Warfare of Art & Law Podcast
Warfare of Art and Law Podcast sparks conversation about the intriguing – and sometimes infuriating – stories that arise in the worlds of art and law with artist and attorney Stephanie Drawdy.
Warfare of Art & Law Podcast
Dr. Michelle Fabiani & Dr. Fiona Greenland on CURIA Lab's Work to Measure the Impacts of Cultural Loss
Show notes:
0:00 Dr. Fiona Greenland discussing collabarative work with Curia co-founder Dr. Michelle Fabiana
2:00 Dr. Michelle Fabiani's background
3:40 Dr. Greenland’s background
5:40 collaboration between Greenland and Fabiani
7:10 overview of Curia Lab
9:40 Informatics, the science of information
11:30 Syrian project - how robust and reliable data on scope of Syrian looting was with a review of remote sensing imagery
15:10 participants in Syrian project
17:20 Syrian project – evidence on whether there is a connection between Syrian civilian fatalities/casualties and cultural heritage looting
22:45 prelude hypothesis
27:50 war in Ukraine and its effects on Ukrainian culture
32:00 how the data is used to inform accountability
34:30 Greenland’s work with Conflict Observatory Ukraine
36:00 Ukrainians’ current restitution, reparations and accountability efforts
37:50 user guides for each area
42:30 Fabiani’s PhD project on Egyptian archeological looting
44:00 current project that builds on PhD project
46:00 impact of technology on their approach, including disinformation
50:00 complications created by AI, including generative AI
54:00 perspective of skepticism required
56:30 online risks and need for mitigation
58:15 how their work speaks to justice
58:30 Miranda Fricker’s book Epistemic Injustice: Power and the Ethics of Knowing
1:02:30 questions of justice for whom
1:05:30 Lauren Stein: question on the role of universities in Curia Lab
1:07:50 hope for their work to establish frameworks of cooperation and collaboration that cherish equal access to knowledge/information that would then lead to equal access to accountability
1:09:10 hope for their work to facilitate a shift to a multi-disciplinary approach
Please share your comments and/or questions at stephanie@warfareofartandlaw.com
Music by Toulme.
To hear more episodes, please visit Warfare of Art and Law podcast's website.
To leave questions or comments about this or other episodes of the podcast and/or for information about joining the 2ND Saturday discussion on art, culture and justice, please message me at stephanie@warfareofartandlaw.com.
Thanks so much for listening!
© Stephanie Drawdy [2025]
When we started this collaboration, it was as two social scientists who were focused on cultural heritage questions. I now understand that we have found other collaborators within cultural heritage who share our interest in and commitment to questions about the social, the political, the economic.
Speaker 2:Welcome to Warfare of Art and Law, the podcast that focuses on how justice does or doesn't play out when art and law overlap. Hi everyone, it's Stephanie, and that was Dr Fiona Greenland discussing CurioLab, which she co-directs with Dr Michelle Fabiani. What follows is a conversation with Dr Greenland and Dr Fabiani, during which they share about how they came to found CurioLab, of their various projects that have brought to life the vision of Curia Lab to provide a flexible platform for the collection, analysis and dissemination of research on cultural heritage, dynamics and community impacts, particularly in the aftermath of loss, and we also look at how they define justice and the mark they hope to be leaving with their work through Curia Lab. Dr Michelle Fabiani and Dr Fiona Greenland. Welcome to Warfare of Art and Law. And Second Saturday Thank you so much for being here.
Speaker 3:Thrilled to be here. Thanks for having us, Thanks.
Speaker 2:Tiffany. So I would love to begin if each of you would kind of share a bit about your background and then how you came to collaborate together to create Curia Lab.
Speaker 3:Sure, fiona, I'll go first, if that's all right.
Speaker 3:I come from a background in cultural anthropology and archaeology.
Speaker 3:Before moving into criminology and criminal justice, I was trying to find my way in terms of where's an area where I felt like I could make a difference, and, through a lot of experimentation, working in collections and working within the legal framework of museums and a very interdisciplinary education, I settled on criminology and criminal justice, as it gave a solid foundation and a very holistic methodological framework that I could use to look at this international intersection of art and antiquities and illicit economies, behavioral patterns, as well as novel approaches to data collection, and so my focus has always been on the illicit antiquities realm and illicit art area.
Speaker 3:But how I have approached it has changed, and I try to come at my work from as broad a perspective as possible. So if there's a field that you can think of, I have probably at least done some studying in it. I do data science. I've studied cultural heritage law, domestically and internationally, economics, sociology and internationally. Economics, sociology, political science, terrorism, environmental science. I try to bring as much as I can to the table, while recognizing that I don't have to be the expert in everything that's very true about Michelle.
Speaker 1:She is the Swiss army knife of cultural heritage studies, where every blade is sharp. I trained first in classical archaeology and my focus was on the Western Roman Empire, and I trained in the formal features of imperial sculpture and did excavations in Italy and in Spain, and I was a classical archaeologist for 10 years. In fact I got my first doctoral degree in that field and along the way I got curious about bigger contemporary system questions. So that included questions about museums and collecting practices as well as the global circulation of archaeological materials, and these questions started to be of interest to me in the late 90s, which was a period before we could say established university interest in these. There weren't the range of museum studies programs yet I found it necessary to retrain. So I did a second doctorate, and this one was in public policy and sociology doctorate and this one was in public policy and sociology, and it was through that program that I developed many of the skills and frameworks that I bring to bear now in the work with Michelle and that Michelle touched on as well. So I would say at core we are social scientists with deep training in archaeology and anthropology.
Speaker 1:And you asked another question, Stephanie, that I want to pick up on, and that is when we started working together. We met in 2015 at a conference in Chicago, and I remember this conversation because Michelle was a PhD student, I was a postdoc. She told me about this project she was doing. I thought this project was wild and what she was doing was more ambitious than anything I'd seen in studies of looting. She was working with a large set of satellite images to understand the relationship between political and economic instability and looting in Egypt uncharted territory and I thought, if this woman knows how to do this, this is somebody I want to work with. And in 2021, we formally established the lab after many years of collaborating projects and different methods.
Speaker 3:Yeah, and that that initial conversation, I think I cracked Fiona down because of a really fantastic article she wrote um bashing the 5 million, 6 million value of the art market. That comes from nowhere, and I just so appreciated how nuanced her framing of it seems like anything she wrote was, and so I just found my way to the conference specifically so that I could sit down with this fantastic researcher and pick her brain about really anything that I could, and we never looked back.
Speaker 2:Yeah, what a powerhouse collaboration you two create. So Curia Lab, let's dive into that. And so we've talked about kind of the seed of it, so would you take it from there?
Speaker 3:So Curia Lab has this idea behind it that there's so much we can do when we try to, when we have first of all data to do it with, and so a lot of what we started with in the lab is working on how can we and where can we find and coalesce data that are relevant for policymakers and for scholars and that are importantly, accessible to these communities.
Speaker 3:There are a lot of data sources that are protected, either because they were produced by federal government or that are part of the intelligence apparatus of any nation, and so there may be very nuanced views of art and antiquities, crimes that we can't actually validate or replicate and therefore we can't build off of repository of data.
Speaker 3:In the criminal justice area, there's the really large open access data set or data system called ICPSR, where you can find data on pretty much any topic you want and as you create data you can house it. That was sort of one. Inspiration is if we could become this clearinghouse, but we also really wanted to incorporate the idea of resilience and how data can help inform ideas about what resilience are, and so Curia in a large extent focuses on this nexus of new technologies, on state government participation in protection and destruction, as well as grassroots efforts. We try to have a very highly ethical, high ethical standard for how we collect and store data, for consideration about how we should be training ourselves and others and how, then, what we produce can help train others to contribute to this work. And I'll pause here to let Fiona also add in I'll tackle the I in our lab name.
Speaker 1:From the beginning, stephanie, we thought about information, the heterogeneous sources of information that go into understanding cultural heritage, archaeological materials, artworks, sites of importance and, of course, intangible heritage. Informatics is the science of information and we recognize in each other a deep found respect for the ways in which librarians organize information, preserve information, preserve the traces of the provenance of information, where does it come from, and making it available and useful for users.
Speaker 1:So, information science shouldn't be confused with data science. These things, of course, speak to each other. But informatics is the focus on storing and making useful in an organized and transparently logical way for a range of users, and I want to home in on this point for just a moment. Is we think collectively about the ways in which museums, libraries and librarians find themselves facing significant funding cuts at this moment in the United States? And just to say, there's a lot that we will be missing if libraries and librarians are not able to do their job of preserving and organizing and presenting information. So informatics of course speaks to broader fields of work. For us in the Currie Lab, it's about how we can capture, preserve, store, organize and make available information about cultural heritage.
Speaker 2:When you were talking about your first meeting together, I believe you mentioned the project that related to Egypt. Would that be one of the first projects that Curia Lab started on? I'd seen a list of projects. One of them related to archaeological looting in Egypt. I believe One of them related to archaeological looting in Egypt, I believe, so I wondered if that was it or if you could expand on that. That was Michelle's.
Speaker 1:PhD project and she should tell you about it because it has continued to analyze the ways in which satellite images shaped the field of then developing knowledge about looting in Syria and in the Syrian civil war. Of course, for many years, but especially in the 2010s, archaeological sites and archaeological materials were significantly damaged, dispersed or destroyed by the war. We knew that the question of looting finance coming out of the sale of archaeological materials was a significant interest to policymakers in the United States, so that study looked at how archaeologists and policymakers and satellite image scientists were using remote sensing imagery to generate knowledge about this set of activities. We really were together, interdisciplinary, multi-methods in the work and asked our participants a range of questions about cultural heritage damage in Syria, arrived at estimates of damage and how that might have been influential in some of the estimates that were then coming out. Michelle mentioned that before she met me, she was aware of my article. It appeared in a couple of different online platforms questioning the figures that had been discussed widely in the press and in policy circles about how much the Islamic State in particular was earning. I saw figures of $6 billion US $7 billion US. This seemed unlikely, but I wanted to understand, with rigor and better data how these figures could have been sustained.
Speaker 1:So that was a separate project that was built out at the University of Chicago, but just to say, our first project together, I would say, was an extension of that set of questions about what was happening with cultural heritage in Syria and the Civil War, and what that NSF grant allowed us to do was go deep into one particularly important source of information that was then in use in some pretty novel and untested ways. We wanted to know how robust and reliable the new forms of information were for assessing the scope of the looting trade, and I'll pass it to Michelle and either of you.
Speaker 2:Michelle, certainly you pick up from there, but I, just before we move on, you referenced participants and I just wondered if you could expand on who they were and this novel information that you're referencing.
Speaker 3:Our participants range from a wide swath. We sort of broke them into those three categories of the technicians who work with satellite imagery and produce satellite and other open sources of data. Then the researchers, who then take any data that they have access to. The specifics of the participants.
Speaker 3:But we tried to intersectional a group as we could, so that means we had people who were really familiar with the algorithms that are used to, for example, do orthorectification, which is the process of pinning satellite imagery or other types of aerial imagery to the surface of the earth so that they can be geolocated and useful analytically. We also had terrorism and counterterrorism specialists. We had people within the government, within policy offices of the government, at different levels. We tried to incorporate archaeologists. We incorporated some criminologists, really as much as we could, to look at this sort of briefly interdisciplinary area that was focusing on archaeological looting in this terrorism and conflict space of an active conflict and understand how things like validity and reliability were being considered in the production of data and what it means to produce data in the context of urgent science.
Speaker 2:I had seen your presentation. I mentioned before we got on the call and you had referenced in it, talking about this Syria project, that Aleppo was the exception. Could you kind of describe a bit about what that means, what your kind of an overview of the information that you had from the other areas and what was it about Aleppo that made it the exception?
Speaker 3:Yeah, so that's actually a second project we have on Syria, where we were really. That project started with this goal of wanting to look at the intersection between civilian fatalities and civilian casualties in Syria and attacks against cultural heritage, trying and this was also funded by the National Science Foundation trying to ask again with high rigor and formal methods to what degree can we say that these two things are related? There have been historical claims that they're related, that some are much clearer than others, but it's an empirical question and in the context of Syria, there have been a lot of calls for action based on the supposed relationship. But in a complex geopolitical environment, calls for action ideally should be evidence-based and we were hoping that we could provide some evidence either for or against calls to action. That particular line of investigation ultimately, if you watch the presentation, didn't work because there are always trade-offs when you collect data and when you are collecting data on an active conflict, which any conflict data source is doing at any given time, you are just processing thousands and thousands of pieces of information constantly and, as it turns out, every worthwhile conflict data set that we found does not provide the actual geolocation, the actual location of where an incident that has associated civilian casualties happened. They instead sort of cap it at the town, which makes perfect sense. It also makes any statistical analysis of the relationship between where civilians have had casualties and where cultural heritage sites have been attacked virtually impossible.
Speaker 3:So we have this evaluation has evolved to then consider other relevant factors, specifically, for example, the relevant the the effectiveness of no strike list designations for cultural heritage. There are a lot of really good reasons to put cultural heritage sites as no strike locations, reasons to put cultural heritage sites as no-strike locations and, assuming that those designations work, we should see a protective effect, meaning that if there is a church that is designated as no-strike, then it is more likely for attacks to also avoid the immediate surrounding area, sort of creating these buffer zones of safety for civilians and also the heritage itself. However, conflict is messy and there are also strategic reasons then to try to target sites that are known to be on lists that you know protected location lists, and so there's also a risk that being on a list of no strike would make a church more of a target than it otherwise would have been. You might see this if an adversary ends up saying well, we know, this church is likely on a no-strike list, which means they can't hit us here, so we're going to harbor here and then the heritage ends up getting damaged or destroyed through other means or through this sort of microdynamics of conflict and operations.
Speaker 3:So you asked about why focus on Aleppo. Well, Aleppo had really intense territorial control battles during the course of the height of the Syrian conflict. It also has very clear neighborhoods. It has a lot of really detailed focus because of the intense territorial control change elements, and so it has both a majority of incidents for cultural heritage attacks. It has a high number of no-strike locations compared to the rest of the country, and then it has really good coverage in terms of data availability. So it's considered special in the context of Syria because so much focus was there. It means that it's a little bit different than the surrounding areas from an analytical perspective, but that makes it a really fantastic case study at the same time, if we're only looking at Aleppo.
Speaker 1:And because of the nuance there, there's just a lot that we can unpack. There's one more piece I'll add to that Stephanie, the second project. As Michelle has just described, it started with curiosity about what's been called the prelude hypothesis. This is the notion that attacks on culture sometimes precede attacks on people, and there's this idea that somehow, by registering moments when an aggressor party is attacking culture, writ large can take different forms, that we can use this to predict when and where attacks on human communities might follow.
Speaker 1:This is a really important hypothesis to analyze because it has been, as Michelle said, it has been presented as the reason for pretext I should say pretext for military action. And so, if it's true, this is a really valuable observation and one that we ought to anchor in data. But if it's not true, this is also important because then probably military action as a pretext isn't going to help. It might even inflict unnecessary damage. Studied Syria because of the availability of a few different data sets, and our contribution was to bring together different kinds of data that hadn't yet been connected or put into a quantitative space-time methods framework. So we worked with the Uppsala Data Conflict Program in Sweden, one of the world's best sources of information about civilian casualties in war, and then we also worked with publicly available reports from the Cultural Heritage Initiative of ASOR, and this was a project that was done with funding from the United States government to try to document damage to cultural heritage sites in Syria.
Speaker 1:So what Michelle just said I don't need to say again, but this has been a really complicated, multi-year study. We cannot prove the prelude hypothesis. That itself is an important finding, but nor can we tell you what exactly is this relationship. So how is it that in Aleppo, a high number of mosques have been killed, and so have civilians. Is that simply explained by collateral damage in a high kinetic warfare environment, or is there something else going on? And so this is why we continue to study it and continue to be really open to conversations with other scholars who are also assessing this really important nexus of questions about violence in different forms nexus is questions about violence in different forms.
Speaker 3:Yeah, we can't say anything clearly like specific about this relationship, but one of the things that we have found that's really, really fascinating is that you can trace some of these territorial changes through the different patterns of attacks on cultural heritage.
Speaker 3:So there are some clear.
Speaker 3:There are specific neighborhoods in Aleppo that are just systematically bombarded, and then there are other neighborhoods that have just very precise, almost looks like someone took paint and did a scatter shot.
Speaker 3:It's one month here, it's one month there, and that changes. And we see that pattern change by type of attack as well as by reported actor, although attributing actor is challenging in a conflict, and so we don't go any more specific than state and non-state, which is, you know, a very broad brush. But you can see this when territory changes, the patterns change and so there is something there, there is a pattern, there is a dynamic at play, a potential for understanding this dynamic, to help us understand why heritage is attacked in conflicts and when and by whom. But the data are very messy. We're working with basically open source intelligence on cultural heritage attacks from at this point now, 10 years ago, and so there's a lot of messiness there that we are very thorough and diligent in trying to account for and to let speak for itself, without us trying to impose any of our own contemporary assumptions onto it.
Speaker 2:Was there a similar hypothesis going into the work you've been?
Speaker 1:culture then yes, and that's not a hypothesis that needs testing, that is so thoroughly documented at this point that we can move to deeper questions, including what are the effects of Russia's all-out assault on Ukrainian culture? That all-out assault has taken multiple forms. Some of those forms include what we would identify with traditional forms of cultural heritage museums, archives, monuments, artworks.
Speaker 1:We could think of the built environment of culture and history. But what's been less discussed in mainstream media, but not less discussed in Ukraine, among Ukrainians, is the more subtle forms of Russia's attack on Ukrainian culture. One example is the assault on sources of knowledge and libraries. So before the full-scale invasion, ukraine had 14,351 functioning libraries. As of December, that number had gone down to just over 11,000. There's been an almost 20% drop and there are communities now across the country where there is no functioning library.
Speaker 1:Libraries matter. They matter because in Ukraine they are community spaces. They often have events, often have a safe, warm place, which, in a country where there are often power shortages to the domestic utility supply, is really important. People can go be warm, find shelter and read a book or just be with each other in safety.
Speaker 1:And of course, along with the attack on libraries has come deliberate attacks on Ukrainian language books. So there have been book burnings, documented across the country, done deliberately, and I mean bonfires of books taken out of people's homes, out of libraries, out of churches, piled up and set on fire. This is not collateral, this is absolutely targeted. Also, in the occupied territories, the removal of Ukrainian language books from schools, from universities, from public facilities and their replacement with Russian language books. So this is important because when we think about how cultural heritage is supported and protected, it's essential that we understand the cognitive underpinnings that give us knowledge about those cultural heritage sites and objects.
Speaker 1:Other aspects of Russia's assault on Ukrainian culture have included the mass appropriation of archaeological materials and artworks from museums. This is something that has been well underway since shortly after the full-scale invasion, but then, sadly, in Crimea, which has been under occupation since 2014,. This has been going on for even longer and there's much more to say about that in the archaeological area. I would just underscore that we have enough information and Ukrainians are working really hard to try to document all of this. What we need to understand is what the effects are and what are some of the pathways to accountability.
Speaker 2:And that was a question that was coming to me was the work that you do to inform accountability. How do you use this information and what does that look like? Are there other examples where you've done that?
Speaker 1:I start by saying I'm not a lawyer, and Michelle and I are equal in this in being committed to thorough, transparent and rigorous description and analysis, with data made available for replication studies data made available for replication studies. I don't advise on policy because I want to be able to take my work and have it be useful to a range of policymakers and lawyers and, stephanie, I've learned so much about this from your past guests who are lawyers with real expertise in art law and cultural heritage law. But I don't want to dodge the question. I do have some thoughts on this and how we might be supportive of accountability in the work that we do. One of the cherished principles of our collaboration through Career Lab is transparency in our methods, and what this means is metadata when do the data inputs come from? What were the transformative processes that turned so-called raw information into data? How did we code things? And then, what were the steps that we undertook to try to account for human subjectivity, which is always a factor when creating data for use. Then also methods what are pros and cons of the methods, and so forth. So this sounds perhaps obvious or fundamental. It's not. It's actually really important because, if we think back to this earlier example of $7 billion supposedly raised through the looting and trafficking of artifacts from Syria. We found nothing in the way of robust and reliable data that was supporting that, and yet that was a figure that had enormous repercussions for how cultural heritage was thought about and still is thought about and especially archaeological materials moved out of a conflict zone.
Speaker 1:In the specific case of Ukraine and Russia, my own beginning in this space started as a cultural crimes investigator for a project called the Conflict Observatory Ukraine, and the Conflict Observatory Ukraine was funded by the United States Department of State as an independent and autonomous research project to document activities that might be prosecutable or investigatable as war crimes. So I worked with a team of archaeologists and anthropologists through the Korea Lab to study, to document, I should say, activities. We were actually part of a much bigger consortium that also included researchers at Yale University and at George Mason University. So I began as an investigator and documenting things through open source information to support our work in these areas. And, stephanie, I want to just come back to your broader question about the ways in which we see our work supportive for accountability. The phase of fighting right now in Ukraine and Russia is such that we might need to wait a bit longer to understand what post-war looks like, but what's important to emphasize is that Ukrainians themselves are already hard at work on restitution, reparations and accountability. This takes the form through documenting alleged atrocities, investigating alleged atrocities and working together, collaborating across different Ukrainian NGOs and community groups, to document not only what happened but what effects they're having. Now I want to be as supportive as possible of this in my work.
Speaker 1:My project ended the conflict observatory. This in my work, my project ended. The conflict observatory unfortunately has been closed and there are no more investigations being offered by that program. But what we are doing now is transferring data to groups and accountability platforms that might be able to take the information forward, and there can be a couple of different ways that that takes. So a thing that Michelle and I think very rigorously about is data transfer.
Speaker 1:How do we actually tell the lawyers, the prosecutors, what the value of the findings is? And because your listeners are typically very familiar with cultural heritage, I can imagine that the kinds of reports and investigations we do would have very familiar and resonant themes, but that's not always the case for a war crimes prosecutor who has not had training in archaeology or anthropology or cultural heritage. So what we can also offer in support of accountability is a user guide, is a framing, is a way of helping them to understand what's in it and why it's important for culture, for communities, for history, even though we don't go so far as saying here's the crime that was committed, because that really is for the next set of specialists to tackle.
Speaker 2:Yeah, and I would say, michelle, I'd love to hear your thoughts on this. And then I'll just throw another question in is this user guide? Would it be specific to each project? Is it something that you're gathering now for? Yeah, okay, I'll let you both take that.
Speaker 3:Fiona, do you want a last thought on that?
Speaker 1:Well, actually, michelle, I was pointing to you because I think you can talk about metadata and data dictionaries, and this is the nuts and bolts of data rigor. Stephanie, before that, I will just say yes, there's a qualitative aspect to a user guide that is case specific and that would be, for example, contextualizing archaeological disturbances in Ukrainian sites and settlements, such that investigators, lawyers, would need to understand some of those specific qualitative elements that wouldn't be necessarily relevant to a set of accountability activities happening in another part of the world. Michelle, what do you think?
Speaker 3:in another part of the world.
Speaker 3:Michelle, how we think about data transfer, these user guides, the framework. A lot of it is making connections to things people already know, but connecting them with the why and the how for cultural heritage specifically. So, for example, one of the things that I do in CuriaLab is I post trainings, and some of them have been how do I look at a satellite image and identify a looting pit? Or how do I look at a satellite image and track conflict developing on a border and know that what I'm looking at is actually a buildup of military vehicles and not Amazon's new warehouse in Belgorod? Right, they can look very similar and one is suspicious and one is less suspicious, but what's relevant? The other thing we do is then incorporate the OSINT right. So my whole, when I teach my students a lot of these user guides, it's about thinking about how the trade-offs in data creation translate to a context. So when we're talking about metadata, we talk about everything from what are the platforms you're using to search, how you're refining that, what are the softwares or versions of softwares you're using to then do the analysis. Where are you storing the information? You can almost think about it like a chain of custody that we're creating that by its creation and the transparency with which it's created lends validity and credibility to the data that is produced. So helping people to understand that component can help them understand some of the inherent value of the information that we're providing. But it's also understanding why is this necessary or relevant for cultural heritage? Osint, for example, is very often applied in the cases of legal accountability for human rights violations, so why is it also equally relevant and important for cultural heritage potential crimes? There's an intersection between human rights violations and cultural heritage crimes. Sort of explaining the context of cultural heritage and how it is used or misused in these variety of contexts. That's sort of what we're talking about here as well when it comes to these user guides. So there's sort of the practical logistic here's what to do with what we've given you and how you can reproduce it, and then why you should care in the first place that this has been produced and what it can then be used to say. And so I'll give an example, if that's okay. That's not in the context of legal accountability but does demonstrate it.
Speaker 3:Fiona mentioned my PhD project Egyptian Archaeological Looting. The goal of that project was to establish a baseline understanding of which archaeological sites in the Nile Delta are most likely to be looted and in response to which kinds of factors, thinking through environmental, like drought or crop failure, economic stress, geopolitical stress, terrorism I looked at all of them, and equally important to that was can I produce a methodological approach to creating a data set on archaeological looting attempts that covers a variety of space and is over time, that can then be used again later on to answer other questions, For example, to do a prediction of where looting is going to appear in the future, based on this understanding of the past? And so the actual dissertation is one giant methodological handbook that includes both what to do and what not to do, because I was a PhD student and I did some things that I realized were wrong but were very understandable as an approach, and so it became as important to talk about what went wrong as what went right. This has then expanded, so I now am looking at eight years currently of archaeological of satellite imagery in the Nile Delta, of archaeological of satellite imagery in the Nile Delta, and instead of looking at just 140 sites, I'm looking at the and I'm going to heavily air quote known universe of archaeological sites in the Nile Delta, and we're using unsupervised classification algorithms to try to generate the same type of data set. Except, this is eight years' worth of data that we can use, and so we can develop a much broader understanding of a baseline.
Speaker 3:And the eight-year period we're looking at includes the onset of drought and climate change-related stress on the environment, the collapse of Egypt's currency, two coups, multiple riots and political violence instances, instances of terrorism and also COVID, so there's every possible stress that the country could experience in this window, which means we have a rare opportunity to actually generate information on what can be done in the future. We're generating usable data that future scholars or policymakers can query. The tool itself will be publicly available and open source so that anyone can take it and use it to help and apply it in different contexts, with associated documentation. And it's all still in this framework of how can we produce information and data that can be transferred for others to also help and make good. So we try to do what we do, with rigor, with transparency, with documentation, to produce a valid result that we can then hand off to more specialized groups that will make use of it for their needs.
Speaker 2:Thank you. Over the course of this time, since your PhD till now, and the rise of technology and innovation, I wonder if both of you, or one of you, could speak to how your rigorous approach towards data, how it has evolved and perhaps become easier for you or more complex, how has the use of technology and innovation impacted your work?
Speaker 1:Has the use of technology and innovation impacted your work? I'm happy to start, Michelle, and then I think you should take on AI and some of the really cutting edge things that you're doing now. Stephanie, one of the key changes in my thinking about rigor is how to handle disinformation and that is to say, the deliberate manipulation of sources of information, and this has been especially important in working on Russia's war on Ukraine and attacks on cultural heritage. And that's because the Kremlin operates a sophisticated program of manipulating information, text, images and objects. And, as we learned in the course of being trained as OSINT investigators, it's not enough simply to capture digital assets in a safe way. We also do need to have a secure chain of custody, as Michelle called it, and that is to be able to trace back to the original source.
Speaker 1:Where did that post come from? Whether it's a social media post, an image online, a claim and I'm not even talking here about deep fakes. I really mean the kinds of text-based claims that are made about sites, objects and past events that are manipulated deliberately to twist a narrative. This was not something I thought about very much in the course of that first project on Syria. There, I assumed that if people had access to correct information, they would want it in order to improve epistemically improve the knowledge formation process, that is to say, give them a better understanding of how I don't know the art market in archaeological materials works, and then they will, you know, be cooperative or interested in how that can transform into other practices or policies.
Speaker 1:So now I'm situated in a space in which that's not true, and this has been an important lesson in thinking about how we use our data, our input and information, how we think about it, how we don't play into some of the narratives that seem to come up again and again, that are problematic in these ways that I have described them. So that is you know. I think that if we want to put like a fine point on it being very careful to rely not just on one source of information but several, Do some due diligence on where the information come from, Triangulate sources of information and then preserve the traces of data, of digital assets, so that other folks working in the same research environment can then also benefit from the data provenance and trace back in their own research findings. I'll pass it over to Michelle for some different angles on this question of methods, and she can also talk about AI.
Speaker 3:It's gotten complicated. It was already complicated. Yeah, I mean it's gotten complicated. It was already complicated, but I would say that I have come to more appreciate new layers to. It was probably coming down the pipeline someday, and so the goal was to build in space for it in the future, sort of lay the groundwork with what I could do now, and then know that at some point in the future, somebody smarter than me is going to develop an algorithm that I can take and use, and that to do any of this kind of work did require rigor and it did require thoughtfulness and intentionality, but it was more thinking about initially. What is the sustainability of these data? Who are the end users? What might they want from it?
Speaker 3:As Fiona mentioned, the landscape of what is considered fact or truth has dramatically changed, so it is more complicated now, because to do this work requires an inherent and intense amount of skepticism. The assumption that I start with is what I'm looking at is not what it seems to be, and then I need it to prove to me that it is, and that can be as simple as you know a well-intentioned news story that is just bringing the picture to repost from someone who posted it five minutes earlier because they're trying to beat the media rush. Posted it five minutes earlier because they're trying to beat the media rush, not doing due diligence and understanding that what that picture is representing is an airstrike in Damascus from five years earlier and is not involving Aleppo at all. It has nothing to do with Aleppo and, in fact, the story in Aleppo is fake. But because they're in this rush, this sense of urgency, their priority is different than my priority. And when it comes to something like algorithms, they can be highly beneficial if you know what they're doing and how they work how they work. But if you don't try to explain the black box or you don't know what they're doing, then you have no context for what they produce.
Speaker 3:And we see outcomes of this both in terms of just you know, as I'm a professor, think about AI generated assignments. You know they're not good, we know that any you know someone asks how do I write a paper? Chat GPT will put something out. It's not going to be a good paper, but it will be a paper, probably. But you also see it in the context of manipulation of things that we consider to be immutable, of manipulation of things that we consider to be immutable.
Speaker 3:Photographs have been edited for a long time, both pre-smartphone, age and post. Think about any filter on Instagram or TikTok or a social media platform that's manipulating the video or the image that they're seeing. But also now think about DALI and generative AI and satellite imagery, a thing that we consider to be an objective form of truth. You can easily add a cloud to cover something strategic and seamlessly blend it in. You don't even have to work that hard. Adobe will do it for you. So that used to be really difficult, that used to be sophisticated, and now it's not.
Speaker 3:So everything that I look at to create a collate and synthesize and produce a valid and reliable data set starts from a perspective of skepticism, and this really speaks to what Fiona was talking about triangulation, different types of sources, not just news, not just imagery. We need as many different perspectives as we can get to triangulate that. The thing actually happened as it said, it happened where it happened when it happened, and all of that then needs to be appropriately archived. This places a much higher burden on the researchers and the analysts who are doing this work. It slows it down, and so it's a double-edged sword. Algorithms can also be highly effective for assisting in speeding up some of that process.
Speaker 3:But the distinction that I'll draw there, when it comes to things like algorithms or AI, is is it an algorithmic tool that you can use and explain, or is it generative or is it generative? Something that's generative is not going to be explainable in a way that is satisfactory from a research or policy level and is, at its core, going to be plagiarizing, so not acceptable as a source, whereas an algorithmic tool, for example, can help to streamline things that would previously take a long time. If I get a data dump of raw satellite imagery, which can be at least 1.5 terabytes, I either have to manually click individual hyperlinked a set of hyperlinks in order to manually download every single file in that data dump, or I can work with GitHub Copilot, which only pulls from GitHub and therefore is considered slightly better of a repository to train on and develop a Python code script that will loop through and do that for me. It takes three days of downloading down to 10 minutes and you can tell immediately if it's worked or not as appropriate. So there's a distinction there and it's always a double-edged sword.
Speaker 3:The other thing I'll say is, in this increasingly technological world, the work that we do in other countries is different now.
Speaker 3:There was always a sense of geopolitical risk involved in doing the type of cultural heritage work and answering some of the questions that we try to answer in active or recently active conflict zones. But now the world is very interconnected online and there's so much data and so much information out there that if we can find information about a potential target of interest, they can likely find information about us. So, just as much as it's important to have a secure chain of custody for any documentation or information, you also have to be aware of your environment online and what risks you're willing to take in order to do the work. So protect yourself, use a VPN at a minimum, try not to do your work at home and you know there are other tools out there that can be helpful, using a virtual machine inside of a computer or, if you have access to it, a SCIF if you're really concerned. There are ways of mitigating these things, but you should be aware of them and the context that you're working in, so that you don't have an unfortunate encounter.
Speaker 2:The complexities. Thank you for laying all of that out, all of it. What was ringing in my ears was this concept that you both had raised, I believe, before the call about, and maybe earlier in the call about, data justice. I would love to hear your thoughts about the work you're doing, how you see it, speaking to justice, not just data justice, but in many layers of it.
Speaker 1:I'm going to flip the question and start with injustice, and in doing so I'm inspired by the work of Miranda Fricker, whose book Epistemic Injustice Power in the Ethics of Knowing has been influential for me and for many of us working in cultural heritage. Frickery is a philosopher and she begins with injustice because she says that we have a clear idea of why something constitutes an injustice if we can first describe and analyze the nature of the wrong that's been inflicted. I say this because I think, before we put the question of justice out there, we have a responsibility to understand deeply what the injustice is or was, and for me that involves listening to different people or communities impacted by what happened, the ways in which that has been played out temporally and spatially, especially temporally, since often with cultural heritage and justice, we're talking about wrongs committed many generations ago, so there could be implications for one's ancestors, in addition to the living community. And then, I think, also being open-minded about the different kinds of institutions who are involved.
Speaker 1:I don't think it's productive to start with a position that museums are the enemy. Museums are complicated, but there are many museums doing really important work and they are staffed by curators, conservators, docents who care a lot about the ways in which they are custodians of objects and stories and also relationships with people with a direct interest in those objects. So to come back to Fricker and just like, why injustice? At this point in my career, I'm thinking specifically about epistemic injustice. Fricker breaks that down into two forms testimonial and hermeneutic.
Speaker 1:Testimonial is the idea that we have equal standing in understanding or having information about something. And testimonial injustice occurs when somebody, for example, isn't regarded as having a legitimate interpretation of an object because they don't have a PhD or a master's in museum studies. And then hermeneutical injustice the other side of this is actually not having access to the resources to make interpretations of one's own social experiences or history. So what I said earlier about the destruction of libraries and books could be seen as an example of hermeneutical injustice if a community is actually cut off from the sources of knowledge that would help them to understand themselves. So I am now working in this framework, because for me it is helpful, in filling in a holistic picture of knowledge, culture and societies, to gather information and different inputs in such a way that we become temporary custodians of this information and that what Michelle and I and our collaborators are doing is effectively serving as stewards and researchers in this moment, who can then be sharing our findings with others who can benefit from it.
Speaker 3:Absolutely. For me, coming from criminal justice, the question of what is justice is very sticky. You know it raises questions of justice for whom? And for me, that resonates in the work that I do in Curia and that we do in Curia in two ways. The first is an ethics of practice the recognition that if I'm going to be working on data about Syria or about Egypt, then that information should be relevant to them and there should be access. So some of that is data access, some of that is sharing results, some of that is trying to form partnerships and to listen and place whatever I'm finding back into a, a local context that is then relevant and adding meaning. So then, the end goal of, for example, any of my analyses as a methodological person is not just the results, but then to go to people, who, to whom it would matter the most, and say can you, let's have a conversation about this and see what do you think of it, what's missing, what, what does this mean to you? But it also then, on the other hand, it for me, raises this recognition that a legally satisfactory resolution in the cultural heritage system or in the legal system in general is not guaranteed to be the same thing as justice, and so that comes back to the question of justice for whom.
Speaker 3:Is it justice for the community of origins? Are we thinking about justice for? You know we've had several episodes about descendant communities justice in the eyes of the legal system. Those are not guaranteed to be the same thing and they may be mutually exclusive. So I don't want to be the person to decide what is justice. As Fiona said, we are transient caretakers of information that we hope will be of use to the people who are seeking justice, whatever that looks like for them. But the minute that I personally decide this is what justice looks like, then I'm limiting who could benefit from what we're trying to work with and what we're trying to produce. Keep that recognition and that complexity in our minds, while not necessarily putting our own interpretation onto it.
Speaker 2:Thank you, Thank you both. Is there anyone who has any questions? Hi?
Speaker 1:I have a question. I'm Lauren, so I know you talked about the institutional support you received from the National Science Foundation and the U. What is the role of the universities that you teach?
Speaker 3:at in Curia Labs. People who are entirely soft funded or reliant on soft funding are jumping from grant to grant. So, for example, with the Syria work, being at an institution like my university allows for more time to dig into these questions and to be as thorough as this work requires, and but also it provides a really, I would argue, valuable training ground for future. Hopefully, people who are at least doing reliable and rigorous work, and then I also host the trainings that Curia puts on at the university. So there's some, there is institutional support from the university, as well as resources that allow us to do our work.
Speaker 1:And Lauren, I would add, students, students, students. We have trained students since the very beginning and some of those students have trained us because they bring important and really awesome knowledge, ideas, questions and capabilities and really awesome knowledge, ideas, questions and capabilities. And so our institutions have provided us with the mechanisms and processes necessary to recruit, to identify, to hire and then support students. So it's been a combination of grant money and university money that's actually paid them, but we have both been committed to paying our student researchers good wages, living wages or above, not the kind of like entry level wage, because their time's important and their ideas are important. Then we also have benefited from university support through libraries. Have I said it already? I really cherish our librarians, computing services and storage. So a lot of in-kind support as well. Thanks.
Speaker 2:Thank you. I would just ask one last question, then, of both you, fiona and Michelle what is the mark that you hope to be making with this work that you're doing?
Speaker 1:I want one of our marks to be establishing frameworks of cooperation and collaboration that cherish equal access to knowledge, equal access to information, because I think that's one of the best ways to get to equal access to accountability. Not all of cultural heritage is, or the cultural heritage work that we do is best framed as a legal problem or a legal question, but almost everything that we do is a knowledge question. In fact, I would say, michelle, that all of it is right. I think another vector for us now will be helping the non-cultural heritage economists, political scientists and sociologists and so forth to understand the importance of cultural heritage in the work that they do, so, even if they don't see themselves as cultural heritage specialists that they recognize that this is more than just a variable. This is a central feature of human communities and ought to be accounted for as such. And I'll end that note and then pass it to Michelle for a final thought.
Speaker 3:I agree with everything Fiona said and, as part of that, this is not unique to cultural heritage spaces.
Speaker 3:This is pretty common in a lot of social science fields and probably hard science fields as well.
Speaker 3:People tend to work in silos and our approach is to say everybody has a place at this table, as Fiona was saying, and that the methods.
Speaker 3:We don't have to wait for other people to come around and say this is how this works. There was a perception in cultural heritage studies for like 10 years that there was no data because nobody had bothered to put it into one Excel spreadsheet, and so Fiona and I looked at that and said, okay, we'll put it into a spreadsheet and we'll do it using methods from other social sciences who have figured out how to do this well. So one of the lasting marks that I would hope that we're able to see is a shift in approach, that there's the recognition that other disciplines have in the social sciences have a role here as well, that there's value in sharing knowledge between countries, between organizations, between disciplines, and that these frameworks and methods can ultimately help us move towards proactive approaches here and that those methods, if we can bring people to the table, there's benefit for everybody in doing that and that ultimately, cultural heritage will benefit, social science will benefit, society will benefit and we can push forward instead of sort of sitting in cycles.
Speaker 2:I so value and appreciate this interdisciplinary, collaborative approach that you both have. It's really really to be embraced and celebrated, and it's rare, I think. So thank you so much for all the work you're doing and this approach that you're taking to it to be here.
Speaker 1:Thanks for the conversation and for the work that you're doing on your podcast throughout your episodes to cast a bright light on different ideas and work being done in the field.
Speaker 2:There will be links in the show notes to learn more. If you are intrigued by this podcast, it would be much appreciated if you could leave a rating or review and tag Warfare of Art and Law podcast. Until next time, this is Stephanie Droddy bringing you Warfare of Art and Law. Thank you so much for listening and remember injustice anywhere is a threat to justice everywhere.