Geography and ethics III: Description as a matter of moral concern

Abstract

This report focuses on moral description. It considers how comparative practices often entangle moral geography in philosophical dilemmas, such as debates over moral relativism. Using examples from geographic work on language, algorithms, the appropriation of anti-colonial critique, and Indigenous data sovereignty, the report shows how and why descriptive praxis matters to accounts of moral geography. As the final installation of a three-part series of reports on ethics and geography, it draws together a broader critique of moral geography's initial distinctions and suggests a more constitutive view of ethics for geographic scholarship.

I Introduction

Resignation is my favourite contranym, a word with two opposing meanings. Someone can be resigned to something beyond their control, like fate. But resignation can also be voluntarily; ending affiliation with an organization because they don’t condemn genocide, for instance. There’s also resignation that cuts across opposing meanings. In this sense, it can be the case that someone must act both because something compels them (perhaps collective or cultural practices beyond their sole control) and because, even if that force is set aside, they would still choose to act because they see no other moral option. These tensions, between ethics that may derive from within or from without, are what Albert Schweitzer (1987 [1949]: 314) had in mind when he wrote ‘resignation is the vestibule through which we enter ethics’.

Schweitzer’s moral resignation was fraught, although his ethic of ‘reverence for life’ was key to the Nobel Peace Prize he received in 1952. Schweitzer’s ethic took shape in the years after he resigned from a successful career in music and theology to retrain as a physician and, in accordance with his critique of colonialism, seek atonement for European sins in what is now Gabon. It is well known that Schweitzer never disabused himself of Eurocentrism or white superiority even as he poured his time and energy into humanitarian efforts (see Barsam 2009; Goodin 2013). Yet his shattered faith in Western ‘civilization’ also led him to expand ethics beyond humans in 1915, decades ahead of environmental ethicists later in the 20^th century. The idea came to him after scrawling disconnected notes for three straight days while on a barge creeping up a river ‘when, at sunset, we were making our way through a herd of hippopotamuses, there flashed upon my mind, unforeseen and unsought, the phrase: “Reverence for Life”’ (Schweitzer, 1933: 124). As he developed his ethic, Schweitzer (1933: 126) argued, ‘a man [sic] is ethical only when life, as such, is sacred to him, that of plants and animals as that of his fellow men, and when he devotes himself helpfully to all life that is in need of help’.

Why start this final report on geography and ethics with resignation? First, I have structured these three progress reports using Smith's (2000) tripartite division of moral geography. The first was about metaethics, and the ways that accounts of what morality is also needed to resign themselves to how anthropogenic impacts on the planet affect the conditions of ethics (Schmidt 2022). The second was about justification, and the normative reasons for why actions are right or wrong. There, I identified an emerging ethics of anti-oppression that isn’t resigned to justification via abstract theories but is instead articulated through lived responses to injustice (Schmidt 2023). This final report takes up the third of Smith’s divisions to discuss descriptive ethics, which is typically defined as the field of inquiry that seeks to understand the moral practices of a particular group. Using Smith’s distinctions, which are often used in philosophy as well, I resigned myself to a structure for these reports that aligned with past and on-going work in moral geography.

However, there is a second reason for starting with resignation, and which has to do with challenging moral geography’s orienting distinctions. Much of geography’s ‘moral turn’ in the 1990s shared Schweitzer’s ethos of seeking to retain and enhance connections to Western moral philosophy and its ideas of an ethical self, but also to do so in view of colonialisms, their unfinished aftermaths, and ongoing sites of violence. Smith’s tripartite division was one way this was done. Others were outright rescue attempts. Sack (1997), for instance, sought to anchor a relational view of ethics in a spatial view of moral personhood that he termed Homo Geographicus. Sack’s target wasn’t colonialism per se but a closely related phenomenon: the fractured moral imagination produced by accelerating globalization. Sack argued that because all action was geographical the discipline could anchor ethics as a project that respected the situatedness and partiality of knowledge in a world of complex connections without giving up on the form of rational reflection characteristic of Western moral philosophy. This type of project has an element of pragmatism to it – Rorty (1989) once described the ‘final vocabularies’ of religion or metaphysics as little more than a series of descriptions and redescriptions of the world – but it also reflects a kind of resignation that can be done without. In this sense, moral geography can and should operate free of the tripartite distinctions through which it originally sought truck with Western moral philosophy. There are more creative and robust opportunities for theorizing moral geography (many philosophers have moved on as well) and for the topic of this essay: describing lived moral geographies.

By making resignation an explicit theme of this essay, I similarly aim to trouble moral geography’s account of how goods or harms are described. There is no obvious first step in developing a geographic practice more sensitive to the demands of description. But there are steps to avoid. These include those that Flanagan (2019: 4) cautioned against, and which take as their reference point WEIRD people – Western, educated, industrialized, rich, and democratic – a group that has dominated accounts of Western moral philosophy, but which is ‘possibly the most unrepresentative group imaginable, less representative than our ancestors when the ice melted at the end of the Pleistocene’. In what follows I argue that if this implicit reference point is abandoned, and with it the reference points for many of the debates over how to describe lived moral geographies, then one upshot is an opportunity for thinking of description as a matter of moral concern. This idea is elaborated in Section II through a discussion of the force of moral description, including the work of morally outsourcing description, such as to algorithms. Section III concludes not with a plea for more geographic engagement with ethics, or my own remarks on where it might go, but on what moral geography is resigned to if it continues to operate in the confines of its interdisciplinary inheritance.

II The Force of Description

Descriptive ethics seek to understand existing moral practices and often utilize comparisons to do so. In this sense, descriptive ethics aim to understand what groups take to be moral or immoral rather than to make normative arguments about what is ethically justified (Frankena 1973). Early work on geography and ethics (e.g. Proctor, 1999), including previous progress reports (e.g. Cutchin 2002), similarly argued that descriptive ethics can help understanding historically and socially situated moral practices. In that early work, moral geography largely recapitulated the distinction of Western philosophy between facts and values. Facts, in descriptive ethics, are claims about a particular group’s moral practices; that is, claims about what is the case for a particular group. Values are at work in assessments of those facts and usually operate through comparisons of two kinds. One is by assessing descriptive facts against a broadly held normative standard to explain the moral weight of different practices. For instance, comparing practices that may be culturally accepted by a particular group, but which others would deem unethical, by comparing them to a broader standard, such as human rights. The other kind of comparison assesses moral practices in reference to the practices of some other group. Neither kind of comparison is wholly satisfactory. For instance, the chauvinism of cultural comparisons is frequently rejected. Likewise, the assumption that broad normative standards are themselves independent of culture is also scrutinized, and where those broader norms are not critically appraised, they have also been charged with smuggling chauvinism into moral debates. Sometimes complaints target the distinction of facts and values itself, such as in Putnam’s (2002) trenchant rejection of the dichotomy.

Given the challenges of comparison, descriptive ethics are often resigned to a set of uneasy choices. For instance, to avoid charges of chauvinism one can appeal to moral relativism and embrace the lack of any culturally independent reference point for ethical adjudication. But as Smith (2000) noted, moral relativism doesn’t necessarily follow from cultural relativism, and geographers can appreciate contextual demands without conflating the two. Another option is to adopt some broadly agreed framework and either accept its cultural disposition or bracket that disposition out in order to make more instrumental use of that framework. For instance, the declaration of human rights emerged in a Western context – the first draft was a transposition of Henri Bergson’s ideas (Curle 2007) – but human rights are nevertheless used strategically by others, such as Indigenous peoples (Carmalt 2018; Tsosie 2012). How successful these strategies are can depend to a significant extent on the power to shape the interpretation and application of these broadly held norms (Laliberté 2015; cf. Moyn 2018). There are also variants of a third option, which is to anchor comparison in some trait common to moral agents. Often, rationality has often been the default. In a previous report, Olson (2018) discussed how other traits – emotions especially – anchor ethics in characteristics other than reason (cf. Valentine 2003, 2004). Many philosophers share the view that classic accounts of rationality are insufficient (e.g. Hoffmaster and Hooker 2009). There are also virtue ethics, which are rooted in traits of character. Using traits or virtues face challenges too. As geographers (and others) have noted, virtues are as culturally embedded as values, so relativism is not actually addressed through them (see Korf 2007). Since each option faces significant challenges, moral geography often faces an impasse that is consistent in many ways with the understanding of descriptive ethics it took from Western moral philosophy.

There is another option, however, and that is to treat comparison itself a site of moral concern. This option incorporates spaces of comparison themselves into its ethical orbit. Stengers (2011: 56) provides a path to this end; she’s also concerned with how universalizing norms versus particularist approaches affect comparative accounts but also argues that comparisons ‘must not be unilateral and, especially, must not be conducted in the language of just one of the parties’. Her argument to make comparison itself a ‘matter of concern’ does not reject the fact/value distinction but instead turns towards the practices, including languages used, that direct attention to some concerns and not others. To be concerned, and to make something a matter of concern, is to focus on how the practices through which people pay attention to what matters to them are not incidental to some object of concern. Rather, those practices are constitutive of how people forge relations – epistemic, ontological, and ethical – through their practices. Stengers (2011: 59) is in this way also confronting the unstated but often powerful research imperative that ‘comparison must be possible’. It is a powerful idea, since even in calls for more ‘humble geographies’, comparison is cited as a key practice that we should get more moral about (Saville 2021). But what if comparison isn’t possible? If it is not, the reference points for description must be sought somewhere other than broader standards or cultural comparison.

Countering the presumption that comparison must be possible resonates deeply with practices of Black and Indigenous geographies (and others) that reshape geographic praxis not because of, or in comparison to existing norms. Rather, those geographies set their own terms and conditions for inquiry (cf. Fujikane 2021; Hawthorne and Lewis 2023; Moulton and Salo 2022). In Stenger’s Stengers (2005) parlance, they have their own ecologies of practice through which they attend to what concerns them. To take description as a matter of moral concern similarly requires attending directly to practices of description. These come in many mediums. Talk and text are prominent, but drawings and co-produced materials are no less important. Gassner (2021), for instance, focuses on drawing as an ethical and political practice in spatial analysis (cf. Brice 2023). Bonfiglioli (2016, 2023) has similarly been seeking to recover the idea of chora, and the idea that Earth is shaped by human practices: images, practices of representation, and narratives of mobility in and with the landscape. Relatedly, to redescribe relations of mobility and care in the context of what Sheller (2023) terms ‘climate coloniality’ requires, in such cases, to also think across different kinds of ethical practices and the cosmologies through which they are situated (cf. Ross 2019). Similarly, McLean et al. (2018) think through the ways that what constitutes social or ethical relations into what Val Plumwood (2008) once termed ‘shadow places’ that are obscured by the force of dominant descriptive (and normative) accounts. Here, then, description is tethered to broader cultural practices such that lifting just the ‘moral’ component out is an arbitrary cut through fabrics of relationality and belonging.

These concerns carry into the typical practice of ethics: the use of language for description. Consider this quote from the cover of the 21 March 2024 issue of the London Review of Books:

‘Memories of Jewish suffering at the hands of Nazis are the foundation on which most descriptions of extreme ideology and atrocity have been built. But these universalist reference points are in danger of disappearing as the Israeli military massacres and starves Palestinians, while denouncing as antisemitic or champions of Hamas all those who plead with it to desist’.

Though not taken verbatim from Pankaj Mishra’s (2024: 10) essay – the cover text combines some closely related sentences – the quote captures a critical aspect of the relationship between moral description and the points of reference that anchor comparative claims. Comparisons of evil, for Mishra, have for decades been anchored in the Shoah as a standard of evil. But the universality of that reference point is at stake not merely because there is more violence in the world but because losing the shared practice of using that point of reference also entails losing the universality of the index that maps horrific harms to moral atrocity.

There is something else the quote from Mishra makes clear: description is not a prelude to analysis. Instead, description is a practice that forges relations to points of reference. In this regard, and to draw on Ogborn’s (2020) analysis of ‘uttering geographies’, the language of description constitutes part of an intended action. Ogborn takes this view from the speech act theory of J.L. Austin (1975). Austin was no stranger to the challenges that confronting evil in wartime presents (see Rowe 2023). But he is more frequently remembered for his speech act theory and those utterances that forge relations in the act of speaking – illocutionary acts – such as making a promise or naming a boat (e.g. ‘I dub thee The Witness’). Illocutionary utterances can also carry moral force, such as in the act of describing Peruvian street children as ‘vulnerable’ or in assertions of what constitutes ‘good’ parenting in South Africa (Aufseeser 2020; Rubin and Parker 2023; cf. Wood et al., 2020). Similarly, the growing number of works on geographies of care pivot on how care is described and situated; care for cities, mobilities, humans, and non-humans among others (e.g. Lonkila 2021; Middleton and Samanani 2021; Power and Williams 2020; Ravensbergen and Schwanen 2024). For instance, Richmond et al. (2024) discuss how the idea of Indigenous repossession requires describing ‘repossession’ not in popular terms (i.e. as what happens if payments are defaulted on) but as a response to colonial dispossession that centres Indigenous relations to land. By contrast, descriptions can run against care, such as when theologically loaded moral descriptions are wielded together with individualizing assumptions about drug users (Williams 2015). Or, as Leshem (2025) examines in cases of ‘no man’s land’, where there are sites and landscapes of radical un-caring. Others see in redescription a source of recovery: Weis (2018) argues that redescribing landscapes as the ghosts of extinct species can evoke more powerful notions of loss, while Ingram (2022) argues that artistic acts of redescription can forge new relations to otherwise ‘orphaned’ or neglected spaces and create the conditions for new ways to ethically value them. Across these cases, descriptive speech acts often carry the force of constitutive moral relations rather than operating as a prelude to acts of comparison.

It is not only humans that utter geographies. A growing number of geographers are concerned with how the algorithms of machine learning and artificial intelligence use comparisons to make ethically relevant descriptions. Amoore (2020: 8) argues algorithms ‘must necessarily discriminate to have any traction in the world’ and that such acts of discrimination are ethical. That is, they compare subjects and data in acts of moral description, such by labelling people or places as threats, or risks, or as disposable or collateral. Amoore’s is a call that Maalsen (2023) argues should be met with an ethic of care for algorithms themselves and the kinds of force entailed by their outputs. There is a resonance here with longstanding concerns of feminist geographers around how the ‘counting’ of different attributes might affect claims about data and inferences from it (Fuentes and Cookson 2020). By contrast, Lally’s (2022) focus is on the socially situated use of algorithms, such as for policing, which places descriptive ethics at the core of decisions and comparisons among a suite of relations involving data, software, and state agencies. Lynch et al. (2022) highlight how automation matters to other ethics of care too, such as in cases where robotic devices deliver health interventions once done by humans and in so doing create spaces of ‘more-than-human intimacy’.

Geographers are not alone in reflecting on what kinds of practices, and with respect to what reference points, matter for description. Nancy Fraser (2022) argues the entanglement of capitalist ethics with non-economic normativities of social reproduction, politics, ecology, and resistance means that values of care, autonomy, self-determination, stewardship, intergenerational justice, and community must be defended through boundary work that recognizes capitalist influences without taking them as a reference point. Fraser’s disinvitation from Cologne University owing to a letter she signed regarding Israel’s actions in Gaza points to the force of description that geographers, such as Agha et al. (2024), have sought to engage amid charged linguistic and political landscapes. But it also points to what Ogborn (2019), Amoore, and others emphasize: practices of description carry normative force in the act of description itself. For moral geography, the ‘imperative to compare’ is not a practice that one needs to be resigned to. In fact, it is a practice that must often be confronted when description is understood as a matter of moral concern.

III Resigned geographies

The force of description, the often unstated ‘imperative to compare’, and the ethical challenges of comparative practices all present themselves as reasons to not remain resigned to Smith's (2000)tripartite ordering of moral geography. There are considerable stakes for moral geography if it remains resigned to this disciplinary inheritance, and I am not the first to note it. A decade ago, Clive Barnett (2014) ended his reports on ethics and geography by shifting emphasis from description to the moral affects that seek to place normativity in everyday life (e.g. Dekeyser et al., 2023). But Barnett’s argument around the importance of affect (and theories of affect) doesn’t resolve the tensions identified above. This is because affects also need to be described – so, some ecology of practices is at work in how moral geographers pay attention to affect too.

One way that affect continues to matter to moral geography is in the appropriation of anti-colonial language for illiberal agendas. Zhang (2023) examines his kind of case, where anticolonial critiques of the liberal international order are used not for the liberatory agenda that shaped them but instead for authoritarian ends. Often, this entails reaffirming essentialized accounts of social and cultural difference rather than confronting them – a practice that has a long featured in propaganda and policy used to produce moral geographies of the international liberal order in the first place (Seemann 2020). The more recent dynamic of co-opting anticolonial discourse, however, is also not entirely detached from the trends identified in post-secular geographies regarding the ways religion is again being asserted into political affairs, and with consequent effects on the moral geographies of state claims (see Cloke et al., 2019). As Koch (2023) points out, descriptions in these cases matter owing to the reference points taken up in accounts of different kinds of nationalism and affective dispositions regarding identity; these can often amount to simplistic or equally essentializing moral geographies (cf. Koch 2019). As one example, Anderson et al. (2023) discuss how outdoor activities are used in projects of disciplining refugees into moral geographies of citizenship in Norway.

Concern over affect, however, does not get purchase on the full suite of concerns over the moral outsourcing of descriptions. For instance, in addition to the challenges of algorithms noted above, there are matters of data sovereignty that are a growing concern for Indigenous peoples. Machine learning and big data in these contexts are key sites of concern regarding Indigenous protocol and the ownership and use of information (Lucchesi, 2020; Brewer et al., 2023; cf. Water et al., 2021). This is especially so given the growing interest in the use of Indigenous knowledges to address contemporary challenges like climate change or biodiversity loss (Williamson et al., 2023). But what constitutes ‘Indigenous data sovereignty’ is a matter that also requires jettisoning the kinds of comparative practices that describe different forms of self-determination in terms consistent with Western ideas of sovereignty (see Rivera 2023; cf. Tsosie 2019). These overlapping considerations are all at stake in practices of moral geography that seek to site data, its relations, and machine learning algorithms that rely on Indigenous knowledge or affect Indigenous relations.

To conclude: across my reports I have engaged multiple strands of moral geography to make a case that the subfield needn’t be resigned to the initial distinctions through which it sought conversation with Western moral philosophy.1 In this final installation, I have made similar case for descriptive ethics and the need to rework geographic praxis without the imperative to compare; whether to standards, or theories, of different social practices. Should non-human agency appeal to object-oriented philosophy for ethics or to place attachment (e.g. Mould 2019; or Lockwood and Heiderscheidt 2023)? One approach to answering this question would be to burrow into the different kinds of descriptive, normative, and metaethical claims such a comparison might evoke. Or, when that bottoms out, to turn towards new geographic genres, such as those anchored in affect. I have argued for something else, something less WEIRD. Something that pushes the discipline towards accounts of, and with, the ways that matters of moral concern are constitutive for geography, for lived geographies, and for those not allowed to describe their worlds freely.