S1 E3: The Duty of Memory

They thought they knew what had value. In 1980, soldiers stormed the headquarters of Radio Haiti, arrested its journalists, and stole or destroyed the equipment—not realizing that the station’s most powerful weapon was its audio archive, which was left neglected and damaged but intact.

In this episode, we hear from two archivists at Duke University who, almost 40 years later, are recovering the content of those tapes and making it available to the Haitian people and around the world.



Featured image: Destruction of the Radio Haiti station by Jean-Claude Duvalier’s forces on 28 November 1980. The photo was taken in February 1986, after staff returned. The desk was Jean Dominique’s, and reels are some of the station’s archives. Courtesy of David M. Rubenstein Rare Book & Manuscript Library, Duke University.

[1] SAKS (Sosyete Animasyon ak Kominikasyon Sosyal): the network of community radio stations in Haiti.

 [2] Andeyò literally means “outside.” Here it has been translated as “excluded,” but it has many meanings: people from the countryside are described as from the peyi andeyò (the outside land); they are moun andeyò (outside people)—meaning geographically, socially, economically, politically excluded.

[3] The full recording and additional background for this clip (2:18-3:18) can be found at  https://repository.duke.edu/dc/radiohaiti/RL10059-CS-0753_01

The copyright in original Radio Haiti material is owned and administered by Michèle Montas and J. J. Dominique. To encourage research, scholarship, and general use of the collection, they have licensed this material for reuse under a Creative Commons Attribution-NonCommercial-Sharealike 4.0 license. This means that you are free to share and adapt this material, but only if you provide proper attribution, use the material for non-commercial purposes, and license anything new that you create from this material under the same license.


Joy Banks: Hello and welcome! I’m Joy Banks, your host on this season of Material Memory.

This is the third episode in a series inspired by the U.N. Declaration of 2019 as the International Year of Indigenous Languages.

So far, we’ve talked to several people preserving content recorded on now degrading or obsolete formats. In episode one, we explored general questions of ethics and access to reformatted content. In episode two, we discussed the importance of understanding the context in which that source content was created.

On this episode, we hear from two staff members at Duke University Libraries in Durham, North Carolina:

Craig Breaden: My name is Craig Breaden and I’m the audiovisual archivist at the Rubenstein Library at Duke University.

Laura Wagner: My name is Laura Wagner. From 2015 to 2019, I was the Radio Haiti project archivist at the Rubenstein Library at Duke.

Joy Banks Narration: Craig and Laura have been working together at Duke on a project to preserve a large collection of magnetic tapes. These tapes were master copies of programming broadcast by Radio Haiti, the first independent radio station in Haiti. Journalists Jean Dominique and Michèle Montas led the station’s operations, speaking out against government oppression while speaking up for human rights and democracy.

Government opposition forced Dominique and Montas into exile in 1980, after which government forces came in and destroyed the station and its documents. In our conversation, Craig and Laura will share more about the recovery of the station tapes and how they ended up at Duke.

The project at Duke has been in several phases. An initial phase of digitization completed by Cutting Corporation and funded through the National Endowment for the Humanities, or N E H, included most of the station tapes; however, that revealed a set of 88 tapes that were so damaged that additional stabilization work had to be done prior to digitization. The processing of these 88 tapes was made possible through a Recordings at Risk grant through CLIR. Let’s hear how Duke worked closely with Michèle Montas and the Northeast Document Conservation Center (or NEDCC) to make all the tapes, even the most severely damaged ones, available to the Haitian people and to researchers around the world.

Joy Banks: Why don’t you talk a little bit about your project and how you both got involved with it?

Craig Breaden: This goes back to 2013, the Radio Haiti project. We were in touch with Michèle Montas, who’s the owner of the station, at that point. Michèle had been looking for a place to put the materials that were part of the Radio Haiti station and that included papers, but maybe more importantly, recordings of the actual broadcasts of Radio Haiti. It’s a large collection: there are thousands of tapes—very difficult to manage for archives in general. Michèle was, I think she, she received sort of varied responses from the places she looked at. I think the most high profile one being the Library of Congress who said, “That’s a great collection; we’d love to have it, but we can’t do anything to make it accessible.” And that was really key for Michèle, to make the collection accessible, for her. Otherwise, you know, what would be the point of putting it into an archive? So, through some mutual acquaintances she got in touch with us, and specifically our human rights archive. And because we had the collection of one of her colleagues, Johnny McCalla, who was the head of the National Coalition for Haitian Rights, that went a long way toward an introduction to Duke for Michèle. And we talked to her over the next, course of the next year and through her, and through the NCHR and through our curator, Patrick Stawski, were able to bring in the collection. And at that point we had to really look at what it was going to take to make the collection accessible. And the first thing we recognized, you know, we, we had to have right up front was language expertise. And at that point we started looking around to try to find somebody who could help us with that. And I’m going to let Laura pick up the story at that point.

Laura Wagner: Yeah. So I came in in 2015. I should say that for about a year before that I had been working on the collection provisionally through the Forum for Scholars and Publics here on Duke’s campus. One of, I think the, the parts of the agreement with Michèle was that there would be a pilot website, a provisional website, that would showcase a curated selection of Radio Haiti audio. So I was brought in to basically create that, and to not so much do the web design for it, but to listen to the audio and describe it and sort of create a narrative and create content for this website. That’s how I got to know everyone at the library, because I’m not trained as an archivist. I’m actually trained as an anthropologist, and I’m a writer, but I had never been an archivist before. And so then the library said, “Are you going to apply for the archivist position?” And I said, “Well, I’m not an archivist.” And they said, “No, we can teach you that stuff, cause you know Haiti and you have this relationship with Michèle and with institutions in Haiti. And so I said, “Okay.” And that’s how I became the Radio Haiti project archivist. So the first thing that we did was create an inventory of all of the physical tapes, and that was before any of them were digitized. And so that was basically the information on the tape box, the title, the date, physical condition of the tape, and other information that was just based on the physical tape without listening to it. And once we, once I did that, we sent them off to our first external vendor, Cutting Corporation in Bethesda, Maryland. And Aaron Coe—who is basically the man behind Cutting Corp.—he digitized most of the audio, but there were a handful of tapes that were the most damaged. They’re flaking, they’re spoking: all that kind of stuff. And at the end, there were 88 tapes that Aaron did not digitize that were going to require more extensive remediation. And that is sort of how CLIR got into it.

Joy Banks: Could you define “flaking” and “spoking”?

Craig Breaden: Basically the problems with tape, they come in two very general varieties, and that is the first one has to do with kind of the first formulation for audio tape, which has an acetate base. What happens is that over time, the base, that acetate base, contracts as it dries out. Actually, if you have a tight reel of tape, it actually causes the tape to spoke. We call it “spoking,” but it really takes on kind of an octagonal sort of shape. The other main problem with tape is, after acetate, the next formulation that came in was polyester tape. Early formulations of it, well up through the 80s, actually, they had a problem with the base taking on water and it would cause this thing called “sticky shed.” So when that hits the head of a tape player, the tape will start flaking. And that was really the problem we had with the tapes that Laura was mentioning—the ones that went up to NEDCC—they had this “sticky shed” problem, also called “binder loss.” There’s some mystery behind it. Sticky shed is kind of an umbrella term for a couple of different problems. So yeah, those are sort of the two basic problems we were looking at with these tapes because they span both eras of, of tape production, both the acetate and polyester.

Joy Banks: And the problem with that then is that when recordings are suffering this sort of age and wear that you can’t just put it on normal playback equipment, right?

Craig Breaden: Yeah, if you were to put that just on a regular deck, you would have all sorts of issues with the tape, just—all that coating, all that magnetic coating—flaking off, and it would also cake up on the play heads and destroy both the tape and the recorder that you’re using, the playback deck. So they’ve developed several different methods for playing those sorts of tapes back. And what’s really the most common one is sort of an alcohol drip, which actually drips onto the tape before it’s transported across the play head to lubricate the tape. Now sometimes, you know, the tape is so far gone that you can’t even do that with it. It’s just starting to fall apart. So it takes, you know, it takes a very steady hand. Also, if the tapes are, you know, at the same time if they’re dirty, that’s an issue. So, for instance, on a lot of these tapes, NEDCC cleaned them as they were actually running the tape over the play head. So they would do all the work in a fume hood. So all the dirt and mold would go up the fume hood as they were transporting a tape across the play head.

Joy Banks: And were they able to get content from all of these 88 severely damaged tapes that you had?

Craig Breaden: From an amazing majority of them, yes. It was really impressive. I want to say like 85 of 88?

Laura Wagner: I think it was almost all of them, yeah. You know, there’s some where the audio isn’t great and it, you know, there’s sort of skipping in places, but for the most part it was really good, really good quality and really, you know, the content was there.

Joy Banks: That draws us into some of the questions then about content, because you only had access to whatever the boxes said. But as these recordings were being digitized, then you were able to discover a little bit more about what the tapes had on them, right?

Laura Wagner: A lot more, because I listened to every single one of them. So those 88 tapes are actually a very small subset of an archive that has ultimately, I think, more than 3,700 tapes, which represents more than 5,300 recordings. So the Radio Haiti archive I think represents probably the most complete set of primary sources on late 20th century Haiti. Starting from, I mean it really starts in the sixties, but the bulk of it starts in the early seventies, up till 2003, with two periods of exile from the end of 1980 to 1986 and then from the end of 1991 until late 1994. Because it’s an audio archive, it represents voices that are not available anywhere else because in order to be a speaker, in order to produce knowledge and share your opinions, and basically be an author at Radio Haiti, you didn’t have to be traditionally literate. And it was also part of Radio Haiti’s politics to represent people who had been historically marginalized in Haiti. So, that’s to say rural farmers, grassroots groups, women who, you know, were selling in public markets in Port-au-Prince, sex workers, spiritual leaders in Voodoo. In addition to the sorts of people that you might expect: politicians and literary figures and, and people like that. So it was this really wide representation of Haitian society. And, of course, it was the first radio station to broadcast the news and serious reporting in Haitian Creole, which is the language of all Haitian people. Whereas French is the language of maybe 10 or 15%, who also knew Creole.

Joy Banks: Can you kind of talk a little bit more about Haitian Creole, and what you know about its origins?

Laura Wagner: Sure. I’m, so I’m not a linguist, so this is sort of what I know. So when people were enslaved and brought to the colony of Saint-Domingue from Africa, from different places in Africa, right, from different ethnic groups, people who had different languages. And one of the strategies of enslavers everywhere was to separate people from their linguistic and cultural groups, so the people couldn’t communicate and couldn’t organize. And that’s how in the Caribbean you had these Creole languages develop. So Creole, Haitian Creole, largely draws on a French vocabulary. I forget what percentage, but the majority of words are from French, although it’s often kind of old French, which I learned as someone who learned Creole and then started learning French and then realized that sometimes I would say things that to French speakers sounded kind of very archaic and charming. But the grammar is very unlike French. The syntax is not the same as French. The orthography is not the same as French. The orthography was actually only sort of formalized in the eighties—in the 1980s. And that had to do in part with under the Duvalier regime, Creole not yet being an official language in Haiti in part to control information, right? To keep information away from the masses. So that, that’s basically Haitian Creole. It’s a really beautiful language with, it’s really expressive, a lot of proverbs, a lot of riddles, expressions, ways of saying things without saying them directly. And you hear a lot of that on, on Radio  Haiti: this sort of vividness and what they refer to as mawonaj, which is also a term that comes from the time of slavery and people who are escaping and evading slavery. But it’s basically “to escape and then to hide.” And so mawonaj in Haitian Creole is when you are sort of saying something, but you’re hiding it behind something else. You know, so you can’t get in trouble for saying the thing that you’re actually saying.

Joy Banks: So you shared two different clips with me, both of which I thought were really very interesting. One of them was the introduction shortly before the station was destroyed and the journalists were exiled.

Laura Wagner: Right, well Jean Dominique was the director, basically of, of Radio Haiti. He was an agronomist by training who became politicized in part through his work with rural farmers in the north. In 1957, the right-wing dictator François Duvalier was elected president. Shortly thereafter, one of John Dominique’s brothers who was in the army tried to overthrow Duvalier, and he was killed in the process along with two other officers. After that John Dominique was no longer able to work as an agronomist and, you know, as an employee of the state, and he ended up becoming a journalist. And he first worked at the original Radio Haiti. And then in the early seventies, he bought the lease to the station and renamed it “Radio Haïti-Inter” and became the director. And around that time, that’s also where he met Michèle Montas, who later became his wife and who was also a journalist at Radio Haiti. She was the one who, she’s actually the one who’s trained as a journalist. She went to Columbia Journalism School. She trained multiple generations of journalists who came through Radio Haiti and, I think, was basically the news director. So François Duvalier dies in 1971, and his son, Jean-Claude Duvalier, becomes president. And Jean-Claude Duvalier sort of makes these nominal efforts at liberalization, supposedly respecting human rights and having increased freedom of expression and freedom of the press. And this was largely because foreign aid depended on doing that. During, especially in the sort of mid to late seventies, Jimmy Carter was president in the US, and so Duvalier had to sort of pay lip service to human rights. And so journalists in Haiti could get a little bit more bold about what they were saying, but they couldn’t say everything. So you have a lot of examples of Radio Haiti talking about resistance to Duvalier via a proxy. So they would talk about the Sandinistas in Nicaragua, and they weren’t saying Duvalier, but they were saying it in such a way that people listening could say, “That sounds a lot like what’s happening here.” We have a lot of broadcasts from that time that are, for example, they talked about the crisis of what were called “boat people.” You know, people taking to sea as migrants, as refugees. And this was very political, because even if you’re just reporting on the facts of, you know, people leaving, it’s implied in that they’re fleeing persecution, they’re fleeing poverty that is intrinsically political, that the reasons that people are poor are connected to the regime and the, you know, dispossession of peasant farmers and, and all these kinds of things. So that’s sort of what they were doing in the seventies. But at the same time they were kind of toeing a line. They were going back and forth. They were saying what they could, but also mindful of not getting shut down. I think by October 1980, which is when this editorial that I sent to you is from, they’re really kind of reading the writing on the wall. They see that Ronald Reagan is about to win the election in the US. The regime is getting increasingly bold in its opposition to the free press. So what happened before that is that Duvalier’s father-in-law, Ernest Bennett, who said, “le bal est fini,” which is “the party is over,” for the independent press in Haiti. So in this editorial, “Bon Appétit, Messieurs!,” Jean Dominique is responding to that statement that “le bal est fini,” and very, very sarcastically. And he’s saying, “I don’t know what party you’re talking about. We weren’t at the party. You’re the ones who have been celebrating. You’re the ones who have been rejoicing.” And in that proclamation, the regime outlawed, prohibited all but the official government press. And so in this editorial, Jean Dominique is talking, he’s addressing the members of the government press, the official press. And he’s saying, you know, “Bon appétit, it’s all yours now because we’re going to be gone, and there will be silence, and we will be gone. So enjoy yourselves.”

[CLIP 1: Le Bal est Fini]

Translation for transcript only, by Laura Wagner:

In the official newspapers, they told us yesterday, “The party is over, gentlemen.” The party is over, well, the party is over! We weren’t at the party, ourselves. We had no masks, ourselves. For us, it was no Mardi Gras. For you, perhaps. We will silence ourselves, by force of circumstance. The courts will make sure of that. For you, the banquet shall resume. Oh yes, the banquet will resume. And this time, you will not hear any discordant sounds, any noise that might disturb your appetites. You will not be distracted from your plentiful feast by the cries of the poor, the screams of the boat people devoured by sharks, the gunshots killing our cane-cutter brothers in Santo Domingo or in Nassau or in La Romana. No, you will no longer hear those discordant, disagreeable noises that might trouble your meal, that might prevent you from celebrating — it will be only silence. A reassuring silence… Total silence. So you may celebrate at ease, gentlemen! At ease! And in that profound silence: Bon appétit, messieurs!

Laura Wagner: And then a month later, Reagan is elected. And almost immediately on November 28th, 1980, the Duvalier regime comes and destroys Radio Haiti, arrests most of the journalists except for Jean Dominique, who—there was actually an order to kill him on sight, and he sought refuge in the Venezuelan embassy. Journalists, human rights activists and, you know, union leaders were arrested and exiled. So what happens then is, they don’t have any warning that that’s about to happen. So the station is just left as it was on November 28. So they—Duvalier’s forces—arrest the journalists. They are exiled, they come in and they just trash the place. They break the equipment, they steal whatever they think has value. It’s filthy. It’s left open to the elements. It’s the tropics, right? We have these pictures that are from immediately after Duvalier falls and they go back to the station and you know, things are just utterly destroyed. But the “Macoutes,” Duvalier’s forces, sort of didn’t recognize—they thought that what had value at the station was the equipment or the safe that they tried to break into. They didn’t realize that the most valuable thing at the station was actually their audio archives. And so the tapes are, you know, on the floor, they are in some cases you know, the reels are broken or, you know, they’re very damaged. But they’re still there. And so when Radio Haiti comes back from exile in [19]86 after Duvalier falls, they recuperate all their archives, some of which are so damaged that, you know, they can’t play them. And so a lot of the 88 tapes that we ended up sending to NEDCC through the CLIR grant were those tapes that were damaged and sort of left for dead in the early eighties.

Joy Banks: So what would you say is the importance of these tapes, is the importance of these audio archives? Why is it so important to preserve these?

Craig Breaden: I’ll give you my, I’ll give you my take on that. As an audio archivist, I have a real sort of feeling that this kind of archive—especially this kind of audio archive—is so important to the history of the late 20th century. And that’s a big general statement. I mean I believe that about audio archives in general, but the one thing about this particular one is its place in Haitian history, and the way that it creates a center of memory for Haiti and Haitian history, which, you know, I think we take more and more for granted that we have, that we just have audio and video of everything now, that it’s hard to remember that there was a time when that really was not the case. And also thinking about Radio Haiti in the context of broadcasting in general. They were there at a real, sort of golden era for broadcasting, after it had sort of gotten its legs after the late twenties into the 30s, 40s, 50s, 60s. It was really hitting its stride in the 70s, and people were really beginning to experiment with the format and what it could do internationally. I think Radio Haiti is a very good example of that. So along with the fact that they saved so much Haitian history, just, you know, by the fact of doing their jobs as journalists, they also are a wonderful example of broadcasting history as well. So I think we can capture things through the audio that we’re not going to capture in any other way, either through the papers or sort of people talking about Radio Haiti in retrospect. Having that audio and having it be, you know, really creating a liveliness and a voice, I mean, literally and figuratively, for the people who were there at the time witnessing what was going on, is not only compelling, but gives us a view of history that we don’t often have of any historical event. So for me it’s incredibly important that this archive was saved. And I say this all the time: I mean, I can’t understand anything that’s happening on the tapes, like literally, but since we first talked about possibly getting this archive, there was no doubt in my head that this was the most important collection historically that I was going to work on. So for me it has importance in a lot of different ways, but those are sort of the core things that speak to me as an audiovisual archivist.

Laura Wagner: The Radio Haiti archive contains voices that appear nowhere else.  Normally, history is written by the powerful. Not only the victors, but the powerful: those who have the ability to write, and who have the power to write and be heard. And Radio Haiti is a case where it was, it’s, it’s, they didn’t realize that they were doing history. In fact, there are many cases where they say on the tape, “We aren’t historians.” But it is Haitian history, recorded by and, in a sense, authored by Haitians for Haitians and, in many cases, by poor Haitians. So like I said, you know, Jean Dominique and Michèle are sort of from the intelligentsia, right? But they had a political project to amplify the voices of people who had been excluded from public discourse. And so I think that in that way, it is a totally unique telling of history. I think that in, in Haiti today, as in a lot of places, they’re undertaking a project of “devoir de mémoire,” which is “memory work,” in a sense. It’s a term that I think kind of came into fashion in France, around the Holocaust in an attempt to record and remember that history so that it’s not repeated. But people in Haiti are talking about it in order to preserve the memory of the Duvalier regime, and what happened to the population during that time, as people from that generation who lived through it, you know, grow older and pass away. And so for those kinds of projects of keeping memory alive, including memory of atrocities alive, this is a very, very special and important archive. I think especially in a place like Haiti where there are so many sequential crises. Duvalier, the Duvalier regime ends, and then there’s this sort of rocky transition to democracy. The first democratic elections in 1990, and then, in 1991, the first democratically elected government is overthrown in a bloody coup d’état that in which a military junta leads Haiti, or rules over Haiti, and 5,000 people are killed. And then after that, after democracy is restored, there’s this sort of long period of violence and what they called “insecurity,” which goes through the, really, through the 2000s, and then sort of just as things I think started to feel more settled, more stable, the 2010 earthquake happened. And so I think that in the face of this kind of sedimented trauma, it is really hard to consciously keep memory of things that happened before you were born alive. And in that way, I think that this is an invaluable collection. Yeah. And in that way too, it also, it contains the voices of so many ghosts, of so many people who died by, you know, from the Duvalier years through the coup years, through insecurity, through the earthquake or just through, you know, the ordinary trauma of having your heart broken over and over again by a country. And so I think that, yeah, there’s nothing, there’s nothing else like this archive out there.

Joy Banks: That kind of also brings us to the other clip that you shared, where on the radio they were speaking about the importance of recording the language of those that are “outside.” Do you want to talk about that a little bit, Laura?

Laura Wagner: Yeah. This is just, it’s just a clip that I find delightful. And so and so I, I like it. It’s just the very beginning of an interview that Jean Dominique is doing with two other journalists, one of whom actually was also at Radio Haiti, named Sony Estéus. And so it’s an interview with Sony and a guy named Ti Jò [Joseph Georges]. They were both leaders of the network of community radio stations, which is called SAKS [1]. And they, it’s in honor of International Creole Day, Creole Language Day, which is I think October 28th, I want to say, it’s in late October. And in honor of International Creole Day, they are talking about the importance of using—of having Haitian Creole radio. And they’re talking about the importance of Creole on the radio. And then at the end of that clip, they also talk about the importance of, you know, well, if Internet ever becomes part of it, Creole also has to be included in that. Although the way that he says is, like, a lot more adorable.

[CLIP 2: In honor of International Creole Day]

Translation for transcript only, by Laura Wagner:

Ti Jò: We saw that language and culture are two things that are linked together. They give communication both quality and a good sense of identity, so that it corresponds to the culture of people in the country.

Jean Dominique: And that is why, on the occasion of International Creole Language Day, I asked Ti Jò and Sony, who are two leading members of SAKS, to speak to us about communication—but, in the country of Haiti, you can’t communicate with the poorest of the poor, you can’t communicate with those who are excluded [andeyò] [2], you can’t communicate with the destitute, if it isn’t done in Creole. That reminds me, if the Internet ever becomes part of the question, the Internet should give itself a little kick in the behind to make sure Creole is included in the web-dot-com-dot-email! [3]

Laura Wagner: So, you know, John Dominique—I’m paraphrasing, but he says something like, you know, “and if the Internet ever becomes part of this”—meaning part of this whole broadcasting thing—”then Creole should give itself a little whack on the behind to make sure it enters into the web.com.dot. email dot. something.” It’s very, very funny. And that’s just one example, but he was extremely funny.

Joy Banks: So the languages that are captured on these tapes: we’ve got Haitian Creole, there’s French. Were there any other languages?

Laura Wagner: Yeah, but minimally, so the vast majority is in Haitian Creole or French. There are occasionally interviews done in other languages, but they were always then translated for a Haitian audience because that’s, that’s really who this station was for. So there are a few interviews in Spanish with, for example, there are some human rights activists from Argentina who come to Haiti and the aftermath of the fall of, or after the coup years, to sort of talk about truth and justice and seeking, seeking justice and reparations for people who were victims during the coup years. And so they’re talking about their own experiences in Argentina. And in those cases, the interview would be in Spanish, but there would be translation. Or in 1980, Jean Dominique interviews Michael Manley, who is at that time was a, a candidate for president in Jamaica. And that interview is in English, but then they translate it. So there are some interviews with politicians from the Dominican Republic, or Cuban doctors, or immigration officials in the US, so I would say that there are a handful of things in Spanish and in English, but with translation to either Creole or French.

Joy Banks: One of the things that I think is interesting about this is the dynamic of accessibility, especially since that was a charge for your project since Michèle was in touch with all of you. And I know that as you describe things in an online environment, languages can sometimes be challenging, especially as you’re trying to capture the way that words should be pronounced to really preserve the language. Did you encounter some of these issues during the processing of these materials?

Laura Wagner: Yeah, I mean, I would say that the problems were more about existing structures and practices not really being set up for doing this kind of work that I knew from the beginning we were going to have to produce trilingual metadata, because this needs to be searchable in Haitian Creole and in French. And in addition to that, it has to be pretty thoroughly described. Because I was always thinking about an audience beyond academia, beyond researchers, and even beyond Haitian researchers. I was thinking about just people, right, who are not necessarily scholars. Will they be able to find this if they go into a search bar and just start looking up these terms or these people, will this information come up? So in that sense, it was important first to really describe the audio. I was often told that I was over-describing the audio. But given that audio isn’t, it’s not skimmable, right, the way that papers are, I thought, if I don’t describe it now, a lot of this is never going to be found again. And so it was doing very thorough description, and then doing that in three languages. I would say that the, the issues with that were, one, finding people who could work with me to do that. I write Haitian Creole very well. I do not write French as well as I write Creole. And while I understand the French recordings, I don’t feel that I can write the descriptions in scholarly, elegant French. So I needed people who could translate the descriptions that I wrote into French. And if I was to have student assistants who were also listening to audio with me, they had to be able to both understand and then write either Haitian Creole or French. And finding people who could do one or both of those languages at the level that we needed was very hard in North Carolina, and really in general, because a lot of people in Haiti don’t learn how to write Haitian Creole properly because it isn’t taught in a lot of schools. And for people who are a little bit older, it wasn’t taught in schools at all. And so people who write Creole very well are generally people who’ve taken the initiative to really learn it themselves. So that was a complication. And then frankly, having the resources needed pay people what I thought was a fair wage for doing this kind of work, which requires such specialized knowledge and skills, and we actually didn’t have the budget to do that. So that was part of it. And then just the systems themselves. I mean Craig can talk more about this, but you know, the Duke Digital Repository—which is sort of the main place that, the main interface for searching the collection, for listening to the collection, for reading the descriptions—it doesn’t support multilingual metadata. So we had to work around that.

Craig Breaden: Yeah, the workaround was pretty—right now it’s pretty clunky. Hopefully in the future it can get a little bit more elegant, but we had to dump all this description into one field. That kind of thing. Whereas if we have a database in the future where we can actually say, you know, in the metadata, the metadata says this is Haitian Creole. That’s important. So you don’t have just a single description field with everything in it. But that’s a technological thing, and that’s one of the leaps we’re making here in, you know, this is one way that this project is actually pushing our digital repository to, you know, be more adaptable to these sorts of collections. Because this is not the only foreign language collection we have. And a lot of times, you know, we define what we can do by the tools we have. So in this case, we’re actually trying to take it the other way. And so Radio Haiti’s been really good at pushing some of these borders. But I think, you know, what Laura speaks to is where the, the difficulty of getting the technology—even, you know, very common technologies—to catch up. Using Excel, using Google Docs, stuff like that, to try to create all sorts of workarounds for this description database that we built. Because, you know, you get down to really arcane old things, I mean just—like Excel doesn’t handle diacritics well. And it spurts out all this weird, these weird characters when you crosswalk it into something else. Whereas Google Sheets, remarkably, does very well with that. But Google Sheets can’t get very big, or it starts to freak out. So we had to do a lot of the work in the Excel, wash some of the data through Google: there’s just all sorts of problems that you don’t think of when you think about a collection like this because you want everything to go well because it’s, you know, it’s a great project and this, that, and the other thing. But when it gets right down to it, these little technologies actually have a lot of say in what we can do and they just haven’t really caught up with where we’re going in a, you know, rapidly digitizing world where we have all this information that can potentially be in common, but some of the technology tools just aren’t there.

Laura Wagner: Yeah. I might just jump in. I heard an expression yesterday, which was, “building the roller coaster as we ride it,” I think describes this pretty well, which is that it wasn’t clear from the outset what the workflow needed to be, and we were sort of constantly redoing the workflow, as we started. Like we started putting the metadata into an Access database, which had all kinds of complications, the main one for me being that I couldn’t, like, copy and paste anything. And so, for example, you might have an interview that is on two sides of a cassette. And in that case, because it’s one interview, I would want to copy and paste all the metadata and then just write, “Part One” and “Part Two.” I couldn’t do that in Access. I had to copy and paste every single field separately. And it was taking a huge amount of time, which was one of the main things that then led us to move everything into a spreadsheet, an enormous spreadsheet. And another thing is spellchecking. So if we were writing in English or French, there are, there’s existing spellcheck technology that could kind of, you know, highlight things that might be misspelled. But that doesn’t work for Haitian Creole. And so for Creole, it all had to be basically spellchecked by hand, and that also took a lot of time and a lot of energy.

Craig Breaden: When we were starting the project, we didn’t have the funds. Like a lot of times these projects will actually build their own databases in-house. The problem with doing that is that the information then becomes very difficult to get out of that database if anybody else down the line needs to use it, and really extract the information meaningfully. So we wanted to be able to build something that would be easily adaptable and accessible. I do want to take one step back and say that we had a moment right at the beginning of the project where we thought, “Hey, what if we transcribed everything?” We actually got a couple of quotes, and I think Laura was super accurate when she said that would be way too low, and I think the quote was $3 million to transcribe everything, or the estimate. And then Laura is like, that just seems really low. And I’m like, yeah, I think you’re probably right. But the other thing about that is, I don’t know how much transcription would have bought us, actually, in the end. Like, since 2015 Duke has been transcribing and captioning anything that goes online, but Radio Haiti got a pass because of the, it would cause, quote unquote, “an undue burden” to transcribe all that Haitian Creole. But the other piece of it is, I don’t know that transcribing would actually help access as much as thorough description in this case.

Laura Wagner: It would, it would be too much. It would actually become, I think, harder to manage probably, or harder to navigate if it were all transcribed. Also it would take millions of dollars and I don’t know how long that would take. I think it would also maybe never get done. The main barrier to access would be that it would actually be impossible, and so it would never be done. Because are you talking about transcription and then translation? Or just transcription? And that’s not only thinking about the language, but also the sound quality. You know, when it’s an in-studio interview, that’s one thing, but if it’s, you know, a field recording and there’s multiple people speaking and the wind is blowing and there’s a rooster and the sound quality isn’t that good, and it’s not, you can’t hear people that easily, then what are you going to do? No, I’m tired just thinking about that.

Joy Banks: Laura, how did you come to be so fluent in Haitian Creole?

Laura Wagner: I, after college, I lived in Miami for a couple of years. I was working for the health department, for with people with tuberculosis and HIV. And a lot of my colleagues, and then a lot of the patients that the health department saw, were from Haiti. So I kind of picked up Creole that way. And then when I started graduate school, I was living in Haiti from, basically, 2009 to 2012. And so it got a lot better there as I was, I was living there and doing research and sort of, yeah, I was living in Haiti for about three years. And I, I go back and forth pretty often.

Joy Banks: I’d like for you to talk a little bit about some of the creative measures that you’ve come up with to provide the access to the recordings to the audiences.

Laura Wagner: I think that this project has made us all think about what accessibility really means. So having things available on the Duke digital repository, and it’s free and you can download it, and you can search it, is accessible in one way. But I think that if people are never going to know about or navigate to the Duke digital repository in the first place, then in a sense it’s accessible in name only. And when I was saying before that I was always thinking about audiences beyond researchers, I, you know, researchers will use the repository. They’ll find the repository, and they will sort of look for what they need. If you’re talking about high school students in Haiti, or just ordinary people—basically the people who were Radio Haiti’s audience, or the children of Radio Haiti’s audience—that is, mostly, not how they’re going to access it. Most people in Haiti use smartphones instead of computers, most of the time. They are more likely to use apps like Facebook, WhatsApp, Twitter, other kinds of chat apps, than to go to a search bar and, and Google something. And so one of the first things that we did was create a social media presence. So I run a Facebook page and a Twitter account for the archive. And the sort of purpose there was twofold. One was to let people know about the collection so that they could follow it and seek things out on their own and sort of keep people updated on the project and its progress. And the other was to continue to share content. And often the way that I did that was around anniversaries of a particular event, or if someone was in the news for one reason or another, often because they had died, but for other reasons as well, I would put up a recording or a photo or something like that. And often that got a lot of traffic and then again, had more people using the archive and knowing about the archive. And so that, that’s one thing that we’ve done. We’ve done a series of flash drive distributions, the first of which was in 2016, where I went with a thousand flash drives that were branded with the little Radio Haiti symbol and the URL of the finding aid and then a little informational postcard. And each of those had about 30 examples, like 30 selections of audio that were, sort of spanned time and subject and type of recording, just things that I thought were interesting. And I distributed to grassroots groups, schools, community radio, cultural institutions, libraries. But you know, I had a thousand, so I could, you know, anyone who wanted one basically: “Here you go.” And then said, “Share these, copy them, copy it to a laptop, share it with your friends, whatever.” And then in the beginning of 2018, I went with a smaller number of flash drives that had about 500 Radio Haiti recordings with the metadata, and those I gave to a smaller, obviously, number of institutions. So the National Archives, the network of community radio stations, the main independent media outlet in Haiti, and our partner FOKAL, which is sort of one of the main cultural institutions in Haiti—it’s funded through the Open Society Foundation. And then in the next few months I will be going to Haiti again and bringing, I think, 120 copies, flash drives with the entire archive and all the metadata, and distributing those, hopefully doing some workshops around how to use the archive, how to use the repository, how to, how to do all this and hopefully a small exhibit and some public events around really launching the official completion of this project.

Craig Breaden: We had two NEH grants, and the first grant was to digitize the main body of the recordings. The second grant was to further fund description and to create a pilot project that would test out what it would look like to put Radio Haiti on YouTube and the Internet Archive. We did that in the fall of 2018, so that Laura could then test it with some people in Haiti to see if that was actually a good thing or a bad thing or if there were sort of mixed responses. It turned out there were some mixed responses, which was really interesting. But the idea is you know trying to, trying to put the archive in as many places as possible, to be able to have it be as accessible as possible. I think, you know, one of my pie-in-the-sky things is to put this up in so many places that people can just download it and do their own thing with it to a certain degree so that it becomes, you know, so it really does go back to the people who originally created it. So, you know, we’ve been trying, I think institutionally, to expand our idea of what access means. And I think that this is a really good example of our trying to look at a collection in a different way, to look at different channels of accessibility in terms of, you know, different populations, different digital infrastructures, and different needs and wants. I mean, there are all sorts of motivations that people have for coming to this collection. And we try to, at least part of my job is, to get it out there and let people use it in the way that they want to use it. And you know, that has, it certainly has its challenges and that’s sort of why we did kind of a very scattershot approach to this. You know, we’ve got it in a bunch of different places. The Duke digital repository, part of it now is on YouTube. More of it might be on YouTube, Internet Archive, the flash drives, the Digital Library of the Caribbean, the Digital Public Library of America will be hosting metadata. So really just taking every opportunity to put it on the platforms we can put it on. Because honestly, the metadata that Laura created is so valuable that if we don’t do that, we’re just cheating ourselves. So we’re, you know, we’re pushing as much as we can to get it out there.

Joy Banks: That’s really exciting. So I wanted to just end with the opportunity for each of you to think about maybe one thing that sparked in you about this project—if it was a particular recording that made you happy, an experience with a user, whatever it might be, but just one more thing.

Laura Wagner: I mean, I think for me what makes me happiest is when, when I go back to Haiti and people are just so excited about this project. When people come and say, I remember this recording, do you have this one? Or they start sharing memories of the, of the radio station and what they used to listen to. And that is really exciting and is what sort of in a sense kept me going because it was a, it was a very long and arduous process. And so, I often felt very far from Haiti, actually very far from people, when I was doing this, and so when I would go back to Haiti and have those encounters, and people would tell me what this project meant to them, that was always really meaningful. Sometimes when I would go, you know, Michèle would say, “You should go see so-and-so.” And one time when I was going, I think in 2016, she told me that I should go visit Charles Suffrard who is, he’s a rice farmer from the Artibonite in Haiti. And he’s also the leader of a grassroots peasants’ organization called KOZEPEP. And he was a very close friend and collaborator and, as he would say, teacher, of John Dominique. And so I did. So I went to Haiti, and I was in Port-Au-Prince, and I called him up, and I introduced myself, and he was really excited and he said, “Okay, come visit me.” And so I got on a bus up to the Artibonite. And when I got to L’Estère, the town that he’s from, he picked me up at the bus stop, took me home. His family fed me lunch including, you know, rice that he had grown. And then he, after I finished eating, he immediately took me to the river where they poured Jean Dominique’s ashes. Jean Dominique was assassinated on April 3rd, 2000, and about a week and a half later, farmers organized a, basically a wake for him and they poured his ashes in the river. He said we did that so that his, that he could fertilize every grain of rice that the Artibonite river touches. And that felt very powerful, and it felt like I was being brought into this story, in the process of doing this archive, that I had also become part of bringing Jean Dominique home, and bringing Radio Haiti home. And that was a very moving experience.

Craig Breaden: I think from, from my perspective, I really liked seeing Michèle’s face kind of light up when we would, when she would visit, and we, you know, we’d show her what we were working on, or we’d explain what we were doing, and realizing I think that a lot of what we were doing was far beyond her expectations. I think when she first started shopping the collection around, she had an idea about putting the stuff out there and having it be accessible. But I don’t think that until we started doing it, that the real weight of it sort of settled in on all of us. And I think that, I like the idea that, you know, we did something that was very hard to do, and we did it successfully, most of the time. Most of the decisions we made about how we needed to deal with the collection, I think we made the right decisions. And I was, I’m very pleased about that. And I think that our donor is very pleased. I think Michèle and Jean’s daughter Gigi are very, very pleased about the collection, and it means a lot having it here. And it means a lot to know that we’ve done right to a collection that really needed to have right done by it.

Laura Wagner: Yeah. You know, Michèle often says to me, she says, “This is way beyond what I imagined.”

Joy Banks: That’s perfect. I just love everything that you both have shared with me. Is there anything else that you feel we didn’t talk about?

Laura Wagner: Yeah. I was just thinking, you know, when I was talking before about all the injustice and all the tragedies and all the other bad things that are, that they covered in this, at Radio Haiti and that appear in this archive, I also want to convey how much joy and creativity is also there. I think it’s hard to convey because so much of it depends on the language, and how things are said, and really listening to it. But I think that for the journalists who were there, it was very exciting. It was it was very meaningful, and I think they also had fun a lot of the time. And a lot of people, a lot of people in Haiti are just really funny. And that’s not incidental, right? It’s not coincidental. It’s that humor is subversive. It’s a way to assert one’s voice and one’s freedom in the face of oppression, in the face of power. And so, I would want people to know, even if they can’t listen to it, that listening to the archive of Radio Haiti wasn’t, by and large, it wasn’t actually a sad experience. There are definitely sad things in it. There are things that were really, that made me cry as I listened to it, but overall it was a really, it was alive, you know. It wasn’t just an inventory of tragedy and oppression. It was much more than that, and much more joyful than that.

Joy Banks Narration: Thanks for listening to our episode. We hope that you’ll join us on Episode Four, when we’ll be visiting with a team working on a collaborative project to digitize audio and audiovisual recordings from the second half of the 20th century that document life on Alaska’s North Slope. Large portions of the recordings are in the Iñupiaq language, and the team is working to transcribe and translate them to make the recordings broadly accessible and useful for studying the language.

More information on this and all our episodes, including show notes, transcripts, information on our guests, and links to their projects and social media, can be found online at material-memory.clir.org.

If you like our podcast, we hope you’ll rate, review, and subscribe.

CLIR is an independent, nonprofit organization that forges strategies to enhance research, teaching, and learning environments in collaboration with libraries, cultural institutions, and communities of higher learning. Learn more about us at clir.org.

To learn more about preservation efforts happening in your area and ways that you can contribute, visit your local library, archive, museum, or historical society.

Material Memory is produced by CLIR, with the assistance of Ernesto Gluecksmann, Christa Williford, Kathlin Smith, and Lizzi Albert. Our audio engineer is Benjamin Green. Our theme music is by Poddington Bear.

I’m Joy and I’ll see you next time on Material Memory.


Behind The Mic


Joy Banks

Show Host

Joy Banks is program officer for CLIR’s grant team. She helps manage communications, outreach, and assessment activities for the Recordings at Risk, Digitizing Hidden Special Collections and Archives, and Cataloging Hidden Special Collections and Archives programs. She is the author of The Foundations of Discovery: A Report on the Assessment of the Impacts of the Cataloging Hidden Collections Program, 2008-2019, published in 2019.


Craig Breaden


Craig Breaden is the audiovisual archivist at the David M. Rubenstein Rare Book and Manuscript Library at Duke University, where he has worked on a variety of collections and projects focusing on time-based media, including the Frank Clyde Brown Field Recordings Collection, the H. Lee Waters Film Collection, the Jazz Loft Project Papers, and the Radio Haiti Records.  He also manages the Rubenstein Library’s oral history collections, including the Re-Imagining Project, the Duke University Oral History Program, Indivisible: Stories of American Community, and many others.  He holds an MA in History from Utah State University and an MLS from the University of North Carolina at Chapel Hill.

Laura Wagner


From 2015 to 2019, Laura Wagner was the project archivist for the Radio Haiti Archive at the David M. Rubenstein Rare Book & Manuscript Library at Duke University. She holds a PhD in cultural anthropology from UNC Chapel Hill, where her research focused on people’s experiences of displacement, humanitarian aid, and everyday life in the aftermath of the 2010 earthquake in Haiti. Her nonfiction writing has appeared in SalonSlatesx archipelagos, and other venues. She is also the author of Hold Tight, Don’t Let Go (2015), a young adult novel about the Haiti earthquake.