Data related to this article:
“Plants and Political Ecology in Sudan.” https://zenodo.org/​communities/​plant-ecology-sudan/​
This collection includes data and datasets cited and used in this article, among them images, texts (transcripted interviews, reflective notes), weather data and georeferenced data from the study area. This material was produced or collected by Enrico Ille in 2016 and 2017 in frame of research on socio-ecological changes in the region around the Third Nile Cataract.

Resources related to this article: General map of the case studies: https://umap.openstreetmap.fr/​en/​map/​walking-on-fire-perambulatory-fieldwork-and-shared_836451. Map data: https://doi.org/​10.5281/​zenodo.7372795.

Introduction

4Second, we had to reflect on the methodological implications of reconstructing destructive events as researchers while also being interlocutors in the societal debate that surrounds them. While we have striven to arrive at a distanced assessment that integrates both natural and social scientific approaches for the purposes of making our contribution to this debate, our interpretations are also born out of the perspectives of a politically interested (and invested) academic (Enrico Ille) and an environmental scientist and activist, chemist and farmer from the study area (Mohamed Salah). Since we were both involved in the situation to different degrees, the political dimensions of reconstructing events—the politics of evidence, so to speak—also included us as part of an ongoing investigation in a public forum.

6While we arrived at the issue from very different angles, we share some degree of positionality: like most of our interlocutors, we are male, middle-class adults who were critical of, and to very different degrees openly opposed to, the Ingaz regime in Sudan that was overthrown in April 2019, but whose influence and power still lingers. We have both had a post-graduate education and share a sense of scepticism of mono-causal accounts of the date palm fires, whether as natural disasters or as politically motivated arson, and are influenced by the conviction that we can contribute to the societal debate through a more highly differentiated analysis.

8Our intention is not to present a systematic contribution to the debate on politics of evidence, but we do make an attempt at what Ginzburg termed “understanding society through clues” with his conjectural paradigm that reality is for the most part opaque in its complexity—for society and researchers alike, we might add—and must be investigated through “certain points—clues, signs—which allow us to decipher it” (Ginzburg & Davin 1980, 27). Disruptive events such as date palm fires are not just an invitation to collect, to use Ginzburg’s terms, the traces they leave, in the sense of both clues to what happened—the traces of evidence—and their aftermath and repercussions—the traces of injury. We are also invited to think about what narrative threads they follow, the general direction in societal developments and conflicts, and how they affect these threads, “the trail they leave” (Ginzburg 2012, 1).

9We use this foundation to illustrate an investigative process that seeks to accommodate methodological complications through a co-reading and shared interpretation of the sources we collected. We approach this interpretative work by means of an account of walking interviews and more broadly of perambulatory fieldwork in the areas where date palms were burned, as a technique of situated exposure to the traces of the events (see “Methods”). What might be seen as a variation of the conventional crime scene visit is not concerned so much with the official process of establishing a legal crime: we revisit the process by which a fire event is socially reconstructed as a political crime or not (see “Interpretation: political violence”). However, this also highlights the social limitations of official investigative processes, for instance where date palm owners declined to lay blame or demand retribution because long-term social relationships were involved. In addition, the explanations we collected were not exclusively focused on human agents, but also allowed for other-than-human causes (see “Interpretation: environmental illiteracy”). Finally, some of our interlocutors made the interpretative effort to contextualise the fires in historical development, as a loss in an extended temporal sense that we ourselves were therefore prompted to consider (see “Interpretation: historical disconnection”).

Research problem

12Forest fire ecology is a broad and complex field of study, not the least due to the breadth of the contexts in which fires occur. In their most biophysical sense, fire studies break down the process to the fundamental dynamics of heat accumulation and heat transfer, the elements that are conducive to or inhibit their spread, burned areas and trees and their characteristics, etc. (Johnson and Myanishi 2001; Cochrane 2009, chapter 2). They not only capture the complex effects of fires on landscapes, including their regenerative and transformative function (Wallace 2004), but also relate them to broader climatic and atmospheric dynamics.

13A number of these dynamics, such as hot, dry winds like the Sirocco in the southern Mediterranean basin (Belhadj-Khedher et al. 2020), are crucial to the fire landscape we will discuss; however, this kind of differentiated biophysical assessment and discussion was mostly absent from the political discourse we looked at. Rather, the interpretation of date palm fires and how they start examined the intended and unintended consequences of human fire-making, which is what we focus on below.

16In a governance environment in which administration through formal institutions is considered to be the norm, their “failure to meet social needs effectively” (Morehouse et al. 2011, 350) and the subsequent “informal activities” and institutional contradictions can be the focus. This may result in a shift to community-based solutions such as community fire brigades (see, for example, Sheridan et al. 2015, 405) or other forms of community-based fire management (FAO 2011). A perceived partly state-driven, partly change-driven decline in fire management systems may form the basis for a call to reconsider the use of fire as a useful part of natural resource management (Russell-Smith et al. 2009). However, even taking an ultimately managerial outlook—for instance by discussing fire as a landscape management tool—a contradiction in the aims rather than the methods becomes the concern of more critical analyses of state-citizen relations, such as between state-centred anti-fire policies and local cultivation practices (Kull 2004).

17This leads towards a political ecological concept of fire governance. If one goes beyond the customary understanding of governance as being about state institutions, for instance, “the institutions and rules that determine how fire is used in an area, and how rights holders […] participate in these decisions” (Walters 2015, 276) become relevant, as does “why, how, and in which political contexts fire regimes change” (Walters 2015, 283). Put even more succinctly, “disagreements or conflicts over alternative fire regimes and worldviews, as well as over prevention, disaster, and postfire policies, are analyzed as the effects of competing visions of different social actors over desired territories and their use” (González-Hidalgo et al. 2014, 1015).

18What brings all these studies together, however, is the presumed benevolence of fire use, in the sense that causing the destruction of human life and property is mostly accidental, and not the aim, when starting a fire. The situation is different in studies on incendiarism. Historical research on arson, for instance, has revealed how fire was used as a form of peasant resistance similar to sabotage, “as a final act of defiance by rural peoples who have lost control over the resources on which their subsistence has traditionally depended; as a clear signal of imbalance in the cultural-ecological equation; and as a symbolic gesture of political protest” (Kuhlken 1999, 360). Fire has also been studied as an anonymous and effective weapon, as an “asymmetric threat” where large-scale destruction can be caused with little effort, and it continues to be used in this way (Deshpande 2009). From a national military-defensive perspective, this becomes an issue of acts of “pyro-terrorism” against the powers that be, although a relational definition as “incendiary attacks […] to advance political or social objectives” (Baird 2005, 2) does not preclude the possibility that these same powers that be are themselves the pyro-terrorists.

19Invariably, interpreting fires as arson for political purposes raises basic epistemological difficulties, as they are naturally sparingly documented or not documented at all, and any later accusations are themselves part of the political process (Dillinger 2006). This has been documented most clearly in the case of the huge fires in Indonesia (Kalimantan 1982–3, 1997–8). The “fierce political battle […] about who was to blame for the disaster” (Barber and Schweithelm 2000, 12) saw a government blaming everybody else—small-scale farmers, probably communists, and nature—while also appointing a major investor in the forestry and agribusiness industry in place of a Forestry Minister who had raised criticisms. The debate also crystallized new voices in the public debate on natural resource extraction in the country. The immediate battleground of claims and counterclaims was quantification of the areas affected: a fixed and unsubstantiated number of 520,000 hectares was claimed by the government, compared with the 5.2 million hectares identified by remote sensing and checks on the ground by the Integrated Forest Fires Management Project in East Kalimantan and actively concealed through censorship (Barber and Schweithelm 2000, 12–4).

20However, this numbers politics was also only one side of a complex struggle. Kalimantan was immersed in existential struggles over diametrically opposed visions for using its land: highly polluting gold mining with mercury in the Meratus Mountains, failing rice harvests for small farmers and government schemes for one million hectare rice paddies, unbalanced blackwater swamps used by water buffalo herders, etc. A local resident has been quoted as summarising the situation as “the end of Kalimantan” (Tsing 2005, 23), an apocalypse during which one can only stand by and watch it happen. Meanwhile, the debate on who started the fires developed its own dynamics of spreading through stories that redrew a landscape of suspicions and politically motivated smokescreens (ibid., 43).

21This reflects both a drawing and redrawing of boundaries, not just in a political sense, but implicitly also among socio-ecological categories such as settlement, agriculture, cultural landscape and wilderness, and their relation to humans as (dis)organizing agents at the so-called wildland-urban interface. When we look at fires, therefore, we see that the line between “landscape fire” (McKenzie et al. 2011) and man-made fire is not only blurred, but also part and parcel of investigative processes that relate this line to intent, and thus to responsibility.

Methodology

Perception

24Another train of thought was more concerned with the dynamics between researchers and participants during shared walks, especially if the walk is viewed as an alternative or enhancing interview technique. Will being outside—even more if the “outside” is the topic of the interview—“offer deeper social meanings” compared with the “more detached emotional distance” of a sedentary indoor interview (Hitchings and Jones 2004, 10, on gardening)? Will reflections on and the analysis of an embodied experience allow “an unhurried observation of the interplay between sensual and objectifying modes of performance as actual happenings” (Lorimer & Lund 2004, 131–2, on quantifying mountains)?

25Those who have included co-walking as one of their research activities ultimately have done so for a wide range of epistemological—and also practical—reasons, resulting in a diverse landscape of perambulatory methods: that is, methods that are influenced or determined by being in a specific state of mobility, in this case walking. Based on their structural rigidity, these methods can be broadly categorised: for instance, into bimbling—a spontaneous walk-and-talk with no predefined route or topic; go-along interviews, which shadow a routine route with or without prepared questions; participatory walking interviews, which involve free movement in locations relating to the topics being investigated, using participants as experts and guides; and the docent method, with specific routes for walking interviews being planned and discussed during pre-walk and post-walk interviews (Kinney 2017, 2; Chang 2017).

Representation

27The debates on perambulatory methods have also struggled with issues of representation: namely how to translate the research process into communicable media, especially beyond a purely textual representation. Proposals have been made to visualize the spatial information obtained from perambulatory research more clearly, “to more directly connect what people say with where they say it” (Jones et al. 2008, 2). This refers especially to the absence of mapping and other orientation techniques in many works on place-specific interviews, including video productions that, if there is no spatial reference and if they are not shot at the same time as the other tasks, do not necessarily provide clarity (Jones et al. 2008, 4). Among the suggested techniques are participatory mapping or sketches of node-to-node walks (see, for example, Emmel & Clark 2009, 14).

28The most relevant technique in this case is the so-called spatial transcript, which integrates the spoken word and the location by relating the recorded interview portions to GPS data (Evans & Jones 2011; Jones & Evans 2012). This technique was developed in an urban planning environment, and while the potential for using it in the context of participatory GIS was noted, it was also acknowledged that the non-representational character proposed by many of the works based on perambulatory methods is somewhat lost by the use of GPS data, which has its own ambivalent implications (Evans & Jones 2011, 851; Jones et al. 2008, 7–8). In fact, it has been argued that qualitative GIS illustrates rather than analyses, and ultimately uses “rather conventional, neo-Euclidean, “physical” conceptualisations of space and location” (Merriman 2014, 179).

Participation

30A fundamental difference in how research is approached can also be seen through the juxtaposition of a mining and a travelling metaphor (Clarke 2005, 166):

In modernist mining, knowledge is buried, waiting to be uncovered by human research via the individual interview, purified through transcription, and then conveyor-belted to validity via correlation with others. In postmodern traveling, knowledge consists in heterogeneous discourses gathered by the traveling researcher, entering into conversations, noticing sites and images along the way, gathering stories and available “collectibles”—stuff of all sorts.

31In both cases, entering an environment with others where there is a low level of control over both inevitably increases the complexities of doing observation and conversation at the same time. Looking at this the other way around, the inevitable messiness of the research process can only be eliminated by increasing control over research participants as “experimental subjects” and over their environment as a laboratory “clean room.” Between these extremes, there are many ways of dealing with messiness and “the tension between formalization and intuition” (Flick 2014, 12), but they are inescapably linked to the power relations between researcher and participant.

32Perambulatory methods have been highlighted as at least opening up the possibility of a more symmetrical “research moment,” especially in case of the more unstructured walks (Anderson 2004, 260):

This practice of talking whilst walking is […] useful as it produces not a conventional interrogative encounter, but a collage of collaboration: an unstructured dialogue where all actors participate in a conversational, geographical and informational pathway creation.

33For some authors, this can lead not only to a co-construction of research, but also to questioning or even dissolving the researcher-participant binary (Moles 2008, 37) and to new performative forms of doing research, such as through collaborative art production (O’Neill and Hubbard 2010).

34Whatever the final destination may be, walking with establishes a multisensory moment of co-presence and sociability that follows “a particular style of movement, pace and direction” (Lee & Ingold 2006, 69), where one’s eyes are on the path ahead, and there is only rarely eye contact (ibid., 79). Here, having a conversation implies the need to find a shared rhythm, physically and mentally, while also being subject to the same conditions of moving over the ground.

35The implications of this “grounding” also become clear in a comparison with techniques where the same ground is covered from a distance, with little or no contact with those who are walking there, far from the “‘micro-geographies’ of spatial relations and meaning, where multiple scales of social relations intersect in the research interview” (Elwood & Martin 2000, 649). This lack of “ground truth” (Pickles 1995) has been central to critical discourses on GIS and GPS, for instance, and to their relationship with political processes and governance (Propen 2005). For one thing, it concerns critiques of GPS-based GIS opposed to the kind of knowledge production GIS embodies as a “polemical (discrete, fixed, overcoded) image” (Leszczynski 2009, 356) that claims to show the world as it is and makes it (instrumentally) available for exploitation. Critical GIS, on the other hand, does not reject the technology: it questions the ends to which it is being used, and how other uses might be possible (Leszczynski 2009, 354–6).

36At this point, all these remarks serve mostly as a reminder that moving through specific environments in a certain way during interviews can heighten the participants’ physical, emotional and social vulnerability, which at the same time increases the researchers’ ethical vulnerability (Hibbert et al. 2014), especially if the interview itself joins a direct confrontation among conflicting social actors or creates a conflictual role for the researcher (who may also be a friend, activist or caretaker, for example).

37The type of guidance established before and during the walk is thus not just about the distribution of knowledge production, but also about the distribution of responsibility. This involves the route to be taken, but just as much the “micro-politics” of the walk itself, where questions of power are multiplied: not only active and passive elements of the conversation and shifts between introverted walking and extroversion by solicited and unsolicited statements, but also the movement itself between paces, leading, stopping, turning, etc. (Palmgren 2017, 2). In this sense, the range of possible perambulatory methods presents a “continuum based on who orchestrates the observed mobility, participants or researchers,” from the participants’ usual routines to the researchers’ pre-defined tours (Kusenbach 2018, 350).

38In this way, the creation of “social synchronicity” (Hodgson 2012: 20) casts the research process into the complex conditions of mobility and social connectivity, the conditions of becoming situated somewhere with somebody.

Positionality

39This situatedness of the walk is highlighted, or even becomes essential, when it takes place amidst tensions or open conflict in the form of practices of discrimination and segregation, or violent confrontation. This can be beneficial for a more symmetrical relationship between researcher and participant, not necessarily in the sense that it eradicates positionality and the various vulnerabilities, but by conveying a more immediate sense of what it means to be in a certain situation and/or position. Rather than removing the researcher’s focused interest—which can overlap with the participant’s interest to a greater or lesser extent—it potentially leads to moments that are “not just about researcher questions and participant answers but also about a ‘more equitable’ and dialogical line of inquiry” (Brown & Durrheim 2009, 926).

41The temporal dimension of this place-making reminds us that methods that only seize the here-and-now overestimate not only the importance of the moment but also the researcher’s ability to grasp it, thereby disregarding an analysis of any elements that are not perceptibly present but still have an impact. Walking together through a landscape adds to each individual’s topo-(auto)biography in very different ways, and also induces more or less shared sensibilities and common notions, “the ambiguous middle ground of the sensory commons through which human and planetary coexistence is coordinated” (Murray & Järviluoma 2020, 235). In addition, the narrative exploration of a place results in an “interweaving of personal biography and individual experiences with collective (social) memories and spatial histories” (Clark and Emmel 2010, 5), an act of story-telling through spatial references.

43This “place responsiveness” takes on a new quality if such spatio-temporal ruptures do not occur over a long period but are manifested in one or several distinct events, such as natural disasters (D’Errico & Hunt 2019). The perceptive and emotional focus suggested, or even driven, by such events may then lead to a conversation that is less about narrative reconstruction in the sense of oral history and more about the traumatic or otherwise disruptive effects of these events on the current state of affairs. Again, this can involve an invitation—or pressure—to join sides in the event that this state of affairs relates to accusations or identifying those guilty of causing, or not preventing or not adequately addressing, these ruptures (D’Errico & Hunt 2019, 7–8).

44This latter dynamic is intensified during or in the aftermath of wide-scale violence, and comes close to the ethical and representational complications of war and post-war documentation (Carrabine 2018): What is its role in the ongoing conflict? What does it include and exclude from its view? What does it convey about the emotional and physical content of violence to an audience that did not experience it? Why should it try to do so? How can it avoid presenting “beautiful excerpts,” thereby aesthetising death and destruction up to the point of offering decorative spectacles or even the commodification of the suffering of others?

45One proposal is to “reconfigure the violence of representation by focusing on the traces of conflict” (Carrabine 2018, 642), the ruins and scars in the landscape, and to balance the detachment of the bird’s eye view—whether during remote attacks or during remote investigation—with contemplation of the ruins and the situation of those left behind to live with this landscape of destruction.

Case studies

46This is the site at which our case studies are located, but they introduce a further complication: the very nature of the violence and destruction that has occurred is being questioned, and how and why the fires started is being disputed. Were they accidents due to a disjunction between fire-using humans and their environment? Were they an act of political violence perpetrated by the Ingaz regime? Or were these “natural” disasters (or merely events) facilitated by negligence because someone has failed to reduce the fire hazards?

47This makes walking interviews akin to crime scene visits, including as an investigation into whether a crime has been committed at all. This is even more the case if crime scene investigation is understood in the sense of an enactment of the production of evidence for use in very specific organisational fields, such as the presentation of police work in courtrooms and the codified legal concepts and procedures they follow (Williams & Wheetman 2013; Wyatt 2014), which also involves social, discursive processes of sensemaking (Brookman et al. 2020). However, we are less interested in ascribing guilt or responsibility than we are in focusing on the processes by which guilt or responsibility is established by those who have been directly affected by the fires.

48We can only indicate briefly here how and why so little of such an investigation in the conventional sense of forensics has taken place at the scenes of the fires, but an overview on the political context of the interpretations voiced by participants will touch on this issue, illustrating the need to also consider questions of loss, detachment and disconnection through “controversial or less consensual stories” and “how these interpretative acts came into being: from whom, by whom, for whom” (High 2011, 219).

Pre-walk interviews

Walking interviews

52The resulting presentations of the walks differ significantly due to differences in how they were conducted and documented, as this pre-configured the kind of narrative reconstruction that was possible. The first walk was relatively clearly structured, with a north-south direction along a single path and a detectable correspondence of photo-position-conversation, with the exceptions highlighted in the transcript. Accordingly, a topo-chronological structure was chosen, including audio and audio-visual elements, that precisely followed the trajectory of the walk and the conversation, with geotagged photos taken in between as place markers.

53The second walk had no audio recording and mostly took place at a distance from the accompanying participants (who were also different from the pre-walk interview partner). Accordingly, the transcript is a chronological but photo-inspired reminiscence of Enrico Ille’s own thoughts and reflections during the walk, with a post-walk analytical process in mind.

55These differences in presentation also reflect varying approaches to perambulatory research methods, which were applied according to the social dynamics that developed around the walks rather than pre-planned designs (see section “Participation”). We therefore treated them as part of the sources we analysed when writing the text, which is why both the pre-walk and the walking interviews contain both explanatory and reflective comments on the dynamics and shortcomings of the developing exchange.

Interpretations and conclusion

Case study 1: Saï

Georeference: N 20°45.441’ E 30°19.171’ [geo:20.757350,30.319517]

Case study 1: Saï (map)

Full screen: http://bit.ly/​3HrF6Vl.

Case study 1: Saï (satellite view)

Source: Google Maps. See Full screen.

Pre-walk interview 1: Saï, 24 February 2017

Pre-walk interview 1: Saï, 24 February 2017 (text)

On ArcGIS Storymaps: https://storymaps.arcgis.com/​stories/​54c0859fe8164ffda5045fd669cdbc16.
On Umap: https://bit.ly/​3XSinbR.
Persistent identifier: https://doi.org/​10.5281/​zenodo.6901021.
Download PDF Version: https://zenodo.org/​record/​6901021/​files/​Pre-walk%20interview%201.pdf.

Walk 1: Saï, 24 February 2017

Walk 1: Saï, 24 February 2017 (text)

On ArcGIS Storymaps: https://storymaps.arcgis.com/​stories/​3f476ced55e04ae6a7a9acdd879751b8.
On Umap: https://bit.ly/​3Jp8L49.
Persistent identifier (text and images): https://doi.org/​10.5281/​zenodo.7347398.
PDF Version: https://zenodo.org/​record/​7361287/​files/​Walk_1_Sai_24_February_2017-v2.pdf.

Interpretation: Environmental illiteracy

57Many aspects of the ongoing interpretations made an appearance here: the ecological conditions of wildfires, the changing relationship to agricultural activities in general and to care of date palms in particular and the government response, which was marked at the least by lack of interest and at the most by an active population displacement campaign.

58In the foreground of Faraḥ’s interpretation, however, there seems to be a perception of an increasing disassociation between farmers and their environment and a certain lack of environmental literacy, such as anticipating wind direction and taking it into account. As shown in the pre-walk interview, this epistemic gap has not only deeper historical roots but also more imminent implications for the future of human settlement in this area, implications that underlie more politically critical readings of the situation, as discussed in greater detail in the case study below.

59As a first step to approaching this disconnection, we looked at the wildfires from a perspective of ecological factors that are not directly influenced by humans. Since this was far from being our methodological homeland, we did this indicatively, based on averaged measurements, and not by modelling microclimatic circumstances and developments. In addition, as we will show, the limitations of the environmental data available to us did not allow for very precise assessments.

Negligence

60An implicit starting point for the accusation that the fires resulted most of all from negligence was a wide range of fire usage practices that remained unmentioned in all the interviews, probably as the presumption was that they were well known. Among the many occurrences of fires in the vicinity of date palms—and at the same time among the examples of the variety of ways uncontrolled fires broke out, are the following:

a) burning the soil to “clean” it in preparation for cultivation, in later fields as slash-and-burn or in other open areas

61This involves not just the cleaning of fields, but also the intentional burning of fallen or cut date palms and other dry organic matter to keep the ground accessible and reduce the nutrients for insects and rodents.

b) charcoal-making (mostly other trees, but date palm wood is used in brick-making and bakeries due to its long heat retention)

c) making tea or heating food (using dry leaves, including date palm leaves, a central aspect of sociability among farmers during the day; here with a metal enclosure to prevent the fire from spreading)

62Other outbreaks come about as a result of small smoking fires; consuming green branches that are ignited to keep away insects; smoking branches used to disperse bees during the honey harvest; embers from kitchen fires that are sometimes thrown out while still warm or hot; cigarette ends that remain alight after being tossed away after smoking; lighters or matches used to heat the ends of hoses to connect with pipes; or burning the ends of plastic ropes to prevent disentangling; etc.

63Faraḥ’s claim that negligence was behind the spreading fires was based on a lack of adequate agricultural and ecological knowledge he perceived. However, as the interview and walking conversation showed, both Faraḥ and Faʾiz saw more behind this than just the inefficiency of individual farmers. They linked it to the composition of the workforce and a more general lack of care of date palms, both of which have to do with migration patterns and generational shifts: agricultural workers now often come from other areas, and have no previous acquaintance with—and occasionally no interest in—maintaining a safe use of fire; meanwhile, large segments of the population have emigrated and have no opportunity to keep up a close, caring relationship with date palms, and often no interest in doing so.

64In other words, this interpretation sees a demographic shift that is reflected in the situation of the date palm environment: occasional workers display a failure to use the old cleaning practices, which involve removing dry matter from palms and the ground. This is exacerbated by a greatly reduced use of this dry matter for roofs, fires, etc. From a cultural landscape point of view, the shift can be seen in the prevalence of disorderly forests in place of intense care for individual trees (in the photograph, this is represented by a well protecting the growth of young date palms):

65From an ecological perspective, this may mean a turn to organic re-organisation, to cycles of opportunistic growth and inevitable death, to date palm populations made up of intergenerational groups of trees, death alongside birth, to the open-ended struggle between insects and plants, to biomass reverting to soil, etc. Without human economic interest in the date palm’s abundant fruit, “destructive” wildfires simply fulfil their rejuvenating function, and the former image merely presents an aesthetic problem when a lack of human ordering is equated with general disorder.

67In several instances, the aftermath of the fires also intensified the shift from horticulture to agriculture, from shady forest to open field. The following Google Earth imagery, which juxtaposes June 2004 and November 2009, show this shift quite clearly on Saï’s northern side, in Saïsāb, excluding the strip of undamaged date palms mentioned during the walk:

68Faʾiz and Faraḥ outlined the broader dynamics they saw behind the movement away from date palms:

• The technological shift from waterwheel to engine, which initiated an unprecedented intensification of agriculture, especially in areas that could not easily be extended;
• The subsequent shift from families to sharecroppers, accelerated by a parallel increase in labour migrants from other, often crisis-stricken, regions of Sudan and in resident labour migrants leaving for urban centres or other countries;
• The incremental complications and fragmentation of land and palm ownership through inheritance;
• A perceived reduction in the productivity and profitability of date palms—for a number of climatic, horticultural and commercial reasons—combined with increases in the prices of agricultural cash crops such as wheat and broad beans;
• Generational lifestyle shifts, together with long-term migration and professional diversification (employment, trade, mining etc.), which accumulated with the other developments to result in a weakening relationship with date palms as a social connector and/or economic stake.

70It would be wrong, however, to take this as a resigned acknowledgement of an inevitable societal development. As Faraḥ’s final remarks during the walk—and Faʾiz additional comments during a later walk—highlighted, the response of the official authorities was considered not just to be subpar but also a part of a more general—and maybe intentional—neglect of the state’s protective duty towards its citizens, an aspect that will be central to the interpretation we present next. At the same time, their criticism implies that they considered this development to be far from inevitable.

Wind patterns

72Another more substantial observation, however, was not only the strong winds blowing during the same period, but also their increasing unpredictability. Faraḥ noted a declining ability to predict wind patterns, but we also considered changes in these patterns as a possible explanation for fire-making getting out of hand.

75Like the lack of clear patterns of wind speed changes over the year, the recorded wind directions offer a picture of great regularity—apart from unusual values for 2016. But what if this “regularity” is simply an effect of the “flattening” we see when everything is averaged out? Considering the time- and place-specific knowledge required when it comes to using fire, and when and where, this flat aggregation seems to miss the point.

77However, this variability is shown throughout the year, and is to some extent more extreme in other months. In January 2005, the daily averages rose as high as 115 km/h (10 January), falling to 9 km/h less than a week later (16 January), and similar extremes can be found in February and April, but not in the months after June. In fact, even the maximum speed remained almost always below 40 until 21 May 2006, which saw an isolated peak of 113 km/h. The following year, the period up to March 2007 saw a similar pattern to the June data throughout the year, slowing down over the cold season until several peaks during April 2007, with maximum speeds of 120 km/h on 2 April and 122 km/h on 11 April, accompanied by averages close to 40 km/h (and 72 on 2 April itself). The whole year exhibited a constant inconsistency of up to 40 km/h shown in the diagram, and even April 2008 passed by without further extremes.

78The following year, wind speeds were slightly “slower,” with June average values mostly below 20 km/h, until September 2008 came with further sudden changes, especially around 12-14 September, when the average shot up from 9 km/h to 70 km/h (with a maximum of 111) and again down to 19. The following months reverted to the previous pattern and a slower cold season seemed to be following, until February 2009 saw a similar sudden outburst on around 24 February (average 86 km/h, maximum 148 km/h). Contrary to the usual pattern, April was among the quietest months that year, and only from 24 April onwards were averages above 30 km/h reached, continuing intermittently into May, peaking on 30 May with 111 km/h and eventually settling into a “normal” June and a rather quiet hot and similar cold season. Up to the end of the available data, in August 2011, the wind patterns remained in this range, rarely rising above 40 km/h and without any further extreme values.

80Two extensions of Faraḥ’s observation of environmental illiteracy are possible here: first, the debate on the interpretations of the climatic reasons for date palm fires is hindered by a lack of differentiated and local data due to the absence of a situated environmental monitoring system. The absence of a network of climate data production means that patterns from the observation centres across a distance of over 300 kilometres have to be used to make quantitative statements about the micro-climate of the case study areas. Even these existing data are not easily accessible to those concerned with the fires and are therefore virtually irrelevant to the societal debate.

81Second, in this case, the predictability or “readability” of an environment is closely related to experience rather than to information systems, and is therefore individualised. This is the entry point for one of the interpretations that focus on the usual uses of fires and find that fires are the result of accidents, and by extension of negligence (ihmāl). This will occupy a great deal of our time in what follows, as it is an accusation that drives, or at least accompanies, almost all the interpretations, although they attribute various responsibilities.

Case study 2: Kajbar

Georeference: 19°56’52.9” N 30°32’33.2” E (geo:19.948033,30.542550)

Case study 2: Kajbar (map and text)

See full screen: https://bit.ly/​3kULqgz.

Case study 2: Kajbar (satellite view)

Source: Google Maps. See full screen.

Pre-walk interview 2: Kajbar, 3 March 2017

Pre-walk interview 2: Kajbar, 3 March 2017 (text)

On ArcGIS Storymaps: https://storymaps.arcgis.com/​stories/​b55c52ba9e664dafa30cf6a942b7f357.
On Umap: https://bit.ly/​3kULqgz.
Persistent identifier: https://doi.org/​10.5281/​zenodo.6901047.
Download PDF Version: https://zenodo.org/​record/​6901047/​files/​Pre-walk%20interview%202.pdf.

Walk 2: Kajbar, 3 March 2017

Walk 2: Kajbar, 3 March 2017 (text)

On ArcGIS Storymaps: https://storymaps.arcgis.com/​stories/​0c6693a744e64bdb8d6e6e6def4f63d3.
On Umap: https://bit.ly/​3wL2UyC.
Persistent identifier (text and images): https://doi.org/​10.5281/​zenodo.7348344.
Download PDF Version: https://zenodo.org/​record/​7351089/​files/​Walk-2-Kajbar_3_March_2017.pdf.

Interpretation: Political violence

82What makes the meandering exchange with Fāris so intriguing is that it arrives at a political economic picture that does not differ very much from what we both had arrived at in our own discussions. However, his understanding did not allow for the ambiguities and intricacies that made the political debate about date palm fires a much less straightforward matter. Both direct accusations and interpretative complexity can arise from interweaving the phenomenon of date palm fires with the ongoing struggles over dams on the Nile, from relatively small ones in the study area to the megadams—the Aswan High Dam and the Grand Ethiopian Renaissance Dam (GERD)—as well as other water-related plans, especially the increasing number of large-scale irrigation projects south of the study area.

Statistics

84In this heated milieu, our own assessment of the extent of date palm fires was only one of several other statistical works. Our efforts faced both a political and an observational problem: distinguishing between increased occurrence and increased documentation. There are virtually no data on fires before 2006, which means that there is no baseline for diagnosing an increase. In addition, when documentation began to be kept, it may have focused more on areas of heightened political mobilisation, as there was a greater interest in publicising events.

86However, the limited amount of documentation means that historical reconstruction requires a major effort. A full evaluation of even just the three events presented in this article means that an investigative effort would take years if it were to be extended to all the other cases (see an example of such research below). Alas, our own initial assessment brought us to 97 events between 2007 and 2017, 55 of which we could trace to specific months and 47 to a specific day. In 9 cases, we have additional eyewitness statements but no further multi-layered investigation similar to that presented in our case studies here. In the case of most events, Mohamed Salah gathered at least general estimates of damage, mostly rough numbers, and our subsequent calculation brought us close to 150,000 date palms affected by fire in that period. As will be seen, this is but a small sample of all the fires that occurred, and it is used here merely to discuss the circumstances under which different forms of evidence were included in the interpretation of date palm fires.

87Over the years, hundreds of articles, posts, comments, lists etc. have been published online. It is not our intention to assess these media materials here, but it is of interest to report the immediacy of the eyewitness accounts as experienced during the walking interviews to their translation into numbers, wider narratives and political claims.

91In 2016, Al-Ḥassan Hāshim’s calculations had increased to about 250,000 date palms. His database gathered information from communal websites, newspaper articles, social media posts, police reports, eyewitness statements, journalists and officials, by telephone or in person. He used an order of reliability, favouring eyewitnesses and marking information that was merely mentioned on Facebook and similar sources as “still under investigation.” Since many villages in the Northern State, including Saï, had their own websites, he watched them too. Finally, he tried to connect to the official statistics but came to the conclusion that they suffered from a low level of reliability: he asked for confirmation at Abri’s civil defence office of the approximately 180 fires in 135 villages he had taken note of, but found that only 13 villages had been registered, with an overall loss of about 50-60 palms. At the same time, he urged people not just to count the affected palms but also to register them with the police in order to have official documentation in future court cases to demand compensation after, or even before, the end of the regime. Unfortunately, he observed, this was not always done.

93The quoted numbers must face another complication, however. A comparison between the burned and the unburned areas in the orchard, as was made during the walk in Kajbar, showed not only recovering (and “cleaned”) palms in the former but also ongoing cultivation beneath them—sometimes in contrast to forest-like unburned and uncultivated areas, which in both cases makes it more difficult to speak of total economic losses.

Accusations

94The politically invested search for reasons and/or perpetrators meant that the ongoing registration of date palm fires differed from a statistical process for documentary purposes or a disinterested description of what had happened. In fact, quantification lay at the heart of the political discourses around the matter, as it was rarely denied that there were different causes for fires, including accidents; it was what made up the majority of cases, and thus the main reason for them, that was a cause for debate. Absent generally accepted forensic processes that clearly related perpetrator, motivation and result, an explanatory vacuum was left that was filled with assumptions, many of which had their roots in existing political positions. Since the perpetrator and the exact reason for fires remained unknown to many people in many cases, there was a great deal of room for interpretation with a political bias.

96Fire had always been used in the groves: mothers and grandmothers made honey and other derivates from dates there; lupin beans (turmus) were cooked with date palm wood to reduce their bitterness before being stored in the home; the pits of dates were ground and used to make cigarettes; agricultural waste and food leftovers were burned under the palms; and fire was even used directly—but carefully—on date palms to treat disease and insect infestations. None of this had led to the spreading of fires experienced since 2005.

Government responses

108What remains unclear is what this means for actual fire management. A 2001 assessment of the situation highlighted a historical line from the colonial to the early post-colonial period, up to the 1960s, when wildfire management fell under the responsibility of Native Administration leaders, “under the supervision of the Range and Pasture Department and in close collaboration with the Forestry Department” (Bayoumi 2001, 116), with a clear focus on Kordofan and Darfur. Forest policies continued, such as a 1989 Forest Law on reserved forests, in which regulations on fire lines and the prohibition of the use of fire in these forests are laid out. However, with insufficient funds and almost no fire management structures, whether at a government or municipal level, the diagnosis looks bleak: “A fire management organization is not in place. […] Early warning, detection and monitoring systems are not available. […] Fire research is absent” (Bayoumi 2001, 116).

110As of 2021, the outcome of these initiatives remained unclear, however, although the consistency and structural solidification of the issue, even without the NCP elites at the helm, appeared in April 2021. Two years after the deposition of Omar Al-Bashir, the news of another date palm fire sounded disturbingly familiar:

Fire causes great losses in Gargud, Dalgo Locality, state officials completely absent 3 April 2021 Nubokeen, Dalgo A civil defence force of Dalgo Locality and citizens of Gargud village are trying to gain control over a fire that broke out in the date palms of the village midday yesterday. Great damage was caused to houses and farms, while the officials of the locality or the administrative unit remained completely absent, notwithstanding the great losses the village experienced and the continuing threat to date palms, houses and animals from the fire. It should be mentioned that another fire had broken out in the village a week before, but it was brought under control by the civil defence force of Dalgo Locality, which suffers from large gap in workers and equipment, although the area faces a risk of date palm fires and in spite of repeated demands directed to the regional government to support and to rehabilitate this vital and important service.

حريق يسبب خسائر كبيرة بقرية قرقود بمحلية دلقو وغياب تام للمسؤولين أبريل 3, 2021 نبوكين – دلقو تحاول قوة من الدفاع المدني من محلية دلقو ومواطني قرية قرقود، السيطرة علي حريق شبّ في نخيل القرية منذ ظهيرة الأمس، مما تسبب في خسائر كبيرة في المنازل والمواشي، مع غياب تام لموظفي المحلية او الوحدة الادارية، رغم الخسائر التي تعرضت لها القرية وتعرّض المزيد من النخيل والمنازل والحيوانات لخطر الحريق والجدير بالذكر أن حريقاً آخر نشب في القرية قبل أسبوع، ولكن تمّت السيطرة عليه بواسطة قوّة الدفاع المدني بمحلية دلقو، والتي تعاني شحّاً كبيراً في العاملين والمعدات، بالرغم من أنها تواجه خطر حرائق النخيل بالمنطقة، وبالرغم من المطالب المتكرّرة لحكومة الولاية لدعم وتأهيل هذا المرفق الحيوي والهام.

111The argument that areas were being targeted, whether by arson or negligence, had been buttressed by Al-Ḥassan Hāshim with reference to other regions, such as the town of Shehibi about 35 kilometres south east of the Gallabat/Metema border crossing between Sudan and Ethiopia, where a fire broke out on 5 May 2015 and was dealt with by firemen from Sudan before the Ethiopian firefighters arrived.

112In this instance, Al-Ḥassan argues that it was not a question of non-volitional negligence, but rather one of intentional negligence that became the object of identifying responsibility. In his view, a government’s first duty is to protect citizens and their property, but the Sudanese government has not begun any kind of investigation—or at least it has not published any results—and there have been no court cases, no compensation, no provision of civil defence equipment, and so on since 2005. The government repeatedly blamed the issue on the negligence of farmers, but if these patterns of blame were followed to their natural conclusion, he concluded, the laws against criminal negligence, which exist, should be used to impose fines and should be buttressed by a law protecting date palms. However, no such procedures were adopted at any level of government he contacted, whether they be local administrations, regional governments, ministries, etc.

Political economy

113This resilience of the notion of a legal order and institutional solutions is an interesting topic, but we will focus here on date palms as an issue for economic policies and political economy. According to Al-Ḥassan Hāshim, date palms began to lose not only their economic but also their central social importance. The generation of Al-Ḥassan, who is now in his sixties, had experienced this in the form of a creeping process since the 1960s. In fact, he viewed the Aswan High Dam as the starting point for these broken relations, as economic and social processes are not independent from each other. Previously, Khartoum had not been on the map of date trade from northern Sudan as it was directed towards Egypt, where a large population meant a large market; ships took date palm products there and brought back oil, sugar, tea and clothing. The High Dam certainly broke this link, but it did more than that: since at least the 1950s, the population around Halfa and Faras had been pursued by what he called the nightmare (al-ḥulm al-muzʿij) of the High Dam, a collusion between the Egyptian and Sudanese governments against the Nubian population that was manifested in the 1959 agreement on the distribution of the waters of the Nile.

115In Al-Ḥassan Hāshim’s opinion, it was by no means a coincidence that the same number of date palms he found had been destroyed were then planted in central Sudan. For him, it was the fulfilment of a systematic marginalisation of the Nubian areas and an empowerment of commercial interests close to the Ingaz regime. This was added to by numerous other issues, or “instruments,” as Fāris termed it: the land grabbing, the pollution from mining, the erosion of islands, the withholding of access to electricity, the impact of the Hamadāb (Merowe) Dam, the separation of River Nile State from the Northern State (to divide the population and their common causes), the leakage of groundwater into archaeological sites (from irrigation schemes), the smuggling of antiquities etc; a combination of ten interconnected instruments of marginalisation and political violence.

Circumstances

118However, Al-Ḥassan Hāshim made it clear that his argument had a political function in the struggle against an Islamist regime whose supremacist agenda against Nubians and Darfuris and others, and whose exploitative schemes for their homelands, he asserted, were globally known. He trusted that building his analysis around the well-known nature of the regime would prevent him from being called unscientific, a charge he clearly valued avoiding. Leaving aside his argument at this point, we can take note of the multiple levels of accusation reported here, where a specific event or phenomenon was treated as only one instance or supporting element within a much broader pattern derived from a history that was presumed to be universally acknowledged.

Case study 3: Difoinarti

Georeference: 19°56’46.8”N 30°30’01.8”E (geo:19.9462,30.4973)

Case study 3: Difoinarti (satellite view)

Walk 3a: Difoinarti, 4 March 2017

Walk 3a: Difoinarti, 4 March 2017 (map and text)

See full screen: https://bit.ly/​3kUCKXw.

Walk 3a: Difoinarti, 4 March 2017 (text)

On ArcGIS Storymaps: https://storymaps.arcgis.com/​stories/​b5f5fe7763a144af9021cf6a11fa4513.
On Umap: https://bit.ly/​3kUCKXw
Persistent identifier (text and images): https://doi.org/​10.5281/​zenodo.7350482.
Download PDF Version: https://zenodo.org/​record/​7350483/​files/​Walk_3a_Difoinarti_4_March_2017.pdf.

Walk 3b: Difoinarti, 19 September 2017

Walk 3b: Difoinarti, 19 September 2017 (map and text)

See full screen: https://bit.ly/​3DtsVWQ.

Walk 3b: Difoinarti, 19 September 2017 (text)

On ArcGIS Storymaps: https://storymaps.arcgis.com/​stories/​eab76daa10384206a945d48cd459f734.
On Umap: https://bit.ly/​3DtsVWQ
Persistent identifier (text and images): https://doi.org/​10.5281/​zenodo.7350537.
Download PDF Version: https://zenodo.org/​record/​7350538/​files/​Walk_3b_Difoinarti_19_September_2017.pdf.

Interpretation: Historical disconnection

119One of the harrowing problems with writing about forest fires is the limitations verbal expression imposes on describing its infernal, unbounded quality, its vicious hunger and power. Equally, it is not easy to come close to the emotions and feelings that arise when something of profound importance is eaten away by these fires in front of one’s eyes. It becomes even more of a challenge when this significance is nourished by multiple temporal layers formed by recent events, childhood memories, social memory and historical narratives.

120We will now take another look at ʿAbd al-Ghannī’s thoughts during the walks in the light of two other interviews conducted on 19 September 2017. Once again, they related to the value of date palms and what was lost through the fires. What transpired were histories of past community prosperity and hospitality, and of past abilities to defend against animosity.

Prosperity

123He was aware, though, that the annual date production had much greater significance for others, especially those who depended on it as a cash crop to pay their debts, such as the often landless and palmless collectors of tugin fenti. This Nobíín term describes dates that have fallen to the ground and are permitted by land and palm owners to be collected by non-owners. During harvest season, groups of people, mostly women, as on Difoinarti, could be observed moving through the fields with buckets or sacks, picking up dates from the ground. The dates collected in this way could amount to as many as 4-5 sacks gathered by one person over the whole harvest season, and thus bring about 3,000 SDG (€441) to a low-income household, depending on the seasonal prices.

124For people like ʿAbd al-Ghannī, however, the current situation was only a shadow of the island’s former prosperous self as a major historical source of dates for the entire Mahas area and beyond. When we looked at the square he described as an “evacuated landscape of wealth” during Walk 3b, his eyes wandered, and he spoke of a place where thousands of sacks of dates used to be distributed to those who flocked to it, a place of plenty that was now empty.

125Another of ʿAbd al-Ghannī’s concerns was directed more towards the inside. Disconnection from history was a result not just of fires breaking out, but also of a slow erosion of the region’s demographic stability and continuity. Being a place people come to was not just a proud tradition on the island of Difoinarti: to him, the guesthouse (masīd) on the mainland, which is shown in Walk 3b, was a symbol of generosity towards visitors and those in need of support that went beyond gaining a reputation. There was a practice, he said, for part of one’s date harvest to be placed there anonymously, a general sense of reciprocity that was diametrically opposed to the minutely detailed calculations of an inheritance.

126However, the over-60s harboured doubts about a future younger generation, as cursory, seasonal visits from Khartoum or abroad—rather than life paths—brought most young people to the area. During our visit on 19 September 2017, it transpired that ʿAbd al-Ghannī had made a written agreement with a friend to plant over 100 palms that already imagined future generations benefitting from it, but also prepared for the possibility that those who did the harvesting then would no longer be trusted friends, unlike those who were planting them now; it was a written manifestation of an eroded trust in continuity.

Defence

127When a smaller fire broke out in 2017, ʿAbd al-Ghannī saw a familiar scenario taking place: the police arrived after the fact, took photos but no samples, interviewed eyewitnesses, left and never reported back to the population. It was not only in his account that this image emerged of public servants who seemed to see no duty to have any active contact with the public, but rather the right to need to be pursued before they reacted.

129In his interview, he acknowledged the various opposing arguments about what happened, from a government plot to force migration to the palm tree owners’ fault due to their negligence (ihmāl). He stated that the geographical breadth of the fire locations, which covered the whole region from Akasha to Dongola, suggested that there was no specific target. But he also did not dismiss the fact that “the government’s” objectives benefited from them, if not through active agents, then because of the lack of response. In his opinion as the man in charge of agriculture for the official Popular Committee, a kind of local administration unit founded by the NCP throughout the country, this corresponded to a situation in which committee members were used as security agents rather than being provided with the means to provide services. Still, he believed, the fires themselves were the result of a general danger—fire breaking out during the hot season—exacerbated by a neglect of safety principles such as pruning and burning dry leaves at a long distance from grass, palms and other trees. This was facilitated by a change in the labour force, which had been made up of friends and relatives of the owners up to the 1970s but which had shifted to outsiders.

Transcript

Northern State
Dalgo Locality
Commissioner’s Office
Number: 1/A/48
Date: 3/10/2016
Dear Brother / Minister of Agriculture, Animal Resources, Fishery and Irrigation
Greetings and God’s Mercy and Blessings Be Upon You
Subject: Dafoi Island Fire
Referring to the subject above, we inform you that the damage from fire on the mentioned island exceeds an agricultural area of 300 feddan.
Accordingly and following our communication concerning machinery, we kindly ask for the provision of agricultural machinery and a motorboat for the island in order to rehabilitate the land to cultivate all sorts of horticultural and field crops.
 
May God reward you with good fortune
[Signature]
Dr. ʿAbd al-Karīm ʿAbd al-Raḥmān
Commissioner Dalgo Locality

133Instead, ʿAbd al-Jabār, a member of the ruling party, found himself moving from office to office in vain. The Agricultural Director addressed him as a fellow NCP member, saying “You know the situation of the party, Brother,” and promised to let him have one pump for each sāqiyyah as a gift from the District Commissioner that he could present in his own name. When the pumps arrived, they were missing the accessories that made it possible to use them. He continued to write, to the Ministry of Agriculture in Dongola, for instance, with detailed reports and calculations of the required compensation. However, during his numerous visits to Dongola, which were paid for by a joint fund provided by the date palm owners to the technical committee, he found himself having to defend the numbers for the damage that had occurred. A year after the fire, he received a letter from the Ministry saying that the Northern State, one of Sudan’s main gold-producing states, had no budget item (banda) that could be used for such a purpose.

134Thus far, this is ʿAbd al-Jabār’s account. The claim of political vulnerability by a party that is globally known for its authoritarian, exploitative rule certainly gave us pause; however, we would like to focus here on the changing forms of governance that shine through when this narrative is juxtaposed with the historical dynamics ʿAbd al-Ghannī referred to during Walk 3b.

136However, the multi-storey houses ʿAbd al-Ghannī had shown Enrico Ille still hinted at the time in the region’s history and the preceding centuries when it had resisted control by major empires such as the Funj/ʿAbdallāb sultanate and the Ottoman Empire, and when local royal courts defended the rule they had established over an indigenous population through fortified “castles.” We suspect that for people such as ʿAbd al-Ghannī nothing could be further away from that age of self-defence than petitions humbly asking for compensation from a government that was suspected of being behind, or at least not excessively concerned with, the events that had prompted this request.

Discussion and conclusions

Interpretation

137This article discusses the phenomenon of frequent date palm fires in Sudan’s Northern State in the 2010s. The origin of these fires was, and remains, a contested issue in the region and among those connected to it through belonging. The fire events led to a plurality of interpretations that in different ways assert a specific overlapping of agency, time and place.

138Rather than merely being a variety of interpretations in the narrative reconstruction of a location-specific event, they represent competitive, or even hostile, evidence-making in the context of a tense political environment: as landscapes in this region underwent radical changes—desertification, pollution etc.—pinpointing anthropogenic influences became complicated by major projects on the Nile in the shape of hydropower dams and irrigation schemes whose effects on the river and the local climate were little studied and little understood. Since these projects were often accompanied by substantial population displacements, they became perceived—and fought against—as an instrument and extension of the authoritarian, destructive and exploitative governance represented by the Ingaz regime (1989–2019). At the same time, areas where further dams were planned were already experiencing demographic decline, a development that itself invited explanations that extended from “natural” socio-spatial shifts to the demographic engineering of a regime that was only interested in the undisturbed extraction of resources.

139In other words, the dispute around the date palm fires concerned not just the story of what happened, but also what was behind it, its deeper nature. Many of the fires were not observed when they broke out, or else the observer did not come forward after the fact. This lack of observation constitutes what is politically the most “productive” circumstance of the situation, as it drives investigative processes that are concerned with assessing guilt with reference to the conflict dynamics beyond (or behind) the specific event. They also informed purposive documentation, as other researchers, activists and journalists gathered evidence—through eyewitness statements, police reports, photos, news and social media items—to support their own reading of the situation, an illustration of forms of political participation that involve non-local communication and evidence-making, such as digital activism and participatory GIS. The result was directional, but not mutually exclusive, discussions in a number of different media: daily conversations, WhatsApp, social media, news media, press conferences, official statistics and reports, studies by political organizations (especially anti-dam committees) and documentary films. This political entanglement of evidence-making was complicated by social ties, as sometimes the names of perpetrators who were known in the community were not made public in order to avoid court cases or other confrontations, thereby creating additional, albeit erroneous, observational gaps that had to be filled by interpretation.

140In our view, the problem presented in this article is a very tricky one, in the sense that none of the competing interpretations can be easily proved or dismissed. In addition, they are embedded in a political struggle in which better (more accurate) evidence does not necessarily create a “better” (more acceptable) argument for those involved, while the destructive effects of recurring fires linger on and raise the stakes. In fact, our own way of doing research was occasionally challenged by other investigating participants, not only for drawing the wrong conclusion, but also for not asking the right questions.

Immersion

142The methods we used to add qualitative depth, especially the walking interviews that form a prominent part of this article, invited their own complications, however. The interviews were not just intended to give us a basic acquaintance with the places we talked about; we also wanted to overcome the linguistic limitations of recreating the material aspects of fires and their aftermath in stationary interviews. We tried to do this by adding a certain degree of co-perception of the items referred to during interviews, which would otherwise have relied on our existing sensory repertoire, or on visual material studied after the conversation.

143The verbal recreation of the event still required an act of speech by the eyewitnesses as a descriptive representation of their memory, with us as their audience. Since interviews are basically a verbalisation (and summary) of social worlds, the participants’ linguistic practices influenced this specific communicative situation to emerge in a certain way. In addition, the purpose of the talks as perceived by their participants had an inevitable impact on what was communicated, and how.

144Meanwhile, an essential difference between perceptions and ideas of all the participants, including our own, remained, inevitably so. The presumption of a unidirectional “easing” of the conversational situation through co-presence, especially as regards its affective and emotional aspects, would simplify the experience: what fills the sensory gaps when past events that have not been shared are described is unavoidably autobiographical—the alarmed shouts when seeing a fire, the panicked running, the bursting flames, the crackling, the blasting heat, chasing the wind as it drives the fire here and there and cutting down palms to stop it—palms that were there many decades earlier when they created shade for children to play under, the same former children who were now holding an axe to cut them down. To recreate this environmental sensibility still requires acts of imagination that will differ to a greater or lesser extent from storyteller to messenger to audience, a “greater or lesser” that cannot be known, at least not without another speech-act that will introduce its own divergence.

145Walking outside also creates new social situations such as “interference” from chance encounters that change the conversation in ways that may not be readily understandable, especially in short-term or initial research. Similarly, it is not necessarily the case that one will be able to distinguish what statements have just been triggered by this specific route at this specific time. This illustrates not only the many things one cannot know about the participant’s lifeworld, knowledge and implicit references—and may never do, even after long fieldwork—but also how the problem of generalisation continues to be a daunting challenge.

148In such a situation, a mundane walk at least has the effect of exposing one to the gaze, questions and expectations of the people who have experienced loss and live with it, rather than merely enhancing one’s sense of “ground truth.” It is a variation of a researcher’s ethical imperative to negotiate access to a social situation and highlights how the ethical boundaries involved in achieving better observation are related not only to the question of general consent but also to issues of control over the content and/or the direction of ensuing exchanges. By stressing these ethical thresholds for reaching an “ideal” interview situation from a data mining perspective, with as few uncontrolled variables as possible, a critical reading of open-ended walking interviews can also serve as a cautionary tale against treating a researcher’s control of interview situations as the key to achieving “good” results. In other words, what makes this research process less rigorous from a certain epistemological position may also be what makes it defensible as a social act of intrusion into other peoples’ lives in the first place, especially in the context of situations of conflict and loss.

149A critical reading such as this must also reflect on the circumstances under which access to a certain social situation has been given, or at least has not been prevented. In our own case, this was complicated in several ways. Our access to eyewitnesses was facilitated by social and political connections that predated the research; we were thus, in a way, “compromised.” We both consider Sudan as our home, although to a different extent: Enrico Ille through multiple autobiographical ties that have been developed since 2005, and Mohamed Salah by birth and a myriad personal experiences, not a few of which have been marked by harsh political struggles. Mohamed Salah is also among those who belong to the region directly affected by the fires we discussed. He is also part of the reintegration of our analytical processes into the societal debate as one of the voices in the political discourse, including from standpoints and observations that are at odds with the interpretations of those who share our general political position vis-à-vis the Ingaz regime. While Mohamed Salah was thus much more directly exposed to the political processes and conflicts we have written about, the different position we hold in relation to ongoing political processes also privilege him when it comes to translating our interpretative work into these processes.

Implication

150Finally, this leads us to the implications of the ways we have represented what we found. An observational document that is “good” in more than a reproductive or reconstructive sense—namely a “good representation”—makes a contribution to the interstitial space between political discourse and academic exchange. In other words, epistemic empathy is not just about how we do research, but also what we do with it.

151The fires left scars that were far from inconsequential: the girl who was injured by embers on Difoinarti is a reminder of that (see Walk 3a). This elicits further questions on traces: was her individual memory of date palms now powerfully associated with the injuries caused to her by the fire? Did it join a social memory whose positive associations with and commitment to date palms was already in danger of being eroded, a community injury reflected in an individual one? Not only is her burned skin a record of all the circumstances that failed to prevent the fire or remove hazards from the post-fire landscape, but it may also be seen as an extension of multiple acts of negligence and—if one accepts this interpretation—of political violence.

152Responsibility is both complicated and focused here. Whether it is community negligence turning on itself or an aggressive political apparatus turning on residents, or whether it is a conflict between intent and result or between opposing intentions, the whole debate and the scars it relates to serve to document the landscape of political violence under the Ingaz regime and how it continues to dominate social, political and even socio-ecological relations. Our article is thus a part of the long-term project of bearing witness to the political crimes of the Ingaz regime, whether starting date palm fires was one of these crimes or not. This is not about opening old wounds—which are actually still quite fresh and continue to be inflicted—but about becoming involved in the situation by paying tribute to scars and what caused them, and the threads they derive from and speak of.

153At the same time, we have applied a principle of mediation here: rather than merely pointing out what is contentious, we also highlight what is consensual. In this case, there are three elements of interpretation: the hazardous landscape left by the lack of pruning; a hot, dry climate that is tending towards becoming hotter and drier; and the lack of an infrastructure of responses to fire (although there are differing views on whose responsibility this is).

155Furthermore, the resilience of date palms, their “strong heart,” often meant an independent recovery, an inherent strength, which existed despite human interest or lack of interest in benefiting from the fruits of their reproductive labour (dates). In fact, their survival triggered less exaltation than might have been expected, as their now burned trunks meant that no workers could be found who would make the tedious and dangerous climb up them, although it would be easier than usual to reach the dates once they were up there.

156There is also a practical implication for understanding negligence not just as a common problem but also as a shared responsibility. Negligence was seen by all our interlocutors as a kind of structural failure of a government that was unable to protect or care for its citizens, although they differed on the question of whether or not this was because of malice on its part. Another form of negligence everyone agreed on was that of the date palm owners. However, if government is understood as a shared responsibility of its citizens, then so is reducing hazards, and in areas with a low population density, state-supported community fire management seems to be a way of sharing this responsibility. A mediatory approach to shared interpretation looks for these points of convergence, which can then be translated back into a societal debate in which we strive, as all interlocutors do, to be heard.