John Finnis's Natural Law and Natural Rights (NLNR) was commissioned by his teacher H.L.A. Hart for inclusion in Oxford University Press's Clarendon Law Series.Footnote 1 In her biography of Hart, Nicola Lacey describes Hart's discomfort with the thirteenth, final chapter of Finnis's book, which is titled “Nature, Reason, God”:
Herbert, Finnis recalled, thought for a long time “about the fact of that chapter,” and ultimately suggested that it be placed as an Appendix to the book. Although Herbert worried that the chapter, which presents a theological basis for the book's argument, might undermine the book as a work of philosophy, he made little comment, leaving the final decision on its position to Finnis.Footnote 2
Hart need not have worried: readers of Finnis's work have themselves done the job of making that chapter a mere appendix. Reviewers typically entirely ignored that chapter's existence or begged off commenting on it;Footnote 3 in either case, they seemed content to evaluate the book without reference to the God stuff at the end. Twenty-five years after the publication of NLNR, we find Brian Leiter commenting, in an article reporting and evaluating current trends within philosophy of law, that:
Over the last quarter-century, we have seen a revival of philosophically substantial versions of natural law theory, versions that can stand apart from the theological premises that have rendered much of the historical tradition irrelevant in the post-Enlightenment world. In the work of natural law theorists like . . . Finnis . . . important aspects of the tradition of Aquinas find a place at the jurisprudential table.Footnote 4
Leiter's judgment of Finnis's work—both that it restored natural law theory to its place in contemporary debates, particularly within analytical jurisprudence and moral theory, and that it did so in a way that is detachable from natural law theory's historical context within broader theological views—is representative.Footnote 5
It is important that in reading the theses of chapter 13 as detachable from the remainder of Finnis's views, Leiter and these other commentators do not see themselves as taking up a hostile stance with respect to Finnis. I do not mean that they aim to endorse Finnis's theism or even to remain neutral about it—obviously the passage from Leiter suggests that, on his view, the fading away of Aquinas's “theological premises” is not much to be regretted. What I mean is that they see themselves describing Finnis's views in ways to which Finnis himself would not take exception. Leiter and other reviewers are not claiming that, contrary to Finnis's own purposes, he has unwittingly shown how the natural law view can be forcefully presented in a nontheistic context. They are putting forward the view that this is Finnis's own understanding of his moral/jurisprudential project, and that in this aspect of it, at least, he has been successful. If Finnis's natural law view is ultimately a failure, the failure is not that it is too tightly wrapped up with false or overly contentious theistic premises, for Finnis has managed, whatever his view's other faults, to present a natural law theory from which the theistic claims and arguments of chapter 13 are detachable.
My aim here is to think through the question of the detachability of Finnis's theistic explanation of the natural law from the remainder of his natural law view, both in NLNR and beyond. I will first rehearse quickly the central theses of Finnis's ethics and jurisprudence, placing those theses in the context of Finnis's chapter 13 theistic explanation; doing so will prepare the way for a discussion of what detachability might amount to and in what ways Finnis's theistic explanation of natural and human law is and is not detachable. This paper aims to be both a contribution to our understanding of the structure of Finnis's position and an evaluation of that position, and my central theses cannot be stated precisely and briefly. But here is an approximation. Finnis's chapter 13 theistic explanation of the natural law as actually presented can be, without too much strain, treated as largely detachable in the way that his readers have by and large supposed it to be. Indeed, I will argue that Finnis's account as actually presented in the “Nature, Reason, God” chapter amounts to no explanation of the natural law at all, theistic or otherwise, and that fact accounts in part for the ease with which Finnis's natural law view can be detached from the chapter 13 theistic explanation. But nevertheless the considerations raised in that chapter militate in favor of a much more thoroughgoing, largely nondetachable theistic account. And it is just such an account that we find Finnis affirming in developing his views after NLNR. If, then, “find[ing] a place at the jurisprudential table” requires one to present a view that is detachable from “theological premises,” Finnis should decline the invitation to dine.
I.
To be a natural-law moral philosopher is, first, to hold that the fundamental reasons for action are certain basic goods whose status as goods is in some way explained by features of human nature.Footnote 6 Whatever the sort of explanation at work here, the relationship between human nature and the primary precepts of the natural law that specify the basic goods and thus the fundamental reasons for action is one of necessity: it is impossible for one to be human and for these items not to be goods for him or her. To be a natural-law moral philosopher is, second, to hold that there are substantive principles of reason that specify how it is appropriate to respond to these goods in cases in which more than one option has some practical appeal, that is, how it is reasonable to respond in cases where there is some reason to perform one action, some reason to perform another action, yet these actions are incompatible; the status of these second-order principles of practical reason is in some way explained by the nature of the first-order goods that fulfill us as humans.
Finnis holds that there are a variety of basic values whose status as goods is self-evident to human reason, the pursuit of which makes human action intelligible.Footnote 7 At different points in the development of his views he has affirmed somewhat different lists of goods. In NLNR, he affirms a list of seven such goods: life, knowledge, aesthetic experience, practical reasonableness, friendship, play, and religion.Footnote 8 He also holds that there are principles of practical reasonableness that rule out certain ways of responding to the basic goods as contrary to reason. Again, at different points in the development of his views he affirms slightly different lists and slightly different accounts of how we are to justify which norms are included in these lists. In NLNR, he affirms a list of nine such principles; these rule out randomness in pursuit of goods, arbitrary preference of some kinds of good over others, arbitrary preference of some person over others, inappropriate detachment from one's plan of life, inappropriate commitment to one's plan of life, inefficiency in pursuit of good, intentional destruction of instances of basic goods, failure to act for the common good of one's communities, and violation of one's conscientious judgment.Footnote 9 These principles are, on Finnis's NLNR view, immediately evident to those with a correct understanding of the relevant features of the basic goods (e.g., agent-neutrality, incommensurability, inexhaustibility, and so forth).Footnote 10
Finnis's natural law view, then, consists of a set of practical propositions concerning what is worth pursuing and how it is reasonable to pursue those ends. But it also contains an account of the necessity and mode of explanation of those practical propositions as well as a view of how these practical propositions are known to be true. So while the most prominent features of his natural law view are its practical features—the substance and structure of its principles—it includes also a set of metaethical and epistemological claims.
NLNR contains an interesting natural law ethics, but it aims primarily to be a book of jurisprudence, and the point of his developing the natural law ethics as far as he does is that it gives content to his views on paradigmatic legal systems and the authority borne by such systems. Finnis takes himself to be following HartFootnote 11 in holding that in offering an account of the nature of law, the most promising methodology is to offer an account of the central, paradigmatic case of law and to understand noncentral cases as deviations from that central case.Footnote 12 The selection of the central case, on Finnis's view, is not evaluatively neutral. This is not simply the boring point that in deciding what features to include in the central case of law, one is evaluating some features as more central than others. It is the more interesting point that the criterion for centrality is itself evaluative: what fixes the central features of law is the perspective of a fully practically reasonable agent. The reason for this is that, as Hart showed us, an account of the nature of law must take up the insider's perspective, the perspective of one who uses the rule as a guide for his or her conduct.Footnote 13
But on Finnis's view, to stop here would be unstable: for once we take into account the insider's point of view, it would be arbitrary to fail to take into account that some of these insiders' points of view are more paradigmatically insiders' points of view than others. Finnis holds that the points of view of those who take the legal order to be a moral ideal and its demands prima facie authoritative are obviously more central cases of the insider's point of view than those of persons who take the law as a guide only out of habit, or thoughtlessness, or fear; and of those who take law as guide to action for moral reasons, the central case is that of those who are correct to so view it.Footnote 14 It follows, then, that since on Finnis's view the correct conception of what is reasonable to want and do is that formulated in his natural law terms, the correct perspective to describe and evaluate legal systems is from the point of view of a natural law ethics.
Given the necessity of the practical principles because of human nature and the self-evidence of these principles to practical reason, it may seem that there is nothing more to be said about the natural law view other than how it is to be applied in difficult cases. But NLNR ends not with applied ethics but with an attempt to locate natural law ethics (and so natural law jurisprudence) within a more embracing explanatory framework:
What further explanations are required? After all, the basic forms of human flourishing are obvious to anyone acquainted . . . with the range of human opportunities. And the general requirements of reasonableness . . . are, likewise, as obvious as the norms of rationality, principles of logic, and canons of explanation that are presupposed in any explanation. . . . [W]ould it not be a mistake to expect any deeper level of explanation of the practical reasonableness of community, authority, law, rights, justice, and obligation, once their explanation has been pursued from practice to self-interest, and thence to the common good which both friendship and rational impartiality require us to respect and favour?
The answer must be: No, we cannot reasonably rest here. There are further practical questions; and there are also further relevant theoretical questions about both the whole structure of norms and requirements of good that has been identified, and the whole structure of explanations already advanced.Footnote 15
As a matter of theoretical speculation on natural law, that there is an objective, naturally authoritative moral law is itself a remarkable fact calling for further explanation; as a practical matter, we seem able to ask whether there is any further point to our pursuit of our good, both individually and in common. In neither case does it seem vain to seek such further explanations; there seems to be,ab initio, neither reason to suppose the existence of a natural law and the absence of any further point to acting on it to be brute facts or simply self-explanatory matters.
The argument of chapter 13, then, proceeds to place the natural law theory of the earlier parts of the book in a more adequate explanatory context by arguing for and relying upon a theistic explanation. Finnis offers a version of the cosmological argument, holding that reason suggests that we pursue adequate explanations of the obtaining of contingent states of affairs and that no such explanation could be satisfactory until we reach an uncaused causing, something whose existence is self-sufficient—Finnis calls it “D”—a being the nature of which includes its existing, so that it could not not be.Footnote 16 Given the reasonableness of belief in a self-explanatory explainer, we can answer the theoretical and practical questions with which the “Nature, Reason, God” chapter begins.
With respect to the disputed theoretical questions, Finnis holds that the set of obtaining states of affairs whose explanation will ultimately terminate in this uncaused cause includes states of affairs whose obtaining is affirmed within his natural law view:
By “D” or “God” is . . . meant ([i]) that which explains the existence of the questioning subject; ([ii]) that which explains the existing of good states of affairs, and the opportunity of making them exist; ([iii]) that which explains our ability to recognize goods, to grasp values, and their equivalent practical principles; and ([iv]) that which explains our ability to respond to the attractiveness of those goods, to the rational appeal of the principles.Footnote 17
Further explanation can be offered for the existence of the natural law on the basis of D's creative causality being responsible for there being agents with the capacity to know the good, be moved by it, and pursue it effectively.
Practically, Finnis holds that without appealing to D—a being that, we may speculate, has a personal mode of existence—the basic goods, with their ineliminable relativity to our interests, whether individually or in common, will “seem, to any thoughtful person, to be weakened, in their attractiveness to reasonableness, by a certain relativity or subjectivity.”Footnote 18 But (and here Finnis allows that he is moving beyond what can be established or defensibly affirmed of D from natural reason) if we were to come to justifiably believe that D's creative causality proceeds from D freely and intelligently, not to serve any one of D's needs (for D is self-sufficient) but as a kind of play, then one could enjoy goods of community with D in freely cooperating in D's activity. Thus one can place one's own and others' pursuit of the basic goods in a wider context that itself has an intelligible point.
Finnis thus holds that both the theoretical and practical further questions that arise with respect to his natural law view—indeed, that would arise on any moral view—should be handled by placing that natural law view within a theistic explanatory context. It is thus far from clear in what sense Finnis's natural law is detachable from the theistic claims that he affirms and employs in that chapter. In order to make a start on dealing with this problem, we will need to get straight, or at least straighter, on what it is for a view to be detachable from some other set of views.
II.
I use the term “detachability” generically to refer to an obvious phenomenon within a theorist's work, that is, that it is frequently possible to pull apart one set of a writer's views from another set of that writer's views without doing violence to that other set of views. (We often hope for detachability when some of a writer's views are odious: we should hope, for example, that Aristotle's conception of the goods of political life will be detachable from his views on women, slaves, and vulgar craftsmen; we should also hope, with somewhat less confidence, that his theory of the human good is detachable from his assessments of the value of power and prestige.) But there are various species within this genus, and in thinking through the claim that Finnis's work has made natural law theory safe for nontheists, it will turn out to be important which senses of detachability Finnis's work exhibits.
Suppose that there is some property in which we are intensely interested—say, being water. One might have an account of water that attempts to formulate and interrelate, clearly and helpfully and truly, various theses about water that answer to our interest in it. Such a theory might include theses about the various states water exists in and how it can be transformed from one state to another, about the various purposes to which water can be put, about how to recognize a substance as water, and so forth. This theory could be of interest in at least a couple of ways: it could be of use to one in managing competently with respect to water, and it could be of use to one in assessing one's own and others' attempts to manage with respect to water.
Now, consider in relation to this sort of theory of water—we might think of it as a “user's manual for water”—the proposition that the property being water is just the property being hydrogen hydroxide. This proposition is true, even necessarily so, but even if its truth were in dispute, we could bracket the question of its truth for the sake of asking whether the adequacy of the user's manual for water is detachable from the proposition that being water is being hydrogen hydroxide. It seems that, on balance, one would argue for its detachability here. In all but the most unusual circumstances of human life, one can perfectly competently comport oneself vis-à-vis water without having the slightest idea that water is H2O; indeed, one can perfectly competently comport oneself vis-à-vis water while denying that water is H2O. One can know water's basic features, how to recognize something as water, and what uses water can be put to without knowing the chemistry of water. It would not render pointless or unintelligible one's adherence to the manual if one were to lack awareness that water is H2O. There may be special cases in which it would be helpful to know that water is hydrogen hydroxide—when taking a chemistry exam or when one is thirsty and facing a number of vials of clear liquid labeled “hydrogen hydroxide,” “muriatic acid,” and so on. But these are indeed special cases, and it is important that they are subsidiary to the interests that govern the inclusion or exclusion of propositions from the user's manual.
Call this a distinction between a user's and a theorist's account of water: the thesis that water is hydrogen hydroxide belongs to the latter rather than the former; and thus the thesis that water is hydrogen hydroxide is detachable from the user's account of water. Now, one might object that in this case detachability is too interest-relative—it depends simply on what particular interests one takes in water (or what particular interests we take in water) whether a proposition is included or excluded from the user's theory and thus whether it counts as detachable from it. If all of us became hot for chemistry, with a burning interest in chemically identifying the substances we ingest, that water is H2O might well enter into the typical user's theory, and detachability of water is H 2O from the user's manual for water would fail to obtain.Footnote 19 But regardless of the force of this objection against the user/theorist distinction with respect to water, it does not apply to Finnis's natural law theory and the distinction between a user's manual of the natural law and a theorist's knowledge of it. For natural law theory is a fundamentally a theory of what is worth taking interest in; we can state the difference between a user's and theorist's knowledge not just by appeal to contingent interests but in a distinction between what-is-to-be-pursued and further truths about what-can-be-known-about-what-is-to-be-pursued.
Finnis's natural law theory is thoroughgoingly practical: it is concerned with what is basically of value and how it is reasonable to respond to these values. What is the relationship between this system of practical propositions and the theistic proposition that facts about God ultimately explain salient features of this system of practical propositions? The relationship seems to be no tighter than that between the user's theory of water and the fact that water is H2O. We can provide an account of what is of interest in acting in terms wholly internal to the natural law theory that Finnis offers; we do not need to advert to questions about God to provide such an account. Those who act on the principles of the natural law while lacking belief in God or while lacking belief that God stands in an explanatory relationship to practical reasonableness can act fully intelligibly; arguably, not knowing that God ultimately explains certain features of the natural law threatens the intelligibility of one acting on its norms no more than not knowing that water is hydrogen hydroxide threatens the intelligibility of one's pouring oneself a glass of water on a hot day. The basic goods and the principles that dictate how to respond to them are self-evident; the attractiveness of the basic goods and the constraint posed by those principles are clear to those who have reflected carefully and accurately on the nature of those goods.
One might object to this characterization of Finnis's position that it does not sit well with his view that religion—a relationship of harmony with the divine—is itself a basic good.Footnote 20 But this is a bad objection. The affirmation that religion is good is not a theistic fact; it does not presuppose God's existence (or, more accurately, that what is good is a relationship with God does not presuppose God's existence). Hence an atheist can consistently affirm that there is no God and that religion is a basic good. That it is good to stand in such a relationship to a more-than-human source of order gives point to investigation into the existence, nature, and will of such a being, even if it turns out that there is none; one's pursuit of an investigation into whether there is such a being, so that if there is such being, one can get right with that being, does not become unintelligible upon the conclusion that there is no such being. (Compare: that knowledge is good does not presuppose the falsity of skepticism. Indeed, it partly explains why the truth of skepticism would be so unfortunate.)
So here is one sense in which one might claim that Finnis's natural law view exhibits detachability from the theism of his thirteenth chapter: one can possess the practical knowledge expressed by the natural law theory without affirming any theistic facts. Just as the success of the user's theory of water does not turn on the outcome of the debate over whether water is hydrogen hydroxide, the success of Finnis's natural law theory as a user's theory does not turn on the outcome of the debate over whether the correct set of practical principles must itself be theistically explained. Call this the user availability sense of detachability and consider the forgoing remarks my defense of the view that Finnis's NLNR view exhibits detachability in that sense.
The first sense of detachability is a practical and epistemological sense. Here is a second sense, metaphysical rather than practical/epistemological. We can say a theory T is A-free if the facts expressed by the propositions of T do not include any A-facts. Whether a theory T is A-free is likely to be a disputable matter in many cases. Here is an indisputable case. The user's theory of water is not hydrogen-hydroxide-free; since the user's theory of water includes the proposition that water is good to drink, and being water is just being hydrogen hydroxide, the user's theory of water expresses a fact about hydrogen-hydroxide: that hydrogen hydroxide is good to drink. Here is a disputable case. Are psychological theories physics-free? If it turned out that any psychological facts to which a psychological theory refers are type-identical with physical facts, for example, then that psychological theory is not physics-free.
There is a very strong case to be made that Finnis's account of natural law theory in NLNR exhibits, by Finnis's lights, even this second sort of detachability. Good appears to be an irreducible property, certainly not reducible to any set of descriptive properties, and it is clear that it is the most fundamental normative property. So it does not seem that the system of propositions expressing what is good for one and how one is to respond to these goods is thereby expressing facts about God.
Does the fundamental character of goodness in Finnis's account entail that the goodness of the basic goods must remain unexplained? Finnis does explicitly offer something like an explanation of the goodness of the basic goods, but the explanation as he runs it appeals to facts about human nature, not about God: he thinks that the natural law chestnut that “were man's nature different, so would be his duties” is true,Footnote 21 because human nature defines human possibilities, and human goods are possibilities for human fulfillment.Footnote 22 But while he thinks that there is this explanatory connection between human nature and human goods—he is, after all, a natural law theorist—he does not suggest that the fact that certain things are good for us is reducible to facts about human nature; he is not that sort of natural law theorist, one who attempts to offer an informative identification of facts about good with facts about human nature.Footnote 23
So far I have described two senses of theological detachability and have tried to exhibit the plausibility of the view that by Finnis's lights both of these are exhibited by his natural law ethics. According to this reading, on Finnis's view the natural law is in practice available to theist and nontheist alike, and the theory of natural law itself includes no theistic facts. So there is both a practical and a theoretical autonomy displayed by Finnis's natural law theory that underwrite the idea that his natural law view is detachable from the theological chapter 13.
Let me now, more briefly, note two senses of detachability in which Finnis's natural law ethics and jurisprudence are not theologically detachable. Both are boringly obvious but have to be made explicit at this point. The first boringly obvious sense in which Finnis does not take his natural law view to be theologically detachable is that Finnis believes that the actual explanation of the truth of his natural law ethics—and the truth of his natural law ethics needs explanation, on Finnis's view—includes theistic facts. The second boringly obvious sense in which Finnis's view is not detachable—perhaps marginally less boring—is that Finnis believes that necessarily the explanation of the truth of natural law ethics and jurisprudence includes theistic facts. Finnis's God is a necessary being and a necessary source of explanation; it is not merely contingently but necessarily true that the truth of natural law theory requires God as its ultimate explanation.
Does the fact that Finnis's theistic explanation of natural law ethics is not detachable in these senses threaten Leiter and others' assessment of Finnis's natural law theory as one that is safe for nontheists? We can approach this question in a more satisfactory way if we first consider a fifth sense of detachability. We owe the canonical formulation of this sort of detachability to Grotius, who writes of his own natural law theory that “What we have been saying [about natural law] would have a degree of validity even if we should concede that which cannot be conceded without the utmost wickedness, that there is no God, or that the affairs of men are of no concern to him.”Footnote 24 The “even if we should concede”—Grotius's famous “etiamsi daremus”—is a raiser of puzzles. It is puzzling in itself because on orthodox Christian views, like Grotius's and Finnis's, what is conceded is a counterpossible. It is even more puzzling when what is conceded to be false plays some sort of explanatory role, indeed a necessary one, with regard to some fact. If I said that the natural law would still have its fundamental content and binding force even if my dog were a stone, that is one thing—after all, what does my dog's material constitution have to do with the natural law? If I say that the natural law would still have its fundamental content and binding force even if there were no God—while also holding that God's existence, nature, and/or will ultimately and necessarily explains the content and binding force of the natural law—then what I have said will be more puzzling, harder to evaluate.Footnote 25 Yet if we want to know whether, by Finnis's own lights, his natural law theory is an available view within nontheistic jurisprudence, then we are required to evaluate it.
It is not clear that Finnis in chapter 13 of NLNR means to reject Grotius's etiamsi daremus. One might be tempted to run Finnis's view in NLNR together with the denial of this etiamsi daremus, but it is just not obvious that this is right. What Finnis explicitly says is that, in fact, what makes it the case that the natural law facts are as they are includes facts about God. It does not follow from this that if there were no God, then the natural law facts would not be as they are or that they would not have a nontheistic explanation. What generates the complications is the necessarily false character of the antecedent. I am again stating the obvious by saying that the truth-conditions for counterpossibles are extremely controversial. Some want to say that counterpossibles are trivially true; others want to deny this. For this to be another sort of detachability exhibited by Finnis's view, one that renders it available to nontheistic ethics and jurisprudence, we need for the counterpossible to be potentially nontrivially true or false.
I would not dare to offer here anything approaching a general account of the truth-conditions for counterpossibles. I will rest content with two points. The first is that regardless of our theoretical inclinations here, it sure looks as though nontrivial truth for counterpossibles is our stock-in-trade as philosophers. We are constantly asking what would be true if Option A were true and what would be true if Option B were true, where either Option A or Option B is necessarily false.Footnote 26
The second point is that there is a strong inclination to ascribe nontrivial truth to counterpossible explanatory claims in certain sorts of case. Here is one. Let A, B, and C be obtaining states of affairs. Suppose that A obtains necessarily, A explains B, B explains C, and insofar as A explains C, it is only through B's mediation. Suppose that God's existing explains the existence of humans, and the existence of humans explains the existence of human conventions. Suppose also that the deistic story is true: once God creates the world, then God's causal role with respect to the universe is complete and the universe runs on its own steam. In such a case, even if it is impossible for there to be anything distinct from God without God's creative act, we would be tempted to hold that the counterpossible “if there were no God but there were humans, there would likely be human conventions” is true—unlike “if there were no God but there were humans, there would not likely be human conventions,” which is false. The relevant sufficient explanatory factor is stipulated to exist even without its necessary causal preconditions; if counterpossibles can be more than trivially true, this is one sort of case in which they would be.
Here is another such case. Let A, B, and C be obtaining states of affairs. Suppose that A is a necessarily obtaining state of affairs, and A's obtaining necessarily explains B's obtaining, and C's obtaining necessarily does not explain B's obtaining. Suppose, though, that while A's obtaining necessitates B's obtaining, C's obtaining is also sufficient to necessitate B's obtaining; it fails to explain B's obtaining only because it is “screened off” by A's necessitation of C. In such a case, it may be nontrivially true that if A were, counterpossibly, to fail to obtain, B's obtaining might still be sufficient to explain C. Here is an example of this phenomenon in the case of mere counterfactuals rather than counterpossibles. That my children were fed breakfast this morning is in fact explained by my giving them breakfast. Had I not been at home this morning to give them breakfast, it does not follow that my children would not have been fed breakfast or that their being fed breakfast would be unexplained; my wife would have done it. My having given them breakfast screened off her giving them breakfast; she didn't give them breakfast just because I had given them breakfast. There seems to be no reason in principle for thinking that this explanatory pattern that can hold for counterfactuals cannot hold for counterpossibles, given the general legitimacy of appealing to counterpossibles.
Now, consider (A) God, (B) human nature, and (C) the principles of the natural law. Theistic facts are, we can suppose, fully distinct from facts of human nature, though no humans exist save for God's free creative activity; and that can be a necessary truth. Nevertheless, the principles of the natural law are immediately explained by facts about human nature: to the extent that the natural law can be explained, it is in terms of the nature of the beings bound by those norms, not immediately in terms of God's creative activity. And for one to be a being with that nature is sufficient for certain things to be worth pursuing for one; there is no possible world in which one has that nature and those things are not worth pursuing. Because this is so, it makes sense to say that, per impossibile, human nature in a world without God would be sufficient for the binding force of the principles of the natural law.
At any rate, here would be a fifth sort of detachability. “If there were no God, then it would still be true that humans are bound by these natural law principles,” if this expresses a nontrivial truth, exhibits yet another sort of detachability. On this view, even though it is true that God explains the natural law, that explanation is, from a nontheistic point of view, nonessential, for Finnis could agree that if there were no God, the natural law that binds us as humans would still have all of its key features. Put another way, the nontheist who affirms Finnis's natural law ethics would not be seizing Finnis's view and reworking it for his or her own purposes; it is, according to this reading of Finnis's view, just true that if there were no God, then the natural law would still be binding. If the nontheist affirms what is necessarily false—that God does not exist—then nevertheless he or she should continue to affirm that there is a natural law and that its features are as Finnis describes them.
Call the three sorts of detachability that are plausibly attributed to Finnis's view user availability, property distinctness, and counterpossible explanatory independence. Call the sorts of detachability that are not plausibly attributed to Finnis's view actual explanatory independence and possible explanatory independence. In a moment, I am going to argue that Finnis's views on these points have moved away from these sorts of detachability, and that the rational pressure to go in this direction is already present in the considerations raised in chapter 13. But first I want to evaluate the claim that Finnis is right in thinking that he has offered an explanation of the natural law.
III.
Finnis's account in chapter 13 does not, as advertised, explain the natural law. This is not a criticism of theism, or of his argument for theism, or of theistic explanations of morality. I say that neither the theoretical nor the practical putative further explanation in fact further explains the natural law.
The D-explanation that Finnis offers for the natural law does not explain the natural law as such but rather why human beings are capable of acting on it. The D-explanation does not aim to explain the goodness of the basic goods—it does not explain why (e.g.) knowledge is good for us—so much as to explain what makes our grasp of this goodness and our pursuit of it possible. After all, the specific states of affairs to which Finnis appeals in laying out the explanatory relationship between D and the natural law are those states of affairs concerning human capacity to know and act on the natural law which we might take to be necessary conditions on humans being genuinely bound by it—“ought implies can”--type constraints. I suppose that one might claim that, insofar as goods must be the sort of thing that is in principle possible to know and pursue, explaining how we are able to know and pursue the basic goods counts as explaining the natural law. But Finnis does not give us any reason for thinking that we must appeal to D to explain even these facts. For that humans are capable of knowing the natural law and that humans are capable of being motivated by it are not contingent facts about individual human beings; they are necessary truths about the kind human being. By Finnis's lights (e.g.), knowledge is humanly fulfilling andhumans are capable of knowing that knowledge is good and humans are capable of being motivated by the goodness of knowledge are all necessary truths about the human kind, and given his criterion for appealing to D as ultimate explainer—that we are to seek out explanations for contingently obtaining states of affairs—these states of affairs fall outside the range of D's explaining.
Why does it matter whether it is true that all of these states of affairs that one must explain if one is to explain the natural law obtain necessarily? Well, first, Finnis begins his discussion of D and the rational pressure toward further explanation by noting that what calls for explanation are contingently obtaining states of affairs. But the natural law consists of necessary truths and so seems to be outside the scope of those facts with respect to which we are rationally pressured to pursue further explanations.
In the face of this objection, Finnis should simply note that it is an error or an omission not to hold that necessary states of affairs may well require further explanation. Even if some state of affairs necessarily obtains, it does not follow that this state of affairs requires no explanation. Some necessarily obtaining states of affairs may have their necessity through some other state of affairs. And even those that have their necessity of themselves nevertheless may need to be explained by analyzing the state of affairs so that we can see why it must obtain. All of that leaves open the possibility that D could have a role in explaining the necessary states of affairs whose obtaining constitutes the existence of the natural law.
One might note that while the natural law is not contingent, the existence of beings that can be bound by the natural law is contingent, and so that can be explained by D. But to explain the existence of beings to whom the natural law applies is not to explain the natural law. The natural law theses that Finnis affirms and that are to be explained concern necessary relations between universals, not particulars; explaining the contingent existence of those particulars does not explain the necessary relations between universals that constitute the natural law. That is what we need and what the “Nature, Reason, God” chapter does not provide. We lack any sense from chapter 13 of how an appeal to D can explain why certain connections necessarily hold among certain universals.
It is also not clear why we should think successful Finnis's attempt to answer further practical questions about the natural law, about the point of acting on its categorical norms given that their status is indexed to mere human needs and interests. There is more than one way to put the problem. Look: what we learn from the theory of natural law given in the first several chapters of NLNR is that in figuring out what is worth doing, the answers that are available to us bottom out in the basic goods. For every craft or line of inquiry and likewise every action and decision have point and intelligible purpose only through their bearing on one or another of these basic goods, and none of these basic goods is reducible to an aspect of or an instrument for the realization of some other good. Practical questions, one might plausibly suppose, ultimately have to be answered by appeal to the basic goods. If there is some further point beyond them necessary to underwrite their goodness, then the basic goods are not so basic, after all.
Put it another way. When Finnis describes the relationship with D in terms of which our pursuit and promotion of the basic goods gain a fuller level of intelligibility—practical, not theoretical—it is in terms of explaining how D's activity in creating, conserving, and directing the world is a kind of play, and we are cooperating with D in this intrinsically good activity, thus being in community with D. These are basic goods, and if there is any trouble about those basic goods being relative to us in some sense, the trouble should reappear here in the explanation.
We are left with a bit of a dilemma. It does seem that Finnis is giving voice to a perfectly respectable worry about the basic goods and their status in guiding our conduct when he raises the question about the placement of our pursuit of those goods in a wider context. And it does seem that this can be a practical concern, meaning that the extent of our dedication to the goods and the constraints of practical reasonableness that we are to honor in pursuit of them may well turn on how these worries are answered. But on the other hand, it is not clear whether any such further answer is really compatible with the structure of correct practical thinking as Finnis describes it.
So far I have made two claims regarding the “Nature, Reason, God” chapter of NLNR. One of these claims is interpretive: the theistic theses of chapter 13 are best read within the context of NLNR as exhibiting a large measure of detachability from the natural law ethics and jurisprudence defended in that work. The other is evaluative: the theistic explanations of the natural law offered in chapter 13 are unsuccessful. For the remainder of this paper, my aim is to make two further arguments, one interpretive, one evaluative. The interpretive point is that in Finnis's post-NLNR work, his accounts of the relationship between natural law and theism have in fact ceased to exhibit the sorts of detachability that his NLNR view exhibits. The evaluative point is that, given the considerations raised in NLNR in favor of theistic explanations, he is right to offer theistic explanations that do not exhibit detachability.Footnote 27
IV.
A theory exhibits property distinctness detachability with respect to some set of properties A if and only if none of the facts referred to by T is an A-fact. When a theory is detachable in this way, one can treat it like a “module,” asking whether that theory can be located within a world in which A-facts obtain and whether that theory can be located within a world in which A-facts fail to obtain. If Finnis's natural law theory exhibits property distinctness detachability with respect to theistic properties, then none of the properties the instantiation of which Finnis's view affirms are theistic properties. It can hence be in principle an open question whether Finnis's view requires a broader theistic context to be explained or whether it is ultimately compatible with nontheism.
The property around which Finnis's view is organized is that of good. And as I note above, I do not see that anything in NLNR explicitly denies or obviously commits him to denying property distinctness detachability with respect to theistic properties. But I want to say two things here. First, it does not take long for Finnis to make explicit claims about the nature of the good that mark it as a theistic property and thus to commit himself to the rejection of property distinctness detachability. Second, this making explicit is true to the trajectory of the chapter 13 argument—the rejection of property distinctness, or something like this, is what the argument of chapter 13 should have drawn him to. The mystery is not that he does not hold a view in NLNR that he holds in work shortly after NLNR; the mystery is why, since the thrust of his argument carries him to the denial of property distinctness, he does not explicitly affirm that view in NLNR.
The first point: the good with which NLNR is concerned is the good of practical reason—what makes a state of affairs desirable, an action worth performing, or an option worth rejecting. It is what is good for us, in light of our determinate human natures. But while it is the humanly good that properly orients practical thinking, to be humanly good turns out to be, on Finnis's view, a theistic property:
If one believes that unqualified goodness—goodness itself—is found in God, one will regard him as the source of the goodness of all of the basic goods. In this perspective, every human fulfillment is a participation in the divine goodness, and every human act is for the sake of divine goodness insofar as one can participate in it by the benefit for which one choose to do the act.Footnote 28
I read this passage as holding that a certain informative property identification is true: to be humanly good is (in part) to be a participation in the divine goodness. (It has to be in part because not every participation in the divine goodness will be part of the human good—there may be other conditions as well, for example, that it be within the range of human possibility.) I read it this way because that is the facially plain reading, and other readings that one might suggest for it that would preserve property distinctness are either very implausible or inconsistent with core features of Finnis's natural law view.
It is the facially plain reading. Finnis does not say simply that everything that is in fact good is a participation in the divine goodness. He says that the divine goodness is the source of the goodness of the basic goods, and that these goods are participations in God's goodness. Participation here means something like “caused likeness”: A participates in B's C-ness if A's C-ness is like B's C-ness and A exhibits its C-ness because of B's C-ness.Footnote 29
One can reject the argument that Finnis's post-NLNR view commits him to the denial of property distinctness detachability by rejecting this reading. But one will need to offer another in its stead. One might opt for a view on which facts about goods are not theistic facts but are simply caused by theistic facts. One might account for the “source” idea in this way. But that would be strange. On this view, God's being good simply causes (efficiently?) the being good of some state of affairs—not that state of affairs obtaining, but its being good. This is simply strange, and I would not know what to make of it.
In order to avoid the claim that being good is a theistic property, one might instead opt for something like a subsumption view. Being like God is the only basic good; things like knowledge, friendship, and so on are good only insofar as they are subsumed under the description being like God. The most obvious problem with this reading is that it is incompatible with the basicality of the basic goods. Finnis is insistent that knowledge, friendship, and so on are irreducible categories of goods, diverse, not to be subsumed under a more fundamental good.Footnote 30
My argument so far is that Finnis's views in NLNR do not clearly commit him to the denial of property distinctness detachability, though within a few years of NLNR he is affirming a view that is incompatible with that form of detachability. But I want to make one further claim, which is that given the chapter 13 argument of NLNR, he is right to make this move. The thrust of Finnis's “Nature, Reason, God” argument calls for him to affirm such a view.
Why? The pressure comes from two directions. It comes, on one side, from the rational interest in more and more complete explanations of obtaining states of affairs. This interest is perhaps most obviously pressing in the case of contingent states of affairs but it extends even to states of affairs whose obtaining is judged to be necessary but which nevertheless do not appear to be self-explanatory. And the obtaining of states of affairs like “knowledge is good” seems to be of this sort. On Finnis's view, that knowledge is good is self-evident. But it is not at all clear that it is self-explanatory—why is it the case that it is good? Why are knowledge and friendship (etc.)—conditions that are very much different—all worth seeking and having? One powerfully unifying form of explanation is to posit that D is not only the self-explanatory explainer of what exists but the self-explanatory explainer of what is good.Footnote 31 D is being itself, D is goodness itself; as all other beings' being is explained by reference to D's being, so all other goods' goodness is explained by reference to D's goodness.
The pressure on the other side comes from what we want to say about D in order for D to be the final explainer. The typical way of putting this is in terms of D's aseity—D is independent of and not dependent on anything distinct from it. So D cannot just be good; it has to be goodness itself. So it is the ultimate standard of what is good, and to be good is just to bear the right relationship—Finnis suggests the notion of participation—to it.
This is a genuine explanation of the natural law, unlike the nonexplanation that is present in the actual chapter 13 argument. While Finnis holds that D can explain our natural urge to pursue the basic goods, our capacity to know them, our ability to realize them effectively, what he conspicuously does not say is that D can explain the goodness of the basic goods. That would be an explanation of the natural law. The appeal to human goods as likenesses of the divine goodness satisfies this role: the link between certain states of affairs related to human nature and their being worth pursuing is that to be worth pursuing is just to be a participation in the divine goodness. This is explanation through unification.
There may be some sympathy for the view that even if Finnis's moral theory is not theism-free, we can still take the liberty of carving out the nontheistic theoretical portions while leaving the rest of the view intact. If this point is equivalent to the point that one can know what is good and how it is reasonable to pursue it without knowing that there is a God and that to be good is to participate in God's goodness, then it is true that nothing that I have said so far about Finnis's view calls that into question at all. But this is just to note the distinction between the user availability and the property distinctness detachabilities—an important distinction but not one that I am concerned with at this point in the argument.
One might, on the other hand, have something else in mind—that the identification of goodness as a theistic property is a theoretical “flourish,” as it were, and not part of the real substance of Finnis's position. I do not have a sufficient theory of “flourishes” to reply well to this suggestion, so I will be content with the following. Consider the metaethical theory proposed by Robert Adams in his Finite and Infinite Goods.Footnote 32 Adams is, with respect to moral norms, a divine command theorist.Footnote 33 But on the nature of the good, he holds that all goods are divine likenesses—to be good is just (in a certain way) to resemble God.Footnote 34 He affirms this view while also holding that atheists can well know what is good and what is not.Footnote 35 If anyone claimed that the proposition that to be good is just (in a certain way) to resemble God is just a flourish in Adams's view, not part of its substance, he or she would be comically wrong, as comically wrong as one would be to say that the idea that humans are ends-in-themselves is just a flourish in Kant's view or to say that the idea that rational activity is constitutive of human happiness is just a flourish in Aristotle's view. But Finnis's considered view that this is what goodness is is relevantly similar to these other cases.
V.
If property distinctness detachability cannot be ascribed to Finnis's post-NLNR view, counterpossible explanatory independence cannot be attributed to that view, either.
Even if x's being a human being is sufficient, in any possible world, for knowledge's being good for x, that is true only because God exists in every possible world. X's being human entails knowledge's being good for x, in the sense that in every world in which x is human, knowledge is good for x. But the explanation always runs through God's existence and nature: that knowledge is a participation in the divine goodness. So the explanation of the natural law is not fully mediated by human nature in the way that would support the interpretation that Grotius's thesis is Finnis's thesis as well.
Very well: but one might say that the counterpossible could be true even if, necessarily, the goodness of the basic goods is immediately explained by facts about God. For it could be true that if (counterpossibly) the goodness of the basic goods were not immediately explained by facts about God, some other facts would “step up” to explain their goodness. But even granting that there may be cases in which counterpossibles are true in virtue of this structure of dependence, the strategy must fail in this case. For this is not a case in which someone who affirms Finnis's view can coherently entertain the idea of some other facts as explanations-in-waiting if there were no theistic facts to do the explaining. For the facts about the good to be explained just are, are identical with, certain theistic facts. The immediacy of the explanation of normative facts by theistic facts is not causal immediacy but the immediacy of identity. Once one affirms Finnis's particular view about the way that normative facts are related to theistic facts, one must hold not only that normative facts are actually explained by theistic facts and not only that they are necessarily explained by theistic facts, but that any nontrivially true claim about how they are explained in counterpossible circumstances includes appeal to theistic facts.
VI.
Finnis's post-NLNR work is best read, I claim, as rejecting the property distinctness and counterfactual explanatory independence forms of detachability from theism. If Finnis's view is to remain detachable from the theistic framework in which it finds a place, then it can only be through its availability to users, even apart from their understanding the connections between the practical truths of the natural law and the truths of theism affirmed in the “Nature, Reason, God” chapter. But I want to call into question the presence and desirability of full user availability detachability in Finnis's post-NLNR work.
I do not think that Finnis has been interested at any point in his career in denying that one can come to know the truths of the natural law apart from having any knowledge of the truths of theism, nor do I think that he has ever been interested in denying that one can know the main lines of the natural law even while denying the truths of theism. It thus seems that user availability detachability should continue to be attributed to Finnis's work. But matters are not so clear. Begin with two points. First, I endorse above Finnis's claim that there seem to be further, practically relevant points about the character of the natural law—“the basic values will seem, to any thoughtful person, weakened in their attractiveness to reasonableness”Footnote 36 by their relativity to us—even though I call into question the explanation provided that is meant to allay those worries. So there still remains a difficulty on the table, even if it is not clear how it is possible to approach the difficulty within the constraints set by Finnis's natural law ethics.
The second point is that we can distinguish two senses of user availability. Let me illustrate this by noting how the question of user availability detachability of this sort arises in the work of Kant. I hope that no one is going to dispute the claim that the moral law is, for Kant, available generally. One does not need to carry out philosophical investigations into the ultimate source of the moral law in order to know that one is under such a law, what the content of that law is, and that it is binding unconditionally. One might say, on the basis of these points, that whatever else Kant wants to say about God and God's relationship to the moral law, it is going to have to be detachable from Kant's ethics in at least the user availability sense. But that would be a bit misleading. For when Kant describes the role of belief in God and immortality within his ethical view, it is not merely as further theoretical explanation. These are, rather, practical postulates, and Kant thinks that, practically speaking, these must be postulated.Footnote 37
What is the force of this “must”? What Kant has in mind is not that it is actually true that whoever understands and wills to act on the moral law has these beliefs, so that it could be shown that one does not grasp or follow the moral law simply by showing that he or she does not believe in the theist's God. What he has in mind is that the practical life of a nonbeliever exhibits a sort of rational instability.Footnote 38 The agent who acts on the moral law presupposes that the goods of happiness and virtue are corealizable, and for this possibility to be live requires God's existence and action. It is thus possible for one to grasp and follow the moral law, but without belief in God there will be something unintelligible about this agent's practical life—he or she will be set, come what may, on the realization of two distinct ends that to all appearances are uncoordinated and even at odds with each other. These propositions must be postulated in order to preserve the intelligibility of the agent who acts on the moral law.
Kant attempts to answer these further practical questions about the moral law not by positing ends beyond happiness and virtue but by bringing to explicitness our presuppositions in acting for these ends and noting how these presuppositions could be true—that is, through divine existence and action. We can take this as an illustration of how Finnis might be able to answer the further practical questions about the natural law raised at the beginning of chapter 13. Here is a way of reframing Finnis's worries, modeled on Kant's distinct concerns. The natural law, as Finnis describes it in the early chapters of NLNR, binds us regardless of our particular desires, aims, goals, perspectives. Its indications of what is worth pursuing are categorical, as are the principles of practical reasonableness that constrain our action for the good; indeed, some of these principles of practical reasonableness are absolute, ruling out certain courses of action as unreasonableness regardless of the circumstances. To be committed to the natural law is to be committed to acting on such norms, come what may. But this full commitment to the demands of the natural law seems at odds with the natural law's being something relative to us—serving our goods within the limited horizon of our judging and acting. So we have an unintelligibility: the natural law demands full allegiance, come what may; but the recognition of the limitedness of our perspective calls for a weakening of that allegiance. To be fully committed to the natural law seems to be less than fully intelligible.
Finnis does not put the point this way in NLNR. But once the point is put this way, it is clear that it would be deeply misleading to say that, by Finnis's lights, whether one is a theist or not is irrelevant to one's use of the natural law as a guide to one's conduct. For once he has raised this set of issues and decided that only a theistic solution will do, he has committed himself to the position that adherence to the natural law is rationally unstable in the absence of a certain sort of theistic stance.
One way to fill this gap is by appeal to a being who fills the role filled in other ethical viewpoints by an “ideal observer” or an “ideal prescriber”—with the difference that D is not a construct of our own limited practical intelligences but a real being who “favour[s] the well-being of everyman, for no other reason than its own goodness”; we may view our fellow humans (and ourselves) as “persons whose good is favoured also by one whose own goodness is unrestricted and whose love is in no way blind but rather is given knowing fully the true worth and all-explaining point of everything, of the existence of every person, and of the history of every community.”Footnote 39 We should appeal to D, to God, on this alternative view, not by way of understanding further human goods implicating D (which goods will themselves be subject to the same practical questions) but by way of underwriting the authority of the basic goods from a perspective more objective and universal than our own.
There seems to be a trend in Finnis's work toward making explicit the role of theistic belief and commitment with respect to adherence to the natural law. In the same paper in which he identifies the human goods with participations in the divine goodness, he argues for the necessity of a religious commitment to organize and sustain commitment toward the basic goods.Footnote 40 More recently, he has argued that there are important practical consequences to secularism, where secularism is that habit of belief whereby one denies belief in God, or in God's active involvement in the world, or in God's fidelity to God's promises.Footnote 41 Here he is very explicit that the natural law view that he has been defending for the last few decades cannot be detachable in the user availability sense, if by this we mean that one can fully intelligibly commit oneself to adherence to the natural law in the absence of theistic belief. Of course he does not want to deny that one can follow the natural law while rejecting theism. But what that shows is that “people are often less consistent, and better, than their theories.”Footnote 42
VII.
None of the five forms of detachability from theism is present in Finnis's mature natural law theorizing, and the impetus for rejecting all five of these forms of detachability is present even in NLNR. It is a mistake to treat the “Nature, Reason, God” argument as a mere appendix to that book, even if the making of that mistake contributes to its receiving the invitation to the “jurisprudential table” that it so well merits.
