This is Water
It is both an honor and a great pleasure to participate in this discussion of Robin West’s brilliant book, NORMATIVE JURISPRUDENCE. There are so many ideas to laud in this work, many of which have been ably raised by other commentators within this conversation. But reflecting upon this work, I have been particularly struck (as I first was as a student) by Robin’s extraordinary capacity to illuminate aspects of our legal landscape that, while foundational and ubiquitous, remain invisible. Robin’s chalk-outline of a missing progressive normative jurisprudence calls to mind a parable told by the too-soon departed David Foster Wallace. In a commencement address at Kenyon College, Wallace told the story of two young fish out for a swim who happen upon an older fish. The older fish says: “Good morning boys. How’s the water?” After the older fish passes by them, one of the young fish turns to the other and says: “What the hell is water?”
In NORMATIVE JURISPRUDENCE, Robin offers us an opportunity to rethink the “water” of our analytic practices. Most significantly, she presents the possibility of a jurisprudence in which normative argument constitutes the “water” of our analytical practice. Normative argument, she argues, should occupy a central rather than marginalized role in our jurisprudence. Moreover, she observes that progressives’ absence from the normative table has relegated our jurisprudential conversations to an unduly narrow and adjudicatorally-obsessed preoccupation with explicating the law that we already have. This positivist analytic jurisprudence (or, in Robin’s helpful Benthamite parlance, “expository” jurisprudence) has consciously and perhaps even aggressively eschewed normative argument to the peril of the project of legal reform and the promotion of social justice. Robin’s point is not that the project of exposition should be set aside in favor of a project of developing normative/critical (or, again in Robin’s Bethamite vocabulary, “censorial”) jurisprudence, but rather that room should be made in the center stage of our jurisprudential tradition for normative/critical/censorial jurisprudence.
Yet within Robin’s rendering lies room for the hypothesis that the agnosticism that we take as a matter of course to be a basic precept of analytic jurisprudence is itself a tacit manifestation of a conception (or, more accurately, varying conceptions) of the good. Robin’s argument raises potential doubts about the capacity of our conventional analytic jurisprudence to maintain agnosticism about conceptions of the good. In this rendering, competing conceptions of the good are the “water” that our various jurisprudential projects are already immersed in. While we may have become acculturated to understanding and explaining the law in a way that is formally divorced from conceptions of the good (e.g. whether wise or not, our tort law is committed to a principle of corrective justice), is it nonetheless possible that we have, all the while, been swimming in it?
Of course this is not Robin’s principal point. Whether or not our existing analytic jurisprudence is capable of the moral agnosticism it formally espouses, Robin would have us draw our foundational moral conceptions (whatever their source or origin) out into the light where they could serve more prospective and ambitious (rather than merely descriptive and thereby modestly – in the service of continuity to past practice – prescriptive) ends. Nonetheless, the question of whether our analytic jurisprudential practices necessarily depend upon a conception (or conceptions) of the good seems to me to be an important one in light of Robin’s thesis. Not only does the question seem to be intimately tied to her overall picture of progressives’ commitment to neutrality that figures centrally in her argument, but if this hypothesis bears out, it strikes me that it has potential to significantly undermine potential pragmatic objections to Robin’s thesis. It is, therefore, a question that I think merits some attention.
The Possibility of Moral Agnosticism within Positivist Analytic Jurisprudence
Robin does a masterful job of making the case that progressive scholars – particularly those writing in the positivist analytic tradition – have purposefully declined to pursue the project of normative/critical/censorial jurisprudence. She makes this point not to suggest that the project of expository legal theory is inherently misguided, but instead to distinguish it from the project that would/should be pursued by our missing normative jurisprudence. But what if our analytic jurisprudence was both necessarily and tacitly committed to various conceptions of the good? If, as a matter of methodology, moral conceptions necessarily undergird positivist analytic jurisprudence, the possibility of analytic agnosticism with respect to conceptions of the good would be called into question. Moreover, if these moral suppositions were tacit rather than explicit, they would in effect be insulated from the very type of considered evaluation that Robin quite persuasively advocates. Thus, in the context of Robin’s thesis, analytic jurisprudence’s tacit dependence on moral suppositions would seem to represents a kind of worst case scenario in which our understanding of what our law is (which necessarily informs, to some degree, our understanding of our law can or should be) becomes bound by a set of implicit, unexamined moral limitations.
To explore this possibility, it may be helpful to first gain some clarity about what is meant by the term analytic jurisprudence. Most scholars operating within the branch of modern positivist analytic jurisprudence that is concerned with understanding substantive aspects of law (rather than understanding the concept of law itself) can generally be said to embrace one of two prominent methodologies: conceptual analysis (as exemplified by Jules Coleman’s explanation of tort law in The Practice of Principle) or, Brian Leiter’s approach which relies on the empirical sciences to explain legal phenomenon (as described in Naturalizing Jurisprudence). Both methodologies are concerned with arriving at the “best” explanation of legal phenomenon, although each pursues the best explanation with distinct tools. Further, both are guided by a set of scholarly norms which help elucidate which is the best among competing explanations. These scholarly norms (e.g. consilience, simplicity and/or parsimony, internal consistency, coherence) are, of course, normative, but they are not moral norms – at least not in the sense that Robin uses the term and the sense that it is used here. So the question remains: do moral norms play a constitutive role in analytic jurisprudence?
With respect to the possibility of moral neutrality in the context of conceptual analysis, Brian Leiter relates Stephen Perry’s worry that conceptual analysis is an inadequate positivist method for explaining legal phenomenon because part of the “background conceptual framework essential for jurisprudence is a view about the function of law… and we cannot specify the function of law without engaging in normative argument.” (NATURALIZING JURISPRUDENCE at 132). Perry is specifically worried about conceptual analysis in the context of understanding the concept of law itself, but it follows that, if Perry is right, conceptual analysis of substantive aspects of our positive law also depends upon a normative conception of the function of law. Leiter posits that this worry may be mitigated by appeal to our intuitions (some functions of law are more fundamental than others), but ultimately he concludes that “[p]hilosophy becomes unsatisfying…when it turns to intuition-mongering and armchair sociology about what is really fundamental to ‘our’ concepts.” (Leiter goes on to make the case that these concerns are best mitigated by a naturalized of jurisprudence (a la Quine), but an exploration of this methodology within the context of Robin’s thesis regrettably exceeds the scope of this already-too-lengthy post).
If Perry and Leiter’s are right, then conceptual analysis may be methodologically dependent upon a concept of law’s function that is itself dependent upon some conception of what law’s function ought to be. If this is indeed the case, then moral agnosticism in conceptual analysis appears impossible. In this rendering, various conceptions of the good influence the explanations of law that are produced by conceptual analysis, but those moral conceptions are subterranean, embedded within the methodology itself.
Of course, this observation, if accurate, does not necessarily undermine the validity of the explanations produced by conceptual analysis (the methodology could certainly still yield true accounts of our legal practices), but it does seem to undermine the methodology’s ostensible commitment to moral neutrality. And it is this methodological commitment to moral neutrality that purports to hold explanatory theories derived from conceptual analysis “above the fray,” so to speak, of the more work-a-day explanatory theories that depend upon overtly stated ethical premises. Because of its espoused methodological moral agnosticism, conceptual analysis as a methodology would seem to free the analyst from defending her conclusions on the basis of what the law ought to be doing. Given the difficulty of articulating and defending what the law ought to be doing (a point that I return to below) this proves to be a significant advantage and is one of the reasons, I suspect, that conceptual analysis has become such a popular method of analyzing the law.
Consider for example, two scholars: Able and Bell. Able is a proponent of conceptual analysis. He understands tort law to be committed to a principle of corrective justice. Able believes that the principle of corrective justice better explains more features of tort law than competing accounts. But in making this claim, Able is not committed to defending corrective justice against competing ethical theories (e.g. utilitarianism). Able is not making a statement about what the law should be, she is merely making a statement about what the law is.
Bell, on the other hand believes that tort law should be committed to a principle of corrective justice and (perhaps incidentally) that such a principle is also consistent with existing tort practice. Bell commits himself to an overt ethical framework: the principle of corrective justice as a just organizing principle of a tort regime. Bell must be prepared to defend his moral conclusion against competing conceptions of the good (e.g. those that prefer the ethical framework of utilitarianism). Able, in contrast, need only defend her conclusion in light of competing explanations of existing tort practice. Between the two, Able would seem to have the much easier task.
Part of the reason that Able’s task seems so much easier is that she is able to select an uncontroversial framework for evaluating her success (those scholarly norms that were mentioned earlier) and she is able to deal with a controlled set of relevant variables or data by particular cases or pieces of legislation. Bell, on the other hand, cannot point to an uncontroversial framework for evaluating his success: we lack a meta-ethical means of resolving inter-ethical conflicts between competing moral theories. Moreover, there are substantially more variables at play within Bell’s task. Any fact or factor that is relevant to justice as it relates to tort practice becomes relevant to Bell’s project. The sheer number of relevant variables makes the task of defending Bell’s proposition seem Sisyphean. In this light, the prospect of definitively defending Bell’s proposition appears illusionary. Ultimately, Bell’s proposition may be permanently consigned to a state of perpetual uncertainty, perhaps winning adherents who are persuaded by the strength of his arguments, but also lacking a satisfying answer to detractors who either embrace another ethical theory or embrace another vision of what Bell’s ethical theory calls for in this particular instance.
In other words, what Robin is asking us to do – i.e. constructing a normative jurisprudence – is hard. It is a much more difficult undertaking than the more manageable project of defending our explanations by the lights of accepted scholarly norms. Indeed, the project of articulating and defending a set of moral criteria and its application to a concrete legal problem is so sufficiently daunting that it is safe to assume that at least some of the forbearance that progressives exhibit about normative jurisprudence is attributable to the worry that normative jurisprudence cannot be done – or at least cannot be done in an analytically rigorous way (or in a way that currently “counts” as legal scholarship, as Marc Spindelman noted earlier this week).
This fear (that normative jurisprudence either cannot be done, or cannot be done well) potentially presents a kind of pragmatic objection to Robin’s thesis. How can we take up her call to arms to create a normative jurisprudence, if the project is, in some sense, fundamentally undoable? A consideration of this fear and its implications in the context of Robin’s thesis follows.
The Anxiety of Assertion
I recently ran into the difficulty of critical/censorial paralysis that Robin has so skillfully describes while commenting on another scholar’s work. As part of a book project, I was asked to comment on Caroline Mala Corbin’s piece “Expanding the Bob Jones Compromise” in which she urges that religious institutions that discriminate on the basis of sex (e.g. the Catholic Church’s exclusion of women from the priesthood) should not receive government subsidies in the form of tax exempt status, government contracts and so forth. While I was certainly sympathetic to her egalitarian objection to the categorical exclusion of women, I was cognizant, too, that other competing values were also at stake in her proposal. Here was a classic clash of competing conceptions of the good: Corbin understood justice to require a particular ordering of equality and liberty in this circumstance, while the status quo (and those that support it) embraced a different ordering of equality and liberty in the same circumstance. This presented me, as a commentating scholar, with something of a difficulty: how to evaluate Corbin’s justice claim in light of the competing justice claim in the absence of a generally accepted ethical theory that could mediate the conflict. In the absence of such a framework, I followed the lead of Brian Leiter and proceeded with the task of employing the usual suspects of ethical theory to the problem:
“There are several possible moral reasons for tolerating politically unpopular conscience-based practices – even when those practices offend our fundamental values. For example, it may be that our concept of justice requires toleration of conscience-based practices even when those practices offend public values (for the Kantians); or we may believe that toleration of value-offending practices enhances human welfare (for the utilitarians); or it may be that only by tolerating practices that offend public values can we acquire moral knowledge about those values. Any one of these reasons may provide an independent reason for pursuing a general policy of toleration of conscience-based practices including those that offend public values.”
From each of these approaches, I drew provisional and conditional conclusions (i.e., if we ascribe to a Utilitarian ethical theory, then we would want to investigate whether the practice in question enhances welfare). The conclusions were conditional because I did not argue that one ethical approach is superior to another, and in this sense my approach followed the pattern of progressive agnosticism that Robin, I think rightly, criticizes. But had I argued in favor of adopting a Kantian theory of justice in this context, others could have quite reasonably offered compelling arguments for favoring utilitarian ethical theory here. We would, at that point, be engaged in a conversation about competing conceptions of the good – and in that sense our conversation might come closer to constituting a framework for ethically criticizing (or defending) the law we have. Yet the conversation might also have become hijacked by well-worn arguments for favoring one ethical theory over another rather than focusing on an application of those theories to the particular ethical problem at hand. Thus if we are to take up Robin’s call to develop a normative jurisprudence (which I believe we should), one helpful tool to have in our normative jurisprudential arsenal would be a set of scholarly norms or methods that would help us move past this type of well-trod impasse.
Moreover, the conclusions I drew in my colloquia with Corbin were provisional, because the sheer number and complexity of factors that were relevant to evaluating these seemingly discrete competing moral claims were potentially overwhelming. I noted some of this complexity in the paper:
“Yet practices that offend fundamental public values, almost axiomatically, are likely (in our collective estimation) to cause harm, and the greater the harm that is likely to issue from offending practices, the greater the pressure that is placed on our moral reasons for tolerating those practices. A practice that involves nonconsensual physical cruelty or injury – say, for example, ritualistic animal sacrifice – places a great deal of pressure on our reasons for tolerating the conscience-based practice. In this example, our moral reasons that favor protecting liberty of conscience generally (i.e. because it is justice-enhancing, utility-enhancing, or epistemologically useful, etc.), are subject to a side constraint: the likelihood that this specific instantiation of liberty of conscience will cause significant harm (i.e. pain and injury inflicted on sentient beings). Depending on the causal nexus between the harm and the practice (as well as other variables including, e.g., the severity of the harm; the type of harm; etc.), our reasons to tolerate may give way to more compelling reasons to protect the competing value of protecting the imperiled interest (e.g. liberty, or welfare or dignity etc.) of those harmed by the practice.”
Thus second set of scholarly norms that should be placed on our normative jurisprudence wish list should allow us to wrangle the deeply complex factual messiness of lived human lives and all of its attendant variables into manageable structures that allows us to apply broad principles to those facts in a way that allows us to draw meaningful, principled conclusions.
It is these types of difficulties, I think, that has, in part, lead progressive jurisprudes to either despair of arriving at a defensible moral conclusion in these contexts (and therefore abandon the project for more descriptive ground), or, alternatively, to clean up the messiness by assuming a can opener (or some such), and thereby move the conversation further and further away from relevance to real world applications. The underlying concern, I believe, becomes one of justificatory regress: because there is a host of relevant variables and a large compliment of general (or generalizable) moral norms internal to any one ethical theory, there seem to be a near-infinite number of justifiable analytic moves that can be made that support (or contravene) almost any moral conclusion. By what principled means are we to select one conclusion over another? I described some aspects of this worry in my commentary:
”This leads us back to the place we began: the uncomfortable collision of a particular instantiation of liberty (the freedom to follow one’s conscience) and a particular instantiation of what Corbin quite correctly describes as a “fundamental public value” (sex equality). … Depending on our moral framework, we are left with some version of the question: is the harm that sex discrimination by religious group inflicts greater than the moral good that issues from permitting others to engage in conscience-based practices with which we disagree? We may attempt to discover a principled means of resolving this conflict: we might for example, articulate a reconciling principle based on the relationship between the likelihood and degree of harm caused by the practice weighed against the degree of harm caused by the elimination of the practice. … We might even interrogate categories of incommensurate harms (e.g. how does the harm of unconsciously internalizing sex-stereotypes compare with the injury that follows from the state using its financial power to indirectly influence matters of conscience?).
However, eventually in attempting to evaluate Corbin’s central moral claim … we must meet with the worry that we are dealing with turtles all the way down. If we are to evaluate the competing moral claims with both earnestness and rigor, we must admit that regardless of the reconciling principle we endorse, many variables are relevant to our analysis. In fact, so many variables are relevant to the analysis that it is difficult to insulate ourselves from the concern that our interpretation of what justice requires is (perhaps outcome-determinatively) linked to our initial intuitions regarding the relative merit of the particular instantiation of liberty at issue as compared to the relative merit of the particular instantiation of equality at issue.”
It is these three concerns – (1) the difficulty of principally selecting among ethical theories; (2) the difficulty of deriving concrete applications from general principles; and (3) the sheer number of (often shifting and contingent) variables that are relevant to applying a general moral principle to a concrete human problem – that, I think, causes progressive analytical types to worry that attempts at normative jurisprudence will result in turtles all the way down.
So what should we do? Throw up our hands and head for more descriptive ground? Robin, of course, would say certainly not. The considerable (and I think, stunningly persuasive) weight of Robin’s argument is that we cannot abandon the project of normative jurisprudence in the face of these difficulties because both the instrumental and epistemological consequences of that abandonment are devastating. Instrumentally, abandonment of moral argument cedes that ground solely to the illiberal-minded and also renders our work as legal scholars irrelevant to the causes of social justice and legal reform. But there is a grave epistemological cost to this abandonment as well: we cannot generate knowledge about the moral content of law (what the law should be) without engaging – and ultimately overcoming – these difficult ontological and methodological obstacles. The project of developing a rigorous normative jurisprudence is certainly a daunting one, but it is by no means an impossible one. I hinted at such a possibility in my discussion with Corbin:
“That is not to say that a principled evaluation of the competing moral claims at issue here is impossible. It is only to say that (a) a cursory application of the three usual suspects of moral argument in favor of an action (i.e. that it is justice-enhancing; that it is utility-enhancing; or that it is knowledge-enhancing) fails to yield an obvious result given the complexity of the variables that are relevant to the analysis … and (b) it is difficult to evaluate the gravamen of Corbin’s moral claim in the absence of such an analysis. In other words, such an analysis is possible (and necessary), but it is a complex undertaking and one that exceeds the scope of this chapter.”
I ultimately arrived at something of resigned conclusion (i.e. the task of properly evaluating this moral contest exceeds the scope of this project) – a kind of intellectual sigh that such projects cannot be undertaken, or at least undertaken well, given the content of our present intellectual toolbox. However, I was not then (as I am now), bolstered by Robin’s optimistic and strikingly persuasive case that abandonment of the project is simply not an intellectually defensible option. We must, Robin contends, find a way to construct an intellectual tool box that is equal to the task.
This, I think, returns us to the significance of the impossibility of moral agnosticism in positive analytic jurisprudence. Robin’s point that we must find a way to engage in normative jurisprudence is made all the more salient in light of the proposition that when we purport to eschew moral argumentation for safer, surer, descriptive ground, we are in fact, still engaged with moral propositions, but our engagement is unreflective. If in positive analytic jurisprudence we are assuming certain moral propositions, then to some degree the integrity of our analyses seems dependent upon these unchallenged assumptions. This is not an comfortable state of affairs by any scholarly lights.
Ultimately, this point about analytic jurisprudence and the possibility of agnosticism occupies only a subscript position relative to Robin’s general project. Nonetheless, it portends, I think, a number of interesting and far-reaching implications each of which tends to strongly reinforce Robin’s claim that to be doing sensible, constructive, knowledge-generating, jurisprudence well we need to have a methodology and vocabulary for engaging competing conceptions of the good.