This text is also available to download as a PDF, as is the entire Capture, Connect, Shift project.

* * *

In Invisible Cities, Italo Calvino describes two vantage points from which to arrive to the city of Despina (Calvino 1974, 54–55):

Despina can be reached in two ways: by ship or by camel. The city displays one face to the traveler arriving overland and a different one to him who arrives by sea…. Each city receives its form from the desert it opposes; and so the camel driver and the sailor see Despina, a border city between two deserts.

Port cities, like Calvino’s Despina, are “borderscapes” (Mezzadra and Neilson 2013, 9) straddling two geographic media: land and water, a liminality that is central to processes of moving and mooring; of circulation and exchange. This dual orientation marks port spaces as constant sites of arrival and departure. Some regulated, some clandestine, these arrivals and departures give a rhythm to life along this threshold of land and sea (Dening 1996). Sometimes, certain arrivals muddle distinctions and temporalities—objects wash ashore; ships run aground; tsunamis churn up rubble from the hidden depths of the sea. In these moments, classifications and distinctions are sought and remade (Tazzara 2017; Hecht 2023, 16n57; Iodice 2025). Flotsam and jetsam, gift and theft, are not transparent descriptors. They rather encapsulate legal and ethical arguments about movement across the threshold, arguments which emerge from within encounters and arrivals (Shryock 2012). What arrives in and departs from ports and littorals is often referred to as “cargo,” increasingly signified through deadweight tonnage (DWT), the unit of measurement that quantifies the weight that a vessel can carry. Cargo, this abstraction, known through categories like “dry goods,” “bulk,” “oil,” and “container,” makes the specificities, geographies, and histories of trade disappear, thus concealing the relations of exploitation, violence, and commodification at the heart of maritime mobility (Baucom 2005; Sharpe 2016; Dua 2024). Here, we focus on a specific sub-category of cargo, referred to often as “toxic cargo,” in order to understand the relationship between evasion and investigation in the making and unmaking of relations across space and time.

Relations across the Mediterranean and the Red Sea appear in the written archive as early as first-century, Roman sailing manuals that listed in rich details a diverse port geography from the Mediterranean to the Red Sea and the Indian Ocean (Casson 2012). From the mid-sixteenth century an imperial infrastructural imaginary sought to cut a sea-lane through the Sinai Peninsula to connect the Mediterranean to the Red Sea and to ease access to the riches of the Indian Ocean. The Ottoman grand vizier had planned to dig a canal between the Mediterranean and the port of Suez in 1568, but the idea was abandoned after surveyors deemed it infeasible (Casale 2010, 135–37). It was only in the mid-nineteenth century, when the semi-private Suez Canal Company took on the task, that the idea came to fruition. The opening of the Suez Canal in 1869 marked a watershed moment in world economic history. The distance one had to travel from London to India was reduced by almost half, and the volume of goods that made their way to the port cities of the Indian Ocean increased exponentially. Empire followed in the wake of the canal’s construction as British, French, and Italian ships entered pre-existing worlds of trade and connection and ports like Massawa, Djibouti, Aden, Berbera, and Bosaso forged new routes and connections that continue to this day.

One such route connected Bosaso in Northern Somalia to ports in Italy between the mid-1980s and the early 1990s. It channeled toxic, and perhaps also nuclear, waste. That route added Somali ports to a list of others around the world—Venezuelan, Romanian, Lebanese, Nigerian, to mention a few infamous cases—that became destinations for transnational toxic waste circulation in the 1980s. What sets the Italian-Somali route apart is that it is the only one that is both known and has not triggered a demand from the country that received the material that Italy excavate and retrieve it. Thus, there is something that is disavowed, acknowledged and evasive, hidden but also visible in these circuits of toxic circulation (Abu El-Haj 2023). These toxic circulations, we argue, emphasize the centrality of relations—corrosive and opaque but nonetheless relations—that are sustained over space and time; their corrosion persevering, through such opacity and “latency” (Liboiron 2021, 100). Through an examination of how relations are made—simultaneously legible and opaque—we reflect on the multiple registers of evasion that structure toxic waste’s transnational transportation. Thinking through the trajectories of this “rubble” (Gordillo 2014) at sea and over land enables us to make visible the sinews of relationality that shape ports and other littoral spaces. We can thus consider the worlds of toxicity and obligation created in the wake of these relations—these arrivals and departures. In turn, our attention to the relations that facilitate toxic waste’s circuit permit us to see how toxic waste attains new, destructive, value.

We join scholars who argue for a rescaling of toxicity from portraits of “chemicals as discrete entities” harming bodies in place (Murphy 2017, 495), emphasizing in our account the transregional relationship of evasion where transparency, secrecy, and history all structure mobility. To “capture [these] forms of violence that are ongoing, accretive, yet also diffuse” (Fennell 2018, 523), we examine how people see each other not just as connected though toxic circuits but also related (Ben-Yehoyada 2017, 12). This entails not only following ships, but also relationships (Hughes et al. 2021)—racialized colonial and postcolonial relations (Kahn 2023)—of cargo logistics and the investigative logics and limits of liberal governmentality structured through the production and circulation of documents (Hull 2012).

Our focus on toxic relations within a transregional frame is also an attempt to work against the kinds of regional silos within which this story gets told. This is a collaboration between scholars of the Mediterranean and the Indian Ocean. Andrea Carnì developed his research interest in the case of the Italy-Somalia route while studying the international trafficking of hazardous and radioactive waste from Italy to developing countries. To understand it, he analyzed some historical and legal primary sources about this illegal movement and he came to study the Somali case in the scenario of the “bad cooperation” (malacooperazione) within Italian foreign policy. Naor Ben-Yehoyada came across this Italo-Somali “bad cooperation” case and the Italy-Somalia route within his work on the 2011–14 trial for the murder of Mauro Rostagno, the sociologist, journalist, and co-founder of extra-Parliamentary left-wing group Lotta Continua (continuous struggle), who was murdered in 1988 after two years of intensive investigative journalistic work in the province of Trapani. Jatin Dua encountered tales of containers washed ashore and barrels buried inland in Puntland during his research in Northern Somalia from 2008–2014 on regimes of risk, protection and piracy in the Western Indian Ocean.

Toxicity and Regulation

The unregulated or undocumented shipping of toxic and nuclear waste is situated at the intersection of two attributes of shipping cargo: the danger of carrying it and the risk in the discovery of the act of shipping it. The first entails shipping infrastructures and marine precautions, the latter—secrecy. These attributes do not always align. Various forms of toxic cargo that do not check both conditions have travelled in the broader period that interests us: the Cold War and its aftermath, throughout the Mediterranean and, more broadly, the European littoral (Paci 2024).

One captain who worked for twenty years in the national shipping company of a country that underwent arms embargo in the 1970s and was involved in other energy export limitations, recalled how before embarking from the company’s main homeport for a shipping voyage that entailed secrecy,¹ 

people would come with a portfolio with documents, which put everything under Liberian flag, with a Liberian home port. They would have gotten it from the embassy. It had all the appropriate signatures. If you needed to take jet fuel to Odessa or arms […] from Zeebruge, or export [dangerous chemicals], you also repainted the ship’s name and home port on the [ship’s] hull.

The “Sparky,” the ship’s communication’s officer also had to report to that country’s navy “every six hours our location, encrypted.” In one case, while the country was at war and under international limitations on what ammunitions could be shipped to it, the ship was about to carry arms back to the ship’s actual home port. When the pilot maneuvered the ship in port, “he knew to take you to an isolated pier, even though I saw on the way many vacant slots in the more central parts [of the port].” The pilot at least, also knew exactly where the ship had come from. “The pilot comes on board, and the first thing that he asks is for glass of [the country’s national drink].”² Similar voyages have caught investigators’ attention over the years. In the late 1960s, Italian ports served in shipping to Israel arms it was not supposed to receive. Italian prosecutors who were tracing the networks of neofascist terrorists came across how Italy’s unique position within NATO made it an actor in U.S.-drawn triangulations with Israel.³

The particular case of transnational maritime toxic waste circulation that we examine here is of the dangerous and unregulated kind. It is unregulated not in the sense that there are no documents about it. There are: forged, fake, false, or misleading. Similarly to exposed cases of breached sanctions (Yıldız 2020, 2021), in the aftermath of those voyages that do get discovered, documents proliferate: investigative, prosecutorial, testimonial, parliamentary (Carnì 2023). What remains crucially undocumented is a certain combination of the sender, the receiver, the composition of the material, and/or the route, rendering the paramount account of the circuit uncertain, and thus accountability for it largely unattainable. This form of undocumentation, we argue, reflects the centrality of opacity in structuring relations.

Since transnationally circulating toxic waste embarks on unregulated journeys because of someone’s need to move them offshore exactly in that manner because of their toxicity, it departs from the wider set of things—dangerous, secret, or both—that anyway travel through ports; as they do from the wider set of shipped goods, which are neither dangerous nor secret (Mellinato 2024). This opacity turns the choreography of investigation and evasion regarding toxic relations into the maritime infrastructure of transnational disavowal (Abu El-Haj 2023). That is so, because while official and independent investigations have revealed enough about this circuit to merit calls for justice and accountability, the dynamics of these investigations and institutional conduct, which have made it impossible to deny the circuit, can still ultimately dismiss it (Wedeen 2019, 164; Abu El-Haj 2023, 257). In our particular case, the dynamic of such disavowal is transnational and international, rather than personal or limited to the state.

This toxic waste circuit has received media and some political attention as soon as it was revealed and, more recently, academic attention as well (Hilz 1992; Hecht 2012; Müller 2023; Carnì 2024). Most famously, Greenpeace coined in 1992 the term toxic colonialism: “the dumping of industrial wastes of the West on territories of the Third World.” In her illuminating The Toxic Ship (2023), which follows the voyage of the Khian Sea, a ship that left Philadelphia in 1986 carrying 15,000 tons of incinerator ash, Simone Müller recounts the globalization of hazardous waste, environmental justice, and environmental governance in the following decades. Müller explicitly focuses on the kind of regulatory aspect that excludes nuclear waste. Her reason for doing so is that “[n]uclear waste […] constitutes a waste category legislatively, culturally, and historically distinct from hazardous waste” (Müller 2023, 6–7). Others have focused more specifically on nuclear waste and its afterlives (Hecht 2012). Studies on toxicity often foreground the particular materiality of toxic waste and the epistemic communities and fault lines that form around it: how bodies and communities engage with “the local, regional, and global forms of power that have made… [their] contamination possible.” This perspective reveals how “bodies are connected to commodity chains, to uneven relations of colonial/postcolonial power, and thus to world systems” (Agard-Jones 2013, 192).

In this essay (and in our broader project), we focus less on the particular substance, but on those cases where the nature of the material is mixed or unverified. First, because such mixing plays a role in the attempt to evade discovery. Second, because such a focus permits a shift in attention from the toxic material to the relationship that sustains its evasive transport, its disappearance and its, ambiguous if any, reappearance. The opacity of transnational waste transport, resonates with, and at times depends on the longer histories and geopolitical worlds of capture and evasion, within which these circuits emerge, become legible and, finally, disappear. We do not claim that this scenario is unique to shipping. We rather suggest that at sea we gain a privileged perspective of the transnational relations, political imaginaries, and colonial or imperial pasts (Ben-Yehoyada 2017; Dua 2024). These relations and histories are crucial to understanding circuits of toxic waste transportation.

A Shipping Circuit

The transnational shipping of toxic waste challenges our idea of global shipping for one important reason. Maritime shipping in the global economy has mostly been understood as the transporting culmination of an economic exchange: the shipment of merchandise following a selling and purchase deal. In this scholarship, shipping routes are made and unmade through transnational circuits of labor and financialization (Khalili 2020; Campling and Colás 2021). Toxic waste shipping, while built on a buyer-seller transaction, contains within it a deep uncertainty. The uncertainty and surprise regarding the ability to unload, bury, and leave underground or on the seabed toxic material means that shipment is an open-ended voyage, which can take the material, and at times the ship, through multiple continents and ports of call. Yet, as we will show here, these detours are not random but rather reflect longer histories and relations. Toxic routes, thus are routes made in media res.

While transnational toxic waste traffic is probably as old as nations, two key years structure the toxic routes we explore here: 1972 and 1986. Until 1972, throwing waste overboard was not regulated. That year, the London “Convention on the Prevention of Marine Pollution by Dumping of Wastes and Other Matter,” which went into effect in 1975, promoted the effective control of all sources of marine pollution (currently, 101 States are parties to the London Convention or the London Protocol).⁴ The second date, April 26, 1986, to be exact, is the Chernobyl Nuclear Power Plant disaster, which triggered a shift in public opinion and institutional treatment of nuclear waste and, by implication, toxic waste as well. In Italy in particular, a national referendum in November 1987 abrogated three laws “concerning nuclear plant siting and the activity of the public electric company Ente Nazionale Energia Elettrica (ENEL) in the nuclear sector” (Orsini 2022, 139). At that point, all traffic of radioactive material became illegal. The various forms of trade and transport that had proceeded more or less promiscuously until that point began to receive more attention. For example, until that point, there were no radioactive detectors in ports in Italy, so it was impossible to turn a ship away because it carried radioactive material. More broadly, the late 1980s marked a rise in public awareness to the movement of objects from urban and industrial heartlands to planetary peripheries (Caraher 2024, 52).

Until the 1987 referendum, nuclear material transit through or from Italy was a transparent affair. Most of these trades had been authorized on paper. As a result, they left a paper trail. There are documents from the Foreign Ministry that authorized all these trades (two examples are Nigeria and Lebanon; Carnì 2025). After the referendum, even that partial authorization was annulled. From that point, clandestine trafficking was investigated when it was discovered. What this means for us methodologically is the prevalence of a rich trove of documentation on transnational waste circuit that centers on Italy’s critical role as the originator and a major site of transshipment of toxic waste both when it was legal and then when it was illegal. This allows us to start mapping the ways in which shipping of toxic waste was categorized as waste or not, critiqued and made legible at multiple locales, from ports to governmental reports as we see below in the voyage of the Radhost.

In January 1987, three ships carrying over 2,000 tons of toxic waste left the Italian port of Marina di Carrara for the Red Sea. The Radhost, the Baru Luck, and the Lynx were contracted by a Milan-based company called Jelly Wax following agreements to bury this waste in Djibouti. In Djibouti, the French Navy blocked the ships and prevented unloading. After several days, they left toward Puerto Cabello, near Caracas, in Venezuela, where upon arrival, unloading commenced. Since this took place in early 1987, before increased illegalization, all loading and unloading was documented, including some wrongdoing and false reporting. Once those were discovered, the Venezuelan authorities forced the Italian company to reload and repatriate the waste. The Radhost made its way to Beirut, arriving in Beirut port on September 21, 1987. At that point, ten months had passed and still the waste had not been permanently deposited. In Beirut, the Radhost started to unload 241,131 tons of material for more than two weeks (until October 6, 1987). Meanwhile, Lebanese press and Greenpeace started mobilizing against the operation. During preliminary research, a former resident of Beirut told us that churches organized to toll their bells while the trucks were passing toward the Beq’a valley so as to encourage people to come out and block them. Following the uproar, the Lebanese government forced the Italian government to intervene, and the material—at least some of it—was repatriated to Italy in 1988.

The Italian government agreed with one caveat: the state refused to take responsibility for an affair that it insisted on considering as involving only companies, not states:⁵

It seems appropriate, in this case, to clearly emphasize that the recovery intervention carried out by the Italian Cooperation in 1988—of the approximately 12,000 barrels then found of residues of the Italian chemical industry Jelly Wax, transferred to Lebanon through the Ecolife Company—was merely humanitarian in nature and did not involve any implicit admission of responsibility on the part of the government, moreover never even aired by the Lebanese authorities. (Braga 2018, 38)

The Italian government’s rhetorical framing choice illuminates both the growing moral recrimination of transnational irregular toxic shipping and the postcolonial context in which it sought to operate as evidenced by the choice to frame this simultaneously as an entirely corporate affair and made legible through the discourse of the “merely humanitarian.” What we have here is not just evasion, but transnational disavowal.

From Evasion to Transnational Disavowal

Shifting our analytical attention from evasion to transnational disavowal, we must retrace how waste dumping became an object of investigation in Italy, what relations the circuit mobilized, as well as the relations that attempted to counter this circuit. For that, let us return a bit in time. Several reports and investigations mentioned Romania as one of the first countries supposed to receive radioactive waste in the 1970s and 80s, particularly the Sulina Peninsula in the Black Sea. Sulina is situated at the end of one of the branches of the Danube River, just south of Romania’s border with Ukraine, which runs along the river’s main, northern branch. As we shall see, Sulina does appear in some of the official Italian authorizations for transport and export of material, however the absence of facilities to process the waste in Sulina meant that any reference to Sulina was really a recognition that waste had been dumped at sea. The 2010 Greenpeace report on Italian toxic ships acknowledges as such in its conclusion: “The Italian magistrate investigating the case confirmed that much of the waste had been exported to Sulina, Romania, on board the ship Akbay-1 and others. Since no waste facility existed in Sulina, it is likely that the barrels were dumped at sea.” Such assumed dumping had been the previous practice as well. Specifically for the Greenpeace report, it connected to “364 barrels full of waste [including solvents, pesticide residues, acetone and lead] washed ashore on the Turkish Black Sea coast” between July-December 1988 (Greenpeace 2010, 11–12).

This mode of documentary practice—of transforming a call for repair by a postcolonial state into “merely humanitarian” action or the invention of Sulina as a destination for containers is a process of transnational disavowal that draws our attention from the specific question of documents’ existence to the epistemological labor of disqualifying, ambiguating, and ultimately dismissing their location, existence, or the meaning of their content (Abu El-Haj 2023, 253; Ben-Yehoyada 2024). In the case of Sulina, the information came from the documents we mentioned, but also from the deposition of a person called Aldo Anghessa. He was interrogated by prosecutor Francesco Neri and his investigating collaborators (one of them was Natale De Grazia, more on him below) in 1995.⁶  After describing to his interrogators the long chain of investigations against him, Anghessa explained the traffic:

The real contents of the traffic are usually hidden under the name of iron material, landfill, inert, and even grain. The inert materials could make people think that the abusive disposal of these substances can take place in quarries safely located in the South ([Italian regions of] Campania, Apulia, Basilicata and Calabria).

With reference to the abovementioned traffic, it is certain that disposal can take place in three distinct ways: a) burial in locations in southern Italy in old quarries or landfills; b) sinking of ships normally in extraterritorial areas (7–8 cases are known); [c)] disposal in third world countries (until the year 1989–90 they were Romania, port of Sulina – where the commander of the port was shot for opposing the unloading of these said materials…

As are various other impresarios of this realm of commerce and shipping, Anghessa was a compelling figure. In the article recounting his passing in the Italo-Swiss newspaper of Ticino region, described him as “self-styled secret agent” (Storni 2020):

One of the most controversial faces in Ticino and Italy from the 1980s to today has died in exile in [Dakar in] Africa at the age of 76—he was involved in various large investigations, often with unclear roles: sometimes informer, sometimes accused, sometimes both.

A more empathetic piece added (Fusi 2020):

The legendary Alpha-Alpha, Lotti-Ghetti Agent, Commander Manfredini. Take your pick from his thousands of successful masks. He died alone, as he always lived, even when he was among others, an actor in a spectacular play, of which he was the author, protagonist, and director, and which had since his youth taken the place of his real existence, his real feelings, his real self.

As he was wearing one of these masks, Anghessa told his interrogators the story about the “gunned down” commander of the port of Sulina, which they duly wrote down. It remained unclear whether he had attained that information in the first person or had read it in the newspapers. For example, already in October 1990, the Corriere della Sera reported on the shifting of waste traffic toward Eastern Europe after the fall of the Berlin Wall. One of the pieces on the topic that day argued:

That the port of Salina [sic], facing the Black Sea, has not been during the 1970s and 1980s a normal commercial port is demonstrated by an obscure episode of 1987. That year, the port manager was put under judicial investigation, arrested, and replaced by a right-hand man of the “royal family” [Ceaușescu].

Why Anghessa turned the investigation, arrest, and dismissal of the commander into being “shot”—we could only assume. In any case, this liberal rephrasing shows how indices of evasion and its context come from sources that, much after the fact, remain invested in maintaining things unclear. Within our attention to transnational disavowal, such a question illuminates the tension between trying to reconstruct what happened on the one hand and trying to reconstruct the epistemic relationship that accompanies exactly such attempts at investigation.

These investigations have taken many shapes over the years. The most stable organizational attention to toxic cargo comes from the longest method for studying the routes and fates of cargo more broadly: that of maritime insurance companies. Since insurance companies act as guarantors and protectors of cargo (Dua 2019, 110 et passim) in the aftermath of loss (and given that many ships were sunk and containers “lost” at sea), a paper regime exists of claim-making by ship owners, shipping companies, and cargo owners where toxic cargo is converted into claims for compensation.

Another form of investigation is by environmental organizations, the most central of which is Greenpeace, as already mentioned. In their 2010 report, Greenpeace researchers construed the object of their study as a network with a global reach, which during 1994–7 expanded its influence. Through company ownership and financial relations, the 2010 reports outlines “a loose network of companies, trusts and individuals” (Greenpeace 2010, 6):

Greenpeace found out that between 1987 and 1996 the network attempted to dispose of hazardous waste in Guinea, Brazil, El Salvador, Lebanon, Nigeria, Mexico, Mozambique, Paraguay, Romania, Somalia and Venezuela. The network’s operative branch in Italy included at least 26 companies handling an estimate 3000 tonnes of waste per day, with a total value of about $4.8 million.

Yet another form of investigation is that carried out by state officials. Here too variety and hierarchy of criminalizing power is the rule. We can examine the pathways of investigation from their end. The parliamentary investigative commissions have operated since the mid-1990s, in various distinct moments. For example, the first commission’s chairperson was Massimo Scalia, one of the founders of the Italian national environmentalist union, Legambiente (then “Lega per l’ambiente,” founded in 1980 out of the antinuclear and environmentalist struggles of the late 1970) and later a deputy from the Green Party (Carnì 2023, 95). The Commissions have been summoning documents from various investigations over the years, by different investigating and prosecuting offices, which had dealt with infringements as broad as insurance fraud, shipping document fraud, waste register fraud, misclassification of hazardous waste, and illegal waste dumping. The investigative bodies varied between local prosecutors of environmental or economic matters, the regional districts of anti-mafia prosecution directorate, and provincial magistrates (pretore). Investigators included pretty much all policing entities: the Forest Corps, the Carabinieri, the State Police, the Coastguard, and Financial Police. Regardless of the particular investigative body in question, no case of a poison ship ever made it to the conclusive phase of criminal investigation and prosecution. While most researchers of this realm—some of them by decree and delegation from state and international institutions—agree that something should have been found, officially speaking, nothing ever was. In a sense, transnational disavowal expands from the logic of concealing testimony to the logic of state. Such logic pilots “ships of poison” to the next leg in their voyages. When they are sunk within the horizon of Italian public attention, that is, offshore Italy, they turn into “ships to lose [navi a perdere].”

A Poison Ship to Lose

The ship Jolly Rosso was one of the vessels leased by order of the Italian government to collect and return the material from Lebanon. On January 10, 1989, while the ship was loading the material in Beirut port, several shots were fired at American journalists who were filming the operations. Eight days later, the Jolly Rosso made it to the Italian port of La Spezia, and in April unloading began. It remained in port until 1990. The ship, which was not deemed seaworthy on December 6, 1990, was suddenly approved for departure less than a day later and sailed to Malta. At that moment, the ship’s owning company changed its name to Rosso (Carnì 2024, 101). The Rosso arrived in Malta on December 12 already under her new name. A day later, after some transshipments, the Rosso left headed toward La Spezia and then Port Sudan. On the next morning, after crossing the Strait of Messina northward, the ship emitted SOS for taking water in. The crew was rescued by a ship transiting nearby, while the ship drifted and then beached at Amantea, on the Calabrian coast.

Several irregularities in the following days raised the suspicions of environmental officials and activists. The entire area was cordoned off, yet subsequent checks concluded that some material had been taken off the ship. Later examinations revealed a sort of opening in the hull from which some material would have been pulled out clandestinely. Yet no legal evidence of such retrieval was ever produced. Finally, and strangely, someone had painted over the company name on the ship’s hull.

The waters offshore southern Calabria are strewn with sunk ships like the Rosso. Dozens of them have been suspected over the year as “navi a perdere” or “ships to lose” (Braga 2018; Carnì 2024). The only one concretely traceable to the routes of toxic waste is the Jolly Rosso, then Rosso. And even in this case, no material was found on it at the moment of its beaching. The Jolly Rosso/Rosso is the only ship of the entire toxic shipping fleet that combines various conditions that make it legible as a toxic ship: it made it back to Italy, supposedly with material that had been sent transnationally. It was identified with the material that it carried until its last journey—making the ship itself an emblem of its cargo, rather than its mere vessel or host. And it beached, as opposed to drowning or altogether disappearing. As a result, the Jolly Rosso/Rosso became a monumental emblem of the entire chain of affairs. Consequently, it received much attention from journalists, activists, and magistrates.

At the same time, no investigation about the beached ship clarified the multiple doubts surrounding the fate of its cargo. On the contrary, investigations proliferated these doubts. The name change from Jolly Rosso to Rosso construed the ship not through its penultimate part in the set of “poison ships, but rather through its ultimate voyage en route to become a “ship to lose.” It is in the folder bearing that title that the ship’s beaching resides in the Reggio Calabria Public Prosecutor’s Office, under her last name, Rosso (Carnì 2024, 101).

In the case of the Calabrian ships of poison, one investigating official was killed in circumstances so obscure that an enquiry into his death was opened by the Italian Parliamentary Commission on waste circuits. Natale De Grazia, mentioned earlier, was an officer of the Italian Coast Guard. In 1994, after he was transferred to Reggio Calabria, he collaborated in a pool of three investigators and one magistrate that looked into the traffic of toxic and radioactive waste. On December 12, 1995, while on a mission to La Spezia with the other two investigators, De Grazia died suddenly, and his autopsy, which spoke of “athlete’s collapse,” raised even more doubts. At least one of his collaborators, who told us about their work, treats De Grazia’s death as the act taken to derail their investigation, which succeeded.

The affair that ended with the beaching of the Jolly Rosso/Rosso reveals how evasion and investigation produce an ongoing transnational relationship of disavowal. At sea, such relationship accompanied the toxic cargo for four years: from the January 1987 departure of the Radhost, the Baru Luck, and the Lynx from Italy, to the beaching of the Rosso in December 1990. Ashore, the dynamics continue through periodic attempted investigations to this day.⁷ This particular form of disavowal requires, therefore, continuous maintenance. In this case, maintenance entails killing, thereby opening up new investigations and further forms of evasion. This goes beyond situated coverups or individual corruption, though plenty of that happens too. This transnational form of disavowal ends up dismissing—ironically through the investigation that produces doubts and dead-ends—the suspicions that triggered it in the first place.

The co-production of evasion and investigation into transnational disavowal also emerges when looking at another circuit, the transportation between Italy and Somalia. What sets the Italian-Somali route apart is that it is the only one that is both known and has not triggered a demand from the country that received the material that Italy excavate and retrieve it. Indeed, what emerges most clearly here is how the colonial and the postcolonial seamlessly move into each other and the multiple scales of disavowals that nonetheless constituted, and constitute, an enduring, albeit toxic, relation.

Italo-Somalia

Italy’s encounter with Somalia was one forged by the sea. In the second half of the nineteenth century, the period preceding and immediately following the construction of the Suez Canal, a scramble for ports and protectorates emerged along coastal Horn of Africa. As nineteenth-century European imperialism encountered non-European others, a central legal question emerged regarding the relationship between Europeans and the “uncivilized.” Within an emerging world of legal positivism, where legal relationships could only exist between two sovereign nations, how might one understand the non-European other? For many, legal relations could only be envisioned between Europeans, citing international law “as a product of the specialized civilization of modern Europe… [that] forms a highly artificial system of which the principles cannot be supposed to be understood or recognized by countries differently civilized” (Hall 1880, 40). Against this position, Jon Westlake in his now classic Chapters on the Principles of International Law (1894) suggested that international law turn to established form of the Protectorate, that had its history in European encounters with Indigenous populations in the Americas. As Westlake noted:

When a people of European race comes into contact with American or African tribes, the prime necessity is a government under the protection of which the former may carry on the complex life to which they are accustomed in their homes, which may prevent European powers for supremacy on the same soil, and which may protect the natives in the enjoyment of a security not less than they enjoyed before the arrival of strangers. (Westlake 1894, 141)

Along the Red Sea coast, the protectorate became the dominant mode of establishing relationality between European powers and coastal sultanates. A tri-partition of the coast between French, British, and Italian Somaliland emerged by 1884 with each imperial power eager to secure Somali port cities as fueling ports and trading entrepots. Here, as elsewhere, protection served as imperial idiom (Benton, Clulow, and Attwood 2018). Italy gradually began to take an interest in the southern region of Somalia, which was dependent on the sultanate of Zanzibar after the signing of a trade agreement on May 28, 1885. Spearheaded by Filonardi, a trader and the consulate in Zanzibar, Italy managed to impose its own protectorate on the Sultanate of Obbia and the Sultanate of Majeerteen which then became part of the colony of Italian Somaliland. In 1893 the Italian government obtained a lease on the ports of Barawa, Merca, Mogadishu, and Warsheikh from the Sultan of Zanzibar and their administration was granted for three years to the Filonardi Company. This model of a company-led empire built around expropriation through key ports continued through the Fascist period (1922–1943) when the colonial government expanded further south into the fertile Jubba valley and established banana and sugar plantations. These plantations were all managed and run by Italian business interests employing, in extremely exploitative and coercive conditions, a large Somali labor force. While Italy was forced to relinquish its colonial domains in Africa following World War II, Somalia was “returned” to Italy as part of a trusteeship agreement supervised by the United Nations until Somali independence in 1960 when British and Italian Somaliland united to form Somalia. Within a decade, a military coup led to the establishment of a single party state under the dictatorship of Siyad Barre, a regime that lasted until the dissolution of the central government in 1991 and the ensuing civil war. This long colonial and neo-colonial relationship between Italy and Somalia, was one built on plunder and overseen by companies. Bananas, cattle, and other resources went to Italy through sea routes. These routes were also circuits of toxic returns.

Suspicions regarding the discharge of toxic waste from Italy along the Somali coast emerged already in the late 1980s and reported in 1992 (Brevik 1992). In 2005, the UN Environmental Program already denounced such a traffic, particularly following information that went beyond toxic dumping per se. It also included a trade in weapons for containers and barrels being undertaken between Italian shipping mediators and people close to the interim government of Ali Mahdi in Somalia.

As the UNEP 2005 report, After the Tsunami, elaborated,

During the Somalia civil war, hazardous waste was dumped in the country by industrialized countries. Somalia appeared attractive for hazardous waste dumping due to:

• Political instability. Since 1991 Somalia lacked a central government to safeguard its long coastlines and territories.

• Availability of dumping sites. There is a general problem of finding suitable dumping sites within the countries generating these wastes as well as high cost of recycling or incinerating. Somalia happens to have abundant sites for dumping waste.

• Low public awareness. The public were hardly informed about dumping of wastes in the country. Besides the people were trying to eke their living in the midst of extreme social problems and poverty created by the war.

The impacts of the December 2004 tsunami stirred up hazardous waste deposits on the beaches around North Hobyo causing some health and environmental problems in the area.

The report specified that “the impact of the tsunami stirred up hazardous waste deposits” in the coastal areas north of Hobyo and in Warsheik:

Contamination from these waste deposits has thus caused health and environmental problems to the surrounding local fishing communities. Many people in Somalia’s impacted areas are complaining of unusual health problems including acute respiratory infections, mouth bleeds and skin conditions. (UNEP 2005, 11)

The UNEP report outlined the preliminary findings from visits conducted by UNEP, FAO, and WHO experts between May 25 and 29, 2005, in some coastal locations in the Puntland territory, including Lower and Middle Shabelle, Lower Juba, Bay, Bakool and Puntland (UNEP 2005, 129). The experts examined coastal areas that would have been affected by alleged waste dumping but not by the findings following the 2004 tsunami. They concluded that “no traces of toxic waste were found” although in the same report it was pointed out that “Somalia is one of the many Least Developed Countries that reportedly received countless shipments of illegal nuclear and toxic waste dumped along the coastline” (UNEP 2005, 134). Despite the lack of empirical evidence—either in 1992 or in 2005—the UNEP report concluded that discharges of hazardous and radioactive wastes had actually been taking place in Somalia since at least the late 1980s, with shipments increasing around the very early years of Ali Mahdi’s interim government—coinciding, therefore, with the Italy-Somalia arms and toxic waste agreements that the UN Environmental Program report denounced in 1992 (Sniffen 1992; Brevik 1992).⁸

In contrast with the Nigerian, Venezuelan, and Lebanese cases, in the Italian-Somali case, toxic waste shipments were more closely tied to illicit intergovernmental agreements. Moreover, in 1992, these shipments were already clearly illegal for more than one reason. The reference to Somalia as a “war-torn country” in the UNEP 1992 (Sniffen 1992, 1) report referred to the relationship between illegal arms deals and the shipment of toxic waste. The Ali Mahdi government was interested in weapons that the international embargo prevented it from attaining (an embargo similar to the earlier 1970s embargo against the home country of the ship whose captain opened this essay). Italian officials were willing to offer such arms in return for the Somali willingness to receive Italian-origin waste. The context, or perhaps the pretext, for such collaboration appeared in the shape of two infrastructural projects: industrial fishing development in the port of Bosaso and the construction of a road from Bosaso to Garowe.

These infrastructural projects, moreover, turned out to play a central role within Italian development cooperation projects. This role illuminates the broader constellation of transnational and inter-state and diplomatic relations within which we should analyze the trafficking of waste and arms to Somalia. Such projects, in fact, were often poorly designed or fragmented, to the point of favoring the interests of local élites or the financiers themselves, rather than responding to the needs of the Somali people. For this reason, too, in the early 1990s, the Milan Public Prosecutor’s Office investigated these affairs, speaking of “bad cooperation [malacooperazione]” for development (Raffaello De Brasi et al. 2006, 7).

Transnational Disavowal

In this complex scenario, in March 1994, Ilaria Alpi, a journalist of the Italian RAI national television company, was killed in Mogadishu with her cameraman, Miran Hrovatin. She was investigating the circuits of arms and waste, mostly radioactive, in a new context. Speaking and writing about these matters in Italy was no longer a forbidden matter. After all, almost a decade of public attention and mobilization made the issue public and overt. But Alpi did not remain at the level of general denouncing. For her audience, she interpreted the toxic circuit within the context of the ongoing civil war between Siad Barre and Ali Mahdi. Particularly, Alpi was using her contacts within the Siad Barre side to learn as much as she could about Ali Mahdi’s involvement with Italy and the United States. In that specific moment in October 1993, the Battle of Mogadishu had taken place, including the downing of the American Blackhawk Helicopter. The United States was supporting Ali Mahdi, whose dependence on Italy and the United States in the war Alpi was exposing in detail. By that point, many Italians had written about radioactive waste in general or on the basis of existing publicized documents. Ilaria Alpi went to interview the brother of the governor of Bosaso, where the ships from Italy were docking and from where the road to Garowe was to be constructed.

Alpi was not only making visible the presence of a circuit but creating a cartography of culpability of U.S. dollars, Italian arms going to specific individuals in parallel to circuits of toxicity that were traversing similar geographies. Toxicity was not only the detritus of war, as has been importantly documented by a number of scholars (Stamatopoulou-Robbins 2021; Vaughn, Guarasci, and Moore 2021; Crane 2023). It was also fueling war and conflict in deeply causal and tangible ways. Like many other industrial production chains, those that produced the material that ended up in Somalia did so as a “byproduct.” Yet such products embarked on an entire circuit that related it to other things—like arms or the interests and funds of international Italian-Somali “mis-development” and “bad cooperation.” Such toxic material attained, in this sense, a new set of values.

Exposing this culpability, challenging evasion, and naming the relationship at the heart of this unequal exchange is what made Alpi a threat in ways that were different from the logics of state responsibility (or lack thereof) in other investigative reports. It is in that moment, as with the case of the Jolly Rosso/Rosso, when the actual relationship becomes documented and spelled out, when investigation can no longer simply produce dismissal, that a different logic—a deadly one—comes to the fore. Alpi and Hrovatin, like De Grazia, had grasped the transnational and local scenario within which the trafficking was unfolding, the actors pursuing it, and the broader institutional and political responsibilities. When investigation is able to “connect the dots” so to speak, it has to end abruptly. As one of the minority reports of the Parliamentary Commission investigating Alpi and Hrovatin’s deaths noted, it is plausible to assume that their articulate understanding of the nexus between arms, waste, and bribes related to Italian-Somali ill-cooperation (as those staged for Italian and U.S. interests within the ongoing civil war) led to these journalists’ deaths (Raffaello De Brasi et al. 2006, 161 et passim). Here, the logic of the report mirrors the tensions and aporias of ethnographic research. To explain this, let us return again to the port of Bosaso.

The port of Bosaso served as an important hinge not only for the operations of evasion but also for the boundary of transnational disavowal. The 2004 tsunami had exposed many barrels that had been thrown overboard in Somali territorial waters. We don’t know whether the Italian ships threw them overboard instead of bringing them on land, or whether the Somali side of the relationship had done so. Perhaps we do not need to know. That ambiguity reveals both sides’ ability to disown their responsibility for the hazardous effects of the toxic or radioactive materials’ presence on Somali environment.

Yet some material definitely made it to land, through the Bosaso Port. That material was buried somewhere—that is, in someone’s territory—in a shifting political geography of protection. In this sense, to receive material from Italy functioned as abaan or protection (Dua 2019), with all the relational and violent implications of such an act. One of co-authors of this essay, Dua, learned of the gravity of this relationship of protection that emerges from harboring someone else’s toxic waste, when he asked a friend in coastal Somalia if he could send him on WhatsApp an image that Ben-Yehoyada had sent him, which showed where containers had been buried in Puntland. In reaction, Dua’s friend said:

Oh, it’s on land? No. You don’t want to go there. If it’s on land it means someone drove it there. Someone buried it there. Don’t even send me the picture. A container that washes up on the beach is different.

The seabed furnished an alibi – for it provides the ability to admit the presence of toxic cargo without the territorial, institutional, and personal responsibilities that the same cargo’s location on land would entail.

Conclusion

The transformation of evasion into disavowal occurs therefore on either side of the transnational, postcolonial, ongoing relationship between various scales and instances of government in Italy and Somalia. Pointing that out does not call for a framing of symmetry or equal culpability between the shipper and the receiver. On the contrary. The “toxic colonialism” that Greenpeace has coined depends on the racialized postcolonial inequality between the parties. It also depends on the graded and proliferating forms of mediation, facilitation, and dismissal by actors and investigators whose official title does not exhaust their role in the chain of events. In our particular case, both the attempts to expose the toxic circuit and the attempts to fudge or avoid them confront an ongoing presence of toxic material in the postcolony and, therefore, the interest of those who accepted the material in the first place to hide their responsibility.

The routes of toxic cargo, dangerous and unregulated, rechart maritime routes through ports and coasts that depend on older relations. While these past relations—colonial in the Italian-Somali case—prefigure the toxic circuit, they are reinforced and reframed through the shipping, concealment, investigation (which deepens the parties’ investment in concealment) and paramount disavowal. This anthropology of evasion has emphasized that something has to remain invisible, submerged, or unclear at the heart of transnational toxic waste disposal. In this sense, the circuits of disposal are circuits of disavowal. And while one form of evasion leads to the killing of the investigators, that violence at times only proliferates further investigation, and reinforces the political import of the search for repair and the justice it is hoped to beget.

Abstract

In the middle of the Cold War, various state officials, politicians, middlemen, and mafiosi operated in transporting toxic (chemical and radioactive) waste from Italian ports to Venezuela, Nigeria, Lebanon, Syria, and Somalia. Public protest in some receiving countries triggered, at least to some extent, repatriation. In some cases, ships carrying toxic material sunk at sea in mysterious circumstances. The circuit stood at the center of numerous international and Italian journalistic investigations, judicial enquiries in more than twenty Prosecutor Offices throughout Italy, and several Italian Parliamentary investigative commissions (with self-described inconclusive reports in 1995, in 2001, and in 2018). Aspects of the circuit interested UN and international environmental organizations as well. We focus on the maritime, transnational, and evasive characteristics of this circuit and its aftermath. Specifically, we emphasize how erstwhile and ongoing colonial and imperial relations mobilized to facilitate the transit and depositing of toxic waste, and how this everchanging circuit reshaped those relations. This is a circuit that is known through investigations, at multiple scales and exercising multiple jurisdictional authority. At the same time, these investigations—we argue—produce and reproduce the very evasive relations and a transnational politics of disavowal that sustained this circuit in the first place. Through a historical anthropology of this circuit and its enquiries, we discuss the co-production of evasion and investigation within the transnational histories and presents of maritime mobility. Toxic circuits thus straddle the geopolitical temporalization that hinges on the end of the Cold War. They also provide a corrosive twin to the historiography of the globalization of commercial shipping and the world, which the container and super-tanker made.