“How It Was”: Semiotic Approaches to Soviet References

Amy Garey

Amy Garey is a doctoral student in the Department of Anthropology, University of California, Los Angeles. Address for correspondence: 375 Portola Plaza, 341 Haines Hall, Box 951553, Los Angeles, CA 90095, USA. amgarey@ucla.edu.

I would like to thank Nancy Levine, Yunxiang Yan, Mariko Tamanoi, Robert Bird, and Pauline Turner Strong for commenting on parts of the paper, as well as Elizabeth Bishop, who cautioned me against the words “historical memory” many years ago. Thanks, also, to three anonymous reviewers and the editors of Laboratorium for their suggestions. Research in Russia was supported by the University of Texas at Austin and would not have been possible without the hospitality of the residents of Kuita.

Keywords: Russia, discourse analysis, comedy, post-Soviet transformations

In a 1969 Soviet cartoon entitled Antoshka (Little Anton), dutiful Young Pioneers implore the layabout Antoshka to harvest potatoes, play the accordion, and generally be industrious: “Antoshka, Antoshka, let’s go dig up potatoes,” they sing. He refuses. When the others gather for lunch, however, Antoshka finds his soup bowl empty. The cartoon’s moral message, “no work—no food,” is the intentional converse of the Marxist catchphrase, “From each according to his ability, to each according to his needs.” In the capitalist 2000s, the sentiment is outdated, if not outright anachronistic. Nonetheless, a young Kazakhstani comedy team brilliantly sampled the theme song of Antoshka in a 2002 performance. Explaining what this musical reference means today and why it might be humorous, requires tracing how Soviet-era indexical valences seep into contemporary discourse.

What should we call such vestiges of Sovietness? Are these citations history? Are they “social memory”? Are they pop culture, or imagined community (Anderson 1983)?

All references to the past have three terms: an event in the world, sensorially apprehended; a private, mental image of that event; and public depictions of it. Historians sift through representations in newspapers, diaries, artifacts, and interview materials in order to understand, through triangulation, a now-vanished moment—to answer the question, “What happened?” However, analyzing the process of remembering requires a different question, “What is happening?” That is, how are references to the past currently created, circulated, and understood? Using data from sots-art visual parodies, Soviet bloc sketch comedy competitions, and nostalgic discourses in rural Siberia, this article examines how historical images are reworked both in everyday interaction and global media contexts. I first describe how Peircean semiotics concretizes the mechanisms linking personal experiences and public representations, then use this lens to examine how two methods of transmitting information about the past—mass-mediated and interpersonal—differ in their implications for meaning making, resignification, and censorship.

Parody, in both visual art and sketch comedy, is an ideal site to investigate processes of resignification (how meanings change) because comedy depends on multiple meanings. For instance, in the joke, “A duck walked into a bar, ordered a scotch, and told the bartender, ‘Put in on my bill,’” the word “bill” has two interactionally salient interpretations. Redefinitions of the image of Stalin or other familiar Soviet signs rarely take the form of such straight puns, and comedy certainly is not the only arena for resignifcation. It happens daily. The final example of this article—analysis of a Buriat shamanic ritual—illustrates some of the ways in which symbols can take on new significance through everyday interaction. This case also underlines that interpersonally transmitted, oral tradition differs from socially decontextualized sources of information about the past (i.e., books and television broadcasts) both in terms of controlling how information is expressed and how meaning gets attributed to given practices.

Prelude: Collective Memory and Its Discontents

Memory studies crosscut all fields that examine the past and its use in the present, including history, anthropology, sociology, and cultural studies. I want to begin by getting my stance on this item out of the way: collective memory does not exist. Scholars will claim that when they say collective or cultural or social memory, what that really means is textual mediation (Wertsch and Karumidze 2009), or “representation,” or “commemoration,” or “narrative” (Allen and Bryan 2011; Connerton 1989; 2009; Roediger, Zaromb, and Butler 2009; Wertsch 2002).[1] However, all of these latter terms are social in nature and exist independently of individual memory. The argument runs like this: because representations are a real and legitimate thing to study, and we say that “collective memory” is the equivalent of representations, then collective memory is a real and legitimate thing to study. But it just does not wash, in part because a theoretical hedge in an article’s introduction does not tend to carry through in how scholars analyze their data. Examples of statements that have appeared in recent work include:

• “The American view of the 1941 attack on Pearl Harbor” (Wertsch and Karumidze 2009:382);
• “…I am interested in the work of social memory, the means through which a group reconstructs, assimilates, and understands its past, and its role in the formation of the group’s contemporary identity” (Cole 1998:610);
• “The notion of forgetting here is extended from something an individual might do in an everyday sort of way to something societies, or indeed civilizations, might do” (Connerton 2009:47);
• “…the memory of Guadalupe is carried in the culture of a people, both consciously and unconsciously, until it gradually becomes embedded in their values” (Rodriguez and Fortier 2007:5).

In each statement, the agent doing remembering or forgetting is a group: a culture, society, or an entire nation-state. In the final example, there is no agent at all, as Rodriguez and Fortier use the passive voice to claim that memory “is carried in the culture of a people.” “Cultural memory” implies that cultures remember; “social memory,” that societies do; and “collective memory” that an unspecified unitary actor does.[2] This is why I oppose the use of these terms, all of which I will refer to as “collective memory” in this essay. While a useful shorthand for some researchers, such a framework would obscure many of the more interesting processes of signification and resignification presented in my data. In order to talk about how interpretations of the past are framed in terms of current circumstances, it is necessary to pinpoint where representations come from and how they are propagated. This is especially true when studying the Soviet Union, where censorship interfered with so many of the data sources that count as “collective memory”: newspapers, history books, and even what informants would divulge in interviews.

The concept of collective memory comes from Maurice Halbwachs, who hews to the intellectual tradition of Durkheim. Durkheim, when speaking of the conscience collective, meant something more vague and metaphysical than those who currently use the terms cultural, social, and collective memory would likely stand behind.[3] But the conceptual leakage between Durkheimian essentialism and social analysis is demonstrated by the excerpts above.

Acknowledging legitimate criticisms of Halbwachs’ group-mind “strong version” of collective memory, James Wertsch incorporated social context into his approach, noting that “cultural tools” such as books, newspapers, and the Internet mediate the relationship between personal and collective memory (2002:13). However, he also defines collective memory as “a representation of the past shared by members of a group such as a generation or nation-state” and gives the example, “Americans can be said to have a collective memory of the [Vietnam] war” (Wertsch 2008:120; 2002:24). While members of nation-states may have had similar experiences, they remember individually, not in groups. And they also do not share representations of the past, exactly. If, for example, everyone of a certain age in the former Soviet Union received identical instruction in school about World War II, they are likely to restate similar narratives because they share information about the past. The representations of the past that they reiterate, however, came from a specific, highly controlled source: the Soviet state. If they then tell their grandchildren these narratives, the social location of the representations changes. Even if individuals repeated exactly what was printed in their 1950s textbooks, the motivations they would have in narrating this past would be quite different than those of the original authors. Their grandchildren would hear these stories in a new social context, one in which they have access to more textual sources of information about World War II and far fewer living ones. One of the reasons it is so important to distinguish between memories and representations of the past is because calling something like Pravda newspaper accounts “collective memory” masks the sources of historical data, the political and economic machinery driving dissemination, and the rationales for silencing dissent.

A special issue of the journal Memory Studies illustrates the confusion caused when scholars apply the psychological term “memory” to complex social processes. In fact, the issue did not deal with memory at all but, instead, focused on “mediated commemoration” of the July 7, 2006, terrorist attacks in London. The editors wrote:

As media technologies continue to evolve and to pervade our daily lives at an astonishing rate, the profound interconnections between media and memory… are becoming ever more entwined. Thus, whilst the remembrance of contemporary events such as 7/7 entails an enormous range of messages, intents and desires, it also involves a diverse range of media and mediations including texts, images, objects, artefacts, bodies and interactions. (Allen and Bryan 2011:264)

Therefore, those in the field of memory studies are already speaking in terms of signs—of messages created, perceived, circulated, and redefined, usually through mass media. Semiotic and discourse-analytic techniques offer a theoretical framework that deals with these issues openly rather than sweeping them all under the “memory” carpet. Memory is, admittedly, a more svelte term than “representations of the past.” But if research is actually about representations, we should be clear about it so that images of the past can be honestly investigated. Granted, scholars for the most part do accomplish what they set out to do with “collective memory.”[4] But this language obscures some of the processes involved in creating historical knowledge.

I will detail the reasons why a seemingly trivial terminological choice has political implications at the end of the paper. But at the outset I wanted to be clear about why I am not using this framework and to ask that the reader keep the distinction between social and psychological in mind as I go through the following examples.

Semiotics: event, narrative, circulation

Vitaly Komar and Aleksander Melamid were classically trained Soviet artists who emigrated to New York in 1978 (Ratcliff, Komar, and Melamid 1989:14). In one of their more popular series, the team made visual puns of the socialist realist style. The first image below was painted by Vladimir Serov in the 1940s to commemorate Vladimir Lenin’s announcement of the beginning of the 1917 Revolution (Ratcliff et al. 1989:129). With only a few changes in the painting’s figurative content, Komar and Melamid parodied this representation of Lenin. They portrayed the leader with the frozen marble features of a statue, reminding viewers both of his ubiquitous presence on town squares and his long absence as a living being.

In these pictures, Komar and Melamid took a familiar, authoritative, and highly serious style and undermined it by embedding elements of the ridiculous. Paintings in the Ancestral portraits series (Figure 3) were done in the dark, dramatic style of socialist realist portraiture, but the disjuncture between genre and subject matter triggers chuckles. The same is true for Bolsheviks returning home after a demonstration, with the added insinuation of acid trip (Figure 4).

Explaining how Komar and Melamid built these puns requires Peirce’s three-pronged model of the sign rather than Saussure’s simpler version. Saussure’s sign had two components: signifier and signified (Saussure 1959). The signifier “dog,” for example, can refer to the signified “furry canine pet” to English speakers. Peirce’s sign model, on the other hand, has three elements: sign vehicle, represented object, and interpretant (Parmentier 1994). The sign vehicle is anything one perceives as meaningful through the human faculties of perception: sight, sound, touch, taste, and smell. The moon, for example, can function as a sign vehicle. Its represented object can be simply “moon,” or “half moon,” or “waxing gibbous moon.” This is where Saussure’s model stops, with a stable signifier (sign vehicle) and an arbitrarily determined signified (represented object). Peirce, however, added the idea of the interpretant, which is like a mental screen. It is the framework through which one chooses to read a sign vehicle or a set of sign vehicles. The image of the moon, or even the collocation of letters “moon,” may refer to the same signified object in the minds of all readers of this essay. But the interpretant, the film of personal significance, allows us to examine what that object means in the context of its occurrence. Even if two individuals agree that a sign vehicle refers to a given object, their interpretations of what that object means vary according to their individual thoughts and biographies. Hence, to one person “full moon” may mean that it is time to plant, to another that a feast day approaches, and to me, for instance, very little. Changes in the interpretant are what allow resignification to occur. Meaning, ultimately, is in the mind. But cooperatively constructed aspects of meaning come from shifting semiotic ground. Semiotic ground—or mutually presupposed meaning—is the overlapping of interpretants.[5] Misunderstandings occur due to unavoidably imperfect interpersonal communication, which indicates that “meaning” is private while talk about meaning (or talk leading to meaning) is what is shared.

In the same way, images of historical figures serve as sign vehicles whose ultimate contextual meaning varies. A Stalin sign vehicle might refer to the same represented object in the minds of most Russians—“Stalin” as a biographical entity—but the analogic associations it conjures are determined by the interpretant (see Figure 6). A Peircean semiotic approach delineates public symbols from individual meaning making while still providing a way to join the two.

In their art, Komar and Melamid habitually exploit the gap between represented objects (Stalin, Lenin, Soviet symbols) and interpretants. The signified object in both a serious socialist realist depiction of Lenin and a parodic one are identical. But the artists manipulate the interpretant via other visual cues. The mechanism of Komar and Melamid’s visual parodies (like that for all puns) is the interpretant.

Paul Kockelman, however, has adopted another, incompatible understanding of the interpretant, arguing that it manifests primarily in nonmental phenomena (Kockelman 2010:3). Interpretants, according to this reading, can include (1) other utterances, (2) changes in attention, and (3) physical responses (such as ducking a punch in the mouth) (Kockelman 2007:378; 2010:2, 7). This may be more faithful to Peirce’s writings, but it is not as analytically useful to think of the interpretant as a “reaction” instead of as a mental screen.[6] There are too many possible physical, emotional, and discursive effects for this to serve a clarifying function. Further, a nonmental reaction, like a shift in eye gaze made in response to an utterance, necessarily comes long after the absorption of the object’s meaning.

Imagine, for example, that I show two people my pet spider Mittens. Subject One holds out her hand to play with one of God’s fellow creatures. Subject Two, however, knows that Mittens is a black widow and screams in horror. The sign vehicle was the same in both cases. This sign vehicle, Mittens, does not cause the two diverse reactions. What the reactions do manifest is the represented objects (cute spider vs. black widow), although in the reverse direction of causality than the one outlined by Kockelman. Apprehending the represented object, via the interpretant-as-screen, is what leads to the reactions. Remember, the entire semiotic triangle is the “sign,” with meaning being the ultimate product of the interaction of all three legs. Someone cannot scream (have a reaction) until after they have understood a given sign vehicle to correspond to a particular represented object. The interpretant is what allows that understanding to occur. Since the interpretant is integral to the sign itself, it cannot occur temporally after meaning has been attributed to the sign vehicle. Thus, a nonmental reaction to a sign, such as a purr or a punch in the mouth (Kockelman 2007:378; 2010:2), falls outside of the semiotic triangle. This indicates, then, that there is not an obvious correspondence between the sign vehicle and the interpretant if we view it as reaction. Instead, the reaction is another sign, a response indexically linked to the first. When Kockelman’s understanding of the interpretant is applied, it ends up reproducing a binary Saussurean framework (see Figure 7).

Peirce really liked triads. So just as his signs had three components, he divided types of signs into three categories: icons, indexes, and symbols. An icon is a sign that resembles its represented object. The painting of Lenin in Figure 1 is an example of an icon. An index is a sign that stands in spatiotemporal continuity with what it represents. So if my roommate asks me where the salad spinner is and I point (perhaps with my index finger) to the sink, this action is an index. A symbol, in Peircean terms, is a sign that only refers to its represented object via tradition or convention (Parmentier 1994:6). All words are symbols, since the relationship between their aural or written form and their signified content is arbitrary (Saussure 1959). A given sign can fall into multiple categories. For example, the onomatopoeic words “hiss,” “thud,” and “whoosh” sound like what they represent, making them both icons and symbols.

I would like to use the distinction between symbols and indexes as metaphors to interrogate the political implications associated with different ways information about the past is transmitted—by those who rely on media sources (symbolic storage) and by those “who were there” and draw on personal memories (indexical storage). Jan Assman claims that, “Cultural memory preserves the store of knowledge from which a group derives an awareness of its unity and peculiarity,” and lists texts, rites, monuments, recitation, practice, and observance as the technologies through which it is maintained (Assmann 1995:130, 129). James V. Wertsch also, in part, defines collective memory as information storage. For instance, he describes Amazon.com as a “cultural tool” used to aid memory (Wertsch 2002:11). And it is that. But Amazon.com is not memory itself, and neither are other print media, monuments, or interview transcripts. Instead of collective memory, or even “shared representations,” what researchers like Wertsch and Assmann describe are different ways of collecting, accessing, and circulating information about the past. When such information is distributed solely in media sources—be those textbooks, TV, radio, newspapers, or the Internet—this is a kind of transfer that is spatially and temporally disconnected from the events described and from people who might have memories of them. This makes it symbolic storage. Oral history, in contrast, is indexical storage. With indexical storage, information about the past is transmitted from one person to another. As Peter Osborne (2010) argues, individual memory is politically powerful; it allows one to bear witness with the authority of real experience. Oral history is also hard to censor (Garey 2011). Scholars such as Serguei Oushakine (2009), Mariko Tamanoi (1998; 2009), and Catriona Kelly (2011) have all analyzed narrative recollections of the past in order not only to piece together past events but also to understand how people deal with contemporary challenges.

Unlike books, the carriers of personal memory have expiration dates. After a generation dies out, the only testaments to their accounts are archival, residing in texts, videos, and recordings (Osborne 2010). These are easy to destroy. The information on Amazon.com is all pixels, which can be erased or altered or blocked with a DNS attack. Similarly, printing presses are run by people who pay for them, either companies or governments. Historical, political, and social circumstances condition what information is available and, therefore, constrain the possibilities for describing the past.

Symbolic Storage: Antoshka

The Soviet cartoon Antoshka mentioned in the introduction, is an example of symbolic storage. Describing how the reference to Antoshka was deployed in a comedy skit requires both an appreciation for individual memory and an understanding of how cultural texts get circulated. Antoshka depicts aspects of Soviet life and ideology, and it was these that the sketch comedy performers parodied in an internationally broadcast competition.[7] They assumed that members of the audience would have personal memories of the cartoon and, therefore, would understand the joke they constructed.

In the same performance that featured the “Antoshka” skit, one actor approached another and said, “At any rate, that’s our cow. And we milk it” (“Mezhdu prochim, eto nasha korova. I my ee doim”). That’s the punch line. To me, this statement was not immediately funny. This is because the setup for the joke existed independently of the performance space. It resided in the interdiscursive links between brief onstage utterances and a larger cultural imaginary (cf. Ricoeur 1991). Decoding this line’s humor requires knowing two things: (1) what prior texts the performance presupposes and (2) the metapragmatic framing that tells the audience how to interpret those texts (Silverstein 1993). I will describe the six skits in this performance in order, explaining how the team Astana.kz from Astana, Kazakhstan, used quotation to take stances on national identity, musical genres, and Sovietness. Like Komar and Melamid, the team’s humor relied on parody. Since Astana.kz spoofed what were once serious utterances, the performers used almost exclusively nonreferential techniques to signal their stances as either mocking or sincere. To put it differently, the team took advantage of a variety of performative signals to guide the audience to the correct interpretant of familiar Soviet signs. In this case, the signs were primarily melodies.

The performance analyzed here was an octofinal elimination round held in Sochi, Russia, in 2002. The entire five-minute act parodies the Russian show Old songs about the most important (Starye pesni o glavnom). Old songs began in 1995 as a New Year’s special which featured contemporary pop stars singing the best-loved hits of the 1950s. In subsequent years, the tradition continued with songs of the 1960s, 1970s, and 1980s (Oushakine 2007). Astana.kz used Old songs as the central conceit of their act, drawing on music from pre-Soviet, Soviet, and contemporary eras. They presented the songs in chronological order, which allowed them to offer critical commentary on the last 60 years of Kazakh history. All but one of the skits cited musical genres. The jokes operated, thus, in multiple interdiscursive frameworks. Macroparody in the form of Old songs structured the microparody of more specific jokes. Dense intertextual allusions make performances like these funny.

However, simply standing in front of a microphone and singing Red Army Choir standards would probably not get many laughs. In order for a prior text to function as parody, performers must metapragmatically mark it as such (cf. Lucy 1993:17). Michael Silverstein writes that, “Insofar as a text represents events, particularly events of language use, the text is explicitly a metapragmatic discourse about such events” (1993:35). In other words, any evaluative stance is metapragmatic because it provides commentary on quoted speech. However, Astana.kz did not ever directly refer to prior texts. Instead, they made their assessments through the manner in which they represented events. In the first skit, an ethnic Kazakh member of the team, in Kazakh national dress, stood in front of two microphones. He began by saying “One, two, three, one, two” (Raz, dva, tri, raz, dva) in Russian into the microphone on the right, presumably to test the volume. He then moved to the mic on the left and said in Kazakh, “One two, one two” (Bir eki, bir eki). Dissatisfied, he went back to the mic on the right and said in Russian, “Ah, be so kind. Make the Kazakh microphone a bit louder.”[8]

The ridiculousness of using one microphone to speak in Russian and another to speak in Kazakh is one level of this skit’s humor. But the primary joke comes from the melody of “bir eki,” which is repeated five times in the course of the skit. Rather than the words, which are not funny on their own, the joke comes from the fact that the melody mimics the one from the Antoshka cartoon. Here is an excerpt from the first verse:

That Astana.kz’s “bir eki” is a reference to Antoshka and not simply coincidental similarity is illustrated by two points. First, the pitch pattern and rhythm of “bir eki” matches that of Antoshka exactly. Second, the performer sang “bir eki” rather than producing it in counting intonation. Just as with English, in Kazakh one would expect short pauses between items in a series: bir, eki, ush, tort. Instead, “bir eki” is governed by one cohesive intonational (musical) phrase.

Why did Astana.kz count in Kazakh when most audience members were Russian? This language disjuncture, in fact, is the basis for the skit. The fact that the audience would not understand the language lent itself to a joke that relied on music. This was meant to emphasize that despite linguistic, cultural, and even religious differences Russians and Kazakhstanis had a common Soviet heritage. Nearly everyone in the audience would have known the theme to Antoshka. Using Soviet-era references both ensured that people in a newly international consortium of post-Soviet states would get their jokes and contributed to the Olympics-like attitude of goodwill these performances encourage. In this act, Astana.kz put themselves in the role of friendly guests by saying, “We have come to see you” (“K vam priekhali siuda”). Similarly, in a 2008 performance the same team exclaimed, “Well, neighbors, let’s have some humor!” and addressed members of the audience as relatives (rodstvenniki).[9]

Not all of Astana.kz’s skits were ironic though. In the second number, the team added new lyrics to a song from the 1949 movie Cossacks of the Kuban (Kubanskie kazaki). Songs and staging from Cossacks of the Kuban were also used as the theme for the 1996 episode of Old songs about the most important (Oushakine 2007:454), and invoking this movie helped Astana.kz construct the superordinate framing of their act. Old, bombastic Soviet music would be easy to parody. Instead, this act came across as serious. This is because the team aligned the denotational text of their lyrics with their performative, interactional text. They used deictic markers such as “here” (suida) and “this happy game” (eta veselaia igra) to anchor themselves in the immediate space of the competition. The team called themselves the “Kazakhs of the Kuban,” thus punning on the slight phonological difference between “Kazakh” (kazakh) and “Cossack” (kazak).

In the third skit, the same performer who sang bir eki in the opening sketch began singing Kazakh music. Before he had finished the first stanza though, the other performers shouted him down. They began talking but were interrupted by the eager singer of Kazakh folk songs, who said, “Okay, okay. You don’t like Kazakh songs. How about I recall a Yakut song instead.”[10] He then proceeded to sing in the style known as throat singing, or harmonic singing, in which two notes are sung at once to produce a chord. It takes great skill. (Throat singing, however, is commonly associated with Tuva, in southern Siberia, not with Yakutia, which is in the north.) The folk performer got cut off again and, this time, physically pushed to the side to make room for the “talented guys.” These guys proceeded to sing part of the Backstreet Boys’ “Show Me the Meaning of Being Lonely”—badly.

One of the performers had begun the boy band skit by saying, “As they say, it’s better to see something one time instead of one hundred… Watch the show, How it was.”[11] The name of the show is a clear allusion to Old songs. Instead of wistfully replaying records hundreds of times however, Astana.kz said that the music to follow could only be tolerated once. They also nonreferentially demonstrated that their enactment of “Show Me the Meaning of Being Lonely” was mockery by exaggerating musical and kinesthetic ineptitude. “Show me your menu, I’m feeling lonely,” they sang, off-key and with ungainly dance moves. When the band asked the sound technician to play back what they had sung, he played the original Backstreet Boys recording instead of their own heavily accented version. In short, How it was was not how it was, at all.

This skit made several points about Soviet history and its post-Soviet glorification. First, there was a great deal of controversy surrounding the Old songs episode that featured music from Cossacks of the Kuban. The film presented a patently false image of collective farm bliss, drawing criticism as early as the Khrushchev’s “Thaw” of the 1960s (Oushakine 2007:454). Thus, the skit could imply that the Russian boy band’s version of the Backstreet Boys is as believable as the picture Old songs paints of the Stalin era. In addition, the skit plays on generational differences. The 1950s are a source of childhood nostalgia for some audience members, but this group of young students had been children only a few years before, in the late 1990s. The team thus commented ironically about the impossibility of being nostalgic for the recent past. It was perhaps also an indictment of the younger generation’s artistic impoverishment. Soviet baby boomers can claim postwar prosperity songs (even if untrue). Generation X gets the Backstreet Boys.

The next skit contained only 31 words, but this made it no less interdiscursively complex. Punning once again on the similarity between the words “Cossack” and “Kazakh,” in this number the leader of a group of Kazakhs went up to a Cossack sitting on a horse and said, “At any rate that’s our cow. And we milk it” (Example 3, lines 11–12). These lines are from a contemporary Russian detective show called Lethal force (Uboinaia sila). One level of this skit’s humor, therefore, comes simply from the phrase’s multiple interactional meanings (as with “bir eki”).

The second layer of this joke’s significance comes from the way the performers oriented toward the speech they quoted. In any scripted performance, the author—the creator of a given text—is separate from both the animator and the principal (Goffman 1981:144). In most plays though, the characters who are animating lines collude with the audience in the pretense that they are the principals of the texts—that is, that they personally align with the words they voice. But in parody, the performers set up a metapragmatic frame in which they explicitly disassociate themselves from primary responsibility for the utterance. This is understood to reside with those who previously voiced the texts. In example one, the imagined principals were the Soviet ideologues who wrote Antoshka. In example two, it was the Backstreet Boys. The principal for, “That’s our cow. And we milk it,” is a bit more complicated. While the phrase originally comes from Lethal force, it has become a popular, nonsensical catchphrase, which has even popped up in commercials. Even if Astana.kz did not affiliate with the utterance’s original meaning, they were nonetheless principals of the nonserious stance generated by the phrase’s many revoicings.

“That’s our cow” was the only skit that did not quote music. In this respect it deviated from the overall frame of Old songs but remained thematically consistent by commenting on Russian-Kazakh relations. It was funny because it inverted power dynamics and, like a funhouse mirror, distorted historical roles. Cossacks were portrayed as the cultured, civilized party in contrast to the Kazakhs who could not tell a horse from a cow. But Cossacks have a reputation for being drunk, rowdy soldiers. Kazakhs were nomads who bred, relied on, and revered horses, which the audience would well know. The team created a gap between presupposed knowledge and comedic representation. This skit, therefore, was an effective dig at colonial arrogance.

As in the boy band number, the final skit also used music from a Western band (The Offspring). But this time the team indicated that their revoicing was meant to be serious, not mocking. Just like in “Kazakhs of the Kuban,” they linked their words to the immediate performance space through verb tense, deictic markers, and self-referential descriptions of their activities. Nearly all of the lyrics described what they were doing at that very moment: “We will joke around all the same,” “we are beginning to play,” “we will continue on [in the tournament].” In this way, the team aligned themselves as principals of the words they animated. In both the final skit and “Kazakhs of the Kuban” the team members were “committed to the words they said” (Goffman 1981:144), whereas they quite markedly were not in the others.

Both of these examples of humor, visual and musical, show pop-cultural references displaced from their original contexts, then transplanted into ones that were inconsistent with the sociopolitical messages of their original instantiations. People do need to have some understanding of what symbols represent or there will be no basis for humor. But this knowledge does not have to come from personal memory. The comedians almost certainly did not have it, as they were too young to have substantive memories of the Soviet era. Yet they designed an elaborate skit that both mocked Soviet nostalgia and allowed the audience to indulge in it.

Astana.kz’s sketch furthered symbolic circulation of Soviet-era songs, and their act was subsequently broadcast on television and posted to YouTube. But some of the most enduring examples of symbolic storage involve few to no words at all. Aspects of the built environment also reference past events in a manner that does not require a person “who was there.” Paul Connerton calls monuments, memorials, and city architecture examples of social memory. As evidence for his position, Connerton cites landscape-naming practices among the Wamirans of Papua New Guinea: “…each stone, each tree, each dip in the ground has a name and a story…” (Connerton 2009:13). These stories must be taught though; they must be told and retold. Memory is not intrinsic to the place. And neither is it a property of artifacts. A monument spurs memory because its presence must be explained. Upon encountering an Egyptian pyramid, its sheer magnificence would cause me to wonder who built it, when, and why. So I would ask people who lived around it, consult written texts, and, as archaeologists do, decipher hieroglyphs and do chemical analysis. This is not memory, thousands of years later. This is attribution of meaning to a monument.

The same holds for other kinds of material culture: street signs, statues, bomb ruins, abandoned buildings. Streets hold no memories. What remains of past regimes—either in tradition or building codes—is information about the past. The fact that something tangible exists, like a zoning law or a massive stone structure, means that it will remind people of the historical conditions of its creation. But cities, streets, and institutions do not have agency.

This brings us to a simple test of collective memory’s methodological feasibility: Where is it? If it is encoded in the minds of individuals, that is not data that social scientists have access to. On the other hand, if collective memory is preserved in texts, interviews, songs, rituals, and concerts, then these are all representations of the past (cf. Briggs 1986). Wertsch writes, “Indeed, my point is that a coherent account of collective memory can be based on notions of knowledge of texts, a line of reasoning behind the notion of ‘textual communities’…” (2002:27). In this definition, having a set knowledge base about particular texts, such as the Bible, constitutes collective memory. But it is not legitimate to talk about collective memory of Jesus. For the same reasons, no aspect of the past can be spoken of in terms of collective memory because the only ways we can access that past are through its semiotic detritus. Collective memory confuses data with the framework used to examine them.

That said, Connerton (2009) persuasively argues that the advent of modern society has affected how we are able, cognitively, to remember. Changes in the structures of social life, the speed of mass media communication, and the destruction of built environments have all increased our reliance on “non-thingy” sources of information: computer chips, television images, reels of microfilm (Connerton 2009:99). Rather than chatting in courtyards, we now watch TV. Further, the time-space collapse facilitated by the Internet encourages undue focus on the continually represented present at the expense of even the most recent past (144). This fosters dependence on mass-mediated cosmologies rather than interactionally supported presuppositions. The former, being more centralized, are far easier to control. “Everything that is solid melts into information,” Connerton writes (2009:124). What he describes is a fundamental shift towards symbolically mediated knowledge of the past and away from indexical communication, or that which is grounded in social relations. In the next section, I illustrate how such indexical storage can be preserved, suppressed, and recovered.

Indexical Storage: Oral History

In the spring of 2003, I conducted ethnographic fieldwork in the Buriat village of Kuita, in rural Siberia, for three weeks. Very fortunately for me, a sociology student I had met, Lida, allowed me to stay with her parents in the village.[14] The couple’s son and two daughters all went to university in Irkutsk, but they often came home for weekends. Lida’s family taught me much themselves and also aggressively introduced me to many other villagers. One morning, as I was making a map of Kuita, Lida’s mother leaned over my shoulder and said, “Why are you drawing that? That’s not important. So is it no big deal that you didn’t interview anyone yesterday? You don’t have much time here, you know.”

Kuita is located in the Ust’-Ordynskii Buriatskii Autonomous Okrug, about 120 kilometers (75 miles) north of Irkutsk, which lies at the southwest tip of Lake Baikal in southern Siberia. Around 570 people lived in Kuita. Most of the residents were ethnic Buriats, but about one hundred ethnic Russians also lived there.

While in Kuita, I was able to observe repercussions of the state suppression of shamanic traditions. People there had not been allowed to openly practice their religion during the Soviet period. One elderly man recalled, “At the time of the Soviet Union it was completely—shamans were absolutely forbidden. But now, please, you can go [to a shaman] every day.” And the village did indeed have an active shaman. But the most visible shamanic practices did not require his participation. For example, when drinking vodka, people sacrificed the first shot to the stove “for God.” And before downing individual shots, it was necessary to dribble (kapat’) a bit on the table with the ring finger of the left hand. I never saw anyone in Kuita, whether Buriat or ethnic Russian, fail to kapat’’ when drinking alcohol. Some people also put a bit of food on the table before eating, but this was not a regular occurrence in the family I stayed with.

The amount of sacrifice required when passing an arshan (shrine; literally “spring”) varied. On the drive out to Kuita, I was told that one must always stop at a shrine, leave something of value (money, cigarettes, or vodka), and tie a piece of cloth to a tree limb at the site. On subsequent trips past an arshan, honking was sufficient acknowledgement of the sacred place. However, I heard stories of the bad luck that befell travelers who failed to honk.

These were simple, pervasive activities that occurred at every meal, on every car ride. But I also witnessed a few more elaborate rituals. It is customary, for example, to slaughter a sheep when a guest arrives. Since I was a guest, my host family killed one for dinner soon after my arrival. “This is how Buriats slaughter sheep,” my host father, Iurii, told me. He made a small slit in the middle of the sheep’s underbelly. Then he stuck his hand into the still-alive animal and felt around until his whole forearm was inside. In this manner, he ruptured the sheep’s aorta.

Lida, who was twenty-two at the time, asked me if I had a camera. I did. “Take a picture!” Iurii said. “Take a picture of how I do this with my hands.” He wanted me to document how to kill sheep “the Buriat way.” He cut deeper into the sheep’s belly and began transferring blood from the aorta and organs into a pot. Then he gave the gallbladder to Lida and instructed her to put it on the roof.

“On the roof?” she asked, looking at him as if he had told her to place it on her head.

“Which one? Why necessarily on the roof? ”

“It’s a custom,” he replied.

Why hadn’t Lida known about this custom? She had helped slaughter sheep in the past. Why was the gallbladder placement so puzzling?

I might have been watching a more “authentic” version of the slaughter than is usual. My host family, who were overall very concerned about what I would write in my “book” about their village, may have felt obligated to perform a more canonical ritual than they generally did, opting to include an element that even Lida, their daughter, had never seen. I later asked about the meaning of the gallbladder, but Iurii either did not know or did not feel like trying to explain it. “It’s just a custom,” he said.

Whatever the gallbladder meant to Iurii was not transferred to Lida. And it may have held entirely different significance for Iurii’s ancestors, even if he faithfully enacted the physical steps of the ritual. But these practices would not have survived at all if they had not been maintained (privately) throughout the Soviet period (Humphrey 1992:380; Fridman 2004). Individual memory is important here because it is very difficult to teach someone healing rituals from books.[15] Possible, yes; but much is lost. As anthropologist Eva Fridman notes, “Because shamanism is primarily an oral religion, so intimately bound up with locale and kinship, it is especially vulnerable to actual physical disruptions through ideological, political, or economic upheaval” (2004:98). Caroline Humphrey, who has conducted ethnographic fieldwork among both Buriats and Mongolians, likewise writes:

In the Stalinist period any ordinary reproduction of identifiably “traditional” social and cultural forms became impossible. It is difficult to convey the extent of the obliteration of Mongolian culture. Though travellers to Mongolia might still observe people living in tents and looking after herds, in reality everything that looked the same was changed because it now had a different place in social relations, in the organization of work and people’s values (1992:379–380).[16]

As opposed to just knowing history, traditions must be learned first-hand. At a time when practicing shamanism earned one a trip to the gulag, it was harder to pass on the kinds of skills that must be taught, transferred from one person to another. The breakup of social relationships was not just an ancillary effect of pursuing wealth and territory but a strategic move to permanently forestall future opposition. There were concrete agents—the Soviet government—who took concrete actions, such as killing, imprisoning, and exiling, that resulted in concrete effects on the ability of a group of people to represent their past (Fridman 2004:31). Not accounting for the location of “collective memory” obscures methods of destroying cultural heritage (and ways to neutralize the political impact of personal memory). Relocation makes those who were once close rely on symbolic transmission (letters, documents) instead of indexical transmission (word of mouth). Information passed by word of mouth is far harder to censor, because it takes much more manpower—and often firepower—to destroy social networks.

People in Kuita did have reason to view themselves as victims of the Soviet regime, and many probably experienced it that way at the time. Several villagers told me that they had been forbidden to practice their religion, and, while no one talked to me about the Stalinist purges, a monument in the center of town testified to the fact that Kuita had been touched by the Terror. But individuals were much more likely to complain about victimization they felt at the hands of capitalism. Most of these comments were framed in terms of the difficulty of economic survival in Kuita. Remarks such as, “Of course, things were better during the time of the Soviet Union. They paid money,” and “Capitalism came. Everything is stricter now. People don’t want to help each other out. Everything is for money,” typified popular sentiment in the village. The Soviet state had guaranteed its citizens education, employment, and medical care, all of which are bitterly missed now, especially in rural areas.

There were, then, dual discourses of tradition, ones of both recovery and loss. Despite talk about the revival of shamanism enabled by the end of the USSR, Soviet nostalgia dominated many conversations: “The Soviet time was, of course, better. Young people were stronger.” “Before, everything was cheaper.” “Before, we were a big country. Now everyone has their own nationalism.” “Before, people only drank a little. And now…” The reference point for the idealized “before” was “before the end of the Soviet Union.” Residents of Kuita appeared to feel wronged more by communism’s sudden collapse than by the fact that it had existed at all.

One Buriat woman chided the older generations for not responsibly handing down traditions themselves, instead of faulting either socialism or capitalism. While speaking to an ethnic Russian, she transitioned from talking about how Buriats had forgotten their traditions to arguing that “everyone has forgotten their own [heritage].” She said that before Russians too had known beautiful folk songs and dances. She also lamented that now children read detective novels instead of Pushkin. She repeated several times that everyone—Russians, Buriats, and Tatars—had forgotten “their own” (svoi).

Many facets of this account of my time in Kuita could be considered collective memory: the sheep slaughter ritual, villagers’ narratives about the past, the monument to Stalin’s victims. Rosy-tinted appraisals of life under the Soviets could be glossed as instances of cultural forgetting. Presented with the same data, collective memory adherents might argue that memories of shamanism were encoded in the rituals of sacrifice. They might say that these practices contained traces of shamanic belief even when those who enacted them had no conscious understanding of those beliefs (cf. Cole 1998; Shaw 2002). Connerton might see the sheep slaughter ritual as a conduit of social memory, and one with a fairly fixed meaning (cf. Connerton 1989:57, 70–71). This, however, would equate ritual performances with memory of pre-Soviet shamanism (cf. Hobsbawm and Ranger 1983). Claiming that memory (as distinct from information) is stored in ritual, discursive, or material form makes it immutable. Tradition is what it was, what it ever shall be. Like Geertz’s winks and twitches, context transforms signs (e.g., the gallbladder) (Geertz [1973] 2000:6). Geertz distinguished “cultural texts,” such as rituals, from culture itself. In his definition, culture is an “ensemble of texts” which are constantly reinterpreted in interaction (452). He advocated capturing the slippery, puzzling property of meaning that allows the same performance to signify different things with thick description. As a practical guide though, this amounts to ethnographic intuition. Silverstein and Urban elaborate on the concept using the language of entextualization. Where thick description stresses the interdependence of “text” and “context,” entextualization focuses on the process of embedding text in context: taking a piece of interaction, rendering it as text, and planting it in social life (Silverstein and Urban 1996:14).

This is what Astana.kz did with their parodies. A cultural text, the tune to Antoshka, was entextualized in a complex statement about the politics of Soviet nostalgia (cf. Nadkarni and Shevchenko 2004). This is also what was at work in the sheep slaughter ritual. Placing a gallbladder on the roof is a kind of cultural text, not culture itself.

Ultimately, a wink is a wink if it is perceived as one. The obvious differences between “real” and parodic performances—even when ritual script, costuming, and participant habitus remain constant—points out the importance of separating cultural texts from entextualization processes. As Asif Agha notes, “…acquaintance with norms also makes possible effective forms of tropic improvisation; and…talk of ‘sharedness’ generally misrecognizes the fractional congruence of models of behavior for the identity of models” (2007:340). Too much attention to the fact that people are doing the same thing can wind up confusing behavior, such as singing or enacting a ceremony, with a calcification of those acts as “the ritual.” Just as the grammar of a language exists only insofar as speakers use it, its edicts renewed utterance-by-utterance, social structure emerges through—and only through—practice (Sahlins 2000). Even people acting in concert, as happens in a church service, do not share interpretations of the past. And any heuristic advantage gained with a holistic approach to “collective memory of Stalinism,” for example, would be severely mitigated by the inaccuracies such overgeneralization would entail. It may be true, following Halbwachs (1980), that memories are based on social frames. But this does not mean that the social frame is based on memory. Framing is a discursive and semiotic product. One can narrate without recalling.

Counterintuitively, this suggests that acknowledging variation is essential to understanding continuity. This is very difficult to do with collective memory for three reasons: (1) because it does not allow for social location, (2) because resignification cannot occur if meaning is encoded in symbols themselves, and (3)because there is little space for individual agency.

Conclusions: Semiotics and the Politics of Censorship

Komar and Melamid’s paintings, a Kazakhstani comedy team, and a Buriat family’s feast illustrate different ways of transmitting knowledge about the past: mass-mediated (symbolic) and individually transmitted (indexical). While all three examples show how the meanings of symbols change, the Buriat ritual demonstrates how this happens (and fails to happen) in a person-to-person, direct generation-to-generation interaction. The different communication practices employed entail different modes of representation and erasure (Irvine and Gal 2000). Censoring the former type requires, primarily, the destruction of records or the pulling down of statues. Eradicating the latter generally involves violence, either wiping out those who remember or severing their ties to subsequent generations.

Control, or agency, is a theme running across all the examples presented here. Alessandro Duranti defines agency as “the causal relationship between participants’ actions and certain states of affairs or processes” (1990:646). Building on this more general model, Paul Kockelman identifies two main types of agency—residential and representational. “Residential agency,” he writes, “is the degree to which one can control the expression of a sign, compose a sign-object relation, and commit to an interpretant of this sign-object relation.” Representational agency is “the degree to which one can thematize a process, characterize a feature of this theme, and reason with this theme-character relation” (Kockelman 2007:375). Much simplified, this means that residential agency has to do with whether and how information is expressed, and representational agency concerns how meaning gets attributed to a given representation. The KVN team members, for example, had a great deal of control over how their references to Sovietness would be interpreted (representational agency). They could rest securely in the knowledge that audience members would recognize the tune to “Antoshka” and understand it as parody. However, they had extremely limited residential agency. They managed their actions in the performance space but they did not run the television channel that rebroadcast it. Nor did they have a say in editing the tape, or in posting the show to YouTube, or in deciding who watched it on YouTube and forwarded it to friends. They almost certainly did not anticipate that their skit would become embroiled in an academic debate on the nature of historical knowledge.

Kockelman’s concept of agency stresses the notion of degrees of control over representations, then ties these to two separate aspects of meaning making: circulation and understanding. Censorship is generally thought to involve only residential agency (governments block information). But tactical resignification, which requires representational agency, has long been part of colonial, missionary, and state-building enterprises. Pagan festival dates in Russia and elsewhere were appropriated for Christian holidays, for example. Stalin razed Moscow’s Cathedral of Christ the Savior in 1931 to build the Palace of the Soviets. And in 2000 the Cathedral was rebuilt. These are also forms of historical erasure. As noted above however, getting rid of personally transmitted knowledge is harder than denying publication rights or even demolishing monuments. Symbolic storage is the easiest to censor precisely because there is the least individual agency associated with it. People rarely meet the authors of the books they read, and it is difficult to get accurate stories about monuments once the generations that built them have died out.

A semiotic approach, thus, brings to the forefront two foundational elements of any history, focusing on: (1) what representations can be made and (2) what they mean. It also provides a vocabulary for illustrating how people orient toward and negotiate the meanings of given historical sign vehicles (Stalin, Antoshka, a sheep’s gallbladder). The same sign vehicles can refer to different semiotic objects, and interpretations necessarily vary from person to person. As a result of these inconsistencies, consensus about how symbols should be read is continually reaffirmed, revised, and adjusted; presuppositional variations drive cultural change. Tradition is, and must be, maintained on an interaction-by-interaction basis. As Sherry Ortner said, history is not “something that arrives, like a ship, from outside the society in question” (1984:143). People make it, one word at a time. Further, data from the Buriat family illustrate how individual memory factors in the maintenance of tradition. History is practice, continually renewed.

Linguistic anthropologists, along with other social scientists, have honed techniques to show how discourses of the past figure in the present (Basso 1979, 1996; Briggs 1986; Irvine and Gal 2000; Shoaps 1999; Silverstein 1993; 2003; 2005; Silverstein and Urban 1996). That said, neither the work of these scholars nor Peirce’s triangle can help us understand “how it was” any better than a skillfully done history. But semiotic frameworks are well suited to elaborating on “what it means.” They allow us to trace the same symbols across multiple temporal frames, media formats, and interpersonal contexts. Semiotic analysis parses what the past is, now; what it isn’t; and why people with the same experiences have, nevertheless, different histories.