Wall, Jesse --- "The 'juriste garcon de café' : An essay in honour of Michael Robertson" [2019] OtaLawRw 4; (2019) 16 Otago LR 47

Last Updated: 16 November 2022

The ‘Juriste Garcon de Café’

The ‘Juriste Garcon de Café’:

An Essay in Honour of Michael Robertson

Jesse Wall*

1. owe Michael Robertson a great debt. As a teacher, and scholar, Michael is responsible for my captivation with the Critical strands of jurisprudence. Whilst I am not (yet) a card-carrying member of either the Pescine Society or the Critical Legal Studies movement, I am nonetheless allured by such scholarship. This particular essay is an attempt to isolate the role that an interpreter’s “ideological convictions” play in the activity of the interpretation of law. The discussion draws upon Robertson’s own exegesis of the work of Stanley Fish, and Robertson’s analysis of the relationship between Fish’s work and that of Ronald Dworkin and the Critical Legal Scholars (principally Duncan Kennedy, for the purposes of this paper). These three commentators – Fish, Dworkin, and Kennedy - are united by two common themes. First, they are all “theorists of the legal background”; they explain how the “background elements of a legal system” – “the principles and policies that lie in back of the rules and texts”

– provide reasons to interpret the law in a particular way. Second, these three theorists adopt an anti-Archimedean methodology. That is, they deny that it is possible for an external, second-order, or meta-discourse (about the practice of interpretation) to guide or assess the practice of interpretation. I will explain here that, if we properly embrace the implications of this methodology, we can begin to understand the role that an interpreter’s “forbidden ideological convictions” play in the activity of interpretation. My only hope is that this essay reflects well on Michael, as a teacher and mentor.

This article discusses how an interpreter becomes constrained by the activity of interpretation and how we can evaluate the interpretive finding. This discussion then enables us to consider Duncan Kennedy’s concern that the role of an interpreter of law is conflicted. His concern is that the interpreter must “take positions on legal questions that have no self-evident legal answers, without offering an explanation of how he can do this without bringing in his forbidden ideological convictions (if he has any) ...”.¹ In doing so, I will consider here the accounts of legal interpretation provided by Kennedy, Ronald Dworkin, and Stanley Fish. Two common themes unite these three commentators. The first theme is that they are all “theorists of the legal background”. They explain how the “background elements of a legal system” – “the principles and policies that lie in back of the rules and texts”² – provide reasons to interpret the

* Senior Lecturer, Faculty of Law, University of Auckland.

1. Duncan Kennedy “The Hermeneutic of Suspicion in Contemporary American Legal Thought” (2014) 25(2) Law Critique 91 at 125.
2. Jeremy Waldron “Did Dworkin Ever Answer the Crits?” in Scott Hershovitz (ed) Exploring Law’s Empire: The Jurisprudence of Ronald Dworkin (Oxford University Press, Oxford, 2008) 155 at 155.

law in a particular way. The second theme is that these three theorists adopt, to a greater or lesser extent, an anti-Archimedean methodology. That is, they deny that it is possible for an external, second-order, or meta- discourse (about the practice of interpretation) to guide or assess the practice of interpretation. For Kennedy, this creates the above role- based conflict for the interpreter. I will argue here that Dworkin and Fish, who embrace the implications of their methodologies, can resolve the conflict. In finding resolutions to Kennedy’s concern, we can understand the role that an interpreter’s “forbidden ideological convictions” play in the activity of interpretation.

Once a judge has settled on a relevant legal norm, the further task remains of determining what ought to be done according to the legal norm.³ The activity of interpretation forms part of the task of finding “rational support for legal conclusions ... as to what is to be done according to law”.⁴ That is, an interpreter must perform the “activity of identifying legal reasons that support a conclusion as to the meaning that is to be ascribed to a legal communication”.⁵ For example, a provision in a criminal code that prohibits “acts capable of assisting suicide” may clearly apply to a physician who fulfils a terminally ill patient’s capacious request for medication that would enable the patient to hasten his or her own death.⁶ It is a question of interpreting what meaning ought to be ascribed to the term “suicide” in the criminal prohibition.

Our initial apprehension is – most likely – that “suicide” should mean “intentionally causes one’s own death”.⁷ This meaning may even seem self-evident. If we were required to apply the prohibition, we might feel constrained to interpret “suicide” in that way. However, upon further reflection, we might question whether “suicide” really ought to be interpreted in that way. Perhaps “suicide” should instead mean “intentionally causes one’s own death as a result of irrational or impaired thinking”.⁸ That would be a more liberal, less paternalistic, interpretation. However, when the initially “self-evident” meaning is called into question, it is unclear how we could affirm this interpretation, beyond re-stating the “self-evident” reasons for the interpretation.

3. John Gardner “Legal Positivism: 5½ Myths” (2001) 46(1) American Journal of Jurisprudence 199 at 224.
4. Timothy Endicott “Legal Interpretation” in Andrei Marmor (ed) Routledge Companion to Philosophy of Law (Routledge, New York, 2012) 109 at 110.
5. Timothy Endicott “Interpretation and Indeterminacy: Comments on Andrei Marmor’s Philosophy of Law” (2014) 10(1) Jerusalem Review of Legal Studies 46 at 48.
6. See, for instance, Seales v Attorney-General [2015] NZHC 1239, [2015] 3 NZLR 556.
7. See, for instance, Frederick C Mish (ed) Merriam-Webster’s Collegiate Dictionary (11th ed, Merriam-Webster, Springfield (Mass), 2014) at 1249; and see also James Fitzjames Stephen A History of the Criminal Law of England (McMillan, London, 1883) vol 3 at 104.
8. Seales v Attorney-General, above n 6, at [135].

The discussion here progresses through four sections. In section one, we encounter the interpretive background that sits behind the foreground of the legal text and enables the interpreter to identify “legal reasons that support a conclusion as to the meaning that is to be ascribed” to the foreground legal text.⁹ In section two, we consider how an interpreter can feel constrained to ascribe a particular meaning to a legal text. We then consider, in section three, whether a theory of interpretation can stand outside of the activity of interpretation itself and guide the practice of interpretation. We are then able to consider, in section four, how it follows from Kennedy’s account that the role of the interpreter is conflicted and how Dworkin’s and Fish’s accounts of legal interpretation can resolve this conflict.

I The Background

When required to interpret a legal text, we encounter the text itself in the foreground of the activity of interpretation. The legal text is the object of this activity; it is what requires interpretation. In the “immediate” background rests a series of rights, principles, and policies (as well as other legal norms, standards, and considerations) that we encounter when we consider the legal context of the text. For example, if we were asked to attribute a meaning to “suicide” in a criminal provision that prohibits acts “capable of assisting the suicide of another person”, we might encounter an immediate background that includes the following elements:

(Lib₁) Contrary to the principle of sanctity of human life, the criminal law does not prohibit suicide;¹⁰

(Pat₁) The decriminalisation of “suicide” by the legislature was not intended to prioritise personal autonomy over the sanctity of human life;¹¹

(Lib₂) Where a patient exercises his or her right to refuse treatment, the principle of personal autonomy is prioritised over the sanctity of human life;¹²

(Pat₂) No legal authority permits any positive acts that are performed with the direct intention of hastening a person’s death;¹³

9. Endicott, above n 5, at 48.
10. See, for instance, Suicide Act 1961 (UK) 9 & 10 Eliz 2 c 60, s 1.
11. See, for instance, R (Pretty) v DPP (Secretary of State for the Home Department intervening) [2001] UKHL 61, [2002] 1 AC 800 at [35] (Lord Bingham of Cornhill), [106] (Lord Hope of Craighead); R (Nicklinson) v Ministry of Justice (CNK Alliance Ltd intervening) [2014] UKSC 38, [2015] AC 657 at [212] (Lord Sumption SCJ).
12. See, for instance, Ms B v An NHS Hospital Trust [2002] EWHC 429 (Fam), [2002] 2 All ER 449 at [22]–[27]; In re MB (medical treatment) [1997] 8 Med LR 217 (CA) at 221.
13. See, for instance, Airedale NHS Trust v Bland [1992] UKHL 5; [1993] AC 789 (HL) 892.

(Lib₃) The only legitimate aim that can be advanced for the prohibition on assisting suicide is the protection of vulnerable people who may be induced to commit suicide at a time of weakness;¹⁴

(Pat₃) The current prohibition on acts that assist another to intentionally cause their own death is consistent with the right to life¹⁵ and the right not to be subject to cruel or degrading treatment;¹⁶

(Lib₄) Respect for personal autonomy implies that an act of intentionally causing one’s own death is harmful only when it is the result of irrational decision-making;

(Pat₄) Concern for vulnerable persons implies that it can be necessary to protect those who are susceptible to irrational decision making by limiting the choices that non-vulnerable people can make;

(Lib₅) Changes in social conditions and social attitudes generate reasons for empowering courts to engage in innovate interpretations of legal texts;¹⁷ and

(Pat₅) The interpretation of legislative provisions ought to accord with the intention of the legislator.¹⁸

As I will explain here, theorists of the background account for how meaning is attributed to a legal text by virtue of the interpretive background. Recall that interpretation is the activity of “identifying legal reasons that support a conclusion as to the meaning that is to be ascribed

^{to a legal communication”.19 The above scheme – Lib}1-5 ^{and Pat}1-5 ^– therefore represents some of the possible reasons that could support a conclusion as to the meaning of “suicide”.

14. See, for instance, Nicklinson, above n 11, at [311]; Carter v Canada (Attorney- General) 2015 SCC 5, [2015] 1 SCR 331 at [78]; Pretty v United Kingdom 2002-III Eur Ct HR 155 at [74].
15. See, for instance, R (Pretty) v DPP, above n 11, at [6]–[7]; Pretty v United Kingdom, above n 14, at [39]–[40]; compare Carter v Canada (Attorney- General), above n 14, at [57].
16. See, for instance, Rodriguez v British Columbia (Attorney-General) [1993] 3 SCR 519; R (Pretty) v DPP, above n 11, at [60]; Pretty v United Kingdom, above n 14, at [54].
17. See, for instance, Joseph Raz Between Authority and Interpretation: On the Theory of Law and Practical Reason (Oxford University Press, Oxford, 2009) at 236–237: “Since to be justified the law must be just to the people it is applied to, equity has an inescapable separate role in the application and enforcement of the law ... the moral need for equity to inform interpretation, combined with the tendency of institutions to develop routines ... generate the fourth major factor fashioning interpretation: the role of courts in the development of the law.”
18. See, for instance, Raz, above n 17, at 236: “To the extent that the law arises out of respect for legitimate authority, legal reasoning must establish the law as laid down by authority, that is it must rest on an interpretation of the decisions of legal authorities which accord with the intentions of those authorities.”
19. Endicott, above n 5, at 48.

A Kennedy

According to Kennedy, “the full body of valid legal norms” as well as “more abstract rights and principles” are all “considered relevant in establishing the meaning of [a legal] norm”.²⁰ But as this schema seeks to illustrate, the immediate background that the interpreter encounters “does not form a coherent body of principles and doctrine”.²¹ Rather, as the critical legal studies movement has sought to demonstrate, the valid legal norms and abstract rights and principles are all a “product of real collective conflict, carried on over a long time, among many different wills and imaginations, interests and visions”.²² As a result, the legal background is “a patchwork of provisions” that embodies “diverse and conflicting commitments of principle”.²³

Hence, an interpreter encounters diverse and conflicting rules, policies, and principles – Lib_1-5 as against Pat_1-5 – when they engage the immediate background. The meaning of “suicide” may become less self-evident. It may be that the interpreter must choose between a broad, paternalistic ^{meaning and a narrow, liberal meaning – based upon reasons Pat}1-5^, “suicide” should mean “intentionally causes one’s own death”; based upon reasons Lib_1-5, “suicide” should mean “intentionally causes one’s own death as a result of irrational or impaired thinking”.

When confronted with competing rights, principles, and policies, our instinct may be to strike a balance between these conflicting elements. However, for Kennedy, these conflicting elements are themselves different attempts to balance social visions. The principles of individualism and altruism are different attempts at balancing egoism as against solidarity. By extension, while engaged in the activity of interpreting “suicide”, the principles of liberalism and paternalism can also be seen as different attempts at balancing concern for personal autonomy as against concern for vulnerable persons who may harm themselves. This means that when principles of individualism and altruism conflict, or when principles of liberalism and paternalism conflict, there is no metaprinciple, or higher- order balancing exercise that can determine which value ought to inform our interpretive task. We cannot, according to Kennedy, “balance” these values, since “[t]he imagery of balancing presupposes exactly the kind of more abstract unit of measurement that the sense of contradiction excludes”.²⁴ There are no larger scales upon which we can place our scales

20. Duncan Kennedy “A Left/Phenomenological Alternative to the Hart/ Kelsen Theory of Legal Interpretation” in Duncan Kennedy Legal Reasoning: Collected Essays (Davies Group, Aurora (CO), 2008) 153 at 159–160.
21. Joseph Raz “The Relevance of Coherence” in Ethics in the Public Domain: Essays in the Morality of Law and Politics (Oxford University Press, Oxford, 1994) 277 at 298–300.
22. Roberto Mangabeira Unger What Should Legal Analysis Become? (Verso,

London, 1996) at 65.

23. Waldron, above n 2, at 162.
24. Duncan Kennedy “Form and Substance in Private Law Adjudication” (1976) 89(8) Harvard Law Review 1685 at 1775.

of justice. Kennedy’s account of the immediate background embodies, as Jeremy Waldron explains, “not rival tendencies but incompatible visions or incompatible theories of what we owe to others and to the collective life of our society”.²⁵ Hence, the interpreter encounters conflicting principles when engaging the immediate background without there being any “higher level concepts” available to him or her to resolve the conflict.²⁶

A further resource is therefore needed. To explain, let us adopt Roberto Unger’s tripartite levels of analysis. As we have seen, the interpreter can address “the ambiguities of rules and precedents” (the foreground) with “recourse to imputed purposes or underlying policies and principles” (the immediate background).²⁷ However, the interpreter must also avoid the “ambiguities of these policies and principles” and can only do so “by appealing to some background scheme of association” (the wider background).²⁸ That is, the interpreter must appeal to some third-level set of values or principles (the wider background) that offers “guidance about the relative reach and the specific content of the opposing principles and counterprinciples”.²⁹ Beyond the immediate background elements of the law, a tertiary level of values, beliefs, and assumptions is invoked to address the conflicting immediate background elements. Then the immediate background elements provide the interpreter with reasons for attributing a particular meaning that ought to be attributed to the legal foreground. The “background” therefore has two layers: the legal norms, abstract legal rights, policies, and conventions that can be located in legal materials (the immediate background); and the values, beliefs, and assumptions of the interpreter (the wider background).

This may seem to loosen any interpretive constraints. The immediate background may be “so riven with contradiction as to be capable of offering spurious support for everything and determinate support for nothing in legal reasoning”.³⁰ As we have seen, to attribute a broad, paternalistic meaning to the criminal prohibition, an interpreter can provide reasons based upon Pat_1-5. Alternatively, to attribute a narrow, liberal meaning, an interpreter can provide reasons based upon Lib_1-5. Critically, the implication is that it is the wider background – the tertiary level of values, beliefs, and assumptions – that motivates either finding.

B Dworkin

It is difficult to deny that the elements of the immediate background may have “conflicting inspirations”,³¹ such as the tension between the individualistic and altruistic impulses or the liberal and paternal

25. Waldron, above n 2, at 167.
26. Kennedy, above n 24, at 1775.
27. Roberto Mangabeira Unger “The Critical Legal Studies Movement” (1983) 96(3) Harvard Law Review 561 at 579.

28 At 579.

29 At 619.

30 Waldron, above n 2, at 155 (footnote omitted). 31 At 169.

impulses. Dworkin suggests that the tension should neither worry nor surprise us, as “any moral vision would be defective if it wholly disowned either [conflicting] impulse”³² and “[n]o general interpretation that denied either one would be plausible”.³³ Rather, an interpreter must nonetheless “assume, so far as this is possible, that the law is structured by a coherent set of principles”³⁴ and “assume[ ] that these contradictions are not so pervasive and intractable ... that his task is impossible”.³⁵

For Dworkin, the interpreter curates the materials that make up the pre-interpretive legal background and uses these to formulate the post- interpretive content of the law. Hence, the interpreter’s task is “to impose order over doctrine, not to discover order in the forces that created it”.³⁶ When the interpreter is confronted by “two or more interpretations of some statute”, he or she “must choose between eligible interpretations by asking which shows the community’s structure of institutions and decisions – its public standards as a whole – in a better light from the standpoint of political morality”.³⁷

Similar to the wider set of background values, where the interpreter must choose between eligible interpretations, the interpreter’s “own moral and political convictions are ... directly engaged”.³⁸ The wider background (the interpreter’s own moral and political convictions) gives interpretive shape to the immediate background (the community’s political record), and this enables the interpreter to give a post-interpretive meaning to the legal foreground (the legal text in question).

Note how the above schema is “rigged”, to an extent, against Dworkin’s account.³⁹ The above propositions – Lib_1-5 and Pat_1-5 – are interpretive propositions. The interpreter, according to Dworkin’s account, would take pre-interpretive materials and then determine their interpretive shape. However, the above schema provides propositions that already have a presumptive interpretive shape. At the same time, propositions Lib_1-5 and Pat_1-5 are all part of the community’s record or past political practice. These rights, principles, and policies are “not just outcomes but modes of decision, patterns of argument, emergent doctrines, and so on”,⁴⁰ which all need to be given interpretive shape when attributing a post-interpretive meaning to the term “suicide” in the criminal prohibition. There is therefore a face value tension between the diverse and conflicting pre-interpretive material – the conflicting arguments and doctrines – that the interpreter must resolve through the process of constructive interpretation.

32. Ronald Dworkin Law’s Empire (Belknap Press, Cambridge (Mass), 1986) at 269.

33 At 270.

34 At 243.

35 At 268.

36 At 273.

37 At 255–256.

38 At 256.

39 Waldron, above n 2, at 169. 40 At 170.

C Fish

Fish is a theorist of the wider background. The “background” in any interpretive exercise is broader than what can be explicitly stated in a schema and includes the tertiary level of beliefs, values, and assumptions. The interpretive background therefore includes the interpreter’s “local context” or “unnoticed and already-in-place”⁴¹ set of beliefs, values, and categories that enables and structures the interpreter’s “thoughts and perceptions”.⁴² For example, we – most likely – would not describe Antarctic explorer Lawrence Oates’s act of leaving his tent and walking into a blizzard as an act of “suicide” because he had gangrene and frostbite and did not want to reduce his companions’ changes of survival;⁴³ we have some deeply embedded assumptions about acts of self-sacrifice being distinct from acts of “suicide”. A multitude of such assumptions is what enables meaning to be attributed to text. Hence, the existence of “the legal rule or text” in the foreground is made possible only by virtue of a background set of shared values, beliefs, and assumptions. Note that, similar to Dworkin, Fish would also consider the above schema to be “rigged” against his account of interpretation, since it presupposes a “context-independent” shape (discussed below) to the elements of the immediate background.⁴⁴

Shared background beliefs, values, and categories between interpreters construct “interpretive communities”. The fact that the meaning of “suicide” is reduced to two plausible meanings is attributable to a substantial set of background assumptions that we share as the interpretive community. Moreover, the fact that there are two plausible meanings can also be explained in terms of some different local commitments and assumptions that can point to different meanings in a text. As Fish explains, “[a]ll language is code-like”, and when we encounter an “ordinary” or “literal meaning” of a text and an unorthodox or technical meaning, we encounter “a tension between two forms of code meaning, one of which is shared by a larger public.”⁴⁵ What “suicide” means depends on our shared background and local commitments that, like a code, unlock the meaning that ought to be attributed to the text.

For the legal interpretive community, the set of background beliefs and assumptions will be discipline-specific. For instance, the background beliefs and commitments will include concern for personal autonomy and particular formulations of “the harm principle”, values about

41. Michael Robertson Stanley Fish on Philosophy, Politics and Law: How Fish Works (Cambridge University Press, Cambridge, 2014) at 213.

42 At 13.

43. See, for instance, Sue Limb and Patrick Cordingley Captain Oates: Soldier and Explorer (Pen & Sword Military, Barnsley, 2009).
44. Stanley Fish Doing What Comes Naturally: Change, Rhetoric, and the Practice of Theory in Literary and Legal Studies (Duke University Press, Durham (NC), 1989) at 329.
45. Stanley Fish “There Is No Textualist Position” (2005) 42(2) San Diego L Rev 629 at 638.

human-rights-consistent interpretations, certain respect for authorial intent, certain commitments to stable and consistent application of legal standards, a particular appreciation for the distinction between an act and an omission, and so on. Members of the legal interpretive community, by virtue of their discipline-specific training, share a set of background beliefs that enables and structures the meaning that they attribute to legal texts. Moreover, when there is disagreement within the legal interpretive community, it will be “disagreement that has a particular ... ‘discipline- specific’ shape”.⁴⁶ Disagreement over the meaning of suicide will be most likely structured by the conflicting principles in the Lib-Pat schema because an unnoticed and already-in-place set of beliefs identifies these principles as requiring attention and resolution.

According to our three theorists of the background, behind the foreground text of the law rests a set of immediate background elements: the application of rights, principles, and policies embodied in the relevant legal material. These immediate background elements are supplemented by a wider background: a “background scheme of association”, “the moral and political convictions” of the interpreter, or the unnoticed “local commitments” of the interpreter. We shall now turn to consider how these layers of conflicting and tension-ridden interpretive materials are nonetheless capable of generating stable meanings and the phenomenon of constraint.

2. Constraint When we first encountered the criminal prohibition on assisting “suicide”, I suggested that our initial apprehension of the prohibition was (most

likely) that it applies broadly to instances where “a person intentionally causes his or her own death”. It is common place for us to settle on such a “clear meaning” – and to feel constrained by this interpretation – when we engage with the legal (and text, generally). In this section, we consider how the foreground text can have a clear, stable, and constraining existence even when the elements of the immediate background are in contradiction with one another, tension-ridden, or ultimately the result of the contingent local commitments of the interpreter.

A Kennedy

Kennedy accepts the “idea that law is sometimes determinate and sometimes indeterminate”.⁴⁷ Where the legal text is encountered as determinate, the clarity of meaning is attributable to the interaction between the legal text, the immediate background, and the wider background that renders only one possible interpretation. The law can therefore have the “effect of necessity”, where the interpreter “experiences the argument for a particular answer to a question of legal interpretation to be so strong that there is no plausible argument

46. Robertson, above n 41, at 220, citing Stanley Fish “Response: Interpretation Is Not a Theoretical Issue” (1999) 11(2) Yale J L & Human 509 at 514.
47. Kennedy, above n 20, at 160.

against it”.⁴⁸ The “combined background” therefore necessitates – for the interpreter – a particular meaning that ought to be ascribed or attributed to the text, since the elements of the legal background provide reasons in favor of a particular meaning and the absence of reasons to the contrary.

For Kennedy, the interpreter may apprehend a legal rule as being phenomenologically determinate. However, this determinacy is due to “the work of the interpreter” – the interaction between the legal text and the combined background – and not the “‘qualities’ or ‘attributes’ inherent in the norm [or legal text]”.⁴⁹ Hence, a legal text is encountered as being determinate because the interpreter: ⁵⁰

... performs the classic phenomenological reduction or ‘bracketing’ [epoche] ... of whether the resistance of the rule to reinterpretation is a result of what it ‘really’ is or merely an effect of time, strategy and skill.

Recall how the elements of the legal background are in inevitable conflict, and there is no metaprinciple with which we can resolve the conflict between the principles and policies that make up the legal background. This means that the indeterminacy of a legal text is always possible; it is always possible to generate – from the conflict- ridden legal background – reasons in favor of different meanings that ought to be attributed to the legal text. A legal text is “indeterminate”, according to Kennedy, “only in the sense that sometimes it is possible to destabilize initial apprehensions through legal work [or the activity of interpretation]”.⁵¹ When a legal text is encountered as determinate, and the interpreter is constrained, this is because the interpreter “was unable to accomplish the strategically desired re-interpretation of the initially self-evident meaning of the norm”.⁵²

Hence, if it was “initially self-evident” to the interpreter that “suicide” means “intentionally causes one’s own death”, and if consultation with the legal background by the interpreter provides no plausible alternative meaning, then the meaning of “suicide” is clear, and the interpreter is constrained. The constraint, however, is the product of the interpreter being unable to “destabilize” his or her initial apprehension through the engagement with the conflict-ridden background. Moreover, it is the interpreter’s wider interpretive background that brackets or closes off possible alternative meanings and possible reasons to the contrary.

B Dworkin

Similar to Kennedy, Dworkin also credits the determinacy of legal text to the interpretive interaction between the foreground and the combined background. According to Dworkin, an interpretation of the law is a

48. Kennedy, above n 1, at 126.
49. Kennedy, above n 20, at 160.
50. At 161, citing Edmund Husserl “The Natural Attitude and its Exclusion” in Donn Welton (ed) The Essential Husserl: Basic Writings in Transcendental Phenomenology (Indiana University Press, Bloomington, 1999) 60 at 65.
51. Kennedy, above n 20, at 166–167. 52 At 160.

correct one if it “follow[s] from the principles of justice, fairness, and procedural due process that provide the best constructive interpretation of the community’s legal practice”.⁵³ An interpretation must meet the requirements of fit and integrity. It must be “structured by a coherent set of principles about justice and fairness and procedural due process”⁵⁴ and must fit with the community’s past political decisions. Constructive interpretation has a necessitating effect: narrowing the possible interpretations of the legal text to those that can satisfy the requirements of fit and integrity.

Yet, unlike Kennedy, Dworkin believes that the law is determinate even when the interpreter encounters the background legal elements that provide reasons in favor of different interpretations. Even when the interpreter encounters conflicting lines of authority, there is a principled interpretation: one that follows from substantive engagement with principles of justice, fairness, and due process. Moreover, even in these “[h]ard cases”,⁵⁵ the law is still determinate in the sense that it can be interpreted – by a “sufficiently resourceful interpreter” – in a way that is coherent with the immediate background.⁵⁶ Such an interpreter would therefore give “suicide” the meaning that puts a community’s practice of prohibiting acts “capable of assisting the suicide of another person” in the best possible light. The requirement that the interpretation fits or coheres with the community’s political record may “eliminate interpretations that [the interpreter may] otherwise prefer”,⁵⁷ whilst the interpreter may also engage his or her own moral and political convictions in constructing the relevant principles of justice. A particular meaning, one that represents the best constructive interpretation of the legal text and legal background, ought to then be attributed to the legal prohibition.

C Fish

Fish would also accept that the law – at times – is clear, compelling, and determinate. If the meaning of the foreground text is determined by background values, beliefs, and assumptions, it will be stable if the shared background of the interpretive community is stable. It is therefore possible to have “a situation in which for all competent members of a community the utterance of certain words will be understood in an absolutely uniform way”.⁵⁸

Similar to Kennedy, the stability and determinacy of meaning ought not be credited to the qualities or attributes of the legal text itself. It is a mistake, Fish explains, “to think that it is the property of the words that produces this rather than a set of uniform interpretive assumptions”.⁵⁹

53 Dworkin, above n 32, at 225. 54 At 243.

55 At 255–256.

56. Waldron, above n 2, at 156.
57. Dworkin, above n 32, at 255.
58. Stanley Fish There’s No Such Thing as Free Speech: And It’s a Good Thing, Too (Oxford University Press, New York, 1994) at 301.

59 At 301.

Hence, an interpreter may therefore encounter the legal text as having a clear and compelling meaning by virtue of a set of interpretive assumptions that isolate only one possible meaning. The text is then indisputably clear if the interpretive community also shares the same set of interpretive assumptions. If a meaning “seems to leap off the page”, this is because it “is a meaning that flows from interpretive assumptions so deeply embedded that they have become invisible”.⁶⁰ However, the clear and compelling meaning, and shared set of assumptions, is a “historical contingent experience”.⁶¹ Indeterminacy and disagreement, over the same legal text, would follow if the interpretive community had a set of conflicting or inconsistent background beliefs, values, and categories.

Whereas Kennedy describes the “destabilising” work of interpretation, and Dworkin the “constructive” work of interpretation, Fish would deny that the text needs to be subjected to a process of interpretation. Interpreters do not encounter a “bare [legal foreground] and then access their background beliefs and combine the two components so as to generate an interpreted meaning”.⁶² This two-stage process would require the interpreter to identify a “context-independent semantic meaning” and then consider whether the meaning coheres with the interpretive context.⁶³ Rather, Fish argues that “it is within some or other context – of assumptions, concerns, priorities, expectations – that what an interpreter sees as the ‘semantic meaning’ emerges”.⁶⁴ It is never possible, according to Fish, to identify a context-independent meaning of a legal text and then “interpret the law in context”. The interpretive context pre-empts the activity of interpretation.

Here, we encounter direct disagreement between Dworkin and Fish on the source of interpretive constraint. For Dworkin, moral and political principles, and convictions, are the background to the legal foreground. They are not, for Dworkin, the background to the interpreter. It therefore follows, for Dworkin, that an unconstrained interpreter becomes constrained by the interpretive task that is informed by these principles. However, for Fish, “there is no need to impose upon legal actors any constraints in addition to those that already enable their thinking and acting”.⁶⁵ Fish believes that the interpreter is already constrained by an in-place background of political and moral commitments, as well as discipline-specific assumptions and categories. The interpreter is therefore doubly “constrained” (and enabled) by his or her background moral commitments (as “a non-law constrained actor”) and by his or her legal training (as “a constrained legal actor”). If we are motivated to search for constraints on the interpreter, or consider it “crucial to

60. Fish, above n 44, at 358.
61. Fish, above n 58, at 301.
62. Robertson, above n 41, at 218.
63. Fish, above n 44, at 329. 64 At 329.

65 Robertson, above n 41, at 254.

identify them”, we must be able to “imagine an interpreter who needs the constraints because he stands apart from any tethering structure”.⁶⁶ Fish argues that this need for interpretive constraints presupposes that the text has a range of “irresponsible or whimsical” context-independent meanings that the interpreter needs to be kept from.⁶⁷ However, this need to identify interpretive constraints is misplaced, explains Fish, since such context-independent meanings are not possible.

Theorists of the background therefore have subtly different explanations for the same, commonplace phenomena of clear textual meaning and interpretive agreement. According to Kennedy, a legal text may be initially apprehended as clear and may remain clear whilst the interpreter is closed off from the fractured and contradictory background. Nonetheless, for Kennedy, the activity of interpretation – the interaction between the foreground text, the fractured immediate background, and the wider ideological background – enables the interpreter to destabilise the clarity of the legal text in favor of disagreement and uncertainty. Dworkin’s account moves in the opposite direction. Regardless of whether the interpreter encounters a coherent set of past political decisions that aligns with substantive moral and political principles or encounters inconsistent decisions and competing principles, the activity of interpretation can “impose order over doctrine”.⁶⁸ For Fish, either progression – from stability to instability or from inconsistency to coherence – presumes that the legal text has some kind of context- independent shape to it. Where there is clarity, there is uniformity of interpretive assumptions; and where there is indeterminacy, there is a fracturing of interpretive assumptions, all of which are background assumptions to both the legal text and the interpreter.

III Methodology

Note how “the phenomenon of constraint” is something that is experienced by the interpreter. For Kennedy, the interpreter “experiences” an answer to a question that has “no plausible argument against it”;⁶⁹ for Dworkin, the interpreter “thinks he has genuine, good reasons for accepting his ... interpretation”;⁷⁰ and for Fish, interpreters “find the meanings that are perspicuous for [them]”.⁷¹ When we theorise about the activity of interpretation, we seek to say something more than reporting the interpreter’s felt constraint. We want to be able to evaluate whether the interpreter was constrained. Whilst we ought not expect an “algorithm for correctness in legal argument”, we may be still “inclined to say”, with Waldron, that: ⁷²

66 Fish, above n 44, at 115. 67 At 115.

68. Dworkin, above n 32, at 273.
69. Kennedy, above n 1 at 126.
70. Dworkin, above n 32, at 77.
71. Stanley Fish The Trouble with Principle (Harvard University Press, Cambridge (Mass), 1999) at 294–295.
72. Waldron, above n 2, at 172.

... something must count as tending to show – or there must be something one can do which counts as attempting to show – that the argument one is making is better than any opposing argument one can imagine being made by an opponent in the same case, drawing constructively upon the same array of materials. I mean something must count as that, apart from one’s consciousness that one’s own argument can be made or that it just is the argument one is making.

We seek, in other words, something with which to differentiate good reasons (in favor of a particular meaning) from bad, beyond the interpreter’s own apprehension that they have identified good reasons rather than bad reasons.

Our instinct may be to try and create some critical distance between the activity of interpretation and our theorising about interpretation. That is, we may try and adopt an Archimedean viewpoint that stands “outside of our ordinary ways of thinking about things” in order to “give us some special purchase on the questions that we find most difficult to address”.⁷³ With enough critical distance, enough separation from the subject matter, the theorist has enough leverage with which to move the subject matter. As Archimedes is credited with saying, “give me something to stand on and I can move the world”.⁷⁴ Archimedean theories of interpretation then use a yardstick, which stands outside the text and the interpreter, to measure the success of the interpretation. For instance, “[W]hat is going through an author’s mind[?]” or “[W]hat was going through the minds of the Framers[?]” are examples of simple questions that Archimedean theories of interpretation may ask.⁷⁵ As we shall explore in this section, Kennedy and Dworkin are anti-Archimedean in their accounts of interpretation, and Fish is also anti-Archimedean in one respect.

A Kennedy

Recall Kennedy’s “effect of necessity” that occurs when the interpreter apprehends only one plausible meaning that can be attributed to the legal text. According to Kennedy, this interpretive constraint in applying the law, or “assurance” that the interpreter is being faithful to the law, is “vulnerable to being undone by a critique that the arguer really and truly did not see coming, because the limits of ‘the situation’ made that possibility invisible to him”.⁷⁶

There may be background principles and policies that the interpreter did not engage with, or did not calibrate adequately or curate properly. These elements may suggest that the faithful application of the law requires a different interpretation of the legal text than what was

73. Arthur Ripstein “Introduction: Anti-Archimedeanism” in Arthur Ripstein (ed) Ronald Dworkin: Contemporary Philosophy in Focus (Cambridge University Press, Cambridge, 2007) 1 at 5.
74. At 5.

75 At 14.

76 Kennedy, above n 1, at 127.

provided by the interpreter. The “correctness” of the “self-evident” interpretation may, in this way, be understood as an illusion.

With regard to whether “the effect of necessity” (or the phenomenon of constraint) is “always an illusion, because no actor is ever ‘really’ bound, or never an illusion because there is always a right answer even if this answer was not it”,⁷⁷ Kennedy makes the following anti-Archimedean claim: ⁷⁸

I tend to think that the question whether in any case the effect of necessity was ‘true’ or ‘real’ is a bad question, since the effect ... is the product of interaction between legal work and the materials, and it is never possible to know whether with another work strategy, other resources, more skill and time, the arguer could have undone or even reversed the effect.

In other words, an account of interpretation cannot stand outside of the activity and tell us whether an interpretation is the right one. The phenomena of constraint, of clarity of meaning, and interpretive agreement cannot be supplemented with an outside affirmation that the interpretive solution is the right one. Hence, for Kennedy, the fractured immediate background means that it is possible to destabilise the interpretive agreement and clarity of the legal text, without there being a means of affirming the “correctness” of the previously “obvious” meaning of the text. As we will explore in the next section, this creates a “role-based conflict” for the interpreter.

B Dworkin

Dworkin himself explicitly defends his anti-Archimedean methodology in response to (what he saw as) “Archimedeanism’s golden age”.⁷⁹ As I will explain here, Dworkin is anti-Archimedean in both his account of interpretation (in particular) and in his account of evaluative propositions (in general). The Archimedeans, Dworkin explains, “distinguish the first-order discourse of the practice they study ... from their own second- order platform of ‘meta’ discourse”.⁸⁰ For example, Dworkin suggests that Isaiah Berlin, in “analyzing what liberty really means”,⁸¹ provides a second-order concept: liberty is “the ability to do what you might want to do free from the constraint or coercion of others”.⁸² Whilst this definition may place “liberty” on a collision course with “equality”, the conflict is addressed by Archimedeans as a question of value, not as a conceptual question. The second-order or conceptual definition then “leaves fully open the question whether taxation is in principle justified in spite of its impact on liberty”.⁸³ Dworkin’s concern is that: ⁸⁴

77 At 127.

78 At 127.

79. Ronald Dworkin “Hart’s Postscript and the Character of Political Philosophy” (2004) 24(1) Oxford Journal of Legal Studies 1 at 2.
80. At 2.
81. At 6.
82. At 5.
83. At 6.
84. At 7.

The Archimedean story ignores the way in which political concepts actually function in political argument. They serve as abstract plateaus of agreement. Almost everyone agrees that the values in question are of at least some importance, and perhaps very great importance ...

The same concern arises for Archimedean theories of interpretation. In analysing what makes interpretive reasons good interpretive reasons, the second-order discourse highlights that we ought to “accord some authority to the person or persons who created the text that is being interpreted”.⁸⁵ This second-order discourse does not cement authorial intent as being the sole value in the activity of interpretation, nor can it explain how a conflict between authorial intent and contemporary social conditions should be resolved. It leaves fully open the question whether “living text” interpretation is justified in principle in spite of its impact on authorial intent.

In contrast to Archimedean theories, Dworkin argues that the practice of interpretation has the same character as theories of interpretation. To theorise about authorial intent is to explain why the intent of the person or persons who created the text is important, has value, and informs the meaning that ought to be attributed to the text. Whilst theories of interpretation will be more abstract, concerning authorial intent in general, and the activity of interpretation will be more concrete, concerning the authorial intent of a specific legal rule, the difference is not substantive. There is no Archimedean standpoint, outside of the substantive or first-order discourse, to situate the theorist and enable him or her to guide or correct the practice of interpretation. As Leslie Green explains, for the anti-Archimedeans, there is “no philosophy of law that has as its object, law”.⁸⁶ Rather, “[t]here is only law which is already a philosophical, because interpretive, subject”.⁸⁷ Interpretation is a first-order discourse that may have “more and less abstract aspects”,⁸⁸ without there being a second-order, external, or Archimedean theory of interpretation.

More broadly, Dworkin is critical of Archimedean moral skepticism. This strand of moral skepticism differentiates between internal and external propositions. Internal propositions are “positive, first order, [propositions] of morality”.⁸⁹ External, or second-order, propositions are “express metaphysical or philosophical opinions about the nature of [internal] moral judgments”.⁹⁰ If we accept this distinction, then our “face value view” is made up of external (or “philosophical”) propositions

85. Ripstein, above n 73, at 14.
86. Leslie Green “General Jurisprudence: A 25th Anniversary Essay” (2005) 25(4) Oxford Journal of Legal Studies 565 at 579.

87 At 579.

88 At 579.

89. Ronald Dworkin “Objectivity and Truth: You’d Better Believe It” (1996) 25(2) Philosophy & Public Affairs 87 at 96–97.

90 At 92.

and is therefore “not part of substantive morality”.⁹¹ Dworkin is critical of Archimedean moral skepticism because it allows the proponent: ⁹²

... to abandon all claims as to their own morality’s ultimate truth or even superiority to other moralities ... while still embracing their morality as enthusiastically as ever before ...

It permits first-order statements without committing the proponent to any metaphysical claims about those statements.

In response, Dworkin identifies a set of “further claims” that the proponent may make in argumentation. The proponent may claim that their “opinions are true” in that “[t]hey describe what morality, quite apart from anyone’s impulses and emotions, really demands”. Further, the proponent may claim that their opinions are “really and objectively true” in that “[t]hey would still be true ... even if no one ... thought them true”.⁹³ These “further claims” are not external or second-order claims, leveraged from some Archimedean fulcrum point. The fulcrum point is not possible. Rather, they are internal or first-order claims that are “either repeating or clarifying or supplementing the original [internal proposition] that [assisted suicide] is morally wrong”.⁹⁴ Hence, the moral propositions that we formulate “are not just subjective reactions to the idea of [assisted suicide], but opinions about its actual moral character”⁹⁵ that are internal to the first-order evaluative discourse.

There are, therefore, two implications of Dworkin’s anti-Archimedean methodology that I wish to highlight here. First, in terms of theories of interpretation, “the correctness of an interpretation is internal to interpretive practice”.⁹⁶ There is no means of measuring the quality of an interpretation that is external to the substantive agreement or disagreement in the practice of interpretation. Second, in terms of the moral and political convictions that may be engaged in the practice of interpretation, such propositions are opinions about the moral character of its subject matter. Such propositions can be repeated, clarified, supplemented, abstracted, and applied in the interpretive discourse, but there is no Archimedean standpoint from which we can retract from the metaphysical nature of these propositions.

C Fish

Fish’s methodology is similar to Dworkin’s anti-Archimedean methodology in one important regard. Fish also rejects that the interpreter can move between ordinary discourse with its “thick texture of particular situations with their built-in investments, sedimented histories, contemporary urgencies”, and metadiscourse that is somehow “purified

91 At 93.

92 At 93.

93 At 97.

94 At 99.

95 At 92.

96 Ripstein, above n 73, at 17.

of such particulars and inhabited by large abstractions”.⁹⁷ Notice, for instance, the similarity between Dworkin’s response to the Archimedean view of liberty (above) and Fish’s account of the principles of fairness, equality, and diversity. “When faced with opposing courses of action”, Fish explains:⁹⁸

... one can ask of the contenders, “Which is most responsive to the imperative of fairness?” or “which most conduces to the achievement of equality?” or “which will promote the greatest diversity?”

... If you are determined to go with the alternative that is fairest, you will first have to decide whether by “fairness” you mean fairness to everyone independently of his or her achievements, failures, crimes, citizenship, gender, sexual orientation, or fairness inflected by at least some of the considerations in my non-exhaustive list. If “equality” is your lodestar, then you will have to decide whether you mean equality of access (a strongly procedural notion) or equality of opportunity (which will take into substantive account the current situation and past history of those on whom equality is to be conferred). And, if “diversity” is your watchword, you will have to decide whether under its umbrella you wish to include pedophiles and Neo-Nazis; if you do not, you will have to think of reasons – and those reasons will inevitably be particular and

historical – for excluding them.

The trouble that Fish has with such principles is that they cannot be applied “without fleshing out your favorite abstraction with some set of the particulars [the principle] supposedly transcends”.⁹⁹ Moreover, “if you refuse this task”, you “merely repeat the [principles] as if they were a mantra”.¹⁰⁰

Yet, despite this anti-Archimedean strand, Fish is an Archimedean. For Fish, theoretical, second-order, or meta- discourse is still possible. It is still possible to have a metadiscourse that is distinct in substance from the ordinary discourse. However, theory (or the metadiscourse), like all other practices, “require[s] an already-in-place background”.¹⁰¹ Since all practices (or discourses) require their own already-in-place background, “all practices are separate and autonomous” from one another.¹⁰² Non- philosophical practices (or ordinary discourses) do not need to “be affected by what goes on in the discipline of philosophy” (or by what goes on in the metadiscourse).¹⁰³

In the Archimedean spirit, it is possible (according to Fish) to “stand outside of our ordinary ways of thinking about things”, and construct a metadiscourse about our ordinary discourse, and for the metadiscourse to be different in substance from the ordinary discourse. However,

97. Stanley Fish “Theory Minimalism” (2000) 37(3) San Diego Law Review 761 at 762.

98 At 762.

99 At 762.

100 At 762.

101. Robertson, above n 41, at 58.
102. At 59–60; and see Fish, above n 46, at 512.
103. Robertson, above n 41, at 64.

because the different discourses require different sets of background assumptions, the metadiscourse cannot “give us some special purchase on the questions [in the ordinary discourse] that we find most difficult to address”.¹⁰⁴ “Unlike other Archimedeans,” Arthur Ripstein explains, “Fish does not aim to provide a guide to interpretation, because he denies that any such guide is possible”.¹⁰⁵ He is “a disappointed Archimedean”,¹⁰⁶ since he places critical distance between theory and practice, whilst contending that it is the gap between theory and practice that means that theory cannot guide practice. Fish would reject the Archimedean label.¹⁰⁷ In particular, he would reject the implication that he views theory, metadiscourse, or philosophy, as “the master art underlying all the other arts”.¹⁰⁸ Rather than underlying all other disciples, theory is a distinct and autonomous discipline that stands alongside all other disciplines. It follows from the autonomy of disciplines that:¹⁰⁹

... you can’t make sense of – or evaluate – the arguments and actions of disciplinary agents unless you have identified and held fast to the institutional purpose in relation to which those arguments and actions have been produced.

The Archimedean spirit therefore remains insofar as ordinary discourse and the practice of interpretation is distinct from metadiscourse and theories of interpretation. Yet, unlike the Archimedeans that Fish would seek to distance himself from, metadiscourse and theories of interpretation have no leverage over ordinary discourse and the practice of interpretation.

For our three theorists of the background, theories of interpretation are either internal to the activity of interpretation or external theories of interpretation that are substantively distinct but unable to guide the practice of interpretation. It follows that if the interpreter seeks “something to count as showing that their interpretation is better than any alternatives”, then they are unable to reach for an external yardstick to measure the quality of their interpretation. At best, the interpreter can measure the quality of his or her interpretation as against standards “internal to interpretive practice”.¹¹⁰

IV The “Juriste Garcon de Café”

In this section, we consider the implications of these anti-Archimedean commitments. For Kennedy, the instability of any interpretation and the absence of external measures on the quality of interpretation create a role-based conflict for the interpreter (or “jurist”), requiring them to

104. Ripstein, above n 73, at 5.

105 At 17.

106 At 17.

107. Fish, above n 71, at 178.
108. Fish, above n 58, at 228.
109. Fish, above n 44, at 512.

110 At 17.

act in “bad faith”.¹¹¹ As for Dworkin and Fish, who I argue better come to terms with their respective methodologies, the role-based conflict does not arise.

A Kennedy

To appreciate Kennedy’s concern, recall his account of the (in)determinacy of the foreground text. A legal text is “‘indeterminate’ only in the sense that sometimes it is possible to destabilise initial apprehensions through” the interaction of the foreground text and the combined background elements of interpretation.¹¹² When a legal text is encountered as determinate, and the interpreter is reluctantly constrained to apply a particular meaning, the interpreter is merely “unable to accomplish the strategically desired re-interpretation of the initially self-evident meaning of the norm”.¹¹³ Moreover, since this phenomenon of constraint “is the product of interaction between legal work and the materials”, it is never possible to step outside the activity of interpretation and determine whether the interpreter was “‘really’ bound” to interpret the law in a particular way.¹¹⁴

For Kennedy, it follows from this account that a three-fold “conflict” is “built into the role of the jurist”.¹¹⁵ First, the role of the jurist or interpreter is “to find and defend his answer as categorically ‘right’” despite the vulnerability of any interpretation to being destabilised.¹¹⁶ Recall that “even the seemingly most well-grounded legal interpretation”¹¹⁷ is “shaky” or “vulnerable to being undone by a critique” that animates the fractured immediate background elements in a different way than the interpreter.¹¹⁸ This is a type of methodological conflict. The interpreter is conflicted between the phenomenon of apprehended stable meaning and the vulnerability of instability. Whether his or her interpretation is a “correct” one is – itself – a question of interpretation. There is no measure, criteria, or method – external to the practice of interpretation itself – that the interpreter can use to affirm the quality or “correctness” of his or her interpretation.

Second, the interpreter confronts a rhetorical conflict because he or she must provide reasons for his or her interpretation but cannot refer to the wider – or ideological – interpretative background. The role of the interpreter is to explain or “argue in good faith for the outcome he thinks legally indicated”.¹¹⁹ However, when the interpreter curates and calibrates the rights, principles, and policies that make up the immediate background, he or she will inevitably draw on the wider background in

111. Kennedy, above n 1, at 133.
112. Kennedy, above n 20, at 166–167. 113 At 160.

114 Kennedy, above n 1, at 127. 115 At 125.

116 At 125.

117 At 125.

118 At 127.

119 At 125.

order to animate the background elements. Therefore, if the interpreter cannot refer to the wider background, any reasons that he or she provides for an interpretation will be incomplete reasons. The rhetorical conflict for the interpreter is to provide persuasive but ultimately incomplete reasons for his or her interpretation.

The metholodological and rhetorical conflicts culminate in Kennedy’s third conflict – that the role of the interpreter is to:¹²⁰

... take positions on legal questions that have no self-evident legal answers, without offering an explanation of how he can do this without bringing in his forbidden ideological convictions (if he has any).

This third formulation encapsulates the previous two conflicts. As we know, since the meaning of a legal communication is the product of the interaction between the foreground text and the combined background, there is nothing inherent in the text itself that can provide self-evident ans wers to questions of interpretation. Moreover, it is the wider interpretive background – or ideology – that generates stability in the immediate background and clarity in the foreground. The interpreter cannot settle on a meaning, or explain his or her interpretation, without reference to this set of convictions. Kennedy suggests that the interpreter (or “jurist”) falls “into the category of Sartrean bad faith”.¹²¹ In the same way that Sartre’s café waiter is “engaged in a theatrical presentation of himself as a mechanical function of his duties”,¹²² the jurist also “denies the role of his ideological predilections in generating the outcome he will justify in the language of legal necessity”.¹²³

B Dworkin

Dworkin can rescue Kennedy’s “juriste garcon de café”.¹²⁴ For Dworkin, the role of the interpreter is not troubled by Kennedy’s methodological conflict for two main reasons. The first relates to how we assert, and argue from, moral and political convictions. Consider how, ordinarily, we do not require a means of affirming one’s own beliefs, values, and assumptions. Recall Dworkin’s response to Archimedean moral skepticism. He does not require the interprter to point to something that “count[s] as trying to show that a given moral proposition is objectively correct, over and above the mere assertion and re-assertion of that proposition”.¹²⁵ All moral and political propositions, as opinions about what morality requires, are vulnerable to being destabilized through argumentation. This vulnerability does not prevent the interpreter from asserting that there are right moral and political propositions that contribute to right answers to legal questions. It therefore follows from Dworkin’s anti-Archimedean methodology that it is overdemanding to

120 At 125.

121 At 133.

122 At 133.

123 At 134.

124 At 133–134.

125 Waldron, above n 2, at 172 (footnote omitted).

expect an interpreter to somehow be able to affirm their own background values, beliefs, and assumptions (the wider background) that animate an interpretation of the law.

The second reason why Dworkin is untroubled by Kennedy’s methodological conflict concerns their shared anti-Archimedean theory of interpretation. For Dworkin, an internal (or anti-Archimedean) theory of interpretation still provides a methodology for the interpreter. Whether an interpretation is a “correct” one is a question that is answered, according to Dworkin, by the requirements of fit and integrity. An internal theory of interpretation can still provide a means with which to measure the quality of the interpretation. But as we know, whether the interpretation is a good one is still a question of interpretation. Hence, even when the interpreter encounters fractured or incoherent background elements, the method remains the same. Dworkin’s method can be “understood in terms of an ethic of trying to portray the law as coherent”.¹²⁶ Even “when judges disagree ... about the best interpretation of the community’s political order”, Dworkin explains, “each judge still confirms and reinforces the principled character of our association by striving, in spite of the disagreement”.¹²⁷ Whether or not our theory of interpretation adopts this degree of “descriptive optimism”,¹²⁸ the contention remains that an internal theory of interpretation still provides a method with which we can measure the quality of an interpretation.

It also follows, for Dworkin, that no rhetorical conflict arises. The reasons that the interpreter ought to provide for his or her interpretation are the ones that show how it puts the community’s past political practice in the best possible light. The wider interpretive background – the interpreter’s own moral and political convictions – may be directly engaged when the community’s own political record is fractured. There is no reason for the interpreter’s “ideological convictions” (Kennedy)¹²⁹ or “own moral and political convictions” (Dworkin)¹³⁰ to be “forbidden” from the rhetorical exercise when such convictions feature in the activity of interpretation.

C Fish

For Fish, Kennedy’s “juriste garcon de café” does not need rescuing. As I will explain here, his response is to positively embrace Kennedy’s methodological and rhetorical anxieties. Fish can offer two main reasons to be unconcerned by Kennedy’s methodological conflict. The first is that it is methodologically overdemanding, albeit in a different way from above. Recall that, for Fish, the background values, beliefs, and assumptions are background to both the law and the interpreter. The interpreter is not only unable to affirm the wider background but also has no reason

126 At 179.

127. Dworkin, above n 32, at 264 (emphasis added).
128. Waldron, above n 2, at 155.
129. Kennedy, above n 1, at 125 and 130.
130. Dworkin, above n 32, at 256.

to try and affirm or measure the quality of his or her interpretation in light of the role that the wider background played in formulating it. It is out of the wider interpretive background – “of assumptions, concerns, priorities, expectations” – that a “‘semantic meaning’ emerges”.¹³¹ The wider background enables as much as it constrains. It is not possible, according to Fish, to step outside this “interpretive context” to evaluate the quality of these assumptions, concerns, priorities, and expectations. To attempt to do so wrongly presumes that a “context-independent meaning” is possible.

Second, even if it were possible, the methodological conflict misunderstands the interpretive task. Whilst Fish, like Kennedy, considers “that the resulting pervasive tensions and contradictions in law can never be tidied up in a philosophically satisfying way”,¹³² tidying up the fractured elements of the law is not a task that interpretation is required to perform. In fact, it is because an interpretation is always able to be destabilised that the interpretation can perform the tasks that the law requires of it.

This reorientation of the interpretive task requires some brief explanation. To start, recall Fish’s methodological stance. Theories of interpretation, and the philosophies of law, cannot guide (or measure the quality of) the practice of interpretation and legal reasoning. As Fish asks, “[W]hy should law submit itself to the judgment and values of a foreign practice like philosophy?”¹³³ Recall that according to Fish, as the disappointed Archimedean, “no non-philosophical practice need be affected by what goes on in the discipline of philosophy”.¹³⁴ The practice of interpretation need not be concerned with the logical or philosophical incoherence in the background elements of the law.

The practice of interpretation is instead concerned with two practical tasks. According to Fish, the interpreter must advertise his or her “actions as following faithfully from general principles of justice, due process, impartiality, and so on” as well as “remaking those principles in accordance with the pressures exerted by present-day exigencies”.¹³⁵ Hence, the practice of interpretation is burdened with two jobs: to satisfy the requirement of “stable legal rules that are not amenable to change by legal interpreters”¹³⁶ and to satisfy “the requirements of community morality and ... the broader social consequences in the circumstances of the present case”.¹³⁷

131. Fish, above n 44, at 329.
132. Robertson, above n 41, at 294.
133. At 292–293, citing Fish, above n 44, at 397–398; and Stanley Fish Professional Correctness: Literary Studies and Political Change (Clarendon Press, Oxford, 1995) at 103.
134. Robertson, above n 41, at 64.
135. Fish, above n 58, at 21.
136. Robertson, above n 41, at 291. 137 At 291.

Whilst “[t]he contest between law’s two basic jobs” cannot be “resolved logically”,¹³⁸ the interpreter can nonetheless resolve the tension rhetorically. The interpreter can “tell two stories at the same time” and tell them “as if they were one”.¹³⁹ The rhetorical maneuver is to show how the “apparently irreconcilable impulses – to be purely formal and intuitively moral” – are both given effect to.¹⁴⁰ Fish is not alone in ascribing to the law these two tasks.¹⁴¹

“Rather than being an embarrassment,” Fish explains, this rhetorical maneuver “is an opportunity”.¹⁴² It is an opportunity for the law “to exercise its resourcefulness”¹⁴³ by ensuring that its rules are stable and responsive to present-day exigencies. For Fish, therefore, the methodological conflict mistakes the task of interpretation. All interpretations need to be “vulnerable to being undone”¹⁴⁴ in order for the practice of interpretation to perform the two, incompatible, tasks that it is required to perform. Fish can then turn Kennedy’s rhetorical conflict on its head. The instability of interpretations of legal texts enables interpreters to “reshape the past, efface the evidence of that reshaping, and then declare [themselves] compelled in the present case by the past [they have] just transformed”.¹⁴⁵ Kennedy’s rhetorical conflict is then optimistically recast by Fish as a rhetorical opportunity to tell two stories

– as if they were one – and perform the practical task that is required of the activity of interpretation.

V Archimedean in the End

There are two common themes to how Dworkin’s and Fish’s accounts of legal interpretation can respond to Kennedy’s conflicted interpreter. First, both accounts help demonstrate how Kennedy’s methodological conflict follows from an Archimedean view of the wider interpretive background. With regard to the moral and political convictions that we consciously argue from, Dworkin would take the position that we do not require external affirmation of this aspect of the wider interpretive background. Such convictions form part of the first-order evaluative discourse. With regard to the hidden and already-in-place assumptions, concerns, priorities, and expectations, Fish would argue that we cannot be expected to somehow step outside ourselves and affirm this aspect of the wider interpretive background.

Second, Dworkin’s and Fish’s accounts help us appreciate why Kennedy’s methodological anxiety does not grasp the implications

138 At 295.

139 Fish, above n 58, at 163–164. 140 At 161.

141. See Michael Robertson “Telling the Law’s Two Stories” (2007) 20(2)

Canadian Journal of Law & Jurisprudence 429 at 430.

142. Fish, above n 58, at 161. 143 At 161.
144. Kennedy, above n 1, at 127.
145. Robertson, above n 41, at 299.

of anti-Archimedean interpretive theory. As we have seen, internal theories of interpretation can still measure the quality of interpretation. The internal theory may be the utopia of putting the community’s past political record in the best possible light. It may be the dystopia of declaring oneself compelled by a self-edited past. Or it may be some modest middle ground of making a good faith attempt at upholding the requirements of the rule of law and substantive justice. In any case, such theories nonetheless provide a method of interpretation (that is internal to the practice of interpretation itself). We can accept that it is “never possible to know whether with another work strategy, other resources, more skill and time”¹⁴⁶ an interpreter could destabilise the phenomenon of constraint. We can accept such “global” or “Archimedean” indeterminacy whilst still applying a theory of interpretation that seeks to provide stability to the law through engagement with the values, beliefs, and assumptions held by the interpreter.

It is through this more modest approach that we can come to appreciate Kennedy’s contribution to our understanding of interpretation. The methodological conflict is best understood as a call for some interpretive modesty. That is, an appreciation that even self-evident interpretations rest upon selective and curated use of a fractured and tension-ridden set of background materials. The background is only rendered coherent, and the foreground only apprehended as self-evident, by virtue of a set of values, beliefs, and assumptions held by the interpreter. At best, the interpreter will be reflectively aware of them. In which case, he or she can only (and need only) argue for their correctness and relevance. But, in doing so, the interpreter can (and ought to) apply the best available internal theory of interpretation. And, in doing so, the interpreter is not acting in bad faith when making a good faith attempt at applying the best available internal theory of interpretation.

Recall our initial puzzle: our interpreter needs to ascertain the meaning of “suicide” in a provision that prohibits “acts capable of assisting the suicide of another person”. Let us suppose that the interpreter’s initial apprehension – that “suicide” means “intentionally causes one’s own death” – is also the interpretation that he or she settles on. In doing so, the interpreter may, for example, elevate the contention Pat₅ – that the interpretation of legislative provisions ought to accord with the intention of the legislator – and infer from Pat₁ that the broad, paternal meaning is the intended meaning. The interpreter may do so knowing that another interpreter may highlight Lib₃, Lib₄, and Lib₅ and find that “suicide” should mean “intentionally causing one’s own death as a result of irrational or impaired thinking”. The interpreter may also appreciate that they reached different meanings because they have different moral and political convictions or unconscious or irretrievable ideologies that motivate them to give certain elements priority. We can appreciate why, following Kennedy’s reasoning, that this situation may create, for a sufficiently reflective interpreter, a methodological or rhetorical anxiety.

146. Kennedy, above n 1, at 127.

These anxieties, I have argued, are misplaced. The wider interpretive background will include conscious and reflective moral convictions. For such convictions, there is no external measure for affirming them as being sound convictions. In ordinary argumentation, the stringency of the sanctity of life principle, the moral veracity of the act/omission distinction, or the absoluteness of the principle of self-determination are propositions that are to be argued over rather than externally affirmed or validated. Their accuracy and relevance to the interpretive task becomes a question of interpretation, which “internal” or “anti-Archimedean” theories of interpretation can still answer. The wider interpretive background will also include hidden, unreflective, or unconscious ideological convictions. For these convictions, it makes no sense to expect the interpreter to step outside and evaluate an “unnoticed and already-in- place” set of beliefs, values, and assumptions. There can be no expectation to hide what is already hidden. If the interpreter is unaware that his or her frame of reference has been largely constructed by Joel Feinberg’s Harm to Others¹⁴⁷ or by the Second Vatican Council, it is methodologically overdemanding – perhaps impossible – to expect the interpreter to step out of this frame of reference and into a more “neutral” one.

What I have sought to highlight in this discussion is that, beyond the fractured and conflict-ridden immediate background of rights, principles, and policies, a wider set of background values, beliefs, and assumptions is engaged in the activity of interpretation. By following Dworkin’s and Fish’s methodologies, I have argued here that this wider background – an interpreter’s “ideology” – may be argued over or hidden in the activity of interpretation but ought not to be understood as “forbidden” from the activity of interpretation.

147. Joel Feinberg The Moral Limits of the Criminal Law: Harm to Others (Oxford University Press, Oxford, 1984) vol 1.