’ comments, but when they occurred, it was most often in relation to excerpt 17:

Excerpt 17: That I should start to think about joining the Communist Party. Well certainly, that was a question of discussion both at home and with my parents. And they said, and they gave themselves as an example, that, in short, if that weren’t to happen then I’d be putting at risk … (The narrator is a dissident)

The situation in the excerpt is clear (the narrator is talking about joining the Communist Party), but it is hard to infer what the narrator’s political affiliation is. Probably for this reason, the statement required substantial reader cooperation, which did not stop at encyclopedic knowledge but led to the construction of possible worlds, from which the status of the narrator was subsequently deduced:

Because there is talk here that not joining the party would be to put oneself at risk. … it’s as though, …, to me it seems that the parents, …, they probably weren’t communists body and soul, that they don’t expressly agree with the ideology so much that they would want to join the Party, but more that they did so for some, I dunno, practical reasons.

Hmm.

For their safety. And that this person, …, actually, …, he’s not speaking about it either with much conviction, about that he wants to talk about it, that he wants to discuss the pros and cons as to why he did or did not join the Party. So, …, I think it was an ordinary person. (Excerpt 17, student C)

This is a statement from an ordinary person who is thinking here about joining the Communist Party; he discusses it with his parents. This is a person who maybe in the end joins, but not as…, …he even adds that he’d be at risk if he doesn’t join, so you can see that there’s no great enthusiasm for this ideology, that it’s just that, when, if, he does it, then it’ll be for existential reasons, that’s the standard situation of an ordinary person, who in the end simply did join the Party, then. But he probably didn’t try to get into some high function, right. Because he just wanted not to have problems. (Excerpt 17, parent D)

that I should start to think about joining the Communist Party, this is a young person from a family in which his parents are either themselves, themselves communists, or people who realize that he’d simply have things much easier in the communist regime, an easier career, if he were to be a member of the Communist Party.

Hmm.

So he wasn’t yet a member of the Communist Party, so he isn’t a‚ “C”[ommunist functionary], definitely not a dissident, because he was thinking about it, so that’s very obvious. “O” [ordinary person]. (Excerpt 17, grandparent Db)

As we can see, interpretative cooperation proceeds smoothly regardless of whether the reader’s identification of the narrator agrees with the actual narrator’s identity. In the case of excerpt 17, none of the 14 respondents correctly identified the narrator as a dissident, as they evidently all considered it very unlikely that a dissident’s parents would try to talk him into joining the Communist Party and that he would weigh his options in this respect. Nevertheless, the following is also a plausible interpretation: the parents are not Party members, and cannot be because of their “bourgeois origins”; they pragmatically advise their son to join the Party, but he does not want to on account of his opposition to Communist policy (in part because of the fate of his parents); later, this opposition leads him into the ranks of active opponents of the regime—dissent. There are also other versions of possible worlds in which the narrator could be identified as a dissident. For example, he became a Communist in the 1960s, but he was expelled from the Party in 1968; in the 1980s, he signed Charter 77 and became politically active. However, as the comments indicate, the respondents considered such scenarios as relatively unlikely. This brings us again to the finding that when confronted with a document of memory, respondents actualized only a part of their encyclopedic knowledge and constructed only the simplest and most probable possible worlds. This serves to reduce the significance of differences between respondents with extensive encyclopedic knowledge based on personal experience (i.e. eyewitnesses) and respondents who have no such knowledge and who cooperate with the text based on relatively limited, indirect sources of knowledge (i.e. students).

Discussion

Several findings emerged out of our analysis of reader cooperation with excerpts from biographical narratives. We found that at every level of cooperation—lexicon, encyclopedia, and possible worlds—readers were able, without much difficulty, to make a semiotic interpretation in order to deduce the narrator’s hypothetical identity. Although they were not always sure they had identified the narrator correctly, uncertainty tended to occur as a result of an excess of possible interpretations rather than a lack of them. There were no striking differences between generations at any level of cooperation. In all three generations, the respondents highlighted similar words and phrases—those for which they possessed the necessary encyclopedic knowledge, and in some cases, they constructed possible worlds of the past that provided a framework for their interpretation. At the level of the lexicon, we found two modes of the reader’s construction of meaning: on the one hand, the readers derived the meaning of the entire text or utterance based on the words in it, and on the other hand, words served as linguistic characterizations of different identities—the readers associated words and phrases or grammatical and stylistic forms of expression with particular narrators. At the level of encyclopedic competence probably the most significant finding was that different respondents actualized different knowledge in relation to actors and situations in the past, but on the whole, this knowledge was consistent. In the older generations, when individual respondents differed from each other in how they cooperated with the excerpts, the difference had less to do with the scope or content of the encyclopedia at their disposal than with what they chose to actualize from their encyclopedia and what knowledge they decide to draw on in their interpretation of the excerpt. This was not the case, however, in the student generation, whose encyclopedia was based primarily on “typological” knowledge, on the characteristics typically assigned to the individual categories of actors and to features of communism. Nevertheless, ultimately, in terms of the political identities assigned to the narrators, the generations did not differ. The construction of possible worlds revealed the same finding as did the lower levels of cooperation: there are no major differences between generations, except perhaps the slightly stronger tendency among the youngest generation to engage in the construction of possible worlds, as compared with the two older generations who were aided in their interpretations by the actualization of their own personal knowledge and experience.

The main question we would like to discuss is the possible reasons for the surprising similarities observed in the reader cooperation process among all three generations. When preparing the research, we, quite on the contrary, expected to find significant differences at the levels of lexicon, encyclopedic competence, and possible worlds between the three generations based on their different degree of life experience with communism. As concerns the ability to receive documents of memory, some conceptions of collective memory as a type of knowledge about the past imply differences between generations of witnesses of the past and those possessing only indirect knowledge of it (Schuman and Rodgers, 2004; Schuman and Scott, 1989). This assumption, however, was only partly confirmed. As we noted in the preceding section, we did find some differences between the generation of students and those of their parents and grandparents, but they were not as strong as the overall transgenerational similarity.

In our attempt to explain this finding, we first considered non-semiotic explanations. The similarity in cooperation seen between generations could be caused by particular kinds of common “family memories” of communism, born of more or less frequent domestic conversations about the topic. This idea had to be rejected, however, because quite surprisingly, students unanimously reported that the occasion of our experiment was only the first or second time they had conversed about the communist past with their parents and grandparents.⁴ Another explanation that can occur to the sociologically minded reader is that the social homogeneity of the sample (families of university students) could cause the overall similarity of textual cooperation. However, the students’ families were heterogeneous with regard to their social and educational background.

In terms of a semiotic explanation, one possible explanation for such low intergenerational variability in the readers’ cooperation with documents of memory can be that the interpretation of the excerpt by respondents was only weakly determined. Perhaps if we had given the respondents the same excerpt to read again at another time, they may have actualized different encyclopedic knowledge and come up with different narrator identities, even if their method of cooperation with the text need not have changed. The variability of accounts, then, would be largely brought about by individuals’ momentary choices among their resources available for making sense of the excerpt. In such a case, the lack of obvious structural differences between generations could be due to (1) the vast number of legitimate interpretations of the past typical for a democratic society, and more importantly, (2) the interpretative openness of (narrative) texts presented in such fragments, which would correspond with Eco’s claim that any text is open to an endless number of semiotic interpretations regardless of selectional circumstances (i.e. the circumstances in which the text is read (Eco, 1984: 19)).

In our view, an awareness of the interpretative openness of a document of memory during its reception has important implications for our understanding of collective memory—it makes the pragmatic aspect of the process more significant: collective memory does not solely comprise a stock of knowledge about the past but also encompasses the competences that enable the reader to accept (or reject) a particular document of memory and interpret it in relation to given circumstances. The search for generational variability in collective memory should not be directed primarily at determining the extent of a generation’s knowledge about the past, but at examining generational competences in relation to the reception and interpretation of existing documents of memory.

A second possible semiotic explanation for the similarity we observed between generations could be that the respondents tended to mobilize only the most easily accessible and shared (noncontentious) parts of the lexicon, encyclopedia, and possible worlds to provide a meaningful interpretation of texts. In other words, the task was very easy, touching merely the surface of collective memory and, specifically, of reader cooperation with documents of memory. This would leave any deep-lying intergenerational differences concealed, much as if we were trying to determine the differences in students’ mathematical knowledge by assigning them the task of adding 2 + 3.

Indeed, some of the excerpts were easy to interpret, and all the respondents managed to grasp them easily and in the same way. For example, upon presentation of an excerpt containing the statement “99 percent of society that voted for us,” the respondents, referring to their encyclopedic knowledge of voting procedure under the communist regime, clearly deduced the narrator to be a communist functionary. Not all the excerpts, however, were this straightforward and easy to cooperate with. Sometimes, the respondent had to mobilize an entire range of cooperation strategies and employ every level of cooperation in order to arrive at a choice of narrator that they could justify. The different choices of narrator identified also suggest complex reader cooperation.

Third, there are also grounds for thinking that generational differences were largely obscured by sources shared in actualizing the excerpt content. Commonality in cooperation with texts need not reside in the nature of the excerpts, but rather in the nature of the “memory” used by respondents when reading. As shown by W. Kansteiner, whom we mentioned in the introduction, the main buttress of contemporary collective memory is formed of fragmented and decontextualized documents produced by the media for the purpose of massive and easy consumption. This gives rise to a kind of “thin” or “low-intensity” memory that is routinely produced and consumed without manifest controversy (Kansteiner, 2002: 189–90). If we accept this hypothesis, then the weak generational division in reader cooperation could be the result of the fact that all the respondents, regardless of their generation, drew primarily on this “low-intensity” memory to determine the identity of the narrators, actualizing other strata of experience and knowledge about the past only in those instances where it offered insufficient support.

Conclusion

We have attempted to demonstrate that collective memory should be studied not only on the level of production of representations of the past, which usually take the form of documents of memory, but also at the level of their reception by members of a certain community of memory. We suggest it is the ability to interact with various documents of memory rather than mere information retrieval from collective or individual stock of knowledge that defines persons as members of a community of memory. From this point of view, historians, in their capacity of people with a vast and profound knowledge of the past, are not always considered to be exemplary members of a community of memory specifically because of their particular (expert and critical) approach to documents of memory.

A semiotically oriented approach enables us to explore differences in readers’ cooperation with textual documents. Inasmuch as reading a narrative text is “joint work of the text and reader” (Ricoeur, 1984: 76), it presupposes various semiotic competences. In our study, we paid attention particularly to encyclopedic competence, which is necessary in order for a reader to make sense of the text in a given cultural context (Eco, 1984). As our findings reveal, both the non-eyewitnesses and eyewitnesses of the socialist period in the study had sufficient available knowledge about life in socialism to interpret the short narrative excerpts in a plausible fashion; eyewitnesses of socialism, however, displayed a greater encyclopedic competence in the sense that they also actualized features of the excerpts that non-witnesses did not—namely, the identification of group-style speech.

Furthermore, regardless of generation, the knowledge about the socialist past used by participants in explaining their specific reading of excerpts consisted predominantly of current mass media stereotypes about this period and its actors: communists, dissidents, and “common people.” Nevertheless, when confronted with a document of memory, this shared stereotypical representation of the past did not always predetermine participants’ interpretive choices when inferring the political identity of the narrator. This finding is in agreement with Eco’s claim of unlimited semiosis (Eco, 1976: 68–69) which, in the context of memory studies, means that a document of memory can be read or consumed differently (or with different interpretive results) within the same dominant image of the past (see also Welzer et al. (2002) for a similar point regarding family memories of National Socialism and the Holocaust).

In summarizing all of the above, we are led to the conclusion that knowledge about the past is not the only or principal component of collective memory but is rather a precondition or a tool for its production. As we have shown, a crucial role in its reception is also played by semiotic competences that determine what memory will be accomplished in an actual situation (see also Bietti, 2010, 2012 for examples of shared memory coordination and negotiation in conversation where participants’ statements about the past can be considered as documents of memory). Such competences may also predispose members of a community of memory to different commemorative activities and to their different experience of these activities. Finally, it is also obvious that similarities and differences found between eyewitnesses and those with only indirect knowledge of the past constitute only one dimension of reception of documents of collective memory, and that there could be many others (as showed, for instance, in Shahzad, 2011).

Footnotes

Acknowledgements

This research was financially supported by Czech Science Foundation, grant n. P414/10/0790. We would like to thank to Zuzana Stehlíková, Tereza Marková, and Michal Rázga for their help with the data collection. We are also grateful to Markéta Fuchs for help with the English.

Notes

Author biographies

Martin Hájek is a Lecturer and Researcher at the Institute of Sociological Studies at Faculty of Social Sciences, Charles University in Prague. His main research interests are methodologies of text analysis (discourse and narrative analysis), morality and normality, sociology of state socialism, and collective memory. Currently, he is involved in the project of multilevel analysis of biographical narratives about life in state socialism.

Marie Dlouhá is a doctoral student in sociology at the Institute of Sociological Studies at Faculty of Social Sciences, Charles University in Prague. She participates in the project of multilevel analysis of biographical narratives about life in state socialism.

References

Assmann

(1995) Collective memory and cultural identity. New German Critique 65: 125–133.

Bellah

Madsen

Sullivan

. (1985) Habits of the Heart. Berkeley, CA: University of California Press.

Beneš

Nigrin

(2010) Sozialismus in tschechischen, slowakischen, polnischen und deutschen Geschichtsschulbüchern. In: Zimmermann

Haslinger

Nigrin

(eds) Loyalitäten im Staatssozialismus: DDR, Tschechoslowakei, Polen. Marburg: Herder-Institut, pp. 343–363.

Bietti

(2010) Sharing memories, family conversation and interaction. Discourse & Society 21(5): 499–523.

Bietti

(2012) Joint remembering: cognition, communication and interaction in processes of memory-making. Memory Studies 5(2): 182–205.

Bortolussi

Dixon

(2003) Psychonarratology: Foundations for the Empirical Study of Literary Response. Cambridge: Cambridge University Press.

Confino

(1997) Collective memory and cultural history: problems of method. American Historical Review 102(5): 1386–1403.

Conway

(2010) New directions in the sociology of collective memory and commemoration. Sociology Compass 4(7): 442–453.

Doležel

(1998) Heterocosmica: Fiction and Possible Worlds. Baltimore, MD: John Hopkins University Press.

10.

Eco

(1976) A Theory of Semiotics. Bloomington, IN: Indiana University Press.

11.

Eco

(1984) The Role of the Reader: Explorations in the Semiotics of Texts. Bloomington, IN: Indiana University Press.

12.

Garfinkel

(1967) Studies in Ethnomethodology. Englewood Cliffs, NJ: Prentice Hall.

13.

Gudehus

Anderson

(2010) Understanding Hotel Rwanda: a reception study. Memory Studies 3(4): 344–363.

14.

Irwin-Zarecka

(2007) Frames of Remembrance: The Dynamics of Collective Memory. New Brunswick, NJ: Transactions Publishers.

15.

Kansteiner

(2002) Finding meaning in memory—a methodological critique of collective memory studies. History and Theory 41(2): 179–197.

16.

Kindt

Müller

(2006) The Implied Author: Concept and Controversy. Berlin and New York: Walter de Gruyter.

17.

McHoul

(1982) Telling How Texts Talk. London: Routledge & Kegan Paul.

18.

Mayer

(2003) Les Tchèques et leur communisme: mémoire et identités politiques. Paris: Editions de l’EHESS.

19.

Nora

(ed.) (1986–1993) Les lieux de mémoire. Paris: Gallimard.

20.

Olick

Robbins

(1998) Social memory studies: from “collective memory” to the historical sociology of mnemonic practices. Annual Review of Sociology 24: 105–140.

21.

Přibáň

(2005) The tradition, codification and interpretation of the civic and the ethnic concepts of the nation: reflections on the temporal dimension of constitutionality and constitutional symbolism. Czech Sociological Review 41(4): 563–582.

22.

Ricoeur

(1984) Time and Narrative. Chicago, IL: The University of Chicago Press.

23.

Schuman

Rodgers

(2004) Cohorts, chronology, and collective memory. Public Opinion Quarterly 68(2): 217–254.

24.

Schuman

Scott

(1989) Generations and collective memories. American Sociological Review 54(3): 359–381.

25.

Schütz

(1970) On Phenomenology and Social Relations: Selected Writings. Chicago, IL: The University of Chicago Press.

26.

Shahzad

(2011) Collective memories: a complex construction. Memory Studies 5(4): 378–391.

27.

Šmídová

(2006) Collective memory of the Czech Germans: golden times of the First Republic. In: Profantová

(ed.) The Small History of Great Events in Czechoslovakia After 1948, 1968 and 1989. Bratislava: Veda, pp. 213–243.

28.

Tileagă

(2009) The social organization of representations of history: the textual accomplishment of coming to terms with the past. British Journal of Social Psychology 48(2): 337–355.

29.

Tileagă

(2011) Rewriting biography: memory, identity and textually-mediated reality in coming to terms with the past. Culture & Psychology 17(2): 197–215.

30.

Tschuggnall

Welzer

(2002) Rewriting memories: family recollections of the national socialist past in Germany. Culture & Psychology 8(1): 130–145.

31.

Vaněk

(ed.) (2009) Obyčejní lidé…? Pohled do života tzv. mlčící většiny. Životopisná vyprávění příslušníků dělnických profesí a inteligence [Ordinary People …? Inspection in the Life of the Silent Majority. Biographical Interviews with Workers and Intelligentsia]. Prague: Academia Publishing.

32.

Vaněk

Urbášek

(eds) (2005) Vítězové? Poražení? Politické elity a disent v období tzv. normalizace. Životopisná interview [Victors? Vanquished? Political Elites and Dissent During “Normalization” Period]. Prague: Prostor.

33.

Welzer

(2010) Re-narrations: how pasts change in conversational remembering. Memory Studies 3(1): 5–17.

34.

Welzer

Moller

Tschuggnall

(2002) “Opa war kein Nazi”: Nationalsozialismus und Holocaust im Familiengedächtnis [“Grandpa was not a Nazi”: National Socialism and Holocaust in Family Memories]. Frankfurt am main: Fischer.