The history of the concern about extinction in environmental ethics

A period from the late eighteenth century to the early twentieth century changed the way extinction was thought about in Europe and North America. During that time, three developments came together. Firstly, mostly thanks to Darwin, the idea of evolution and extinction as its key component replaced creationist ideas that distrusted the very possibility of extinction, if it was not outright denied. Reasons for this former denial are historical, as the influential Greek philosophers Plato and Aristotle, unlike some popular writers of the time, paid no attention to fossil bones of no longer existing vertebrates (Mayor, Reference Mayor2011). For them, species could not disappear except locally – a belief in line with the principle of plenitude dating back to Plato (Lovejoy, Reference Lovejoy1936), which states that all possible lifeforms necessarily exist and cannot disappear. Figuratively, the extant kinds form a ‘great chain of being’ and each kind in this chain is there necessarily; thus, the chain is unbreakable, and nature is in balance. This idea prevailed in pre- and early modern Europe and was endorsed by, for instance, Linnaeus as late as the eighteenth century.

Secondly, given the importance of the notion of extinction to evolutionary thinking, extinction was not seen as a matter of concern but rather the opposite: extinct species were too weak to survive, or an extinct species was replaced by another, thus maintaining the balance in nature. Both cases – through weeding out the weak and introducing novel species – were regarded as improvements in natural evolution (Barrow, Reference Barrow2009; Sepkoski, Reference Sepkoski2020).

Thirdly, despite this, and most relevant for thinking about the ethics of extinction, a concern about extinctions as universal and irreversible events started taking shape (as early as Marsh, 1864/Reference Marsh2003) and was broadly reflected in new legislation such as the Sea Birds Preservation Act of 1869 in the UK. This historical change is well conveyed by Aldo Leopold (1949/Reference Leopold1989, 110), when he stated that ‘[t]he sailor who clubbed the last auk thought nothing at all. But we, who have lost our [passenger] pigeons, mourn the loss’. Leopold (Reference Leopold1989, 210–211) also already noted that economic considerations fail to provide reasons for protecting those species that lack economic value. Instead of extending economic valuation, as has been proposed later by economists, Leopold saw conservation as action against extinction that also rests on non-anthropocentric ethical ideas. Nevertheless, a systematic philosophical discussion on the ethics of extinction evolved much later with the formation of the academic field of environmental ethics since the 1970s.

Early books and articles on environmental ethics did not explicitly discuss the ethics of extinction. The discussion focused on the flipside of the coin: Why protect endangered species (Elliot, Reference Elliot, Mannison, McRobbie and Routley1980; Gunn, Reference Gunn1980; Rescher, Reference Rescher1980)? One could see here a connection with new legislation, such as the US Endangered Species Act (1973), but the discussion broadened to also include the significance of diversity in nature more broadly (Naess, Reference Naess1973; Norton, Reference Norton1987), to incorporate the possibility of multiple extinctions in a relatively short period of time (for mass extinction, see Sepkoski, Reference Sepkoski2020) and to introduce a new concept, biodiversity (Wilson, Reference Wilson1986). For example, some ethicists have tried to develop a naturalistic account of the human propensity to value species and their diversity (Callicott, Reference Callicott1984, 305, Reference Callicott and Callicott1986; or, more recently, elaborated E.O. Wilson’s biophilia hypothesis, Baxter, Reference Baxter2007; cf. Takacs, Reference Takacs1996).

Arguably, one of the earliest contributions on the ethics of extinction is Richard Sylvan’s (then still known as Routley) famous thought experiment of the last person on earth (originally the ‘last man’), firstly formulated in 1973, which can be made relevant for thinking about extinction. Sylvan writes:

Later in the same article, he also specifically considers ‘vanishing species’:

He asked whether species or nature more broadly are valuable beyond their potential usefulness for humans. What Sylvan’s famous case suggests is that the permissibility of extinguishing species – as he argued, allowed by traditional Western ethical systems that conceive of nature as something to be used by humans as they please – is an example of what he called ‘human chauvinism’, now better known as ‘anthropocentrism’ (for a critical discussion of the ‘last person’ argument, see Peterson and Sandin, Reference Peterson and Sandin2013). Sylvan’s thought experiment has successively been modified and discussed by many other philosophers such as Mary Ann Warren (Reference Warren, Elliot and Gare1983) and Robin Attfield (Reference Attfield1981, Reference Attfield1983). The latter engaged in this type of argument to scrutinise the possibility of human extinction in the context of nuclear armament and ‘ecocatastrophe’. He aimed to show that non-sentient living nature such as trees has morally relevant interests and that causing their extinction by the ‘last man’ would be morally wrong, even if it would not affect any other humans.

Writing at the same time as Sylvan, Joel Feinberg (Reference Feinberg and Blackstone1974) addressed the ethics of species protection from a different argumentative angle. Because he argued that only (human and nonhuman) individuals with interests can have rights, he rejected the idea that species have a right to exist (for an opposing view, see Staples and Cafaro, Reference Staples, Cafaro, Cafaro and Crist2012). Therefore, he claimed that humans have a duty to protect endangered species, but the ultimate addressees of this duty are future humans, not the species themselves (see also Elliot, Reference Elliot, Mannison, McRobbie and Routley1980). Likewise, Nicholas Rescher (Reference Rescher1980) depicted species conservation ‘as a quintessential humanitarian task’ and John Passmore (Reference Passmore1974) considered wanton eradication of species to be acts of vandalism.

In brief, very early on, there were opposing opinions about why human-caused extinctions are morally wrong, while there was largely agreement on the issue that anthropogenic extinctions are wrong (with exceptions such as the smallpox virus). The similarity between these different views was that they regard extinction as a loss of value, either for humanity or for the universe as considered from a human-independent perspective. Independently of whether one endorses an anthropocentric (i.e. human-focused) or non-anthropocentric rationale for species conservation, both give rise to a range of resulting questions that were formulated in the literature of environmental ethics early on, such as: Are species individual entities or life processes that come to an end with extinction (Rolston, Reference Rolston1985, Reference Rolston and Nierenberg1995) (section “Defining extinction”)? What are acceptable costs for protection efforts (Norton, Reference Norton1987; Naess, Reference Naess and Rothenberg1989), and relatedly, should all species, even the most harmful ones, be conserved (Rolston, Reference Rolston1985; Johnson, Reference Johnson1991) (section “Future generations”)? Are species as collectives more valuable than nonhuman individuals (Russow, Reference Russow1981; Johnson, Reference Johnson1991) and does that therefore justify the culling of individuals that threaten the existence of native populations (Warren, Reference Warren1997; Varner, Reference Varner1998)?

Defining extinction

What constitutes an extinction is a complex subject matter. However, for the purpose of ordinary concerns about extinction and for how moral philosophers discuss its moral relevance, most theorists’ primary concern lies with anthropogenic species extinctions – while treating ‘extinctions’ and ‘species extinctions’ synonymously. Moreover, while many theorists put an emphasis on extinctions on a mass scale (such as Bendik-Keymer, Reference Bendik-Keymer2014; Cafaro, Reference Cafaro2015; Panagiotarakou, Reference Panagiotarakou2016; Baquedano Jer, Reference Baquedano Jer2019), some theorists are also explicitly concerned with the extinction of individual species (such as Rolston, Reference Rolston and Nierenberg1995).Footnote ² Accordingly, there are four conceptual components found in many accounts of the ethics of extinction.

Firstly, there are different ontological positions on the meaning of ‘species’, that is what a species is (see Delord, Reference Delord2007; Maclaurin and Sterelny, Reference MacLaurin and Sterelny2008; Ereshefsky, Reference Ereshefsky and Zalta2022), which then also have different ethical implications (such as in Hull, Reference Hull1978; Gruen, Reference Gruen2011; Smith, Reference Smith2016). Moreover, because different kinds of species conservation practices change the genetic makeup of the target species (via conventional breeding programmes or synthetic biology) and/or influence their behaviour, it is also necessary to consider (1) whether the resulting population still belongs to the same species (which in turn influences whether we consider it extinct or not) and (2) whether that matters morally (see Preston, Reference Preston2021). Thus, there are different conceptions of species and these impact how extinction is defined (see Maclaurin and Sterelny, Reference MacLaurin and Sterelny2008; Godfrey-Smith, Reference Godfrey-Smith2014; Korsgaard, Reference Korsgaard2018a).

Secondly, the term ‘extinction’ can also be understood in different ways (see Delord, Reference Delord2007; Tanswell, Reference Tanswell2022), which is of ethical relevance because (1) it frames what falls within the ethics of extinction and, in turn, (2) the way extinction is constructed in controversial cases is in itself subject to ethical scrutiny. When speaking about extinctions as something morally problematic, many (not all) authors will refer to ‘final’ extinctions, if not specified further in works of practical ethics. Thus, an emphasis is put on ‘global’ (all individuals of a species are gone) as opposed to ‘local’ (loss of a population) extinctions. However, because a species can be extinct in several senses of the term, some authors also specifically problematise, for example cases where there still exist some (or even many) living individuals belonging to the species of concern. That is, it is not finally extinct, but it may be ‘functionally’ extinct. Drawing on the example of the whitebark pine as a case of a functional extinction, Christopher Preston (Reference Preston2021, 4, italics in original) argues in this regard that ‘[t]he concern with this type of extinction is not that you will never see individuals of the taxon again. The concern centres on the anthropogenic reordering of relationships. Not a loss of life, but a loss of arrangement’. That is, regarding this type of extinction an emphasis is put on changing ecological relationships rather than on the circumstance whether some individual members of the species are still alive.

Thirdly, many positions will argue – in different ways – that there is a moral difference between anthropogenic and non-anthropogenic extinctions. In Holmes Rolston’s III words, ‘[n]atural extinction opens doors, anthropogenic extinction closes them’ (Reference Rolston2012, 139). Independently from whether Rolston’s view on the matter is convincing, some people will intuitively find it morally objectionable that a species in the form of all the individuals belonging to that species abruptly disappears due to human actions (such as in Sylvan’s thought experiment, see section “The history of the concern about extinction in environmental ethics”) and thereby also ending an evolutionary process, instead of taking issue with natural evolutionary processes in themselves. This distinction not only assumes that anthropogenic and non-anthropogenic extinctions are conceptually different (Aitken, Reference Aitken1998), but usually also relies on the judgement that human-caused extinctions are morally more problematic than non-anthropogenic extinctions. So, for many authors it is relevant to know whether human causation is at play for the ethical assessment of the extinction in question (for epistemic barriers to assessing this causation, see Tanswell, Reference Tanswell2022) and that, in turn, is closely related to judgements about moral responsibility (Oksanen, Reference Oksanen2007). Contrasting views consider all species extinctions to be similarly morally problematic (i.e. similar ‘badness’) and thereby downplay the moral relevance of anthropogenic causation by adding ‘wrongness’ (Powell, Reference Powell, Beauchamp and Frey2011).

The fourth conceptual distinction needs to be made between mass extinctions and non-mass extinctions, with mass extinctions differing from an ordinary extinction event in several ways (see Bendik-Keymer and Haufe, Reference Bendik-Keymer, Haufe, Gardiner and Thompson2016). Because of these differences, evaluative judgements about the current ‘sixth mass extinction’ event on earth seem to be often based on the intuition that mass extinctions are morally more problematic than single extinction events. For example, consider how different terms such as ‘environmental atrocities’ (see Card, Reference Card2004) or ‘ecocide’ (see Baquedano Jer, Reference Baquedano Jer2019) can be potentially linked to extinction. While many might think that single extinction events added together are as bad as a mass extinction event involving the same number of species lost, some authors argue that a mass extinction is worse – for example due to additionally involving a ‘planetary shift’ (Sandler, Reference Sandler2021a).

Involving loss of value due to the loss of species

Because species extinctions involve the loss of species, the most obvious place to start is to think about whether this loss is in itself problematic (see Passmore, Reference Passmore1974). A common argumentative route in this regard is to argue that species are valuable and, thus, their extinction involves a loss of something that has been important to someone. For example, certain species might be instrumentally valuable because they are useful for human purposes (see section “Leading to the loss of valuable biological variety”). However, species might also be valuable independently from their actual or potential use value. That is, in contrast they might have intrinsic value as entities that we value in themselves.Footnote ³ Let us briefly introduce three (non-exhaustive) versions of this position (for further examples, see Callicott, Reference Callicott and Callicott1986 [for a critique, see Lo, Reference Lo2001]; Gorke, Reference Gorke2003; for a more general critical analysis of holistic accounts of species value, see Agar, Reference Agar2001).

For one, authors such as Rolston have argued that species – not understood as classes but as ‘living historical forms’ (Rolston, Reference Rolston1985, Reference Rolston2012) – are objectively intrinsically valuable (‘some values are objectively there—discovered, not generated, by the valuer’. Reference Rolston1988, 116), which means that with each species that disappears, we lose something of value even if we fail to recognise this loss. In his words,

However, only anthropogenic extinctions are subject to moral evaluation (a duty to avoid extinctions) in Rolston’s account. He relies for this purpose on the analogy between ‘natural death’ (i.e. a natural extinction) and ‘murder’ (an anthropogenic extinction), which leads him to assess each anthropogenic extinction as a ‘superkilling’ (Rolston, Reference Rolston and Nierenberg1995). That is, on Rolston’s account the metaphor of a superkilling illustrates the badness of extinctions by putting an end to an evolutionary ‘story’ and precluding its future evolutionary possibilities. Other authors such as Lawrence E. Johnson have spelled out similar ideas, but with less emphasis on evolution and rather in terms of the species’ interests and their resulting moral standing (see Johnson, Reference Johnson1991).

Criticism of accounts such as Rolston’s or Johnson’s position, in turn, has been aimed at different aspects of their theories. Two primary areas of debate stand out. Firstly, the plausibility of the underlying theory of objective intrinsic value in Rolston’s account is a matter of debate (for a critique, see Elliot, Reference Elliot, Mannison, McRobbie and Routley1980; Callicott, Reference Callicott1992). Against the background that intrinsic values were intensely discussed in the field of environmental ethics at the time when Rolston published his account (see O’Neill, Reference O’Neill1992), it should be noted that other authors such as Callicott (Reference Callicott1984, Reference Callicott and Callicott1986) defended alternative value accounts relying on subjective values or human sentiments as the basis of a non-anthropocentric environmental ethic. Secondly, Johnson’s position relies on the idea that species can be considered to be akin to living beings (i.e. it relies on a specific species concept – as a living entity – that has been widely criticised not only by individualist perspectives but also by other holistic accounts; see Smith, Reference Smith2016) and, thus, also be morally considerable in the same way. However, whether that analogy is convincing and whether species have morally relevant interests are also a matter of debate (for a critique, see Sandler and Crane, Reference Sandler and Crane2006; Sandler, Reference Sandler2012). Consequently, in the more recent literature other arguments in favour of the moral value of species that are not based on such an understanding of species are more widespread.

For example, secondly, Ian Smith argues in favour of the view that a species has an ‘intrinsic good’ that ‘consist in its abilities to flourish’, based on a Hennigian species concept that sees species as ‘historical individuals’, which, in turn, gives us reasons to preserve species on Smith’s account because, amongst other things, ‘[a species’] flourishing is a species’ organisms continuing to reproduce successfully (that is, the organisms producing fertile offspring) and the species remaining safe from extinction’ (Reference Smith2016, 14). Smith takes it for granted that ‘species taxa are real’ (Reference Smith2016, 3–4) but subspecies are not. Therefore, only species are of intrinsic value. Critics have seen this presumption, in turn, as a failure to properly consider subspecies and higher categories in taxonomy (Burbrink et al., Reference Burbrink, Crother, Murray, Smith, Ruane, Myers and Pyron2022).

Finally, Ronald Sandler understands species as ‘groups of biologically related organisms that are distinguished from other organisms by virtue of their shared form of life’ (Sandler, Reference Sandler2012, 6, italics in original). What distinguishes organisms in this way, according to Sandler, are, for example, their ways to acquire energy (what they eat), how they move (and whether they move at all), how they reproduce and so on. Because Sandler argues against objective values being applicable to species (understood in this way), he points out that species can be still valuable in other ways. For example, some species are subjectively (finally/non-instrumentally) valuable (i.e. valuable because people actually value them) in two ways: firstly, in the form of ‘preference value’ as the ‘value that something has because people have a preference for it’ (Reference Sandler2012, 23). That value is also sometimes called ‘existence value’, which according to Espen Stabell (Reference Stabell2019, 180) ‘can be understood to involve self-regarding as well as other-regarding and nonanthropocentric preferences’. Secondly, species can also have ‘integral value’ as the ‘value that something possesses when it is valued in a way that flows from one’s worldview of core value commitments’ according to Sandler (Reference Sandler2012, 24).

This last account of the value of species neither relies on species having objective value (like Rolston), nor on species having interests akin to living beings (like Johnson), nor on the premise that species are being able to flourish (like Smith), and thus, it relies on less premises that might be controversial. However, in contrast to those other accounts its scope is much narrower because it concedes that only some anthropogenic species extinctions are lamentable by involving the loss of such value and only if this value is actually held by some humans.

Resulting from harm to nonhuman individuals

Non-anthropocentric individualist accounts within environmental ethics – that is sentientist (i.e. all sentient beings are morally considerable; Singer, Reference Singer1975) or biocentric (i.e. all living beings are morally considerable; Taylor, 1986/Reference Taylor2011) accounts – are only indirectly concerned with extinctions. What primarily matters to such positions is either the harm, impediment to well-being or injustice that is borne by individual nonhuman beings, and thus, the relevant harm involved in extinctions is not found on the species level (Nussbaum, Reference Nussbaum2011). Yet, it is often argued – for different reasons – that these two levels of analysis are closely related (Agar, Reference Agar2001; Regan, Reference Regan2004; Taylor, Reference Taylor2011; Korsgaard, Reference Korsgaard2018a; Donoso, Reference Donoso, Rivera-Lopez and Hevia2019; Baard, Reference Baard2022). For example, according to animal rights theorist Tom Regan the preservation of the biotic community follows from showing ‘proper respect for the rights of the individuals’ (Regan, 1983/Reference Regan2004, 363) that are part of that community.

As an example of emphasising the current anthropogenic mass extinction within an individualist framework, Anna Wienhues (Reference Wienhues2020) argues in favour of a biocentric theory of interspecies justice. Within this account, anthropogenic extinctions function as indicators of potential previous distributive injustices to nonhuman beings, due to the unjust deprivation of habitat that can ultimately lead to species extinctions. Within this context, extinctions are not injustices in themselves, but an outcome of interspecies injustice – metaphorically speaking she argues that ‘each anthropogenic extinction is a memorial for past injustices’ (Reference Wienhues2020, 157, italics in original).

A commonality of all of these individualist accounts is that they are only indirectly concerned with extinctions and cannot explain why an extinction is in itself morally lamentable (Benton, Reference Benton1993). Moreover, such accounts usually cannot morally distinguish between saving a specimen that belongs to a near-extinct species and one that does not belong to such a species (Sober, 1986/Reference Sober and Elliot1995; Baard, Reference Baard2022). If ethical relevance is placed solely on individual entities rather than species, then the extinction status of the species the entity belongs to does not necessarily count ethically, which could be more of a problem for some individualist accounts (e.g. Singer, Reference Singer1975; Cochrane, Reference Cochrane2012; see Korsgaard, Reference Korsgaard2018b), than for individualist accounts that allow for a degree of value pluralism (as opposed to everything of moral relevance being reducible to, e.g., individual well-being) and/or that allow for a degree of ethical pluralism (e.g. Wienhues, Reference Wienhues2020).

An alternative solution is presented by individualist accounts that conceptualise the individual’s interests as extending to ‘all other organisms that are relatives’ in terms of its species (understood following Mayr’s biological species concept) (Agar, Reference Agar2001, 150). Based on Nicholas Agar’s (Reference Agar2001) account, that consequently allows for an individualist argument that proposes that individuals that are part of endangered species and therefore rare are more valuable than their plentiful counterparts. According to Agar that follows because the value of an individual is not ‘a function only of its own goals. Its demise affects the other-directed and other-requiring goals of its conspecifics, depending on how plentiful its species is’ (Agar, Reference Agar2001, 150), which includes, for example, the ability of individuals to mate and successfully reproduce.

Yet, non-individualist accounts consider a theoretical focus on the individual to be a weakness. As Johnson puts it, ‘[a]n ethic that deals adequately with the issues of extinction must not only avoid being anthropocentric, it must avoid being atomistic’ (Reference Johnson1991, 170, italics in original). That might make one favour again one of the earlier introduced values of species accounts, but an account like Johnson’s faces in turn a range of difficulties in its own as already indicated above.

Leading to the loss of valuable biological variety

Besides accounts that consider the loss of parts of biodiversity to be problematic because of the moral relevance of species or nonhuman individuals, other perspectives consider species extinctions to be problematic due to the involved loss of biological variety. While biodiversity loss does not necessarily follow from species extinctions and vice versa, they are clearly conceptually related and practically implicated in each other. Problematising the loss of (bio)diversity, in turn, is a notable topic in the context of species extinctions because it has been institutionalised on an international level via the Convention of Biological Diversity (established in 1993).

Although the convention text begins with the recognition of ‘the intrinsic value of biological diversity’ (CBD, 1993), it is not obvious what role the intrinsic value in the CBD plays, as it appears only in the preamble. The recognition of intrinsic value is also followed by stating that the parties are conscious of ‘the ecological, genetic, social, economic, scientific, educational, cultural, recreational and aesthetic values of biological diversity’ (CBD, 1993). This shows that there is ambiguity in how the value of biodiversity is to be understood. Moreover, the larger normative international legal framework and even the CBD itself mainly endorse the established resource-focused language and practice (see Oksanen and Vuorisalo, Reference Oksanen, Vuorisalo, Casetta, Vecchi and Marques da Silva2019). Here, again the question arises about how biodiversity-focused claims about value should be interpreted.

So, the extinction of species via the resulting loss of biodiversity seems to involve a loss of value, but reasons for why the loss of variety might be problematic can rest on different argumentative grounds (Oksanen, Reference Oksanen1997; Callicott, Reference Callicott, Garson, Plutynski and Sarkar2017; Heinzerling, Reference Heinzerling, Garson, Plutynski and Sarkar2017; McShane, Reference McShane, Garson, Plutynski and Sarkar2017; Odenbaugh, Reference Odenbaugh, Garson, Plutynski and Sarkar2017; Baard, Reference Baard2022). Moreover, biodiversity can be conceptualised in various ways (CBD, 1993; Norton, Reference Norton, Sarkar and Plutynski2008; Faith, Reference Faith, Garson, Plutynski and Sarkar2017, Reference Faith and Zalta2021; Sarkar, Reference Sarkar, Garson, Plutynski and Sarkar2017; Burch-Brown and Archer, Reference Burch-Brown and Archer2017) and normative emphasis can be put on different kinds of diversity (such as ‘phylogenetic diversity’, Palmer and Fischer, Reference Palmer and Fischer2021). Generally, there are two common (not exhaustive) argumentative routes to choose from, which do not preclude each other but are often found in combination to justify the protection of biodiversity.

Firstly, biological variety as in the variety or abundance of, for example, genes or species might have instrumental value beyond the (instrumental) value of the species themselves (e.g. Maclaurin and Sterelny, Reference MacLaurin and Sterelny2008). While that is a common premise in conservation biology and environmental ethics that can justify why the extinction of many species is problematic due to the loss of valuable variety, different argumentative grounds for such instrumental value such as biodiversity’s contribution to ecosystem functioning or its (potential) agricultural and pharmaceutical benefits have also been critically discussed (see Maier, Reference Maier2012; Deliège and Neuteleers, Reference Deliège and Neuteleers2015; Newman et al., Reference Newman, Varner and Linquist2017; Morrow, Reference Morrow2023). Independent of such critiques, the loss of variety might not be problematic for instrumental value reasons in all circumstances (e.g. consider the loss of variety of certain pathogens), which means that not necessarily all extinctions involve a loss of instrumentally valuable variety.

Yet that needs to be kept distinct from the instrumental value attributed to species themselves (such as their ‘incalculable instrumental value’, see Smith, Reference Smith2022), although not all species will be instrumentally valuable for human purposes either (see section “Future generations”). In so far as these, as well as biological variety, are fundamental grounds for the possibility of human life on earth, a mass extinction can be understood as humanity’s ‘autodestruction’ constituting ‘a wrong in its own category. The wrong is the wrong of putting an end to humankind and all that is of value in it. This is a cataclysmic wrong’ as put by Bendik-Keymer and Haufe (Reference Bendik-Keymer, Haufe, Gardiner and Thompson2016, 7, italics in original). Thus, for such an argument the ‘massness’ (see section “Defining extinction”) of the current extinction event is central and, thereby, it avoids the problem that some individual species might not be instrumentally valuable for human purposes.

The second option is to consider the intrinsic value of biological variety as something that might be valuable in itself – independently of whether it is beneficial for humans or other lifeforms (Mikkelson, Reference Mikkelson2022). There are different versions of what this entails (see McShane, Reference McShane, Garson, Plutynski and Sarkar2017). For example, deep ecologist Arne Naess famously stated that ‘[r]ichness and diversity of life forms (…) are also values in themselves’ (Naess, 1986/Reference Naess, Light and Rolston2003, 264). Providing a more nuanced view on the intrinsic value of biodiversity, Gregory Mikkelson (Reference Mikkelson2011, 186) points out that ‘variety’ is not the sole determinant of value in neither Næss’ deep ecology, nor in Mikkelson’s own preferred richness theory. Both invoke additional determinants such as ‘harmony’ (Mikkelson, Reference Mikkelson2011, Reference Mikkelson2014). To Mikkelson, more related to extinction, ‘anthropogenic species extinctions are decreasing the overall harmony of life on Earth’ (Reference Mikkelson2011, 191).

Philosophical claims about biodiversity’s intrinsic value also get external support. As already mentioned, conservation biologist Soulé considered biotic diversity to have intrinsic value (Reference Soulé1985, 731) and also the Convention on Biological Diversity refers to intrinsic value (for a critical discussion of both, see Baard, Reference Baard2022). Moreover, studies about the environmental values of the public also report intrinsic values being supported (e.g. Leiserowitz et al., Reference Leiserowitz, Kates and Parris2005; discussed by both Mikkelson, Reference Mikkelson2014; Odenbaugh, Reference Odenbaugh, Garson, Plutynski and Sarkar2017) and strong pro-environmental attitudes confirmed empirically have the possibility of generating subjectivist non-instrumental values (see Lo, Reference Lo2006). However, different proposals concerning the intrinsic value of biological variety have also been subject to critical analysis (Sarkar, Reference Sarkar2005; Kraut, Reference Kraut2011; Maier, Reference Maier2012; Newman et al., Reference Newman, Varner and Linquist2017), such as concerning whether intrinsic value can attach to biological variety itself or whether that is better understood in terms of intrinsically valuing parts of biological variety such as species and individual beings (McShane, Reference McShane, Garson, Plutynski and Sarkar2017).

Moreover, in addition to valuing species, nonhuman individuals and biological variety, further (additional or alternative) adjacent reasons grounding the badness of species extinction can be found in the literature. To mention two examples broadly within the realm of biodiversity-focused arguments, that includes, for one, arguments pertaining valuing the ‘irreplaceable design that is embodied by the individuals’ part of a species (Cline, Reference Cline2020, 46, italics in original). The point being that after a long evolutionary process, ‘[e]ach of the wonderful contrivances of nature are (…) unique and irreplaceable. Such irreplaceable design has special value that we recognise when we experience awe and wonderment towards the sophisticated structures and systems and strategies exemplified by living organisms’ (Cline, Reference Cline2020, Reference Elliot, Mannison, McRobbie and Routley59). According to Brendan Cline, that shifts the focus of concern away from the species themselves towards such ‘irreplaceable design’. Secondly, the loss of biodiversity might also be considered problematic due to a lack of respect towards nonhuman living beings and active nature in their natural otherness. Such respect provides us with reasons to protect biodiversity and, consequently, species variety (Wienhues and Deplazes-Zemp, Reference Wienhues and Deplazes-Zemp2022). Subsequently, human-caused extinctions can be interpreted as a lack of such respect.

What is common to all these types of biodiversity-focused arguments – regarding the instrumental value of variety, the intrinsic value of variety and pertaining alternative concepts (harmony, design, otherness, etc.) – is that for them the focus of moral concern shifts from species extinctions towards biological variety or those other concepts. So, like the individualist arguments (section “Resulting from harm to nonhuman individuals”) extinctions are again not problematised because of the loss of species themselves, but rather due to what this loss means for other valued entities or other matters of moral concern.

Future generations

A further reason for why species extinctions might be morally problematic concerns our duties to future human generations (see Feinberg, Reference Feinberg and Blackstone1974; Sandler, Reference Sandler and Gardiner2021b). In essence, if we believe that the interests of future generations put current generations under an obligation to not frustrate these interests, we have a reason to protect species and biodiversity that future generations might (intrinsically and instrumentally) value and, thereby, avoid ‘irreversible loss’ (Spiekermann, Reference Spiekermann and Gardiner2022). While there might be considerable uncertainty about what species, ecosystems and/or ecosystem ‘services’ future generations will actually value and/or need, this ultimately involves again an assessment of different arguments pertaining to the value of species and biodiversity (as seen in sections “Involving loss of value due to the loss of species” and “Leading to the loss of valuable biological variety”). Moreover, considerations about duties towards future generations in terms of species conservation provide a bridge between the ethics of extinction and the ethics of sustainability (see Norton, Reference Norton2003; Armstrong, Reference Armstrong and Gardiner2021).

In addition to these human-focused arguments, we can also consider non-anthropocentric arguments pertaining to the future (building on some of the arguments found in sections “Resulting from harm to nonhuman individuals” and “Resulting from harm to nonhuman individuals”; see also Nolt, Reference Nolt and Gardiner2021; Whyte, Reference Whyte and Gardiner2022). There are two main argumentative routes open in this regard. Firstly, one can consider potential duties towards future nonhuman beings or species in so far as the extinction of other species will affect them negatively (and consider cases where species might benefit from the extinction of other species; related to this, see Palmer, Reference Palmer and Arnold2011 and Cripps, Reference Cripps2013 for reflections about how certain species will benefit from climate change). Like how in intra-human intergenerational ethics a lot of thought has been given to the ‘nonidentity problem’ (see Norton, Reference Norton1982; Parfit, Reference Parfit1986), also here we face a nonhuman nonidentity problem (Palmer, Reference Palmer and Arnold2011), in so far as anthropogenic species extinctions will affect which individuals will come into existence in the future. That is, the extinction of some species will have a knock-on effect on which individuals of still remaining species will meet and mate, which in turn will influence which individuals will live in the future. The underlying philosophical problem is that those future individuals would not have come into existence without the extinction (which, in turn, is relevant for underlying considerations about how the ‘harm’ of extinction is understood). Secondly, the case of extinctions also leads to a ‘non-existence problem’ (Wienhues, Reference Wienhues2020) in so far as nonhuman individuals, species and ecosystems in the future will not only be affected by prior extinctions, but many nonhuman individuals will simply never come into existence.

Can anthropogenic extinctions ever be justified?

Some readers might assume that all anthropogenic extinctions are morally problematic, but does that judgement hold up in all cases? Consider these two examples from species conservation practice. Firstly, concerning ‘local’ (as opposed to ‘global’) extinctions consider restoration-based reasons in favour of the extinction of a population of an ‘invasive’ species (for critical discussions, see Rawles, Reference Rawles2004; Odenbaugh, Reference Odenbaugh2022; Thresher, Reference Thresher2022). Secondly, other conservation practices such as captive breeding and release programmes have in turn been criticised on the grounds of not allowing for a ‘death with dignity’ (Chessa, Reference Chessa2005). While in this case the extinction might still seem morally problematic, such programmes might not present an appropriate response. In both cases, it is possible to claim that an extinction can be all-things-considered better than an alternative course of action, for instance, on ecocentric or sentientist argumentative grounds (which both, in turn, can also be engaged to criticise the moral desirability of such extinctions).

Also, consider the case of disease-transmitting species such as certain mosquitoes that transmit malaria, dengue fever, zika and so on, which is the most discussed case of potentially justified extinction. The global health burden of these diseases is very high (WHO, 2017; Greisman et al., Reference Greisman, Koenig, Barry, Mastroianni, Kahn and Kass2019), which provides a good reason to consider whether the eradication of such disease-transmitting species would be an appropriate means to address this problem (Fang, Reference Fang2010; Bates, Reference Bates2016). While it is an empirical question to decide whether this constitutes an approach that is likely to succeed while minimising the risks in comparison with alternatives, this also generates a range of ethical questions to consider, such as regarding the moral standing of individual nonhuman animals, ethical aspects of risk and the intrinsic and/or instrumental value of species (see Pugh, Reference Pugh2016; Wienhues, Reference Wienhues, Hall and Tamïr2021; Callies and Rohwer, Reference Callies and Rohwer2022). A particular emphasis found in the literature on bioethics concerns questions of research ethics, risk and consent regarding genetically engineering species such as mosquitoes in this context (see Resnik, Reference Resnik2014; Emerson et al., Reference Emerson, James, Littler and Randazzo2017; Resnik, Reference Resnik2017; Neuhaus and Caplan, Reference Neuhaus and Caplan2017; Patrão Neves and Druml, Reference Patrão Neves and Druml2017; Meghani and Boëte, Reference Meghani and Boëte2018). Notably, even some authors who consider anthropogenic extinctions to be highly morally problematic acknowledge exceptions for disease-transmitting species such as malaria-transmitting mosquitoes (e.g. Rolston, Reference Rolston and Jamieson2001; Smith, Reference Smith2016).Footnote ⁴ While the use of synthetic biology such as gene drives remains controversial (Preston and Wickson, Reference Preston, Wickson and Dressel2019), the health burden of vector-borne diseases constitutes a weighty reason to take them into consideration.

Intended species extinctions are also considered for other – even more controversial – purposes, which include the case of predation raised in animal ethics. While some animal ethicists consider wild animal suffering to be of serious moral concern (Horta, Reference Horta2017), a few authors go as far as arguing that there are reasons speaking in favour of the extinction of predatory species. Footnote ⁵ The benefit hoped for would be the reduction of (non-anthropogenic) suffering in the ‘wild’ caused by predation (see McMahan, Reference McMahan2010, Reference McMahan, Chignell, Cuneo and Halteman2016). Such proposals to alleviate ‘wild’ animal suffering (by a range of means aimed at ‘dewilding’ or ‘redesigning nature’, see Duclos, Reference Duclos2022; Kianpour and Paez, Reference Kianpour and Paez2022) are in turn not shared by other animal and environmental ethicists on more general terms (see Palmer, Reference Palmer2010, Reference Palmer2015; Donaldson and Kymlicka, Reference Donaldson and Kymlicka2011; Aaltola, Reference Aaltola, Oksanen and Siipi2014; Hettinger, Reference Hettinger2018), as well as specifically on the question of removing predatory species (Vincelette, Reference Vincelette2022). Reasons include the difficulties of implementing such proposals and more principled objections towards such interventions into ‘nature’.

Risk of extinction

For the purposes of moral theorising, ‘risk’ refers to the likelihood of a loss or harm taking place, which makes ‘risk’ a value concept referring to an unwanted outcome through human actions, in addition to an epistemological dimension where that outcome is likely to happen with a specific probability (Hansson, Reference Hansson2013). If that probability is unknown, it is referred to as ‘uncertainty’ rather than ‘risk’, which often motivates a precautionary approach. With respect to species extinction, there are two main ways of discussing risks and extinctions: the first focuses on the risk of a single species facing extinction, and the second pays attention to broader systemic and accumulative risks, either in terms of mass extinction or in terms of a collapse of the biospheric system as we know it.

Regarding single species, risk is a premeditative concept that enables us to respond to near-extinction, such as when a species extinction is predicted to happen in the light of best ecological modelling, where the predicted quantity of extinctions has often been referred to as ‘extinction debt’ (Hanski and Ovaskainen, Reference Hanski and Ovaskainen2002; Kuussaari et al., Reference Kuussaari2009). Against this background, deliberations on employing concepts and ideas drawn from ecological sciences and (meta)population biology are taking place within environmental ethics and the philosophy of biology, which includes topics of ethical relevance such as decision-making under uncertainty and the applicability of the precautionary principle, land-use planning, protected area networking and prioritisation (Norton, Reference Norton1987; Sarkar, Reference Sarkar2005).

Risks have value dimensions, and what value is justifiably ascribed to a species will play a part in how it is to be prioritised (see section “Why (or when) might anthropogenic extinctions be morally problematic?”). Most efforts to save near-extinct species require prioritising: global and national Red Lists of Threatened Species are examples of such a conservation policy, as is Bryan Norton’s ‘triage model’ (Reference Norton1987, 258). Yet, prioritisations come with a range of philosophical problems. According to Norton, many of those challenges arise due to the difficulties of reducing the values of species to a single scale (Reference Norton1987, 255).

A broader perspective to the risk landscape emerges when we consider what kinds of systemic risks each extinction gives rise to (which is related to some of the concerns introduced in section “Leading to the loss of valuable biological variety”). A common argument rests on an analogy between species extinctions and the loss of rivets on an airplane wing (Ehrlich and Ehrlich, Reference Ehrlich and Ehrlich1981).Footnote ⁶ It aims to show that the ethical significance of extinction is due to its potential effects on ecosystems (the loss of too many ‘rivets’), which in turn are prerequisites for human and nonhuman survival. When conjunct with uncertainty regarding the number of rivets, or species, that an airplane wing, or ecosystem, can dispose of without collapsing, it becomes essential to preserve species and avoid their extinction (for a critical discussion of the analogy, see Sarkar, Reference Sarkar2005).

This expresses a form of precautionary principle, which has been discussed both critically and approvingly from different philosophical perspectives (Sandin, Reference Sandin2004; Sarkar, Reference Sarkar2005; Hourdequin, Reference Hourdequin2007; Newman et al., Reference Newman, Varner and Linquist2017; Tanswell, Reference Tanswell2022). The precautionary principle is also a part of the preamble of the Convention of Biological Diversity (CBD, 1993) where it is stated that lack of full knowledge ought not to justify inaction, and it is mentioned in the framework of the Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services (IPBES n.d.). The complexity and uncertainty pertaining to how ecosystems work (Regan et al., Reference Regan, Colyvan and Burgman2002; Sarkar, Reference Sarkar2005) provide support for a precautionary principle concerning environmental decision-making. Yet, some authors consider its policy implications problematic due to leading to logical contradictions by potentially recommending two opposing actions (e.g. recommending the development of and simultaneously the prevention of developing genetically modified organisms), failing to account for possible benefits, neglecting that precaution has costs and for lacking conceptual clarity (Newman et al., Reference Newman, Varner and Linquist2017; critiqued in turn in Baard, Reference Baard2022).

De-extinction

What should be done after extinctions have already occurred? Besides more conventional species conservation practices, there is a broader discussion on the ethics of developing and using biotechnologies for conservation purposes and beyond (Ehrenfeld, Reference Ehrenfeld2006; Basl and Sandler, Reference Basl and Sandler2013; Kaebnick and Murray, Reference Kaebnick and Murray2013; Oksanen and Siipi, Reference Oksanen, Siipi, Oksanen and Siipi2014; Preston, Reference Preston2018). Setting these larger concerns aside, more technology-focused conservation approaches – if justifiable in certain cases – should be seen as complementary rather than alternatives to conventional conservation practices (Sandler, Reference Sandler and Gardiner2021b). Accordingly, the subject of debate concerns whether there are certain cases where such technological means are appropriate and desirable at all.

For example, a topic that has recently been gaining attention is the evaluation of the possibility of engineering species (via gene drives, cloning, etc.) either with the purpose of avoiding their extinction, such as by genetically adapting them to new climatic conditions or by increasing the genetic diversity of a population (see discussions by Palmer, Reference Palmer2016; Rohwer, Reference Rohwer2018; Sandler, Reference Sandler2019; Preston, Reference Preston2021; Sandler, Reference Sandler and Gardiner2021b; Welchman, Reference Welchman, Bovenkerk and Keulartz2021), or as means to eradicate invasive species (Thresher, Reference Thresher2022).

Related to proposals in favour (or critiques) of altering species to avoid (or cause) their extinction, a second body of work focuses on how and whether methods such as back-breeding, cross-species cloning and genetic engineering can be employed to appropriately address extinctions once they have occurred. Thus, the possibility of de-extinction by bringing a species ‘back’ from extinction has captured the imagination of many environmental philosophers (amongst many others, see Cohen, Reference Cohen2014; Oksanen and Siipi, Reference Oksanen, Siipi, Oksanen and Siipi2014; Turner, Reference Turner, Oksanen and Siipi2014; Diehm, Reference Diehm2015; Minteer, Reference Minteer2015; Campbell and Whittle, Reference Campbell and Whittle2017; Kasperbauer, Reference Kasperbauer2017; Oksanen and Vuorisalo, Reference Oksanen and Vuorisalo2017; Sandler, Reference Sandler2017; Browning, Reference Browning2018; Preston, Reference Preston2018; Katz, Reference Katz2022; Sandler et al., Reference Sandler, Stabell, Baylon, Lundgren, Weisbeek, Yelle and Zaba2022; for an alternative narrative, see also Odenbaugh’s, Reference Odenbaugh2023 recent review of the ethics of de-extinction).

This topic has become the centre of a lively debate in the ethics of extinction, which is noteworthy in so far as de-extinction is only a minor subject within the broader field of conservation ethics and debates (see also section “Extinction studies”). Of concern are, for example, different moral values and goods that might be supported or hindered by de-extinction projects (e.g. Cohen, Reference Cohen2014; Haught, Reference Haught2017; Smith, Reference Smith2017; Welchman, Reference Welchman2017; Rohwer and Marris, Reference Rohwer and Marris2018), which also include anthropocentric reasons in favour of the de-extinction of certain species, such as their potential value for ecosystem functioning, their economic value or their scientific value.

Non-anthropocentric attempts to provide a qualified justification for this approach include providing a list of criteria for its employment (e.g. Kasperbauer, Reference Kasperbauer2017) or by trying to make the idea useful for other normative purposes such as whether de-extinction might be a way of enacting a reparative duty towards nonhuman species or individual animals after extinctions have occurred. While thinking about de-extinction in terms of reparation or restitution might be intuitively appealing and constitute an argument occasionally engaged by conservation professionals, concerns about de-extinction as a form of reparation or retribution are shared by many authors (such as Oksanen and Siipi, Reference Oksanen, Siipi, Oksanen and Siipi2014; Diehm, Reference Diehm2015; Lean, Reference Lean2020; Welchman, Reference Welchman, Bovenkerk and Keulartz2021; yet for an argument in favour of a prima facie obligation to re-create species, see Jebari, Reference Jebari2016), while more might be said in favour of de-extinction as ecological restoration (Turner, Reference Turner, Oksanen and Siipi2014). Overall, while some authors argue that de-extinction can be ethically permissible in certain circumstances (Sandler, Reference Sandler2019), the debate on the desirability of de-extinction has been rather sceptical. Three common (not exhaustive) critiques of de-extinction fall into the following three categories.

Firstly, it has been frequently pointed out that the resulting species is not the same species as the previously extinct species, but only constitutes a copy (which concerns the species’ ‘identity’, Blockstein, Reference Blockstein2017; Siipi and Finkelman, Reference Siipi and Finkelman2017; Lean, Reference Lean2020). Actual ‘resurrection’ is not considered to be possible – or only possible by drawing on a very specific species concept (Oksanen and Siipi, Reference Oksanen, Siipi, Oksanen and Siipi2014) and related metaphysical commitments to resolve the ‘resurrection paradox’ (Delord, Reference Delord, Oksanen and Siipi2014) – which leaves any de-extinction proposal wanting (Diehm, Reference Diehm2017) and reintroduces the abovementioned considerations about the ontology of species (Beever, Reference Beever2017).Footnote ⁷ However, other writers argue that de-extinction technologies should not be understood as aiming to re-create extinct species in the first place (Lean, Reference Lean2020), which weakens this critique aimed at de-extinction while also reducing its intuitive appeal.

Secondly, concerns about animal welfare have been raised (Gamborg, Reference Gamborg, Oksanen and Siipi2014; Oksanen and Siipi, Reference Oksanen, Siipi, Oksanen and Siipi2014; Kasperbauer, Reference Kasperbauer2017; Turner, Reference Turner2017; Browning, Reference Browning2018; Browning and Veit, Reference Browning and Veit2022), which regards, for example, the welfare of the animals used for laboratory testing, of the animals that are (re)introduced into ecosystems and of the animals already living in the target ecosystems. Distinct from but related to such worries are also concerns regarding the ecological risks of (re)introducing species into ecosystems and, thus, what this means for ecosystems beyond the de-extinct species in question (Oksanen and Siipi, Reference Oksanen, Siipi, Oksanen and Siipi2014; on the more general relationship between biodiversity conservation and resurrecting species, see Oksanen, Reference Oksanen, Oksanen and Siipi2014).

Thirdly, we can find a range of worries raised against de-extinction on a more abstract level of analysis concerning the human–nature relationship (Kohl, Reference Kohl2017; Katz, Reference Katz2022). That is linked, in turn, to worries about naturalness (Gamborg, Reference Gamborg, Oksanen and Siipi2014) in terms of artificiality and authenticity (Katz, Reference Katz2022; discussed in Campbell, Reference Campbell2022; Lean, Reference Lean2022; Preston, Reference Preston2022; Reydon, Reference Reydon2022; Turner, Reference Turner2022, but for an argument against this concern, see Campbell and Whittle, Reference Campbell and Whittle2017) as well as concerns about instrumentalist ‘techno-optimist’ conservation strategies more generally as exemplified by de-extinction (Diehm, Reference Diehm2015).