Can Nature Hold Rights? It's Not as Easy as You Think

Abstract

The Rights of Nature movement has recently achieved significant successes in using legal personhood as a tool for environmental protection. Perhaps most famously, the Whanganui River in Aotearoa New Zealand was accorded legal personhood in 2017. These kinds of development have attracted plenty of scholarly interest, but few have scrutinized a foundational underlying question: Can natural areas, such as rivers, or other non-sentient natural entities actually be legal persons?

The case of the Whanganui River is an example of the direct legal personhood model: it purports to grant legal rights to the river directly. Some other jurisdictions have set up legal persons to administer rivers, without declaring the rivers themselves to be legal persons: the indirect legal personhood model. This article offers legal-philosophical arguments for why legal personhood cannot be attributed to rivers directly.

Normally, legal persons can hold claim-rights and be legally wronged. Some legal persons, such as human adults, can also be held legally responsible and exercise legal competences by entering into contracts. Natural entities cannot do any of these things. Hence, they cannot be legal persons directly; rather, their putative direct legal personhood will collapse into indirect legal personhood. Hence, treating natural entities as direct legal persons amounts only to a legal fiction. Such fictions may be justified for symbolic reasons. However, if environmental protection requires setting up a legal person to protect a natural entity, such protection in most cases can be realized without claiming that the natural entity itself would have become a legal person.

1. INTRODUCTION

The Rights of Nature movement has been growing in prominence over the last decade or so. In their comprehensive analysis of this movement, Craig Kauffmann and Pamela Martin note that ‘Rights of Nature’ here can mean two distinct things. Firstly, in a broader sense, the phrase can refer to the legal philosophy of Earth Jurisprudence, which advocates a complete restructuring of Western law to ‘prioritize ecosystem functioning’.¹ Such a restructuring need not necessarily rely on the concept of rights.

Secondly, in a narrower sense, the phrase refers to ‘[l]egal provisions recognizing ecosystems as subjects with rights’, as a particular way of legally codifying Earth Jurisprudence principles.² One famous example of this approach is the Whanganui River in Aotearoa New Zealand, which has been accorded legal personhood.³ This article is focused on Rights of Nature in this latter sense. I will scrutinize a foundational question underlying the legal recognition of Rights of Nature: Can non-sentient natural entities⁴ hold rights or be legal persons?

It is increasingly taken for granted that rivers and other natural entities can be legal persons and right holders. This is an instance of a prevalent, deferential anything-goes approach.⁵ According to this approach, more or less any entity can hold rights and be a legal person if the legislator or some other appropriate legal actor decides to endow that entity with rights or legal personhood. This line of thinking has been encapsulated by Natalie Stoljar as the claim that ‘[a]nything can be a legal person because legal persons are stipulated as such or defined into existence’.⁶ Thus, civil law countries could simply introduce a new category of legal persons, such as ‘environmental persons’, in their civil codes. Common law countries – as well as Nordic legal systems, which lack civil codes – would need to use some other legislative strategy.⁷ The anything-goes approach has venerable roots in Western legal theory. For instance, John Salmond propounds in his classic treatise Jurisprudence that ‘the law might, if it so pleased, attribute the quality of personality to a purely imaginary being’.⁸ This article is primarily a contribution to that Western jurisprudential discussion.

There seem to be two primary arguments for the anything-goes approach. According to what I call the deference argument, it is up to the legal authorities – legislatures, supreme courts, and so on – to determine which entities can be legal persons. Therefore, more or less anything can be a legal person. Another argument mounted in support of the anything-goes approach is the corporate argument: given that law can treat invisible, intangible and fictitious entities like corporations as legal persons, surely it can treat virtually any entity as a legal person? This argument may be bolstered by mentioning other peculiar examples of legal persons, such as the putative legal personhood of ships, Hindu idols, and so on.⁹ I will argue against these two arguments, as well as against the anything-goes approach: not anything can be a legal person. Whether natural entities can be legal persons is a question that requires justification; the legislator cannot simply determine it by fiat. Even though some legislative arrangements and judicial pronouncements ascribe legal personhood to natural entities themselves, they may not succeed in actually endowing the area or entity with legal personhood.

In their analysis of the role of legal personhood in river management, Erin O'Donnell and Elizabeth Macpherson compare two models, which I will term indirect legal personhood and direct legal personhood. Australia and Chile have made use of the indirect model by creating legal persons that are tasked with managing certain rivers. In Australia, for instance, the state of Victoria has set up a legal person entitled the Victorian Environmental Water Holder ‘to efficiently manage Victoria's environmental water entitlements to improve the values and health of Victoria's aquatic ecosystems’.¹⁰ On the other hand, the Aotearoa New Zealand model is an example of the direct legal personhood model: it purports to grant legal rights to the river directly by setting up a legal person entitled ‘Te Awa Tupua’.

I will argue in this article that legal personhood or rights cannot be attributed to rivers, or other natural entities, directly. Natural entities, such as rivers, cannot be wronged, held responsible or exercise legal competences, which is why they cannot be legal persons. Rather, their putative direct legal personhood will collapse into indirect legal personhood: the Whanganui River arrangement is better understood in terms of, say, a foundation or a trust, rather than as conferring legal personhood on the river itself. However, this analytical criticism of such direct legal personhood arrangements does not necessarily entail that such arrangements should not be adopted: they can still be justified as legal fictions with symbolic value. To understand these justifications, we should consider the different rationales behind the Rights of Nature movement.

1.1. Rationales behind Rights of Nature

I can discern two primary rationales behind the movement in seeking to extend legal personhood to natural entities. I will term these the ‘instrumental rationale’ and the ‘symbolic rationale’.

The instrumental rationale treats legal personhood as a tool for environmental protection. Someone motivated instrumentally will think of the ‘toolbox’ of Western law, and which tools in that box could be used to protect the environment or natural entities. ‘Tools’ associated with legal personhood are then seen as having the potential to serve these ends. Christopher Stone's article on the standing of trees is a classic example.¹¹ Stone notes a number of legal implications resulting from the fact that natural entities are not treated as right holders. For instance, he brings up the issue of standing. It may be difficult to prevent and remedy the pollution of a river through lawsuits if the doctrine of standing recognizes only anthropocentric interests in the river. Such interests are often fragmented: though the various interests pertaining to the river may be substantial in aggregate, none of the interested parties alone may have sufficient incentive to, say, sue the polluter.¹² Ascribing legal personhood to the river may serve to bring together these disparate interests. Similarly, in a legal system that prioritizes constitutional rights, granting such rights to natural entities allows them to better ‘compete’ with other holders of similar rights on an equal basis: for instance, when conducting judicial review of a law prohibiting the pollution of a river, a court may weigh the potential polluters’ constitutionally protected property rights against the constitutional rights of the river.

The instrumental rationale applies to both direct and indirect legal personhood arrangements. However, the symbolic rationale relates primarily to direct legal personhood arrangements, such as those of the Whanganui River. Legal language matters, and the language used in statutes, constitutions and other important legal texts is often value-laden. Consider a legal system that has ‘marriages’ for heterosexual couples and ‘civil partnerships’ for same-sex couples. These two forms of legal union could give rise to exactly the same rights and duties, the only difference being the label. Regardless, one might argue that such a ‘separate-but-equal’ policy is discriminatory. Thus, the values embedded in, or expressed by legal terminology may make a difference, even when not reflected in legal norms.¹³ Somewhat similarly, recognition of the rights of nature or natural entities in statutes, constitutions and treaties may carry much symbolic weight. Recognizing the rights, or legal personhood, of natural entities may be seen as a challenge to the Western anthropocentric worldview, and is often an attempt at accommodating the worldview of some specific non-Western people.¹⁴ An example of the latter is the Whanganui River in Aotearoa New Zealand, where the Indigenous Māori people consider the river to be a spiritual entity. Hence, declaring the river a legal person serves to respect the Māori worldview. For instance, James Morris and Jacinta Ruru ask, rhetorically, whether legal personality is ‘a vehicle for recognising Māori relationships with water’, and reply: ‘The beauty of the concept is that it takes a Western legal precedent and gives life to a river that better aligns with a Māori worldview that has always regarded rivers as containing their own distinct life forces’.¹⁵

The instrumental and symbolic rationales may also come together. Given the strong symbolic force of rights, legal provisions framed in terms of the rights of nature would send the message that nature is not to be exploited.¹⁶ Presumably, then, legal personhood arrangements that purport to confer legal personhood and rights on natural entities, as with the Whanganui River, would be taken more seriously by officials than arrangements that refrain from such explicit conferral, as with the Australian and Chilean cases.¹⁷

The main aim of this article is not to challenge either of these rationales. The instrumental rationale may certainly be justified; reasons such as path dependency may very well require using existing legal institutions for new purposes. In Mostyn v. Fabrigas,¹⁸ the Mediterranean island of Minorca was treated as part of London for purposes of the lawsuit in order to secure justice for a wrongfully imprisoned individual. Frederick Schauer analyzes this case as a legal fiction:

That conclusion [that Minorca is a part of London] was plainly false and equally plainly produced a just result, and thus Mostyn v. Fabrigas represents the paradigmatic example of using a fiction to achieve what might in earlier days have been done through the vehicle of equity.¹⁹

Another pertinent historical example is the treatment of ships as legal persons by United States (US) admiralty law. The purpose of the doctrine was to ‘address the quagmires of maritime trade’.²⁰ For instance, rather than suing ship owners residing on the other side of the Atlantic, the doctrine allowed for ships to be sued directly. Similarly, in some jurisdictions it may be practically necessary to confer legal rights to nature or certain natural entities in order to secure environmental protection – even if the arrangement were, from an analytical point of view, best treated as a legal fiction.

As regards the symbolic rationale, finding ways to reconcile Western law with Indigenous worldviews is a worthwhile goal. However, such reconciliation need not lead to adopting such a worldview. One can respect the worldview of others without adopting it; and this is the crux of the argument. In this article I argue that a theorist or legal scholar should not simply take legislative and judicial pronouncements at face value. Rather, they should seek to analyze such pronouncements with the help of best available theories of law, legal concepts, and reality more broadly.

The article is structured as follows. Firstly, I discuss differences in how scholars often approach legal and moral concepts respectively: legal concepts are frequently taken to be an ‘easier’ case than their moral counterparts. I argue against this understanding, and then move on to address the deference argument. After this more general treatment, I present my theory of legal personhood and use that theory to show why the corporate argument fails. Finally, I present my positive account of how the domain of legal personhood should be determined, and why natural entities cannot be legal persons.

2. LAW EASY, MORALITY DIFFICULT?

Many philosophers and legal scholars think that difficult moral questions become decidedly more straightforward when transposed to the legal realm.²¹ I will illustrate this phenomenon by sketching and comparing two discourses: one having to do with the moral status of natural entities, the other with their legal status.

The moral status of natural entities has been a topic of careful study in environmental ethics.²² Are some natural entities intrinsically or ultimately valuable? Can natural entities possess moral rights? If so, then which rights do they possess? And so on. Questions surrounding the topic of moral rights can roughly be divided into three categories, with different levels of abstraction and evaluative commitments. An evaluative commitment is a commitment regarding what is good or bad, morally worthy. Thus, a stance on whether some natural entities are intrinsically valuable is an evaluative commitment.

Conceptual questions have to do with the analysis of concepts, such as the concept of a right. Typically, rights theorists take rights to be correlated by duties, and identify as the function of rights to protect either autonomy and choices (will theory of rights) or well-being and interests (interest theory of rights).²³ These conceptual questions can be answered without taking on very many evaluative commitments – though some evaluative commitments are more or less unavoidable.²⁴

Now, contrast these conceptual questions with two further questions. Can some natural entities hold moral rights? This query builds partly on the conceptual question: if rights presuppose autonomy, natural entities may be excluded from the domain of right holders on conceptual grounds – assuming they are not autonomous in the required sense.²⁵ However, answering the ‘can’ question may also take a different form. According to many interest theories, rights can be held only by entities that are of ultimate value, which is why determining who or what can hold rights is a value-laden inquiry.

This idea builds upon Joseph Raz's distinction between three types of value: instrumental value, intrinsic value, and ultimate value. Something is of instrumental value by virtue of its good consequences: a computer may, for instance, have instrumental value for its owner. Something that is of intrinsic value is valuable regardless of its consequences, but it is still valuable to someone or something.²⁶ Art can be intrinsically valuable to human beings: its value does not depend on its consequences. However, both instrumental and intrinsic value derive from ultimate value. Raz provides the example of a man and his companion dog. The relationship between the man and the dog may be instrumentally valuable in bringing about good consequences – but also intrinsically valuable regardless of its consequences. However, according to Raz, intrinsic value must still be derived from ultimate value. Raz holds, anthropocentrically, that the well-being of the man is the ultimate source of intrinsic and instrumental value here. In other words, the intrinsic value of the relationship derives from the ultimate value of the man. However, Raz notes that the well-being of the dog could be thought of by some as such a source of value, too.²⁷ I count myself among them: I think only sentient beings have ultimate value. In Section 5.2, I will expand upon why sentience is so central for rights holding and legal personhood. At any rate, this distinction between intrinsic and ultimate value is useful. It allows us to say that nature, or natural entities, may lack ultimate value but regardless have intrinsic value based on its value to the ultimate value of sentient beings, which is connected with it in various ways. This understanding of value is sentiocentric – that is, it prioritizes sentient beings.

Thirdly, and finally, assuming that some natural entity N can hold moral rights, what moral rights exactly does N hold? Answering this question will, again, require further evaluative commitments and normative reasoning. One may, for instance, need to weigh human property rights against the interests of some natural entity not to be exploited.²⁸ In order to do this, one must not only take a stance on whether both humans and the natural entity are ultimately valuable, but also a stance on, say, whether the rights of human beings can override or trump those of a natural entity, or vice versa.

If we turn to legal rights, we might expect a relatively similar argumentative structure, with an initial conceptual question (‘What are legal rights?’), followed by the ‘can’ question (‘Can natural entities hold legal rights?’), and finally by the ‘what rights’ question (‘What legal rights do natural entities hold, or what legal rights should they hold?’). However, many seem to think differently with regard to legal rights: whereas the moral rights of natural entities are a difficult and controversial topic, legal rights are supposedly easier. One can certainly find sophisticated and rigorous treatments of the legal rights or legal personhood of natural entities, but these are often focused on whether some natural entities should be given legal rights. However, an important antecedent question is seen as immaterial in the legal sphere: many scholars seem to think that certain conceptual features of law and legal rights entail that the ‘can’ question may be skirted over. It is often simply taken for granted that natural entities – regardless of their moral status – can hold legal rights or be legal persons.²⁹

3. THE DEFERENCE ARGUMENT: ‘IF LEGAL AUTHORITIES SAY SO, THEN …’

One of the arguments in favour of the anything-goes approach is the deference argument. According to this argument, anything can be a legal person or right holder because the legislator or some other legal authority can freely determine the domain, extension and intention of any legal concept – or, at least, the concepts of legal personhood or rights – and legal scholars should simply take such determinations at face value. Michael Hartney has endorsed this line of thought very explicitly:

The ultimate touchstone … of all legal statements (and of the meaning of legal terms) is … the acts (and especially the utterances) of these legal authorities. … No argument, philosophical or otherwise, can ever prevail against the pronouncement of the competent legal authority.

This is especially true of the term ‘right’. The utterance of a legal authority (legislature, official, court) that a right is being conferred is conclusive evidence that a legal right has been conferred. Whatever legal authorities say is a legal right, is a legal right, whether this agrees with what philosophers would say about moral rights. If a statute says that trees have rights, then trees have certain legal rights, whether we consider this to be morally defensible or even morally possible.³⁰

I disagree with the deference argument. I have elsewhere distinguished different types of deference with regard to rights.³¹ According to what I call ‘domain deference’, the legislator (or other relevant authorities) may freely decide the domain and range of right holders in that jurisdiction – that is, who or what can hold rights.³² Even more extreme is ‘conceptual deference’, according to which the authorities may also freely decide the meanings of legal terms and concepts.

It is not wholly clear which form of deference Hartney subscribes to,³³ but especially conceptual deference has problematic consequences. For instance, consider the work undertaken by legal theorists on ownership. There is a long-standing debate on whether ownership is best analyzed as a bundle of rights, as dominion over an object, or as something else.³⁴ However, conceptual deference would entail that the legislator could simply decide which of these analyses is correct, by legislative fiat. But could the legislator not have overlooked some important features of ownership? Of course, judges and legal scholars would have to consider the legal import of such a legislative definition when engaging in legal interpretation, but scholars could surely also provide alternative analyses of ownership, with hopes of better explaining the institution. I will expand on these two different modes of inquiry – legal interpretation and analysis – below.

Tomasz Pietrzykowski endorses a somewhat more tempered account with regard to the domain of right holders: legal personhood ‘can be granted to anybody or anything, provided that under binding legal norms the entity on which it is conferred can be treated by others as a holder of separate rights or duties’.³⁵ The proposed domain of right holders can be capacious or narrow, depending on how the condition – that the entity ‘can be treated by others as a holder of separate rights and duties’ – is interpreted. Pietrzykowski relies here on the legal philosophy of Leon Petrażycki:

Leon Petrażycki observes that each sensation of legal obligation comprises the image of the holder of the right, which corresponds to another's obligation. This is why subjects of rights need not be real-life entities; it is sufficient that the subject of a particular imperative-attributive emotion believes that such entities are capable of holding rights. This explains why rights are sometimes ascribed not only to humans but also to gods, natural environment objects, spirits of the dead, future generations etc.³⁶

Petrażycki's theory is certainly a substantive account of rights, and one that might provide some reasons for domain deference. In the next section I discuss the role of legislative pronouncements and definitions in legal scholarship, distinguishing two different modes of inquiry: legal interpretation and analysis of the results of interpretation. I take the strengths of Petrażycki's account to lie in the former mode rather than in the latter. However, the modern analytic theories of rights – which do not ascribe rights based on imperative-attributive emotions – are better suited for the latter task.³⁷ Rather, what I take to be the best theoretical approach to rights involves, firstly, analyzing rights down to their components, employing the Hohfeldian analysis of rights,³⁸ and then constructing an account of rights that is theoretically optimal.³⁹ Most modern rights theorists do not think that an analytic account of rights should try to accommodate all of our everyday linguistic intuitions about rights. Rather, since the folk notion of a right is unclear and muddled, an analytic account should try to improve upon it.

3.1. Can Legal Authorities Be Wrong?

Institutions make various claims about reality. For instance, according to the dogma of the Catholic Church, the Pope is infallible on certain matters of faith and morality. The Pope has used this authority to declare that the Virgin Mary was assumed body and soul into heaven, an event labelled the Assumption of Mary. Should an external observer therefore conclude that if the Pope declares X – ‘X’ being some issue falling within faith and morality – then it follows that X? This would be a strange conclusion to anyone who does not share the teachings of the Catholic Church. We arrive at very fundamental questions about metaphysics, epistemology and objectivity: Is there an objective reality? Even if there were such a reality, can we gain objective knowledge about it, or are there rather only different points of view?

Even postmodernist approaches typically do not deny the existence of an objective reality, but postmodernism is often associated with the denial of our ability to gain any objective knowledge about it. Rather, all we have access to are points of view and discourses. In a way, the ‘really question’ is unnecessary for a robust postmodernist: it is not meaningful to ask whether the Assumption of Mary really took place, because the objective reality is unreachable in any case. However, I am not a postmodernist – I think we can ask the ‘really question’. Answering such a question would mean scrutinizing the premises of the Assumption of Mary, using the best relevant scientific theories and knowledge.

Authoritative legal materials often contain declarations about reality, too. A constitution could make reference to a god or, hypothetically, declare that the Supreme Court is infallible.⁴⁰ Now, the question is how such declarations should be understood: are they actually purporting to describe the objective reality, or are they rather masked expressions of legal norms? For instance, Brian Bix remarks the following with regard to the notion of death:

When the argument in newspaper editorials and at the meetings of hospital ethics committees is framed in terms of the question, ‘Is this patient really still alive?’ the various terms used (‘alive,’ ‘really alive,’ ‘delayed death,’ ‘limbo between life and death,’ etc.) are often only place-markers for positions on difficult ethical questions—for example, ‘Should this patient continue to receive certain forms of treatment?’⁴¹

In this case, putatively conceptual questions actually boil down to questions about legal and ethical norms, and their applicability to the case at hand. Hans Kelsen makes a similar point when discussing Article 347 of the German Commercial Code, according to which ‘a good which is not in time returned to the sender has to be treated as if it had been approved and accepted by the receiver’. Kelsen contests that the law would be pretending that the receiver really had approved and accepted the goods. Kelsen analyzes the provision instead in terms of its legal consequences:

Article 347 is, just like any so-called ‘fiction’ of the legislator, nothing but an abbreviating expression. The law simply wants to attach the same legal consequences to one case as it does to another. To phrase this in a separate norm would be too cumbersome, too laboured; or maybe the second case was not even considered in the first place. It would be superfluous to repeat all the rules which have already been set down for the first case. The legislator can rest content with declaring that in the second case the same rules apply as in the first case. It is a misunderstanding to suppose that this effect would be achieved by forcing the person applying the law to accept the idea that both cases are alike, i.e. that they do not differ as a matter of fact. That they are ‘legally’ the same simply means that despite a natural difference in fact the same legal consequence is supposed to follow.⁴²

Kelsen's analysis can be applied quite straightforwardly to the putative legal personhood of ships as well. Even if similar legal rules apply to the ship as to legal persons, this does not automatically mean that the ship in fact becomes a legal person. Rather, one may side with Justice Holmes, who described the situation as ships being ‘treated as if endowed with personality’.⁴³

Let us take a rather extreme example. The corrupted elite of some country set up a legal person entitled ‘the Great Spaghetti Monster’. They claim that the monster is a god who deserves government funding and tax breaks. Furthermore, the elite claim they can act as mouthpieces for the will of the Great Spaghetti Monster, and they therefore appoint themselves as its representatives. It is obvious to more or less everyone that the whole arrangement is merely a scheme to channel money into the pockets of the elite. How should we understand the arrangement?

It may be helpful to distinguish two relevant modes of inquiry here: legal interpretation, and analysis of the results of said interpretation. A judge or a doctrinal scholar will typically be interested in understanding what legal consequences – legal norms – flow from some legal arrangement, such as the Spaghetti Monster arrangement. Such an interpretation may require her to take an ‘internal point of view’ to the claims of the language. Proper understanding of the norms flowing from the arrangement may be impossible without imagining that the Great Spaghetti Monster really exists.⁴⁴ Of course, it might not be necessary for the judge to actually believe in the monster. Rather, the judge may occupy something akin to what Matthew Kramer denotes the ‘simulative perspective’: she ‘speaks or writes from a certain point of view without being committed to it’ and ‘will frequently elaborate the implications of the set of beliefs which [she] is expressing’.⁴⁵

However, what about a legal theorist or philosopher who is not seeking to ascertain the legal norms of a legal system, but rather to provide the best possible explanation and analysis of said norms as well as the pertinent legal terms and concepts? Should they take it for granted that the Great Spaghetti Monster is now a legal person?⁴⁶ The only credible answer to this question is negative. Instead, an explanation of the Spaghetti Monster case should be informed by the best available theories of reality, law, and the relevant legal concepts. With the help of these tools, the scholar can appraise the case at hand critically.

A legal theorist or philosopher, in seeking to explicate legal concepts and explain the content of some legal system, in most cases should take as the basis for the analysis the way in which terms and concepts are used in authoritative legal materials (such as statutes and precedents), but the theorist or philosopher could certainly also depart from such usage. For instance, I have argued elsewhere that even though animal protection legislation does not typically make any mention of rights, it is perfectly reasonable to understand it as endowing animals with rights: we have duties towards animals which, in the case of humans, would entail rights.⁴⁷

After careful consideration, a scholar might, of course, conclude that the monster really is a legal person. He might, for instance, employ a theory akin to that of Petrażycki, but this takes us to the realm of substantive theorizing, rather than the relatively formal view underlying the deference argument.

The Great Spaghetti Monster is, of course, a frivolous example, used here to pump our intuitions and to make clear the two relevant modes of inquiry: (i) legal interpretation, conducted by judges and doctrinal scholars to ascertain the content of legal norms, and (ii) the exposition, explanation, and analysis of said norms. If we apply this distinction to Rights of Nature provisions and judgments, we can note two interrelated things. Firstly, interpreting such materials may very well require taking on the relevant metaphysical assumptions underpinning them, but such assumptions are not necessary for the subsequent analysis of the norms. Secondly, this distinction also provides us with a perspective on the anything-goes view when invoked by, say, courts. For instance, in a well-publicized – though since overturned – case, the High Court of the State of Uttarakhand in India asserted that:

for a bigger thrust of socio-political-scientific development, evolution of a fictional personality to be a juristic person becomes inevitable. This may be any entity, living inanimate, objects or things. It may be a religious institution or any such useful unit which may impel the Courts to recognise it. This recognition is for subserving the needs and faith of the society.⁴⁸

The Court here affirms the anything-goes view, all the while emphasizing the fictional nature of ‘juristic personhood’.⁴⁹ Though the phrase ‘legal fiction’ is ambiguous, the Court's reasoning may be understood as engaging in a doctrinal language game, focusing on whether there are doctrinal limits to these kinds of legal fiction. If understood as such, there is no necessary contradiction between the Court's claim and mine. The law can treat the river as if it were a legal person in order to achieve certain purposes, much like the law can treat one as having accepted a parcel if one does not send it back within a reasonable time. However, these effects can be achieved without assuming that the river really is a legal person, or that the recipient really has accepted the goods. On the other hand, the fact that some legal personhood arrangements turn out to be fictional does not mean that legal personhood would in all instances amount to a legal fiction. The law can treat some behaviour of the recipient as if they had accepted the goods, but in some cases they have really accepted the goods. Similarly, the fictional nature of some legislative or judicial ascriptions of legal personhood does not entail that all such ascriptions are fictional. Some entities can really be legal persons.

Now, can a natural entity really be a legal person? As noted above, I take this to be a substantive jurisprudential question, and it is this substantive realm where the rest of this article will dwell. I will shortly present my doubts as to why legal personhood, properly understood, cannot be extended to natural entities. However, before then I will address the corporate argument. In the course of so doing I will also present my theory of legal personhood.

4. LEGAL PERSONHOOD AND THE CORPORATE ARGUMENT

Let us firstly note some central features of the Western doctrine of legal personhood. Traditionally, the main types of legal person have been divided into two categories: (i) born human beings (natural persons), and (ii) corporations (artificial persons). The latter includes business corporations, as well as other incorporated entities, such as universities, foundations, and so on.⁵⁰ I will focus on certain ambiguities of the term ‘corporation’ below.

Legal personhood is most commonly understood more or less interchangeably with being a subject of legal rights and/or duties; civil law jurisdictions typically use ‘legal subject’ interchangeably with ‘legal person’. Thus, one is a legal person if, and only if, one holds at least one legal right and/or duty. I have elsewhere termed this understanding as the ‘orthodox view’ of legal personhood.⁵¹ However, I have challenged the orthodox view: the two notions of ‘holding legal rights/and or duties’ and ‘being a legal person’ should be distinguished.

It is encouraging to see that, in their analysis of Rights of Nature, Kauffmann and Martin arrive at a similar conclusion as I have reached in my more general treatment. Kauffmann and Martin distinguish the ‘Nature's Rights Model’ from the ‘Legal Personhood Model’.⁵² The former involves giving unique rights to ecosystems, whereas the latter involves extending to them what Kauffmann and Martin label ‘human rights’. While I am somewhat wary of their choice of the phrase ‘human rights’,⁵³ I believe my theory is more or less in line with what Kauffmann and Martin mean by legal personhood.

According to my Bundle Theory, legal personhood is a cluster property: it consists of a number of incidents of legal personhood. Not all of these incidents always come together. In this respect, legal personhood resembles ownership, understood as a bundle of rights: not all of the incidents of the bundle are always present in all instances of ownership. Though possession of an object is an incident of ownership, one can possess an object without owning it. Similarly, one can hold some rights and duties without being a legal person.⁵⁴ However, being a legal person presupposes that one can hold certain types of rights and duties. I will return to this issue in Section 5.

4.1. Two Ways of Understanding Corporations

As mentioned, the existence of corporations is often taken to illustrate the fact that more or less anything could be a legal person. This is understandable, given that corporations are non-physical entities. If even non-physical entities can be legal persons, then – the argument goes – natural entities certainly can. However, in fact, this corporate argument relies on a theoretical confusion, stemming from a systematic ambiguity pertaining to legal personhood: when discussing the notion, it is easy to conflate bundles of rights and duties with the entities that hold said rights and duties. For example, when discussing Apple Inc., one may be referring either to the purely legal entity created by filing articles of incorporation, or to the collectivity: an organization with a company culture and a hierarchy. These two aspects of a typical corporation come apart in certain atypical cases, such as one-person corporations, which are not collectivities.

Given the number of confusing aspects that arise from the conflation of these two senses of ‘legal person’, I have opted to give them different labels. Thus, an entity that is the subject of an array of rights and duties – sufficient to constitute legal personhood for its holder – is, in my terminology, a legal person. On the other hand, I denote the array of rights and duties itself a ‘legal platform’. Hence, when Matti starts a one-person corporation, he is not creating a legal person, but rather merely a legal platform (see Figure 1).⁵⁵ Now he has two platforms that attach to him: one is his natural legal platform, which is the result of his status as a natural person and which he usually cannot get rid of; the other is his corporate legal platform, which he can, for instance, sell.

Figure 1 Matti's Legal Platforms

Let us now consider two possible ways of understanding the notion of a corporation. The US Supreme Court has famously described corporations as existing ‘only in contemplation of law’.⁵⁶ Following this line of thought, we can understand the corporation as a corporate legal platform: the corporation is simply a legal platform – a bundle of rights and duties. In this case, the corporation is certainly a legal construct. However, if we adopt this definition, we cannot then use the example of the corporation to support the anything-goes view. If the legal person is (by definition) a legal platform – a bundle of rights and duties – it does not follow that anything can be a legal person. Rather, what follows is that only bundles of rights and duties can be legal persons. We can here draw an analogy with other legal concepts that do not involve imposing statuses on objects. For instance, consider the concept of debt. It is a category error to ask, ‘Who or what can be a debt?’ because debts are simply legal relations, rather than statuses imposed upon objects.⁵⁷ Similarly, if ‘corporation’ and ‘legal person’ are defined as bundles of rights and duties, it is more or less meaningless to ask who or what can be a legal person.

The alternative is to denote by ‘corporation’ the entity that actually holds the bundle of rights and duties that comprise the platform. In this case we need to specify the parties that hold the rights and duties constituting the legal platform. This is essentially a question of corporate ontology: is the corporation some kind of a group entity, capable of holding rights and duties distinct from its members, or should the corporation be understood as reducible to the individuals involved? This is a difficult substantive question – and also, to my mind, unavoidable. As Eric Orts notes: ‘To ignore organizational ontology is actually to adopt one or another ontological view unconsciously, ignorantly, or manipulatively’.⁵⁸ I think that the most central case of corporations is that of incorporated collectivities – that is, collective agents that have been given the status of a legal person.

Many ‘traditional’ theories of corporations put weight on one of these two aspects. For instance, so-called fiction theories often emphasize the legally created nature of corporations, thus approaching them as corporate platforms. Real-entity theories posit, as the name implies, that corporations are incorporated collectivities with an existence outside the law as well.⁵⁹ I myself understand corporations as incorporated collectivities. However, we need not pick sides with these theories here; we should simply recognize that we cannot adopt both approaches at the same time, and that neither understanding of the corporation lends support to the anything-goes approach.

If corporations – and legal persons more generally – are understood as legal platforms, the notion of a corporation is restricted to the normative realm. None of the readers of this article can be legal persons because rights and duties cannot read. Instead, the readers can only have a legal person – an array of rights and duties – attributed to them, much like they can have debts.⁶⁰ Furthermore, the fact that we can name bundles of rights and duties ‘Apple Inc.’ or ‘Te Awa Tupua’ does not say anything about who, or what, holds the rights and duties constituting these platforms. On the other hand, if we understand corporations as incorporated collectivities, we have then reached the conclusion that human collectivities – not only human individuals – can be legal persons. From this conclusion it does not plausibly follow that anything could be a legal person. Rather, human collectivities resemble human individuals in a number of relevant ways. As a number of philosophers have recently argued, many such collectivities in fact can function as agents and as loci of interests, separate from the agency and interests of their individual members.⁶¹ Hence, their legal personhood is prima facie quite plausible rather than being a stretch of the legal imagination.

Over the course of this rather lengthy discussion, I have countered two arguments that support the ‘everything-goes view’: the deference argument and the corporate argument. I have thus hoped to establish that not everything can be a legal person. I can now, finally, move on to present my positive view of which entities I take to be potential legal persons.

5. WHICH ENTITIES CAN BE LEGAL PERSONS?

A positive account of potential legal personhood needs to rely on an understanding of what it means to be a legal person. My account is based on the distinction between passive and active legal personhood.

Active legal persons are treated by the law as agents: they can typically decide about their affairs and be held legally responsible for their actions. Some scholars, especially those with connections with sociology of law, seem to think that legal personhood necessarily entails legal recognition as an agent. For instance, Gunther Teubner suggests that the legal personhood of animals would mean their entry as new ‘juridical actors’.⁶² I disagree. Legal personhood can be purely passive, as with infants. Passive legal personhood consists of the kind of incidents that infants are endowed with in a typical Western legal system. Compare a human infant with, say, a non-human animal. Though non-human animals are not legal persons in typical Western legal systems, they do indeed hold some legal rights. For instance, animal welfare legislation protects animals from some forms of maltreatment. It is perfectly reasonable to say that the animals thereby gain correlative rights against maltreatment.⁶³ However, such rights are highly limited in scope and stringency.

On the other hand, a human infant – let us call her Lucy – is endowed with the whole array of passive incidents of legal personhood. These incidents fall within two groups. The substantive incidents determine Lucy's ‘primary’ rights vis-à-vis others and the state. These include the fact that she has strong forms of protection pertaining to her life, liberty and bodily integrity; that contracts can be entered into in her name; that she is able to own property; and that she herself cannot be owned. The remedy incidents, on the other hand, determine what legal means Lucy (or rather her representative, on her behalf) can resort to if her substantive incidents are not respected. Let us say that someone infringes Lucy's right to bodily integrity. The harm Lucy has undergone may be given legal recognition, meaning that the tortfeasor may be liable to compensate Lucy. She also has legal standing, meaning that the tortfeasor can be sued in her name. Finally, if the infringement of her rights is serious enough, she may also be deemed the victim of a crime. Thus, not only does Lucy enjoy much stronger protection than virtually all non-human animals, but she also has access to legal personhood-related tools that are usually unavailable to animals, such as legal standing and tort law. These incidents are set out in Table 1.

Table 1 The Incidents of Legal Personhood

The incidents of passive legal personhood consist primarily of claim-rights. A claim-right is always correlative to some duty, and it is the ‘direction’ of that duty: the holder of a claim-right is the party to whom the correlative duty is owed.⁶⁴ What constitutes a claim-right is highly contested. The main theories that have sought to answer this question are the interest theory and the will theory of rights. According to the interest theory, claim-rights, when respected, further the interests of the right holder, who stands to benefit from the fulfilment of a duty.⁶⁵ On the other hand, according to the will theory, the holding of a claim-right comprises the capacity to control the duty of a third party, for example, the standing to sue the duty bearer if the latter does not comply with her duty.⁶⁶ I side with the interest theory. Its strengths are particularly pronounced when explaining passive legal personhood. The will theory cannot explain why, say, newborn infants can be legal persons, as they cannot, for instance, make decisions about whether to enforce their rights. The interest theory can – without requiring the attribution of agency to the right holder. Now, the will theory is not per se incompatible with my theory of legal personhood, given that I do not equate legal personhood with right holding. However, an account of legal personhood based on the will theory would then need to explain passive legal personhood in terms other than claim-rights. The will theory of rights, on the other hand, finds more success in explaining active legal personhood.

Compare Lucy with Diana, who is an adult and deemed to be sui iuris – that is, not under the power of another. Diana's situation is different in at least two respects. Firstly, Diana may administer her own rights and duties; she normally does not require a representative. Secondly, Diana may be held legally responsible in various ways. She is thus endowed not only with passive but also with active legal personhood. This aspect of legal personhood requires a certain degree of agency: for instance, the capacity to operate with notions such as contracts requires some understanding of the legal-normative world.

What follows from the Bundle Theory is that one may hold a limited number of rights (and duties) without being a legal person.⁶⁷ The line between a ‘mere’ right holder and a legal person is not clear, as is characteristic of cluster concepts. For instance, the path of women to ‘full’ legal personhood in many jurisdictions has been gradual, and there is no single legal development that would have turned women from legal non-persons to legal persons. Rather, women were gradually endowed with an increasing number of the incidents.

We have now distinguished (i) passive legal personhood, (ii) active legal personhood, and (iii) legal platform. The interrelations of these three elements are presented in Figure 2, which represents the legal personhood of an infant, Lauri. The infant has a legal platform that is in his name. However, he is unable to administer his rights and duties. Rather, they are administered on his behalf by his representative, such as a parent. There is thus a tripartite division of elements to keep in mind:

• • the platform itself;

• • the passive party, who is the holder of the rights contained in the platform;

• • the active party, who administers the platform and represents the passive party.

Figure 2 The Tripartite Division between Passive Legal Personhood, Active Legal Personhood and a Legal Platform

This division is illustrated in Figure 2.

In the case of an adult human being of sound mind, the passive and the active parties are the same individual. However, they come apart in the case of passive legal persons, who are instead represented by an active party.

Employing this trifurcation, a legal personhood arrangement that putatively grants legal personhood to a natural entity could be interpreted in at least the following ways:

1. 1. Passive legal personhood:

   1. (a) The natural entity itself (the land, water and so on) is the passive legal person, represented by an administrator.

   2. (b) The passive legal person is a collectivity whose specified joint interests are represented.

   3. (c) There is no passive legal person qua beneficiary: the administrators of the legal platform are simply tasked with carrying out the purpose of the entity.

2. 2. Active legal personhood: further to 1(a) above, the natural entity itself administers its legal platform and bears duties.

I next address the active legal personhood of a natural entity before moving on to passive legal personhood.

5.1. Active Legal Personhood of Natural Entities

The building blocks of active legal personhood are duties and competences.⁶⁸ Adults of sound mind are, on the one hand, held responsible for their deeds and, on the other, able to administer their legal platform.

The active legal personhood of natural entities is quite clearly implausible. For instance, consider the problem that is occasionally brought up with regard to the duties of rivers. If the Whanganui River floods and causes harm, should the river be held responsible? Kauffmann and Martin note that one issue with the legal personhood model is that ‘it creates legal liabilities for ecosystems. … The idea that ecosystems have the same responsibilities to humans that humans have toward each other is absurd. We cannot hold a river liable for flooding, or hold the climate accountable for damage caused by sea level rise’.⁶⁹

The idea is indeed absurd if we actually take the natural entity or ecosystem to be the actor itself. Genuine attributions of responsibility to, say, the river would involve questions such as mens rea and intention. Did the river intend to cause the harm? Could it have foreseen it or prevented it? It does not make sense to ask these questions for the simple reason that rivers are not intentional entities. On the other hand, if we do not attribute legal personhood to the river itself, the idea of the duties of, say, Te Awa Tupua becomes more meaningful. It is perfectly meaningful to think that compensation for damages caused by the flood should be paid from the legal platform Te Awa Tupua, by its administrators, as a type of vicarious liability.

Attributing legal competences to rivers is problematic as well. Richard Tur envisages a contract with a river, ‘stating that if the river rose, it would be given x goats and y of the other things the society deemed valuable’.⁷⁰ Now, again, a jurisdiction or an individual may certainly label this event as a ‘contract’, but is that really an appropriate label? Do most of the usual features of contracts apply? How do we, for instance, apply typical contractual doctrines such as force majeure, mistake, and so on? Can we establish that the river could not have foreseen the drought that prevented it from rising to the level required? The same applies to competences more generally: they are meaningful only when applied to entities with a sufficient level of agency – beliefs, desires and intentions, including the capacity to form such states relating to rights and duties.⁷¹ Contracts, of course, can be entered into in the name of, say, Te Awa Tupua. Someone could, for instance, agree to clean plastic pollution in the river in exchange for money. However, it would clearly not be the river that would be expected to log into its bank account to pay the money owed. Rather, the obligated party would be those empowered to act in the name of the legal platform.⁷² Hence, such contracts would not entail active legal personhood for the river.

There is a rather interesting imaginary ‘case study’ entitled ‘terra0’. Created by the artists Paul Seidler, Paul Kolling and Max Hampshire, the idea is described in the whitepaper as follows:

Can an augmented forest own and utilise itself? terra0 is a self-owned forest; an ongoing art project that strives to set up a prototype of a self-utilizating piece of land. terra0 creates a scenario whereby a forest is able to sell licences to log trees through automated processes, smart contracts and Blockchain technology. In doing so, this forest accumulates capital. A shift from valorization through third parties to a self-utilization makes it possible for the forest to procure its real exchange value, and eventually buy (thus own) itself. The augmented forest, as owner of itself, is in the position to buy more ground and therefore to expand.⁷³

This idea is certainly interesting. It may also have various doctrinal complexities that are outside the scope of this article. The question I am interested in here is whether we can literally say that the forest is doing any of the things that it is supposed to be doing – and it certainly is not: if we wish to find an agent here, it is either the algorithm or, say, its programmers, not the forest.⁷⁴

On the other hand, it would be a mistake to completely reject the legal personhood of natural entities merely because natural entities cannot be active legal persons. Animals can certainly be passive legal persons regardless of whether they qualify for active legal personhood. I will address this aspect of legal personhood next.

5.2. Passive Legal Personhood of Natural Entities

As already mentioned, there are certainly indirect legal personhood arrangements that involve all types of natural entity. Such arrangements do undoubtedly have all sorts of legal consequence, often imposing various duties on human beings – such as, the duty not to pollute the natural area, and the liability to be sued by the relevant legal person if the duty is breached. However, is the duty really owed to the natural entity itself? Is it really the natural entity that is wronged if the duty is breached?

To reiterate my position, I hold that one must be able to hold claim-rights – which I will simply call ‘rights’ – in order to qualify as a passive legal person.⁷⁵ I also think that the interest theory of rights best explains such rights. Hence, a right is correlated by a duty, the performance of which will serve the interests of the right holder. It might therefore be reasonable to think that having interests is sufficient for a capacity to hold claim-rights. According to the interest theory, having interests is certainly a necessary condition for the holding of rights. Consider some obvious cases of entities that do not have interests, such as numbers or the gust of wind that passed by my house yesterday at 6:25pm. It is quite clear that such entities can have neither interests nor rights. However, the notion of interest is still somewhat elusive. What kind of entities, then, can feasibly be said to have interests?

Theorists occasionally distinguish between two senses of ‘interest’.⁷⁶ The term, firstly, can refer to something one is interested in. Such an interest presupposes a basic kind of agency: one is interested in something and presumably works towards attaining it. However, ‘interest’ may also mean what is in one's interest. Though these two senses may be intertwined, it is the latter sense that is more relevant for the interest theory of rights. So, how exactly should this notion of interests be fleshed out? Scholars have offered more and less expansive notions of interests. A relatively, though not extremely⁷⁷ expansive conception of interests is offered by Charles Taylor, who contends that individual organisms may have ‘a good of their own’:⁷⁸ ‘All organisms, whether conscious or not, are teleological centers of life in the sense that each is a unified, coherently ordered system of goal-oriented activities that has a constant tendency to protect and maintain the organism's existence’.⁷⁹

According to Taylor, being a teleological centre of life imbues an organism with ‘a good of its own’ or, in other words, interests. Taylor's overall argument is somewhat surprising. He rejects the moral rights of organisms such as plants by employing a will theory of rights – plants cannot insist on their rights – but, regardless, affirms that plants could hold legal rights by virtue of having a ‘good of their own’. Thus, his account of legal rights is closer to an interest theory or a ‘soft’ will theory, which allows for rights through representation.⁸⁰ Setting aside this incongruity, we can note two things. Firstly, even Taylor seems to deny that ecosystems and natural areas could hold legal rights. Secondly, we can still ask whether being a teleological centre of life really is sufficient for the capacity to hold claim-rights.⁸¹ I would argue against this conclusion because of the intimate connection of claim-rights and wronging.⁸² Consider the relationship between an infant and his legal guardian. Let us say that the infant owns some money, which the administrator is legally obligated to administer in a manner that is beneficial for the infant. If the administrator contravenes this duty, she has wronged the infant. This notion of standing-to-be-wronged is central in explaining the relationship between the guardian and her principal, and it does not apply to plants: plants cannot be wronged.

The idea of standing-to-be-wronged is ‘baked into’ an even narrower conception of interests. According to this conception, interests presuppose sentience, which means having a point of view of the world and subjective experiences. Bonnie Steinbock, for instance, notes that sentient beings have ‘a stake … in their own well-being. It matters to sentient beings how you treat them. This simply does not apply to plants. Plants are totally indifferent to treatment that furthers or hinders their well-being’.⁸³ This view is embraced by scholars such as Peter Singer, Ronald Dworkin, and Joel Feinberg.⁸⁴ Many of the broader views, which ascribe interests to non-sentient organisms, suffer from certain issues, as identified by Katie McShane. For instance, in her taxonomy, McShane describes one account of interests as ‘teleological organization’ having to do with the ‘self-maintenance’ of an organism.⁸⁵ However, such an ‘internal-teleology’ account of interests may lead to rather surprising conclusions: for instance, even flames and hurricanes would fit the bill and would therefore have ‘interests’ and ‘good of their own’, as Sune Holm notes.⁸⁶ For reasons such as these I find the narrow conception of interests to be the most plausible. However, adopting this narrow view of interests does not necessarily affect the overall argument: many who adopt a broader notion of interests do not think that all kinds of interest give rise to the capacity to hold rights.⁸⁷

Christine Korsgaard offers a way of reconciling, to some extent, the Taylorian view with the narrow conception of interests, by explaining from an Aristotelian point of view why plants cannot be wronged even though they have a good their own. She notes, firstly, that things can be good for some X in two different ways. In the functional sense, a whetstone may be good for a knife and gasoline good for a car, because they help these objects in performing their functions:

When we use the concept of good-for in this way, we refer to activities or conditions that maintain or promote the ability of the knife or the car to function well. But of course we do not mean that they are good from the point of view of the knife or the car, for knives and cars do not have points of view.⁸⁸

Things can be good and bad in this functional sense for plants – and for animals, too. However, things can also be good or bad in the final sense, and the ‘point of view’ just referred to is important for this good. According to Korsgaard, ‘[w]e call something “good” in the final sense when we consider it worth having, realizing, or bringing about for its own sake’.⁸⁹ She notes that animals also have a final good because ‘an animal experiences her own condition, and the things that affect it, as good- and bad-for her’.⁹⁰ Korsgaard goes so far as to claim that ‘[t]he final good came into the world with animals, for an animal is, pretty much by definition, the kind of thing that has a final good’.⁹¹

These features of sentient animals explain why we can have duties for animals, and why we can wrong them; but other organisms do not have such final goods because things cannot be good or bad for them in the final sense. Korsgaard remarks that ‘[p]lants seem to have a good of their own. They do not exist for our benefit’.⁹² However, the good of plants is only final in an explanatory sense: ‘the explanation of what is good about the sunshine and rain seems to end with the plants themselves—it does not depend, the way the good of artifacts does, on some other good’.⁹³

It is thus possible to hold that plants have a good of their own, and still deny that plants are of ultimate value and that they can be wronged. Hence, by contravening my duty to take care of some plant, I have not harmed the plant in any more robust sense than if I harm someone else's car.

To sum up, assuming that (i) the passive legal personhood of some X requires that X has the capacity to hold-claim rights; (ii) claim-rights can be held only by entities that have a stake in how things go for them and that thereby can be wronged; and (iii) only sentient beings can have such a stake, then natural entities cannot be passive legal persons.

I should make two caveats here. Firstly, there are some exceptions and difficult cases pertaining to the sentience requirement for passive legal personhood. For instance, some human beings – such as permanently comatose people and anencephalic infants – who are broadly understood as right holders and legal persons, are not sentient.⁹⁴ Such cases are exceedingly difficult, and one possible conclusion is that such individuals cannot hold rights or be legal persons, in spite of their being broadly understood as right holders. On the other hand, some human (and potentially non-human) collectivities are a special case. There are collectivities that can function as group agents – working collectively towards a shared purpose – or as loci of the interests of the affected sentient beings.⁹⁵ Such groups can hold rights and be legal persons in spite of their not being sentient beings, though this capacity depends on the fact that they consist of sentient beings.

Secondly, it is worth reiterating that the argument here does not mean that the Rights of Nature arrangements would somehow be devoid of legal validity. They are certainly valid law, with legal consequences. The question is simply how this law should be analyzed, described, and classified: what are the exact legal consequences following from these arrangements, and to whom do they result in rights and duties?

6. CONCLUSION

This article has scrutinized whether natural entities can hold rights or be legal persons. I have argued firstly against the anything-goes view, and the two main arguments mounted in its support: the deference argument and the corporate argument. I have sought to disarm the deference argument by distinguishing the role that legislative language and concepts play in legal interpretation, as opposed to their role in theorizing. The corporate argument, on the other hand, relies on a conflation of legal person and legal platform. Once this conflation is exposed, the argument loses its power.

I have then argued that whether something can be a legal person depends on whether it can be held responsible; whether it can exercise legal competences; and whether it can hold claim-rights and can therefore be wronged. Natural entities, such as rivers, are not capable of any of these things, and therefore cannot be legal persons.

Now, I may, of course, be wrong. Perhaps there is an explanation for how non-sentient organisms, or even inanimate objects and ecosystems, can be wronged. Or perhaps passive legal personhood is not, under the best theory, connected with wronging and claim-rights in the manner described. Perhaps my overall theory of legal personhood is mistaken. Such questions, however, are a matter of substantive argumentation rather than something that the legislator could simply decide. Addressing such questions is just as unavoidable here as it is with regard to the moral rights, or moral personhood, of natural entities.

The ambitions of this article have been theoretical clarity and preciseness. Hence, the purpose has not been to argue against the idea of employing legal personhood or legal subjectivity for environmental purposes. Rather, I have simply asked whether such arrangements really mean what they are occasionally purported to mean. Is a natural entity really, directly, a legal person following a legal personhood arrangement that purports to give it legal personhood? I have argued that the answer is negative: such arrangements ultimately amount only to indirect legal personhood. There are a number of ways of understanding Rights of Nature arrangements on indirect terms, and the best analysis will depend on the context. Some arrangements can be understood as a locus of the interests of the local collectivity consisting of human and/or non-human animals. Other arrangements might be best conceptualized as a foundation or a trust, not really representing anyone at all. All of these options are open to the scholar trying to make sense of these legal personhood arrangements. In the case of the Te Awa Tupua arrangement, I have argued elsewhere that the administrators of the Te Awa Tupua platform owe their duties towards a collective beneficiary, consisting of the sentient beings that depend on the river in one way or another.⁹⁶ It is this collective beneficiary that the administrators represent rather than the river itself.

However, even if we reject the notion that rivers and other natural entities could be legal persons, this does not necessarily mean being opposed to the political and legislative goals of the Rights of Nature movement. Enacting legislation that treats some natural entity as a direct legal person is a type of legal fiction, but such legal fictions may be justified in the light of the instrumental and symbolic rationales. The symbolic rationale, in particular, can justify the fiction of purporting to give direct legal personhood to nature or natural entities. However, in most cases the instrumental rationale can presumably be served by indirect legal personhood arrangements. If environmental protection requires setting up a legal person to protect a natural entity, such an arrangement is likely to be achieved without claiming that the natural entity has become a legal person. In some cases the fiction of direct legal personhood for natural entities might, of course, be required for some doctrinal or other reason. However, even if the legislature or judiciary were to establish direct legal personhood for some natural entity, legal scholars need not take at face value the ontological assumptions underlying such arrangements.