Herding Katz: Rewilding,paradox and domination

Abstract

Eric Katz has recently claimed not only that rewilding is inherently paradoxical, but also that its paradoxes reveal rewilding's implication in the very mindset of anthropocentric domination against which it is floated as a partial solution. In this paper, I argue that rewilding need not in principle be committed to a pernicious anthropocentrism. With the assistance of an important distinction between ‘synchronic’ and ‘diachronic’ wildness, I firstly argue that rewilding need not be viciously paradoxical in any unequivocal sense. I then suggest, with the aid of Henry David Thoreau's account of synchronic wildness, that rewilding might rather be geared to inculcate hypersensitivity to nonhuman otherness particularly conducive to an anti-domination mindset. Hence rewilding may remain a live tool in responding to the challenges that characterise our shared world.

Introduction

Eric Katz is no flag-waving rewilding advocate. Katz's main reservations about rewilding, expressed in several recent articles (Katz, 2023, 2024), are twofold. Katz firstly claims that rewilding initiatives are inherently paradoxical since they require human intentionality and management to generate and sustain the alleged nonhuman autonomy upon which their ‘wildness’ (and ‘wildness value’) depends. By establishing these paradoxes, and thereby allegedly revealing rewilding projects to be more fundamentally top-down human management initiatives, Katz offers a second, and more important reservation: that rewilding is geared to expunge nonhuman autonomy, and is thus insidiously connected to a mindset of anthropocentric domination. According to this mindset, on Katz's account, the ‘nonhuman’ meanings (e.g. ‘wildness’) and values (e.g. ‘wildness value’) that we humans seek can always be restored within, or imposed upon, the nonhuman world by the right kind of anthropogenic actions. Our guilt about ecological devastation is thereby assuaged by the possibility of rewilding because a ‘technological fix’ (Katz, 2003: 390) is always possible, and virtually nothing is ever truly lost. Echoing his well-known objections to ecological restoration, Katz's criticisms of rewilding amount to the claim that rewilding advocates commit themselves to an impossible task and, in so doing, distract themselves from deeper reflection upon the problematic anthropocentrism in which their initiatives are unwittingly implicated, and which are partly responsible for the degradation of wild places, and other facets of the environmental crisis, in the first place.

In this paper, I explore both reservations but focus on the latter claim about rewilding's connection to anthropocentric domination. I contend that Katz may be correct that some autonomy is regrettably lost when an ancient forest, for instance, is felled. However, this loss of autonomy is irrecoverable only in the sense of what, following Steven Vogel (2015: 108), I call its diachronic wildness, which tracks the ‘trajectory of the autonomous development’ (Katz, 2024: 242), over time, of an individuated token of wildness (here: a specific forest, forest ecosystem or individual tree). However, the rewilding advocate may legitimately seek not to restore or create an authentic facsimile of that historical forest or tree (as per what Hilderbrand et al. (2005) call ‘Carbon Copy’ restoration), but greater wildness in the synchronic sense of an increase in the qualitative autonomy of something here and now. Thus, I argue, rewilding can, in principle, be teased apart from any vicious paradox that would implicate it in the mindset of anthropocentric mastery about which Katz is rightly concerned. Moreover, I suggest, insofar as rewilding so characterised is conducive to relinquishing the dream of anthropogenic control, it may serve to arrest the domination mindset which is Katz's (and my) main concern.

The plan is as follows: I firstly outline how increasing wildness qua nonhuman autonomy is central to the concept of rewilding, appropriately understood. I then explore why Katz takes rewilding to be doubly paradoxical, before responding to the strongest version of that alleged paradox with reference to a pluralist conception of wildness that appears friendly to some aspects of Katz's later scalar account. The final section draws upon Jane Bennett's account of synchronic wildness in Henry David Thoreau's work to explore how rewilding might help to inculcate in human beings the requisite humility to arrest our mindset of anthropogenic domination.

What is rewilding?

Katz (2023) characterises his analysis of rewilding as ‘an attempt to determine its philosophical meaning’, an endeavour which, for Katz, is crucial since (for reasons I explore below) ‘if we get the meaning wrong … we will awaken one day to a world which is totally artefactual’. Given that Katz aims to demonstrate that rewilding necessarily involves anthropocentric mastery, however, Katz must approach ‘rewilding’ as a fundamental natural or social kind, which admits proper and improper extensions. Katz thus provides a conceptual analysis of what rewilding is, in Sally Haslanger's (2000: 33) ‘descriptive’ sense, whereby, ‘the task is to develop potentially more accurate concepts through careful consideration of the phenomena, usually relying on empirical or quasi-empirical methods’.

When approached descriptively, however, rewilding's meaning is contested. A commonly cited definition comes from the widely consultative International Union for the Conservation of Nature report:

Rewilding is the process of rebuilding, following major human disturbance, a natural ecosystem by restoring natural processes and the complete or near complete food web at all trophic levels as a self-sustaining and resilient ecosystem with biota that would have been present had the disturbance not occurred. This will involve a paradigm shift in the relationship between humans and nature. The ultimate goal of rewilding is the restoration of functioning native ecosystems containing the full range of species at all trophic levels while reducing human control and pressures. Rewilded ecosystems should – where possible – be self-sustaining. That is, they require no or minimal management (i.e. natura naturans [nature doing what nature does]), and it is recognized that ecosystems are dynamic. (Carver et al., 2021: 1888)

Conceptual analyses via literature review, however, generate competing conclusions, ranging from Dolly Jørgensen's claim that rewilding is a fuzzy or ‘plastic word’ (2015: 485) lacking definitional precision, to Jonathan Prior and Kim Ward's claim that rewilding's alleged plasticity belies a necessary focus on generating or increasing wildness qua ‘non-human autonomy’ in Mark Woods's sense, which involves ‘animals moving about, plants growing, and rocks falling … because of their own internal self-expression’ (Woods, 2005: 177; in Prior and Ward, 2016: 133). I will spend no time exploring the minutiae of rival claims here, since, ultimately, what (rightly, I think) matters for Katz is that, in the final analysis, the operative or ‘mainstream meaning of rewilding’ (Katz, 2024: 235–236), in Anglophone rewilding contexts, focuses on the generation of spaces substantively independent of human control, by which a partial comparison with ecological restoration might be grasped. Such an account of rewilding's meaning would preserve its definitive commitment to promoting ‘self-willed’ land, pivotal to rewilding's conceptualisation by early Indigenous American advocates (e.g. Vest, 1985). It would also tally with subsequent uptake by heterogeneous figures including Dave Foreman, Roderick Nash, Michael Soulé and Reed Noss, right through to Carver and his colleagues, who focus on increased nonhuman autonomy and the repudiation of anthropogenic control in the final two sentences of their definition (De Vroey and Obst, 2025).

More must be said, however, about how I disambiguate rewilding and ecological restoration. As Laura J. Martin demonstrates in her comprehensive history of restoration, Wild by Design, Katz (2012) errs in taking restorationists to be necessarily committed to the full recreation of a natural area concomitant with some historical baseline. Notwithstanding the inherent unpredictability of secondary succession, ecologists have known for decades that changes in severely disturbed ecosystems render their full repair or recreation impossible (Hilderbrand et al., 2005; Martin, 2022: 130). Moreover, Martin (2022: 172) notes, ‘historical fidelity did not become a widespread restoration goal among ecologists and environmental organisations until the 1980s’, then led by a questionable ‘design by distance’ agenda to erase ‘invasive’ species and thereby restore ‘pristine’ precolonial landscapes, at least in the USA. The Carbon Copy stipulation soon fell out of favour, and is notably absent from the Society for Ecological Restoration's (SER) revised definition in 2002: ‘The process of assisting the recovery of an ecosystem that has been degraded, damaged or destroyed’ (SER, 2004). What is ‘restored’ is here left unspecified and may refer to: ecological integrity, communities, functions, services, narratives, human–nature relationships, interspecies justice and more besides.

In the SER definition, as Martin (2022: 11) emphasises, ‘the verb “to assist” is key … Restoration, in other words, is an attempt to co-design nature with nonhuman collaborators’. Whilst explicitly resisting total, dominating control (and thereby undercutting Katz's (2012: 74–75) criticism of the analogy), Martin argues, contemporary ecological restoration is analogous to the sympathetic gardener, working with nonhumans to co-generate spaces deemed more appropriate by some as-yet-unspecified standard. ¹ Understood as such, whilst they are arguably co-extensive (Carver et al., 2021), ecological restoration typically involves–and, perhaps, should involve (Hilderbrand et al., 2005)–a commitment to (co-)management, which rewilding need not necessarily share. Indeed, insofar as rewilding's focus is on maximising nonhuman autonomy, it appears somewhat incongruous with the gardening analogy. ² Hence, whilst some full-blown initiatives labelled ‘rewilding’ (such as the Oostvaarderplassen encountered in section ‘Rewilding’s paradoxes’) employ human management and aim to restore historical landscapes, these aspects are inessential to their specifically ‘rewilding’ (rather than idiosyncratic ‘restoration’) focus and often appear (as in this paper) to be detrimental to their ability to rewild substantively (i.e. permit extensive autonomy, and, hence, ‘wildness value’). A given initiative might thus score well in terms of its focus qua ecological restoration, but poorly qua rewilding (and vice versa). ³

If the foregoing is on the right lines, then, even if the gamut of practical ‘rewilding’ initiatives might demonstrate some fuzziness around their multifaceted aspects and foci, insofar as rewilding is intelligible as a concept, manifestable in practice, its mainstream meaning appears to be bound to the very generation and sustenance of nonhuman autonomy that Katz thinks paradoxical and, therefore, impossible.

Rewilding's paradoxes

The epistemological paradox

Katz (2023, 2024) claims that rewilding so understood is doubly paradoxical. The epistemological paradox is that rewilding seeks to restore or create ‘spaces of non-human autonomy – that are supposedly “wild” and free of human meaning’, which are nonetheless grasped only ‘through human thought’, and according to the ‘human concepts’, in opposition to which those spaces gain their meaning as ‘wild’ (Katz, 2024: 241–242). In this context, Katz elaborates, any ‘wild’ nonhuman entity ‘must succumb to a human framework of meaning – even if this meaning is a mere opposition, a non-distinct “other” than human’ (Katz, 2024: 242). Hence, Katz suggests, the epistemological paradox is vicious in respect of its link to anthropocentric domination, since the rewilding advocate cannot even think – let alone engender – something's wildness on its own terms.

Contrary to appearances, and on pain of contradiction, the claim here cannot be that merely by employing human concepts to understand or characterise landscapes such as the remote Amazonian rainforest or Arctic tundra, one necessarily overwrites their wildness qua nonhuman alterity (i.e. their epistemological autonomy). Katz (2023) himself uses the alleged paradox to advocate for the preservation of extant ‘robust nonhuman landscapes … where nonhuman alterity is possible’, over rewilding (where it is not), as a ‘primary goal of any meaningful environmentalism’. If merely thinking of the remote rainforest as ‘wild’ (or ‘nonhuman’ or ‘natural’) were sufficient to anthropocentrically overwrite its meanings, then, especially given Katz's Wittgensteinian commitment that the meanings of such words are fixed by their use in human linguistic contexts, the preservationist would fall foul of the paradox no less than the rewilding advocate (Katz, 2012: 92).

Moreover, the claim that merely thinking of something's nonhuman ‘otherness’ is enough to anthropocentrically negate it is simply implausible, even for the radical postmodern and social constructionist thinkers whom Katz associates with ‘anti-realism’ (2024: 242). Indeed, few such thinkers would recognise themselves in Katz's characterisation whereby ‘the meaning of nature is something that humans create’ (2012: 90). As John D. Caputo has shown, for instance, even Jacques Derrida is not committed to the de facto ‘idealism’ implicit in this version of the charge. Derridean deconstruction, Caputo argues, ‘means to complicate reference, not to deny [or create] it’, and does so ‘motivated not by subjectivism or scepticism but by a kind of hypersensitivity to otherness, by a profound vigilance about the other of language’ (Caputo, 1993: 455). Derrida, like many other underserving victims of the straw man charge of anthropocentric correlationism, does not deny the existence of something genuinely extra-human which partly delineates the meanings that we really do find in the things to which we attribute labels such as ‘wild’ or ‘natural’ (see Golumbia, 2016).

Rather, deconstruction's main tasks in this context are ethically non-anthropocentric ones, broadly in keeping with Katz's project: to generate critical self-reflexivity about the value-laden and exclusionary ways those terms or meanings have been and are used (e.g. by dualistically associating ‘wildness’ with ‘disvaluable’ landscapes, often colonially, with reference to Indigenous practices and identities), with concrete consequences (see the longstanding land-grabbing against Indigenous communities (Cronon, 1996; Scheidel et al., 2023)); and to stress that we have no unfettered epistemic access to ‘wild’ nonhuman things wholly independent of our linguistic, existential and ontologically hybrid situations. This refusal to enframe (the meaning of) ‘wildness’ entirely independent of chains of (in Derrida's case, linguistic) reference might perhaps be termed ‘anti-realist’, but only in virtue of its commitment to a kind of benign ontological (but not ethical) anthropocentrism, which, by eschewing naïve realism or representationalism, militates against a mindset of anthropocentric domination which is our main concern. ⁴

Similar things might be said about more explicitly ‘environmental’ social constructionists such as William Cronon (1996) and Vogel (2015), who, although sceptical about the term ‘nature’, nonetheless identify a real – ‘wild’ – resistance to the anthropogenic creation of artefacts through human labour, which they caution against reflectively overwriting. Likewise, the self-identifying postmodernist Donna Haraway argues that ‘the “real” world’ – roughly akin to what Thoreau calls ‘Pure Nature’ – ‘neither speaks itself nor disappears in favour of a master decoder … In some critical sense … the world encountered in knowledge projects is an active entity’ (Haraway, 1988: 579–593; in Bennett, 2002: 71).

Hence, even if the attribution of terms like ‘wild’ is in some sense epistemologically paradoxical (or aporetic) in the context of rewilding, it need not be viciously so, and it is difficult to see at this juncture – especially in virtue of Katz's own commitments – how the preservationist may evade the same paradox. Indeed, if the above is on the right lines, it may be in taking more seriously the complications of approaching wildness from within an ontologically anthropocentric perspective that the rewilding advocate (or preservationist, for that matter) might disrupt their tendencies toward anthropocentric domination.

Nonetheless, there is another possibility available to Katz, which runs as follows. Whilst there is no special problem involved in trying to think past or extant ‘wildness’ from a human perspective, rewilding raises a novel issue whereby its advocate must imagine and project what a future ‘wild’ space should look like before enacting the rewilding process. Hence, Katz might claim, it is in the planning and intending ‘wildness’ that the rewilding paradox becomes vicious. However, this formulation collapses into what Katz calls the physical paradox.

The physical paradox

The physical paradox holds that ‘rewilding projects endeavour to create spaces in which nature can develop freely without human interference, but inevitably pursue this ideal through human interference in nature’ (Katz, 2023). Katz's key claim is that ‘wildness’ qua ‘non-human natural autonomy’ (2024: 239) cannot be generated or recovered via intentional intervention of any sort, since any attempt to do so (e.g. to rewild) would necessarily overwrite the autonomy of the physical product of that intervention.

Katz (2023, 2024) uses the Dutch Oostvaarderplassen (OVP) rewilding initiative as his exemplar. He notes human management from the outset when polder land was mechanically recovered from the sea in the 1960s. Ostensibly unsuitable for human cultivation, the recovered wetland became a key refuge for declining bird species and was more substantially modified under the guidance of Franz Vera in the 1970s and 1980s, who introduced large grazing herbivores, including Konik horses, and Heck cattle (themselves the product of selective ‘back-breeding’ in the 1920s and ‘30s to mimic extinct indigenous aurochs). These species were introduced to prevent wetlands from reverting to forest less in keeping with the area's historical geography and to keep the OVP friendly to its avian inhabitants. Intervening years have seen sustained human intervention via therapeutic hunting (the area cannot incorporate natural predators) and emergency feeding of starving animals in lean winters. Although seemingly wild, the OVP thus remains ‘a land area that incorporates a human purpose for natural non-human processes’ (Katz, 2024: 233) in terms of the make-up of the polder ecosystem, and the shape that ecosystem is permitted to take over time.

It is tempting to think (following Vogel, 2015: 110–112) that the OVP's need for human intervention is contingent upon poor design and/or its idiosyncratic commitment to restoration of something akin to historical wetlands, which other rewilding (and ecological restoration) initiatives need not share. However, Katz rejects this viewpoint on the basis of an ontological distinction between: ‘(a) the outcomes of human-independent ecological and evolutionary processes’ (i.e. ‘nature’, which Katz defines as: ‘that wild other realm separate from human plans and projects’ (2012: 71)); and, ‘(b) the products of human intentions and design (i.e. “artefacts”)’ (Sandler et al., 2022: 104). Furthermore, if, as Katz claims, on Darwinian grounds, the ontological distinction is based upon the ‘lack of intrinsic function of (a)’ (Sandler et al., 2022), then any ‘natural values’ (including any ‘wildness value’, over and above things like ecological stability or species biodiversity not necessarily bound to extra-human autonomy) grounded in that lack of intrinsic function cannot be possessed by things given an intrinsic function (i.e. the products of human intention – even if that intention is to ‘let the polder be’). And this function apparently cannot ‘wash out’ over time, since the allegedly ‘autonomous development of natural entities will follow a different path because of the initial conditions of the human project’ (Katz, 2024: 242). Hence, any human-led rewilding initiative – which must apparently intend and produce ‘artefacts’ with intrinsic functions that they would not otherwise have – are paradoxical in related ontological and axiological senses. We cannot make them wild, or generate their wildness value, without contradiction.

How does the above relate to rewilding's complicity in an anthropocentric domination mindset? Katz claims that ‘domination is the opposite of autonomy … Any human activity that limits autonomy is thus on a spectrum whose extreme end is domination’ (2024: 244). Rewilding, on Katz's account, even at its most benign, is geared toward the erasure of nonhuman autonomy via anthropogenic management and control of natural processes, as privileged human (i.e. anthropocentric) purposes demand. And, while Katz is attentive to the fact that rewilding initiatives can be guided by heterogeneous purposes (e.g. biodiversity conservation, tourism revenue, entertainment), their guiding intention is always allegedly ‘some kind of human interest or purpose’ (Katz, 2012: 77), albeit sometimes only indirectly. Rewilding, Katz claims, is thus both anthropogenic and anthropocentric, and thereby detrimental to a mindset of openness to the ‘authentic’ nonhuman autonomy and wildness value that motivate it. Moreover, Katz concludes, ‘the acceptance of rewilding as a valid environmental policy acknowledges that nature and natural landscapes no longer exist and that the entire world is an artefact produced by human management and control’ (2024: 246). Since, unlike Bill McKibben (1989), Katz does not think the world yet artefactual in an all-encompassing sense whereby we have definitively ended ‘wild nature’, Katz thinks we should have serious reservations about advocating for rewilding rather than nature preservation.

Wild(er) Katz

The nature/artefact binary

In responding to Katz's identification of rewilding's alleged paradoxes, it is crucial to emphasise the importance of his critique of the binary claim that ‘either the non-human world has its own self-sustaining autonomy or it does not’ (Katz, 2024: 239). This claim tracks Katz's aforementioned ontological binary between wild ‘nature’ (lacking intrinsic function) and ‘artefact’ (its intrinsic function determined by intentional human action). ⁵ Hence, insofar as rewilding projects invariably begin as human-contrived projects, and are therefore concerned with hybrid landscapes involving more-than-human entanglements or assemblages (as Prior and Ward (2016) concur, albeit not with Katz's conclusion), it seems that they just cannot be or become autonomous in a nonhuman sense (and therefore ‘wild’ in the mainstream sense) as the product of directed human action. This is the physical paradox. Thus, if wildness value is inherently connected to nonhuman autonomy, which is apparently binary, then when something loses its nonhuman autonomy, it also irrecoverably loses its wildness value.

My objections to the more fundamental ontological claim about the nature/artefact binary notwithstanding (see below), Katz's contentions have some intuitive merit, at least at the level of specifically individuated ecosystems or organisms. Consider the Amazon wilderness region, of which a third has been lost since 1992 (MAAP, 2024). The portion deforested for gold mining or agribusiness cannot now be rewilded or restored without losing some of the meaning and values it had – or would have had if it been left alone to continue to exist under its own steam – in the first place. This is partly why we bemoan its felling, even when new plantations follow: the plantation just is not the previous forest (or sapling the previous tree), and so any specific iteration of wildness value attached necessarily to the continuation of that particular thing in virtue of its trajectory independent from human intervention is forever lost. That specific forest or tree is gone and with it its wildness in the diachronic sense of the trajectory it would have assumed had it not been interrupted by anthropogenic remodelling. We cannot get it back, no matter how technologically powerful we are, or which other ecologically orientated values we maximise. As with individual human life, regrettably, we cannot get that specific wildness back when it is gone, which is partly why Katz is right that preservation is an essential tool in any serious environmental toolkit.

Note, however, that, there remains a practical problem of ontological individuation here which parallels that faced by ecocentric viewpoints: how can one authentically disambiguate a specific ecosystem or tree from its wider relations other than as a scientific abstraction? Katz needs to be able to do this, since, simplifying things somewhat, if he cannot, he must presumably consider the Earth to be one hybrid ecosystem or superorganism, which would come at a severe ontological cost. Given the huge and well-documented impact of intentional anthropogenic activity, and given his own nature/artefact binary, it would be difficult to see how Katz could avoid McKibben's conclusion about the end of nonhuman nature. Moreover, if that Earth-ecosystem or superorganism is already artefactual in Katz's sense (i.e. largely managed according to human-prescribed internal functions), there is little (if anything) in principle to pick between rewilding and ecological preservation because ‘human management, control and meaning’ (Katz, 2024: 259) becomes endemic to all ‘systems’, not just rewilded or restored ones. None could generate or preserve ‘a space free of human influence and control, a space of robust non-human autonomy’ (Katz, 2024: 260) in Katz's sense, whereby wild nature would persist diachronically.

Katz might assuage this practical concern through, for example, a robust account of how to ontologically individuate ecosystems, and/or clarifying exactly how and when intentionality causes something to become an artefact (I deal with this below). However, more pressing is a substantive conceptual issue with whilst Katz must wrestle. Whilst this aspect of Katz's binary might problematise certain Carbon Copy restoration initiatives which trade implausibly upon the resurrection of something's diachronic wildness, it is not decisive for rewilding (nor ecological restoration more broadly). ⁶ This is because, as previously iterated, unlike that Carbon Copy restorationist, the rewilding advocate's aim is not (or not necessarily) to restore the diachronic wildness of an individual token T. The rewilding advocate can accept that a given token of diachronic wildness (pertaining to that forest, or that tree) has been anthropogenically lost. But that regrettable fact does not entail that some present or future iteration of the landscape cannot be unparadoxically wild (or wilder) in some other sense by which rewilding brings about an increase in the present or future forest's qualitative autonomy (e.g. by sympathetically replanting and then leaving that portion of the forest well alone, or simply standing aside to ‘let nature take its course’ independent of direct human contact, as per Lamlash Bay ‘no-take zone’ (Stewart et al., 2020), or New Zealand's marine reserves, which prevent direct human interference (Ballantine, 2014)).

Katz would likely respond that this is exactly the point: rewilding is paradoxical because its meaning is just too close to Carbon Copy restoration; it requires the causal restoration or implementation of meanings and values (‘wildness’ and ‘wildness value’) forever lost because of their necessary link to a-functional nonhuman autonomy, which cannot be intentionally (i.e. for Katz, functionally) instituted without paradox. On this line of argument, whilst rewilding advocates do not think themselves able to restore a specific token of wildness, they nonetheless aim to causally establish present or future wild tokens, and so aim, similarly implausibly, on Katz's account, to revive diachronic wildness anthropogenically terminated in that space. This is also presumably why Katz takes the problems associated with the OVP to haunt all rewilding initiatives: it is not most fundamentally that the OVP requires ongoing human intervention that prevents it from being authentically wild, but that the OVP is the physical product of rewilding plans which rob any subsequent iteration of its a-functional wildness. If Katz is correct that a specific kind of non-fungible diachronic wildness just is what we refer to when we think about wildness and its value, then it may be that rewilding is viciously paradoxical.

Wildness by degree

This question about whether the ‘wildness’ in rewilding is properly reducible to diachronic wildness in Katz's all-or-nothing sense then becomes pivotal. In recent work Katz, however, complicates (and, perhaps, rejects) the nature/artefact binary, hitherto grounded entirely in something's causal history, and for some good reasons (albeit ones that generate problems for his analysis of rewilding). Note that when I expounded Katz's account of rewilding's link to anthropocentric domination, I commented that, in recent works (e.g. Katz, 2012, 2024), nonhuman autonomy and human dominion become poles of a ‘spectrum’ along which well-meaning ecological intervention might variously be situated.

Katz's primary motivations for reverting to a spectrum account appear to be threefold. Firstly, Katz (2012: 76; 87; 91–6) is keen to avoid being aligned with the anthropocentric (and, following Val Plumwood (1993) I would add, Eurocentric, and androcentric) ‘wilderness myth’, arguably operative in McKibben's post-natural hypothesis, by which pristine nature is implausibly corrupted and becomes mere artefact through human contact of any sort. Whilst I reject Katz's residual assumptions – especially the thought that any human contact necessarily and regrettably diminishes wildness – his thought process is somewhat intelligible. In the US context, Katz thinks, even genuinely wild things may bear to some degree the marks of human–nature relationships that precede the mythical ‘terra nullius’ encountered by colonial settlers.

Secondly, and relatedly, Katz claims that wildness is bound to its ‘internal grammatical relation’ (2012: 92) with artefactualness, which, when explored in a Wittgensteinian context, does not require something to be untouched to be wild in the descriptive sense which, for Katz, underlies proper conceptual analysis. The aforementioned forest or tundra remains largely wild even when it bears the occasional human footprint or settlement, or when its flora and fauna reflect anthropogenic climate change (e.g. large predator migration (Katz, 2023)). The OVP, on the other hand, is much less wild in virtue of its originary anthropogenic inauguration and its substantial functional reworking to suit human intentions, even if some aspects, such as greylag geese colonisation, were unplanned (Lorimer and Driessen, 2014), making the OVP wilder than a very closely managed safari park.

This final point also reflects a third reason Katz gives in response to Vogel's criticism that planned human procreation is paradigmatically ‘natural’, suggesting that creating something's manifesting human intentionality is not sufficient to make it an artefact. It is the extent to which its progenitors retain dominating control over something's functional development which are the marks of artifice (and poor parenting) (Katz, 2012: 82–83; Vogel, 2015: 107–108). Overall, by reverting to a modal continuum Katz aims to do justice to the intuition that, although something was or could be more autonomous (wilder) than it currently is, that concession alone does not make it wholly lacking in wildness qua nonhuman autonomy.

Katz's spectrum, however, generates issues for his analysis of rewilding. The first is a lack of conceptual clarity. In all the aforementioned works, Katz equivocates over two possible interpretations: (a) where the continuum covers all entities, and thus concerns the degree of artefactualness and, correlatively, wildness, in any given space (Katz, 2012: 89); and, (b) where the continuum relates only to artefactual entities, which cross the threshold by being ‘an intentionally created new entity with a new function’, but might nonetheless possess more or less nonhuman autonomy (i.e. ‘wildness’), and, thus, be more or less prone to dominance (rather than mere management) than others (Katz, 2012: 89, 2024: 238). The ambiguity is clear in statements like the following: ‘Any human activity that limits autonomy is thus on a spectrum whose extreme end is domination’ (2024: 244).

For Katz's critical analysis of rewilding to get off the ground, he needs to plump for (b). There needs to be something intrinsically artefactual (i.e. defined ontologically by the inescapable ‘presence of human management and control’ (Katz, 2024: 239)) about a space for it to preclude the possibility of generating authentic wildness and therefore serve as an in-principle objection to rewilding. Option (b), however, appears inattentive to the lesser degree of wildness – and greater artefactualness – that Katz identifies in the ringfencing and preservation of ‘national monuments that are wilderness areas, such as the Giant Sequoias’ (2012: 96), for example, relative to their more definitively autonomous ancient antecedents. And, by Katz's own lights, retaining these Sequoias' ‘wildness-by-degree’ appears to be essential to a properly Wittgensteinian understanding of the sense in which rightly we use the term ‘wild’ to describe them. Indeed, speaking originally in the context of ecological restoration which later Katz takes to be conceptually substitutable for rewilding in this context, Katz says the following: ‘[rewilding] projects may be more or less artefactual because of the kind and amount of new functions that result from the [rewilding] activity’ (2012: 89, emphasis mine). It seems, therefore, that Katz's spectrum concerns not only the degree of dominance manifested in a given artefact, but also the degree of artefactualness in any given space. ⁷

It seems, therefore, that Katz really – and, in my view, rightly – intends to affirm a). However, as Ronald Sandler and colleagues note in a parallel argument concerning de-extinction, this option militates against an in-principle objection to ecological interventions like rewilding (and restoration): ‘By endorsing a continuum conception of naturalness/artefactualness Katz allows that designed ecosystems and organisms could have some degree of naturalness and so natural value …. The products of [rewilding] even though they are the result of human design and intention, might possess some natural value’ (Sandler et al., 2022: 105). Whilst rewilded landscapes might have different meanings to historical ones in virtue of the lost diachronic wildness of a given token T (e.g. the historical forest), this point alone is not enough to generate the requisite value distinction that would speak definitively against rewilding. Rewilding a deforested space cannot bring about that lost diachronic wildness of token T, but it seems that even Katz is (or should be) open to the possibility of generating a space synchronically wilder – and, thus possessing more wildness value – than the devastated former-rainforest, through rewilding. In this respect, as Simon Hailwood argues, the proper analogy for rewilding would not be the attempted reanimation of a corpse, but the intentional release of one's long-held captive, who subsequently possesses significantly greater autonomy (with respect to their human captor), even if, for the liberated captive as much as the rewilded landscape, ‘there can be no “reversion” (even on geological timescales) to the state that would have obtained had there never been a human presence there at all’ (Hailwood, 2019: 131). ⁸

Given the starting point at which rewilding begins, and setting aside any non-essential additional commitment to Carbon Copy restoration et cetera, it seems that the key point for the generation of wildness and wildness value here is about rewilding well (i.e. maximising extra-human autonomy in hybrid environments, in balance with other important commitments), rather than badly (i.e. by dominating entities in those hybrid environments labelled ‘wild’). ⁹ Thus, since wildness qua nonhuman autonomy (and wildness value) is irreducible to a specifically Katzian kind of diachronic wildness, it seems that there is no obvious paradox involved in rewilding. Moreover, even if there is (e.g. in virtue of the ‘anti-realism’ associated with the epistemological paradox), that paradox need not be vicious.

Rewilding and domination

Above, I outlined why rewilding does not appear to be viciously paradoxical. But this was not my only task. I also intended to use this analysis to respond to the claim that rewilding is bound to a mindset of anthropocentric domination. To do this more thoroughly, it will be useful to return to Katz's claim about the nature/artefact spectrum: ‘[rewilding] projects may be more or less artefactual because of the kind and amount of new functions that result from the [rewilding] activity’ (Katz, 2012: 89, emphasis again mine). If the foregoing is on the right lines, what Katz appears to – or should – have in mind is that something's degree of wildness depends not only on its causal history, but also upon the extent to which it manifests freedom from anthropogenic domination (hence: ‘amount’), and the extent to which it reflects ethically anthropocentric concerns (hence: ‘kind’). Again, this seems to me broadly right. Returning to concerns about the limits of the OVP's wildness for illustration, one might raise objections to the current extent of anthropogenic control (e.g. via culling), and the fact that it is now littered with bird hides and frequented by jeep safaris to facilitate anthropocentric enjoyment (Lorimer and Driessen, 2014: 177). But, again, if the foregoing is on the right lines, these issues are remediable in this, or other, initiatives, in the requisite sense to bring about a greater degree of non- or extra-human autonomy – at least in the synchronic sense – in those spaces, in a manner not aligned with anthropocentric domination.

However, even if Katz were friendly to my contentions about rewilding initiatives increasing extra-human autonomy by degree, he might nonetheless have two objections to my proposal that rewilding need not thereby be aligned with anthropocentric domination. He might first reply that it is not possible to reduce the ethical anthropocentrism of rewilding initiatives since they are always allegedly produced for some human benefit (Katz, 2024: 228). However, Katz's supporting arguments for this claim are extremely thin, relying primarily on assertions like the following: ‘the real reason we undergo the restoration [of bamboo forests] is for the benefit of those human beings who like to see pandas’ (2012: 77). I see no obvious reason why some human beings who happen to like pandas might not also just not think rewilding (or restoring) bamboo forests, and thus generating habitats friendly to (relatively) wild pandas, to be the right thing to do (e.g. because synchronic wildness is valuable). Furthermore, as Simon P. James (2006) has convincingly argued, even considering a course of anthropogenic action virtuous by no means reduces its motivation to a matter of anthropocentric self-interest. Thus, as I noted in my discussion of the epistemological paradox, although some degree of ontological anthropocentrism may be inevitable for the rewilding advocate, further argumentation notwithstanding, this concession need not entail ethical anthropocentrism (pernicious or otherwise).

Katz's second prospective reply concerns the compatibility with a descriptive analysis of my synchronic use of the term ‘wildness’ in the rewilding context. My proposal is somewhat similar to Vogel's (2015: 110), who writes:

To call [something] wild would be to say that, in it, forces are currently operating independently of humans, although it will have been human action (and human purpose) in the first place that led to those forces operating. To produce a wild artefact might mean to (intentionally) put natural forces into action and then (intentionally) to let them go in ways that are fundamentally unpredictable and outside one's control.

Katz (2012: 83) is unconvinced that extant environmental projects typically ‘let things go’ in the requisite sense. However, for reasons already explored, I contend that this is rewilding's distinctive focus: it is centrally concerned with the extent to which projects might cede control to the relatively wild spaces rewilding fosters. Katz's later spectrum – which understands nonhuman autonomy and domination to be matters of degree – implies some sympathy with this viewpoint. Additionally, Katz (2024: 242) himself says the following:

I do not deny that natural entities and processes can develop autonomously after human intervention and management activities have ceased, although the time frame for this autonomous development will vary greatly depending on the extent and type of human influence on a particular landscape.

The commitment I elicit from Katz's work also speaks to his more plausible objection to Vogel's proposal. Katz takes issue with Vogel's claim that, since everything is apparently equally wild in the aforementioned sense (e.g. dilapidated shopping malls give way to vegetal life in uncontrollable ways), a homogenised ‘wildness’ ends up meaning ‘simply, reality’ (Vogel, 2015: 114–115). Katz's objection to Vogel's claim that ‘every artefact … is wild’ (Vogel, 2015) is that leaves open no space for something (e.g. our remote rainforest) to be significantly wilder than something else (e.g. City Center Mall). This homogenising of wildness, Katz (2012: 91) claims, both robs ‘wildness’ of its normative force and does not cohere with the term's descriptive or grammatical limits term (I cannot ordinarily say ‘that plastic fork is really wild!’ without hyperbole).

However, unlike Vogel, I do not advocate for a flattening of wildness. It seems to me important to recognise scalar wildness, roughly for the reasons, Katz gives, albeit that Katz provides poor conceptual tools for the task (is calling Lamlash Bay an artefact, for instance, commensurate with that term's grammatical limits?). ¹⁰ Furthermore, for the reasons given above, I do not reduce wildness to synchronic wildness. Rather, I adopt a conceptual pluralism about ‘wildness’ (at least as it is used in contemporary Anglophone rewilding contexts), and argue that this pluralism underwrites the multiple senses of wildness qua nonhuman autonomy with which Katz is concerned. It is not that diachronic and synchronic wildness are unrelated or must target entirely different ‘objects’. Diachronically wild forests are likely very synchronically wild, for instance. But synchronic and diachronic wildness come apart, reflectively speaking. ¹¹ To demonstrate both that this account of wildness is compatible with its proper contextual extensions, and to outline some specifically anti-domination potential in advocating for this use of the term, I turn to Jane Bennett's account of ‘wildness’ in Thoreau's work.

Thoreau's wildness

On Bennett's account, Thoreau is no naive Romantic, but an early postmodern thinker attentive to the heterogeneous hybridity of the more-than-human world in which we participate (Bennett, 2002: 59, 68). Thoreau characterises ‘wildness’ in a thoroughly synchronic sense as a kind of distributive property or power, that shows up in one's experience of something's uncanny, uncontrollable and excessive capacities, and by which it might be distinguished by degree from the domesticated. ‘The most alive is the wildest’, Thoreau remarks: ‘Not yet subdued to man, its presence refreshes him’ (1980: 114). ‘Wildness’, for Thoreau, Bennett explains, ‘is the unexplored, unexpected, and inexplicably foreign dimension of anything. It is more easily “fronted” out of doors, but it resides even within the self’ (2002: 19). Thoreau's is the kind of synchronic wildness that we invoke when we think twice about hiking in the Amazon or camping unaccompanied in the Colombia Mountains (‘those places are too wild!’). However, we see this kind of wildness in various outdoor settings, by degree, in each of the aforementioned rewilding initiatives (even the OVP witnessed the unanticipated nesting of rare white-tailed eagles, attracted by cattle carrion (Lorimer and Driessen, 2014: 175)). It is, I want to suggest, the kind of synchronic wildness that rewilding done well seeks to engender and increase, at least in part.

Speaking also to some of Vogel's concerns, this kind of wild autonomy, again by degree, remains in more industrial or more domestic scenarios. Take, for example, Thoreau's example of bee keeping. One might move hives, or remove honey periodically, but the bees remain somewhat wild insofar as one cannot fully constrain or predict their actions, partly because of their irreducibly agentic or quasi-agentic internal natures. ‘The keeping of bees’, Thoreau says, ‘is a very slight interference. It is like directing sunbeams’ (1973: 22). Similar insights about largely domesticated felines are implied through the telling idiom about the foolishness involved in attempting to herd cats. Indeed, Thoreau spends a good deal of time discussing the wildness within humans (‘the bog in our brain and bowls, the primitive vigour of Nature in us’ (1949: 43)), which recent research about the efficacy of dietary fats on human intentions (e.g. the intention to attack or violently dominate others) only serves to cement (Bennett, 2010: 41).

Crucially, however, unlike in Vogel's account, this kind of wildness is not homogeneous across all hybrid spaces, and so better coheres with the grammar ordinarily employed in rewilding initiatives. Wildness does not require pristine wilderness. The remote Amazon rainforest is somewhat wilder, in Thoreau's sense, than the feline with whom I co-habit, but there is nothing that conforms fully to human intention (i.e. fully domesticated or artefactual). Furthermore, as Thoreau (2007: 179) reminds us, this notion of the ‘wild’ has an etymological precedent in the verb ‘will’, reflecting the claim that wilder things are more apt to do as they will. It is not created by linguistic fiat.

However, we should not oversimplify things. On Bennett's account, Thoreau does not invoke a prediscursive wildness that might overlook the kernel of truth about the inevitability of our ontological anthropocentrism, revealed through Katz's epistemological ‘paradox’. Rather, Thoreau foregrounds that (identifying) properties like ‘wildness’ require a certain hermeneutical awareness about our own irreducible epistemic and ontological contributions to those phenomena. In this respect, Bennett suggests, Thoreau views wildness as also partly about a certain mindset, sensibility or phenomenological openness to the shock or wonder of that which exceeds our control. And it is through this wonder that wildness best reveals or expresses itself, should we be able to slacken the intentional threads which attach us to the inattentiveness of everyday domesticity to recognise it (Bennett, 2002: 60). Speaking about his experience of phosphorescent wood, Thoreau (1972: 181) writes:

I did not regret my not having seen this before, since I now saw it under circumstances so favourable. I was in just the frame of mind to see something wonderful, and this was a phenomenon adequate to my circumstances and expectation, and it put me on the alert to see more like it.

This is important. Thoreau suggests that, in our ordinary urban or domestic settings, ‘we are inclined to lay the chief stress on likeness and not on difference’ (1980: 264). Drawn into the expectation of predictability, technocratic measurement and control associated with the ordinary mindset and discourse of a pre-Heideggerian ‘They’, we often become habitually inattentive to ‘the extraordinary and inexplicable, the surprising and perplexing dimension of things’ (Bennett, 2002: 78). Recognisably wilder spaces (ancient forests, perhaps, but also those more substantially ‘other’ places rewilding generates), however, appear to be particularly adept at reminding us of ‘the pesky remainder produced in the wake of human attempts to organise or combat or know or love nature’ (Bennett, 2002: 53). Although this need not be their functional purpose, substantially rewilded spaces, so understood, also appear to be particularly conducive to inculcating an appetite for what, on Thoreau's behalf, Bennett calls ‘microvision’: a phenomenologically-oriented backgrounding of one's domestic mindset, which inculcates hypersensitivity to the wild otherness of things. This includes the everyday unpredictability of things such as frogs and forgotten toxic refuse, even, in our more mundane urban settings, that might arrest our dominating appetites and tendencies toward unreflectively mastering them (Bennett, 2002: 26, 58). After all, what microvision reveals about wildness is something like the following:

In the most obvious sense, if one cannot presume to master one's own body, which has ‘its’ own forces, many of which can never be comprehended, even with the help of medical knowledge and technologies, one cannot presume to master the rest of the world, which is forever intra-acting in inconceivably complex ways. (Alaimo, 2007: 250)

So, where are we? As the foregoing suggests, rewilding may not only be unparadoxical (or at least not viciously paradoxical), the mindset behind it (the attempt to ‘let things be’ in their extra-human otherness) and the mindset generated by non-instrumental experiences of those spaces (an onto-epistemic openness to wildness, and the subsequent questioning of pernicious prejudices), appears to be markedly geared away from anthropocentric domination, at least in principle. Rewilding, conceived as such, an extension of Bennett's analysis suggests, might connote not only a geographic or ecological ‘deterritorialisation’, but also a deterritorialisation – a partial rewilding, even – of oneself (Bennett, 2002: 97). In my view, little could be further from a mindset of anthropocentric domination than that. None of which is to say that rewilding should necessarily be promoted above other environmental foci or ecological interventions, but rather than we should consider it a live tool in the task that we set ourselves in responding to a precious, but somewhat devastated, shared world.

Does anything need to be called wild?

In closing, I want to briefly address the titular ‘elephant in the room’ in Kyle Whyte's (2024) essay: ‘Why Does Anything Need to be Called Wild?’ As previously intimated, what Whyte calls ‘wild concepts’ have a troubled history. Dualistic demarcation of ‘nonhuman wilderness’ has dehumanised Indigenous communities and erased Indigenous ways of life, especially in North America. However, as Linde de Vroey and Arthur Obst demonstrate via a genealogical analysis of the ‘wildness’ term in European contexts, it too has been associated with ethnocentrism and exclusionism (e.g. via ‘Western Medieval Forest or Wilderness Law’ (2025: 7)). A mere terminological shift from ‘wilderness’ to ‘wildness’ cannot, therefore, inoculate rewilding against colonial connotations. Indeed, all ‘wild concepts’ risk presupposing a Eurocentric purview, absent from Indigenous experiences more often elucidated in terms of interspecies ‘treaties’ or ‘covenants’ (Whyte, 2024). ¹² Thoreau acknowledges this to a degree when he writes that, to the Western coloniser, the forest may be ‘a drear and howling wilderness, but to the [Indigenous person] a home, adapted to his nature, and cheerful as the smile of the Great Spirit’ (1980: 323). Whyte (2024: 72, 81) is, however, more critical:

Numerous Indigenous peoples of what's currently called the US never really used anything like wild concepts to describe their relationships with land, water, plants, animals, and ecosystems. They were more concerned with respecting and enacting specific relationships of interdependence within ecosystems and with nonhuman beings, flows, and entities … When Indigenous persons have used [the term] wild, it's just very different depending on the community … No one hangs their hat on the significance of wildness or the wild as major philosophies. Rather, they are English language words that, on occasion, help to express particular meanings if the interlocuters understand the context.

Whyte's criticism of any attempt to rehabilitate ‘wildness’ (and thus rewilding) is that wild concepts are those we could seemingly do without.

In response, I acknowledge deeply problematic historical (mis-)uses of the term and agree that it lacks the definitive axiological clout one attributes to it when one thinks that calling something ‘wild’ immediately settles the question of its value, or necessitates its immunity from responsible management. Other values and commitments are always at stake. Note, however, that this paper has sought to cast light on synchronic wildness, which appears to be the predominant focus of mainstream rewilding initiatives, and admits of wildness-by-degree. This is important because disambiguating diachronic and synchronic wildness might undercut some key issues that wilderness sceptics like Whyte (2024) and Cronon (1996) identify. Whyte, for instance, rightly problematises the settler colonial identification of ‘savage’ landscapes, along with its more specific misidentifications pertaining to ‘wild’ (‘uncultivated’) rice, which silence and denigrates Indigenous land management. ‘Wild concepts’, Whyte argues, thereby ‘have the effect of removing Indigenous peoples from the histories of ecosystems and nonhuman beings, flows, and entities’ (2024: 77). But this issue appears to be primarily a diachronic one pertaining to the abjection of longstanding relationships, in which, especially if her practice avoids reference to any alleged ‘precolonial wild baseline’ (per the Carbon Copy myth), the rewilding advocate need not participate. Things (‘invasive’ species, even) can be synchronically wild, without requiring that they bear no mark of human interdependence. Likewise, things need not be maximally synchronically wild to be valuable.

Moreover, Whyte (2024: 79–80) himself acknowledges more a positive use of the term ‘wild’, commonplace among Indigenous peoples, which appears strikingly similar to my account of synchronic wildness. It is identified by a relationship of opposition-by-degree with the ‘confinement’ quality of industrial agribusiness and animal domestication and is partly behind claims like the following: ‘Sometimes deep reciprocity is needed, a constant tending of landscapes. Other times, true respect or even trust requires that we leave things alone’ (Whyte, 2024: 81, emphasis mine). Nonetheless, Whyte thinks we should abandon ‘wildness’ because its value is mere convenience. I argue that wildness – at least in a Thoreauvian synchronic sense (contra Cronon's (1996: 8) accusations) – identifies an important feature of the more-than-human world which both underpins that conversational convenience, and may sharpen our practice-orientated decision-making foci. Furthermore, engendering synchronically wilder spaces might inculcate attention to the (value of) wildness hidden in our everyday encounters. Even Cronon (1996: 24) appears to be with me on that. I will end by quoting him at length:

Wilderness gets us into trouble only if we imagine that this experience of wonder and otherness is limited to the remote corners of the planet, or that it somehow depends on pristine landscapes we ourselves do not inhabit. Nothing could be more misleading. The tree in the garden is in reality no less other, no less worthy of our wonder and respect, than the tree in an ancient forest that has never known an axe or a saw – even though the tree in the forest reflects a more intricate web of ecological relationships. The tree in the garden could easily have sprung from the same seed as the tree in the forest, and we can claim only its location and perhaps its form as our own. Both trees stand apart from us; both share our common world. The special power of the tree in the wilderness is to remind us of this fact. It can teach us to recognize the wildness we did not see in the tree we planted in our own backyard. By seeing the otherness in that which is most unfamiliar, we can learn to see it too in that which at first seemed merely ordinary. If wilderness can do this – if it can help us perceive and respect a nature we had forgotten to recognize as natural – then it will become part of the solution to our environmental dilemmas rather than part of the problem.