Discourse Analysis of the Historically Audible: A Cultural-Historical Approach to Sound Recordings from Colonial Contexts

Kalibani, Mèhèza

Discourse Analysis of the Historically Audible: A Cultural-Historical Approach to Sound Recordings from Colonial Contexts
^{(Version 1.0, 13/12/2024)}

Author: Mèhèza Kalibani
ORCID iD Mèhèza Kalibani: 0000-0002-2352-8524
Institution (Mèhèza Kalibani): Stiftung Historische Museen Hamburg
Published: 13/12/2024
Received: 11/07/2022, Accepted: 02/08/2024, Last updated: 13/12/2024
How to cite: Mèhèza Kalibani, Discourse Analysis of the Historically Audible: A Cultural-Historical Approach to Sound Recordings from Colonial Contexts, ‘Listening: Cultural Histories and Comparative Phenomenologies,’ special issue, Musicologica Austriaca: Journal for Austrian Music Studies (December 13, 2024)
Tags: Colonial contexts; Colonial knowledge production; Discourse analysis; Early sound recordings; Ethnomusicology; Phonogram archive; Sound archives

Abstract

In 1877, the invention of the phonograph enabled a new hearing practice that created a bridge between spaces and times. For the first time in human history, it was possible to record sound and replay it independently of its original source. The phonograph was used worldwide, including in regions where people were considered “primitive” according to the Western ideologies of the time. Today, these early recordings of non-European musical traditions are stored in European archives. From early on, they have been studied by scholars or used for cultural projects in museums. A look at the collections of many historical sound archives clearly shows that the colonial era (for Germany especially from 1900 to 1914) was the golden age of this collecting practice. Thus, these sound recordings embody a certain “sensibility” due to their relation to colonialism. In fact, a considerable part of the recordings was produced under hegemonic power relations. But how are they heard today, what stories and discourses do they transmit, and how do we deal with them? Close listening, listening to history, collective listening, and listening to the silences are some of the theoretical-methodological approaches developed in recent years in the context of dealing with historical sound recordings. In this article, I will introduce these approaches and highlight their advantages and gaps. In order to bridge the gaps thus identified, I will introduce “discourse analysis of the historically audible,” a cultural-historical approach to historically sensitive sound recordings. Discourse analysis of the historically audible is a cultural-historical approach to sound recordings from colonial contexts which can facilitate the past-present dialogue between former colonized people and former colonizers.

back to index

Paasche Ostafrika 6 is the title of a sound recording held at the Berlin Phonogramm-Archiv (BPhA) and stored as Paasche Ostafrika 6 in its inventories.^[1] In this recording, one can acoustically discern a male voice singing a song that is, according to the documentation, in the Nyamwezi language, which is spoken today in Tanzania.^[2] The Berlin Phonogramm-Archiv currently holds more than 16,000 wax cylinder recordings similar to Paasche Ostafrika 6, made between 1899 and 1954. This archive was founded in 1905 by the German psychologist Carl Stumpf (1848–1936) and was then mainly concerned with the collection and study of so-called traditional music of the world.^[3] At that time, other similar archives had been established in Europe, for example in Vienna, Paris, and Budapest. The Vienna Phonogrammarchiv of the Austrian Academy of Sciences (PhA), which was the world’s first sound archive (founded in 1899), also holds thousands of early sound recordings. In response to the question “What do we hear through these sound recordings today?” one might tend to say focusing on the acoustically audible. But focusing on the acoustically audible itself happens to be criticized even in sound studies, as Hein Schoer, Bernd Brabec de Mori, and Matthias Lewy do in their article “The Sounding Museum.”^[4]

Apart from the audible sound itself, it is possible to listen to other (hi)stories through historical sound recordings. These other (hi)stories are not necessarily perceptible by the ear. In this case, listening also means considering the documentation of these sound recordings, contextualizing them in their cultural and historical origins, or reflecting on their usability today. For instance, the recording Paasche Ostafrika 6 has more to tell than just the Nyamwezi song it offers to our ears. Paasche Ostafrika 6 was recorded with a phonograph in 1906 by the German colonial officer Hans Paasche (1881–1920) in the German East Africa colony. Paasche is known for his involvement in the brutal repression of the Maji-Maji Rebellion, which rose against the colonial administration in 1905–1907.^[5] Because of his role as a colonial officer and producer of sound recordings, and considering the period in which the recording was made, Paasche Ostafrika 6 belongs to sound recordings from colonial contexts. Like Paasche Ostafrika 6, the vast majority of the holdings of the BPhA and the PhA—two of the most important historical sound archives in Europe today—were recorded in similar contexts, in European colonies during the era of colonialism. For the BPhA in particular, a major part of its holdings were recorded by German colonial officers between 1900 and 1914, in what were then formal German colonies.^[6] Some of the recordings were already presented to museum visitors for listening during this period.^[7] Even today, historical sound recordings from former German colonies are used in exhibitions or art projects. However, scholarly research has questioned the authenticity of such recordings, portraying them as the legacy of a violent colonial history.^[8] Therefore, when working with such recordings today, we have to ask the following questions: How do we or should we deal with these recordings? Where and for what purpose could they be used? What (hi)stories can we learn from them, and what methodological approach could help us to identify these (hi)stories?

This article offers a methodological approach on how to deal with these historical sound recordings today. I will first elaborate on what I consider to belong to the category of sound recordings from colonial contexts within the confines of my approach. Then I will review three of the several existing methodological approaches, highlighting their gaps. Finally, I will introduce the “discourse analysis of the historically audible” (DAHA), which fills some of the gaps encountered in the three prior approaches. DAHA is a five-step model explicitly developed for the study of historical sound recordings, especially those that issued from colonial contexts. Through its different steps (finding the topic, building the corpus, context analysis, deconstruction of the audible, and discourse analysis of the audible), it offers a scheme on how to deal with these recordings in terms of the content that can be analyzed as well as the methodology that can be adopted for the analysis. It considers different aspects of the recording, from the context of its production to the contexts of its use in the past and present, including the perspectives of the source communities as well as the cultural aspects of the recording.

1. What Does “Historical Sound Recordings from Colonial Contexts” Mean?

In its guidelines for German museums on care of collections from colonial contexts, the German Museums Association defines the term colonial contexts as follows:

Colonial context … is initially regarded as circumstances and processes that have their roots either in formal colonial rule or in colonial structures outside formal colonial rule. At such times, structures of great political power imbalance may have arisen both between and within states or other political entities. This created networks and practices that also supported the collection and procurement practices of European museums … Colonial contexts, however, also led to the emergence of objects and depictions which reflected colonial thinking. Common to colonial contexts is an ideology of cultural superiority to colonised or ethnic minority populations … and the right to oppress and exploit. This also raises doubts about the legality of its use to justify acquiring collections.^[9]

According to this definition, the main characteristics of colonial contexts are unequal power relations and a self-perception of cultural superiority by those in power. These characteristics have accordingly influenced the procurement of objects from non-European regions for European museums.^[10] In December 2019, cultural studies researcher Annette Hoffmann obtained a similar perspective by stressing that historical sound recordings are important alternative sources for colonial history in an interview on her exhibition War and Grammar.^[11] She referred to sound recordings made in German prisoner-of-war camps (POW), consisting of a total of 1,022 wax cylinder recordings archived at the BPhA and 1,651 shellac recordings at the Lautarchiv of the Humboldt University in Berlin (LA).^[12] These recordings were made between 1915 and 1918 by the Royal Prussian Phonographic Commission, which was founded in 1906 and devoted mostly to recording prisoners of war in German POW camps.^[13]

In contrast to the definition by the German Museums Association and the perspective of Annette Hoffmann, in this article I define sound recordings from colonial contexts not just on the basis of the power relations between recorders and recorded people but also on the basis of the period and location of the recording. Therefore, sound recordings from German POW camps are not a priori included in this category for the propose of this article. I only consider sound recordings made by Western scholars, colonial officers, and other actors in occupied territories as coming directly from colonial contexts. This does not cast doubt on the fact that both kinds of collections (from POW camps and from colonies) were made in similarly asymmetric power relations. Indeed, in both cases, Western “collectors” (meant here as producers of the recordings, i.e., scientists, ethnologists, musicologists, missionaries, colonial officers, etc.), supported by colonial power, determined the rules and recorded what they wanted to record; the oppressed POWs and colonized people had to follow these rules and make their voices available.

Therefore, sound recordings from German POW camps rightly belong to the category of “sensitive collections” as defined by cultural studies researcher Britta Lange. The POWs were recorded as “subaltern” within hegemonic power relations, without being given the opportunity to say whether they agreed or not.^[14] For this reason, the recordings from German POW camps require sensitive care today. Nevertheless, I do not explicitly consider them in this article as coming from colonial contexts. On the one hand, some people whose voices were recorded in German POW camps were Africans or Indians from French and British colonies, drawn into the First World War through colonial relations. On the other hand, the same collections also include voices of POWs from Europe, among them Catalans and Russians who did not have a “colonial origin.”^[15] It would thus be imprudent to consider all recordings from German POW camps a result of colonialism, and the Lautarchiv itself a colonial archive, as Annette Hoffmann does.^[16] Considering these sound recordings and the Lautarchiv itself as “colonial,” without taking into account the various origins of the actors whose voices were recorded, somehow creates the impression that the recorded people were captured and brought into POW camps for the purpose of recording. Of course, the reverse makes more sense: their voices were recorded because they were kept captive there. German ethnomusicologists of the time thus found another occasion in their attempt to record everything they could.

Despite the different situation of making recordings, the motivation for recording in POW camps and colonies is similar, and both can be understood as contributions to knowledge production about the “others.” However, we should avoid declaring the very indirect sub-context of colonialism in the case of the recordings in POW camps as the main context of their recording practice.

Last but not least, the limitation of the timeframe for defining historical sound recordings from colonial contexts carries significant importance. Colonialism’s formal impact began before the nineteenth-century territorial occupations, and for some it has not ended yet, as its effects persist to this day. Examining sound recordings from colonial contexts not only provides insights into object histories but also delves into histories of violence, in the same way looted material objects from European colonies do.

In Germany, for instance, ethnological museums thrived between the 1880s and 1920s, especially from 1900 to 1914,^[17] in the same way the Berlin Phonogramm-Archiv did. The practice of collecting from non-European areas was facilitated by the colonial occupation, and acquiring these objects often involved violence. Consequently, the collection of historical sound recordings drastically diminished in most European colonies after the formal colonial occupation ended, as seen with the BPhA. The collection history of sound recordings is intricately linked with the history of violence during colonialism. To preserve a clear understanding of the violence associated with the ideology of colonial rule and occupation, which constituted an integral part of the collection history of sound recordings in occupied territories, it is essential to maintain a specific focus on sound recordings made in these conditions when discussing sound recordings from colonial contexts. Expanding the concept of colonial context to include sound recordings from prisoner-of-war camps, where even Europeans were recorded, risks downplaying the recognition of the particular violence that accompanied the act of collecting/recording in the colonies.

In the same way, Albert Gouaffo, a Cameroonian Germanist and cultural studies researcher, argues that such general perceptions, particularly when it comes to artifacts from ethnological museums, tend to diminish the intensity of colonial violence in the occupied colonies within the public discourse. Moreover, these perceptions avoid prioritizing the objects that originated directly from the colonial occupation when it comes to handling the cultural heritage from non-European territories in European museums.^[18]

By prioritizing the recordings that originated from the direct colonial rule and the colonies, I acknowledge the particularity of the oppression and exploitation that occurred in these territories during the recording situation. In this respect, I define “historical sound recordings from colonial contexts” as sound recordings made between 1877 (invention of the phonograph) and 1960 (official end of colonization in most of the occupied territories) by Western actors in occupied colonial territories and held today by Western sound archives. This definition embraces thousands of recordings, including voices with songs, proverbs, linguistic-related recordings, music, and so on presented by colonized people. In the following, I discuss the question of how we can deal with these sound recordings today.

2. Some Methodological Approaches to Historical Sound Recordings

2.1 Specificities and Challenges Surrounding Historical Sound Recordings from Colonial Contexts

What does scholarly analysis of sound recordings from colonial contexts entail? What questions can they answer today? What procedure can we use to answer these questions? What specificities do they embody as historical sources? Are they authentic cultural elements from the past or just constructions? How do the (hi)stories and discourses contained in these sources become perceptible?

These questions have not yet been answered in a satisfactory way, because historical sound recordings have long been neglected in the study of (colonial) history.^[19] Nevertheless, they are very important for understanding history, as proposed by music historians Daniel Morat and Thomas Blanck in their article “Geschichte Hören” (Hearing history).^[20] Morat and Blanck argue that historical sound recordings can “‘bring history to life’ and make it possible to experience it in ways other than texts and images.”^[21] The Cameroonian political scientist Achille Mbembe has listed three different universes of the archive in terms of perceptivity: the tactile universe, the visual universe, and the cognitive universe.^[22] As archival materials, sound recordings thus represent the acoustic universe of the archive. In comparison to written documents, the type of source used most often in historical scholarship, sound recordings give the impression of making history present in a much more immersive sense. But as exciting as they may be, these sources must be used with reservations. Morat and Blanck rightly point out that recordings from the past do not always represent what they are thought to be: “they are—just like written or pictorial sources—specific excerpts from the past that were recorded under very specific conditions, with specific techniques and intentions.”^[23]

One main problem with historical sound recordings from colonial contexts is their documentation: most often, written sources, like recorders’ notebooks, do not reveal enough to allow clear comprehension of the conditions and intentions of their production. Accordingly, attempts to reconstruct the original recording situation usually do not lead to satisfying results.^[24] And in most cases, it is also difficult to assume whether these recordings are authentic sources of culture or just constructed components from the past. Do they represent what they pretend to be according to their producers? Was their content created under natural cultural circumstances or just for the recording? Did the recorded people participate freely in the recording process or were they forced to do so? These questions should be considered when it comes to defining the authenticity or construction of the content of the recordings.

In any case, historical sound recordings from colonial contexts remain important sources from the past and can be useful in deconstructing many aspects of history and the cultural past if appropriate methodological approaches are applied in studying them.

2.2 Listening to Archival Silences, Close Listening, and Listening to History: Three Methodological Approaches and their Gaps

In the following, I will introduce three methodological/theoretical approaches to the study of historical sound recordings.

The first approach was developed by Britta Lange, who considers historical sound recordings from German POW camps to be sources of colonial history. Lange extends the concept of “sensitive collections” to historical sound recordings.^[25] In her article “Archival Silences as Historical Sources,”^[26] she argues that historical sound archives withhold a lot of information. However, this kind of information in particular needs to be studied in order to understand the archived recordings in their full meaning. Lange understands silence not only as a discursive but also as an acoustically tangible practice:

Silence is not only discursively named, but also practised as real and audible or inaudible—in sound, in the audio medium. Hence, the parable about silence becomes to some extent a realisation of the moral message, both in the spoken content and on the level of the acoustic. The story about silence figures as a semantically interpretable and acoustically audible parable—and perhaps also as a moral tale for historians: to understand the listening-to-silence of the sources as a possible political act and not to suspend this through overwriting.^[27]

Lange determines five different forms of archival silences that can be analyzed in order to understand historical sound recordings. The first one is structural silence. The author starts from the premise that historical sound collections are incomplete, arguing that the largest part of the past, such as voices, biographies, and events, has not been captured in sound and that the sound archive remains silent about them.^[28] The second form is technical silence, which consists in the fact that the sound archive is predominantly mute at all times, although the peculiarity of the sound archive itself consists in sounding. The recordings always need to be activated by a person in order to start sounding.^[29] The third form of silence is content-related silence. According to Lange, in the context of the recording situation of historical sound recordings with asymmetrical power relations (here from POW camps), many metaphors were used by people whose voices had been recorded because many words were not allowed to be said due to censorship. Here the silence is not absence of sound but is transformed or replaced by codes that convey hidden meaning. Study of historical sound recordings focusing on the silence by paying attention to words not explicitly said should allow the researcher to decipher the metaphors used in the recordings.^[30] The fourth form of archival silence is the silence of presentation. Historical sound recordings do not expose their recording situation. Lange recommends not to regard any unusual attitude in the presentation by the recorded person as a pure coincidence. Focusing on the unusual (disturbances, questions, coughs, laughter, sighs, incidental remarks, or afterthoughts) helps to understand the recording situation though the performative silence of the speaker.^[31] The last form of archival silence discerned by Lange is political silence. Similarly to content-related silence, it helps to discover the unsaid regarding the recording contexts. Deciphering this unsaid involves doing explicit research on what was authorized and what was forbidden by the political power in the context of the recording situation. One should ask which topics were taboo, what was censored, and how this censorship was practically applied.^[32]

It is Britta Lange’s merit to have identified that the recordings—which are supposed to sound by nature—carry many unsounding aspects and can be better understood if these stories are identified. According to Lange’s concept of listening to the archival silences, the main task of the researcher dealing with these recordings is basically to make the silences audible. However, this task remains difficult. Lange does not offer a step-by-step methodology for achieving this goal. For example, it is not clear how exactly one can identify these different forms of silence and which guiding questions researchers must ask in this regard.

Instead of listening to silence, Annette Hoffmann suggests in her “close listening” approach listening to everything one can hear from the recording, to as many audible features as possible—including noises in the background, the noise of a rotating cylinder or scratched record, or the recordist’s announcement.^[33] Accordingly, the sound recording itself takes the form of a document whose paratext should not be ignored. According to Hoffmann, close listening, by virtue of its principle of listening to the recording entirely, can be helpful for understanding the recording situation. In this way, conclusions can be drawn about the recording situation.^[34]

Similarly to Lange’s listening to archival silences, Hoffmann’s proposed close listening approach does not specify a scheme by which the sound recordings should be heard. Close listening does not offer a step-by-step method of interpretation for the individual, non-explicit elements that can be heard while one is listening to all one can hear. In this respect, the approach does not necessarily allow one to guess the recording situation, the history of archiving, and the authenticity of the content from the recordings.

The sound historians Daniel Morat and Thomas Blanck have also worked, and even more deeply, on the question of how to conduct scholarly analysis of historical recordings. They offer a model of analysis for sound documents that can be applied to scholarly research on historical sound recordings as well as to teaching in the field of history. The model consists of five steps for the critical study of historical sound recordings. The first step is to examine the external structure of the recording. In this step, one should clarify what, exactly, is known about the recording and what kind of recording it is: location of the recording, content, media form, purpose of the recording, situation of the recording, and so on.^[35] The external structure is followed by analysis of the scenography. This step examines how a historical recording is meant to sound according to the intention of its recordist.^[36] As in the approach of close listening, one should focus in this step on the audible itself, considering aspects of speaking that are deliberately used as rhetorical features, such as pauses, changes in speaking, speed or pitch, and so on.^[37] After the analysis of the staging, listening to history analyzes the performance. It aims to clarify how the recording sounds and which elements are different from the script of the scenography.^[38] The next step is the analysis of mediality. The point here is to treat the sound recording not only as sound but also as a medium. According to Morat and Blanck, the storage medium can let only certain parts of the sound event though and can additionally modify them or even add its own sounds to the deliberately recorded ones (such as the hissing, crackling, and popping of the shellac recorder).^[39] Morat and Blanck assume that the historically recorded sound has undergone a transformation and is thus not identical to the sound that has survived until today. In this step of the analysis of mediality, one should specifically look for filter processes to identify and reveal them.^[40] The last step of the listening to history approach is called context analysis. Here, the sound recordings should be put in their historical context and compared with other sound documents. The aim of context analysis is to encircle and break down “the sound via its relationships to texts, images, bodies, apparatuses” (“den Klang [einzukreisen und aufzuschlüsseln] über Beziehungen zu Texten, Bildern, Körpern, Apparaten”).^[41]

Morat and Blanck’s approach, whose purpose is to make history audible, is more comprehensive than the approaches suggested by Lange and Hoffmann. The approach is structurally well designed and offers concrete guiding questions that support the analysis of some aspects. Fundamentally, what is missing in this approach is primarily the contextualization of the usefulness of historical recordings outside of research and teaching. In their analysis of performance, Morat and Blanck also suggest asking if the written script the recording sequences follow is different from the content of the sound recording, checking if the audible content sounds different from the script, and then drawing a conclusion about the authenticity of the content to determine whether the content of the recording truly represents what the author of the recording claims it does. What Morat and Blanck do not take into consideration is that this could lead to a false conclusion, especially if the written information accompanying the sound recording is a “postscript” containing errors that was written more as a report on the recording than as a script defining the recording process and content. In this case, the audible content itself could be culturally or historically authentic, whereas the script is problematic, unlike Morat and Blanck's assumption that the audible content is problematic when it does not match the script. In most cases of historical sound recordings, it is not known whether the script (if it does exist) was written before the performance and inspired it or whether it was transcribed after the performance. Morat and Blanck’s assumption also only concerns a script that is known and understood by the researcher. Their approach assumes that the content of the recording is tangible, that is, linguistically understandable. But what if one acoustically perceives the content of the recording but does not understand the recorded language and its meaning? Or what if a script is not available in the first place? How is the performance analyzed? Morat and Blanck seem not to have reflected on these questions. Especially for historical recordings of non-European music, it is obvious that the researcher does not always understand the spoken or sung language.

The three approaches introduced here do not consider the potential usefulness of the results of a study on historical sound recordings for the source communities. How does the researcher discern the usefulness of these recordings for a certain group of people outside the realm of teaching and research? Is it possible to decipher the content of the recordings at all? How can the researcher obtain certainty whether this content is authentic or not? On what basis can the “doubtful” be distinguished from the “credible” in this regard? What cultural practices can we glean from them today? What questions regarding these recordings can be answered by the current state of cultural practices in the source communities? These questions still remain but will be addressed in the discourse analysis of the historically audible (DAHA) approach, which is explicitly suited for the study of historical sound recordings from colonial contexts.

3. Discourse Analysis of the Historically Audible (DAHA)

The methodological approaches introduced above constitute only a part of a multitude of contributions to the study of historical sound recordings. These approaches provide a versatile toolbox for examining some aspects of these recordings today, but many questions still remain open. There are gaps in these approaches when it comes to using them to analyze historical sound recordings from colonial contexts as defined in this article.

These approaches consist in listening to the archival silences, listening to history, and listening to all audible components from the recordings, respectively. I believe that these three and many other goals can also be achieved by focusing on another concept: discourse. The concept of discourse is what DAHA is all about. As a concept that unites linguistic-aesthetic properties as well as social practices, discourse is a suitable subject of study on sound recordings from colonial contexts. The French philosopher Michel Foucault defines the term as “a set of statements which belongs to a single system of formation”.^[42] The concept of discourse is neither limited to what can be said, nor is it only about its linguistic aspect. Rather, the concept of discourse includes practices.^[43] DAHA therefore aims at studying the discourses on, from, and about sound recordings from colonial contexts. By focusing on discourse, we can find answers to many questions within the framework of the study of sound recordings that the outlined approaches do not reflect on. DAHA itself contains several aspects from these approaches and schematically follows the methodological approach of historical discourse analysis developed by the historian Achim Landwehr.^[44] However, historical discourse analysis deals with written sources, while DAHA applies to acoustic sources. Accordingly, DAHA follows a five-step model similar to the steps proposed by historical discourse analysis: topic finding, corpus building, context analysis, deconstruction of the audible, and discourse analysis of the audible. The approach makes it possible not only to grasp the content of recordings but also to understand their history and contemporary significance. Nevertheless, DAHA does not claim to be complete but is rather meant to serve as a guide for studying historical sound recordings from colonial contexts.

3.1 Finding the Topic

This step is the first one when we come across a historical sound recording from colonial contexts, listen to it, and decide to work on it in order to better understand it. The basic requirement for the researchers is to doubt and to question what they can hear, that is, to doubt that the recording is really what it appears to be. We should scrutinize everything that sounds normal to our ears, be skeptical about the authenticity of the recording, and question the self-evidence of our own perception as well as the sources on the recording directly available to us. Unlike Morat and Blanck in their analysis of external structure, finding the topic is not concerned with clarifying the unrecognizable (location, actor, author, and genre, i.e., external source criticism) but rather with doubting the recognizable (all information that appears as evidence). Doubt is important for a critical and “rational” approach to the historical sound recording because it arouses interest and raises questions for scholarly analysis. The step of finding the topic determines the direction any further analysis of the sound recording will take. The following questions could be useful and guide us in this step: What do we hear on these recordings (not only in terms of audible content but also in terms of their medium and aesthetics)? What can we not hear on these recordings? What questions do the recordings raise? Are we satisfied with the available information on the recordings, or do we feel that we need more information in order to understand what we can hear? If so, what kind of information do we lack? Which parts of the available information seem superfluous? What is doubtful? What is the meaning of the audible content? These questions can determine the issues and the topic of our research.

For the recording Paasche Ostafrika 6 mentioned above, for example, the Berlin Phonogramm-Archiv generally provides some primary information along with the sound recording on request: we know the name of the producer (Hans Paasche [1880–1920]), the language of the song (Kinyamwezi), the location of the recording (the German colony East Africa [today Tanzania]), the status of the singer (one of the luggage carriers during a so-called colonial expedition led by Hans Paasche), and the year the recording was made (1910).^[45] In this step one could ask questions such as: Is Hans Paasche really the producer of the recording? Why is the song sung in the Kinyamwezi language? What does the content mean? Is the acoustically audible really only a song in the Kinyamwezi language? Since this is the case for that recording, why can we hear a total of three different voices in the recording? Why is the quality of the audible content better at the very beginning than in the following minutes? Who might have been present during the recording but did not or could not make his or her voice audible?

3.2 Building the Corpus

The step of corpus building brings together different sources for the study of the historical sound recording that are capable of answering the questions from the first step (finding the topic). In addition to the sound recording itself, such a corpus usually includes many other sources (archival sources, documentation on the recordings, texts, further audiovisual sources, artifacts, social practices, sources on colonial history, etc.). In this attempt to pool useful sources together, we must consider all discourses—including the colonial discourse—in which statements on the sound recording have been made and are still made. Similarly to the approach of historical discourse analysis, the corpus building of DAHA answers the questions of where, in what way, and with what means anything has been stated in connection with historical sound recordings from colonial contexts and what truth claims are produced by these statements.^[46] In this step, the researchers’ prior knowledge on the topic as well as their intuition about more and less pertinent sources are important, similarly to the corpus building of Landwehr’s historical discourse analysis.^[47]

For the recording Paasche Ostafrika 6, for instance, this step could consist in building a corpus of sources including the documentation on sound recording and its technical process, the collecting practices in German colonialism, the history of German colonialism in East Africa, the writings of Hans Paasche and his biography, the linguistic works on the Kinyamwezi language and culture, the literature on Paasche’s sound recordings and other similar recordings from German colonies, the history of the archive (BPhA), and the conversations between the archive and Hans Paasche. It would also be interesting to find people who speak the Kinyamwezi language and who might know the cultural area where the recording was made.

3.3 Context Analysis

The task in the next step, context analysis, is to introduce the contexts in which the historical sound recordings from colonial contexts originated, that is, the context of the historical material that was pooled together as its corpus. This step involves reflecting on the situation surrounding the production of a sound recording and on why it was then stored at a certain location (archives, museums, etc.). All contexts related to the production of the recording and its history should be considered: the history of technology, the history of recording practice, colonial history, the history of the institution that houses these recording, and so on. Inspired by Landwehr’s historical discourse analysis, the context analysis of DAHA takes four contexts into account: the situational context (answering the question of who does what, when, and where or what happens when, where, and under which circumstances), the medial context (what kind of medium carries the recording; what makes it a source from the past), the institutional context (the institutional framework in which the recording was made, transported, and stored), and the historical context (the overall political, social, economic, and cultural situation and the interaction between them in the production contexts of the recording).

After one gets to know the corpus of Paasche Ostafrika 6, in this step one can accordingly retrace all contexts that led to the recording, focusing on the ideologies and cultural theoretical assumptions behind the collecting practice; the technical context of the phonograph and its accessories (wax cylinders) and the rise of German ethnomusicology; German colonialism in East Africa, the colonial rule, and the subaltern status of the colonized people in German colonies, especially in German East Africa; the expeditions and their meaning for colonization and the collecting of objects for German museums and the BPhA; and many other contexts.

3.4 Deconstruction of the Audible

The aim of this step is to highlight all convincing and doubtful statements that derive from the discourse produced through and on the recording. The content of the recording needs to be deconstructed first and then made understandable. Therefore, it is important to understand the language and culture in which the content (song, proverbs, music, speech sample, etc.) is presented or to work with native speakers from this culture. The approach of collective listening developed by the cultural anthropologist Irene Hilden can be helpful in making information available in as many perspectives as possible in the deconstruction of statements, especially when it comes to exchange with people from source communities. As the name suggests, collective listening advocates listening to and discussing sound recordings collectively in a group. Its main advantage is the consideration of the qualities of plurality and intersubjectivity.^[48]

The deconstruction of the audible also deals with the profile of the producer of the recordings, with his or her life and work, motivation and own perception of the recording and its content, and authority and reputation within the colonial system, which could have influenced the recording situation. A source-critical analysis of diaries, travel reports, and expedition reports of the recordist and his or her network allows for assumptions and logical-deductive answers about the recording situation. At this point, the focus on the unsaid (Lange) and the analysis of the external structure (Morat and Blanck) can be helpful in answering some questions. Close listening (Hoffmann) or modality analysis (Morat and Blanck) can also help in reconstructing certain aspects of the recording situation. The deconstruction of the audible highlights the linguistic, cultural, socio-political, and religious aspects and meanings of the content (text, saying, instrumental music, etc.) as phrased by the source communities. In order to understand the content and its function, the researcher should rely not only on sources such as transcriptions, translations, or descriptions from the archival documentation of the sound recording but also on interviews with people from the source communities (especially with elders) and, if feasible, with relatives of the recorder. The aforementioned approach of collective listening could also be applied at this point. Since dealing with sound recordings from colonial contexts implies an acquaintance with the cultural practices of the source communities, the researcher could also work with translators, linguists, historians, language specialists, and anthropologists from the source communities. In practice, the methodological approach of deconstruction offered by the Germanist and cultural studies researcher Kokou Azamede can be recommended. Azamede proposes a model with 24 categories for the deconstruction of colonial photographs from Togo for didactic purposes.^[49] The procedure consists in describing elements from a colonial photograph in detail and then writing a commentary or interpretation of the image based on the description.^[50] This model is also suitable for deconstructing the content of sound recordings from colonial contexts. Since each object has its own specificities, it is not necessary to apply all 24 categories in the deconstruction process. They can be edited, supplemented, and adapted to the sound recordings as needed.

For Paasche Ostafrika 6, this part concerns Hans Paasche's life and work as a colonial officer and his contribution to the colonial rule, as well as his cooperation with the BPhA and other colonial officers. Furthermore, one could analyze his own works and statements on the colonized people and deduce the region of the colony where the recording was made, while at the same time considering the acoustically audible from the recording. One could analyze the documentation from the Berlin Phonogramm-Archiv and Paasche's correspondence with the archive, reading between the lines in the extant documents to detect statements related to the content of the recording as well as the recording situation. One could then use Azamede's deconstruction method to deconstruct the recording itself. In this part, one could work with experts of Kinyamwezi language and cultural experts in Tanzania in a collective listening to understand the meaning of the song; relocate the context of its usage, meaning the cultural situation in which it is sung, if known; identify similar songs and rhythms and their cultural contexts in Tanzania; analyze the performative presentation of the song in the recording (Does the singer sound happy, sad?) in order to deduce the recording situation (Did he sing freely or does he seem to have been obliged to sing?); and detect different audible—linguistic and semantic—components from the song (proverbs, metaphors, explicit and non-explicit sayings, etc.). One could then take all this collected information as a basis for writing an appropriate interpretation or comment on Paasche Ostafrika 6 and its cultural and historical meaning and try to answer questions raised in the "finding the topic" step.

3.5 Discourse Analysis of the Audible

Discourse analysis of the audible is the last step of DAHA. This step is devoted to answering further questions arising from the questions developed at the beginning of the analysis (finding the topic). The approach refers to concrete statements from the discourse produced about the recording or through it. It is the result and conclusion of all previous steps. Like the discourse analysis of texts in Landwehr’s historical discourse analysis,^[51] this step of DAHA aims at the constructions of meaning and identity formation in their historical change. The following questions may serve as a guide through this step: What (from the recording) was used when, how, by whom, and, above all, for what purpose and in what context? What was the intention of the discourse produced by the recording? Did it aim to exclude, discriminate, convince, or show social differences?^[52] As a conclusion to this step, the researcher answers whether, and if so to what extent, the intention that motivated the acquisition of the recording is similar to, or different from, the intention that motivated the acquisition of other objects, ethnographic or otherwise, from colonial contexts. Finally, this step reflects on the role this recording could have played in both former colonizing and colonized countries and on their didactic, cultural, artistic, and scientific potential. Hence, it reflects on the question of whether, how, and with whom these recordings should be shared today.

For Paasche Ostafrika 6, one could, for instance, focus on the question of whether Paasche presented these recordings at the several talks he held in Germany after his experience with colonialism. If so, what did he say about them and about the colonized people, and what was his intention? One could try to find out what Paasche's motivation in recording this specific song was and to what extent the practice of phonographic recording influenced his career and vice versa. One could also analyze when and to what extent the recording has been offered to the public for listening in the museum or archive, what the intention was, and what the effect and reactions were. Last but not least, one could analyze how important the recording could be today for the source communities in which it originated as well as for the descendants of the recorded people, if identified, and finally, whether or how they have access to the recording.

Concluding Remarks

To sum up, the methodological approach of discourse analysis of the historically audible takes several aspects into account. Not only does it allow for constructing a biography of acoustic objects; it also analyzes linguistic, colonial-historical, musicological, and cultural-historical aspects of sound recordings from colonial contexts. It questions the recordings and lets them tell their (hi)stories and biographies. It offers a dialogue between past, present, and future, between actors, sources, and researchers. This dialogue emerges from the inclusion of source communities in the analysis as part of the sources (finding the topic). It enables source communities to get in contact with voices of their ancestors and at the same time opens spaces to reflect on sharing, repatriation, and restitution issues. DAHA makes it possible to consider not only the acoustically perceptible sound but also many other sources. As shown in this article, the recording and its documentation from the archive are not enough to engage deeply with different aspects of the recording.

DAHA shows that retracing the history and contexts of sound recordings from colonial contexts does not only involve writing about this topic or story. Many other topics and stories are attached to the sound recording from colonial contexts, and they cannot be ignored. In the end, conducting scholarly work with them, especially in the humanities, means identifying these different topics and stories in small formats and analyzing them in the same work.

It is important to stress that not all steps of discourse analysis of the historically audible must always be followed in the proposed order in any engagement with historical sound recordings from colonial contexts. The order of the five steps in this model could differ from one recording to another depending on the research questions. However, to achieve good results, the researcher should consider the guiding questions proposed in the method and also try to answer as many of them as possible.

In conclusion, as emphasized in this article, DAHA proves to be particularly valuable for analyzing sound recordings from the defined colonial contexts. These recordings were primarily made by Western scholars and colonial officers using early sound recording devices such as phonographs and gramophones. Nevertheless, the steps within the DAHA methodological approach can also be effectively applied to the analysis of sound recordings from other contexts, particularly those associated with a history of violence, even if they do not originate from the colonized areas and the formal colonial era. By following the DAHA steps in examining such recordings, we can trace the embedded violence in the sounds and gain clarity on the asymmetric power relationships as well as on the often silenced or misunderstood content of the recording. The DAHA approach provides a systematic and comprehensive framework for navigating the complex histories and power dynamics in sound recordings and for understanding the implications of marginalized groups and communities and cultures in the history of sound recordings. Moreover, DAHA empowers scholars to engage in ethical and critical research, acknowledging the responsibility to address the historical legacies of violence that persist today. By applying this methodology, we can contribute to a more inclusive and truthful representation of the past, ensuring that the stories and perspectives of those who have been oppressed and silenced are rightfully acknowledged in the study of sound recordings.