Click to return home

The Logic and Antilogic of Secret Rights

D.H. Kaye(*)

1988 by David H. Kaye; the Minnesota Law Review Foundation.
This article appeared in the Minnesota Law Review, vol. 72, February 1988, pp. 603 - 625.


     Students of jurisprudence are a contentious brood. Since the time of Socrates, they have been challenging one another's theories of justice. Faced with interminable philosophizing, one is tempted to dismiss the entire jurisprudential enterprise as a snare for the clerisy and a delusion for the masses. If centuries of theorizing have not produced definitive conclusions, then it may be time to transcend this mode of reasoning--to leap into hyperspace, as it were. Professor Girardeau Spann, who recently described the hyperspatial perspective on constitutional law, (1) now provides a detailed and multifaceted argument in support of this move in his essay, Secret Rights. (2) In it, he boldly asserts that all existing theories of individual rights are necessarily unsatisfactory (3) and promises a new paradigm that 'reconceives analytical use of language and de-emphasizes analytical reliance on logic.' (4)

     One of the most intriguing features of Secret Rights is its reliance on logic to attack logical argument itself. I believe that this trendy deconstructionist (5) tactic fails. To support my skepticism about Secret Rights, I shall describe selected aspects of the analysis and indicate its weaknesses. Part I examines [-----604-----] Spann's claim that all extant theories of rights are logically incoherent and shows that one can avoid the problems identified by Spann without abandoning logic. Part II analyzes Spann's critique of deductive logic as a component of theories about rights and argues that logic has its uses. Part III discusses Spann's proposed nonlogical paradigm for handling rights and questions whether this approach is preferable to a logic-based analysis. I conclude that while there is far more to political and legal argument than deductive logic, logic should continue to play at least a supporting role in theories about rights.

I. ARE RIGHTS INCOHERENT?

     As indicated above, Spann attacks nothing less than all existing theories of individual rights. He comes closest to defining individual rights when he refers to them as 'individual interests . . . beyond the reach of societal abrogation, even when abrogation would serve the collective good.' (6) In other words, rights trump other kinds of interests. (7) In a society that purports to be governed by majority rule, rights are therefore antimajoritarian. (8) Spann believes that such antimajoritarian rights are incoherent, (9) presumably because they entail contradiction or paradox.

A. THE ANTIMAJORITARIAN DILEMMA

     In Spann's view the antimajoritarian dilemma arises because recognition of a right by the judicial system, as an institution created by the majority, makes the putative right 'simply an interest endorsed by the majority.' (10) As such, 'classification as a right becomes superfluous.' (11) To escape this dilemma, one could adopt the familiar theory that the majority does not prefer the specific result of the case but prefers instead a procedure that rejects an outcome-specific preference whenever a right is at stake. (12) Spann maintains that our majoritarian system [-----605-----] bars this route. He writes that 'as long as the majority continues to favor the procedure that it has authorized, any outcome produced by that procedure necessarily corresponds to the preference of the majority.' (13)

     The argument that an apparently antimajoritarian result is essentially majoritarian rests on an undifferentiated definition of antimajoritarian. (14) If we admit, however, that two types of majority preferences exist, the dilemma disappears. For instance, we may call the majority view on whether at a particular time and place the government should ban a given book a Type I preference. We may also call the majority's unwillingness to overthrow an institutional framework, which precludes the banning of the same book at the same time and place, a Type II preference. Some rights theorists would suggest that when the Type II preference frustrates the Type I preference, the individual has a meaningful right. (15) Thus, Spann's complaint that a right lacks antimajoritarian content (16) because it reflects a Type II rather than a Type I majoritarian preference hardly establishes that the right 'lacks logical coherence.' (17) [-----606-----]

     We may easily escape the antimajoritarian dilemma because it trades on ambiguity in the term majoritarian. At the crux of Spann's argument is his insistence that, as a matter of definition, 'the majority prefers to do whatever it does,' including doing that which it does not prefer to do. (18) As with any argument that smacks of troublesome self-reference, (19) we can solve the dilemma by refining the terminology to prevent the apparent self-reference. (20) Type I and Type II terminology accomplishes precisely that result. (21) 'Adherence to a rational analytical paradigm' (22) does not cause Spann's dilemma; an equivocal use of the term majoritarian does. Consequently, the [-----607-----] bold claim that one must step 'outside of a rational paradigm' (23) to believe in individual rights is unwarranted.

B. THE LIBERAL DILEMMA

     Spann presents a class of liberal theories of rights as alternative formulations of the antimajoritarian theory of rights. (24) These liberal theories, according to Spann, build on a distinction between policies and principles but 'succeed only in producing another analytical dilemma.' (25) In a slight variation on the argument just considered, Spann maintains that a political procedure (such as a court) cannot apply countermajoritarian principles because all political procedures (including courts) by their very nature are majoritarian.

     Again, we may respond by distinguishing among political organs and maintaining that courts are expected to apply principles rather than policies in adjudicating claims about rights. In the alternative we might support one political organ, such as a parliament, but contend that it is capable of recognizing when a right (an argument based on principle) should prevail over a more politically expedient result (a policy argument).

     Now we can resolve the putative dilemma by referring to an argument of policy as a Type I claim and an argument of principle as a Type II claim. A court or a parliament can recognize Type II claims and give them priority over Type I claims, not because there is a Type I justification for that result in the individual instance, but because the court or parliament is committed to respecting rights.

     Once we recognize this distinction, much of Spann's argument reduces to the concern that in many instances majoritarian political influences (26) may improperly affect decisions about rights. On this point Spann's reading of the history of the protection of individual rights is defensible. Throughout this less than happy history, rights arguably have been more often honored in the breach. It is quite a jump, however, from an unpardonable gap between theory and historical practice to a conclusion that the principle-policy distinction is logically deficient.

     To bridge this gap Spann insists that even a decision maker who succeeds in deciding on a basis other than momentary majority [-----608-----] preference nevertheless decides on the basis of personal preference. (27) Such preferences, Spann contends, are political and 'potentially biased considerations of expediency rather than actual neutral principles.' (28) Hence, the liberal theorist is condemned to a world of preferences that are inherently unprincipled because they cannot be divorced from political interpretation. (29)

     Once again Spann's argument smacks of definitional fiat. Spann treats all personal preferences of decision makers as political, thereby ignoring the possible distinction between a personal preference to act expediently and a personal preference to abide by principles. Spann comes close to recognizing as much when he considers the counterargument:

     [e]ven if all decision making is political in some definitional sense, there is nevertheless an important difference between the everyday political expediency used by Congress in deciding whether to grant a special interest tax exemption, and the more austere type of decision making engaged in by the Supreme Court when it makes individual rights determinations. (30)

     Because he phrases this riposte in terms of judicial protection of individual rights, Spann says that this counterargument fails unless one can show that judicial decision making protects individual liberty more successfully than legislative activity. (31) As we have seen, however, the critical distinction is not so much which political organ decides a case as it is whether that organ relies on Type I or Type II arguments. If a judge follows the Type II preference, abiding by principles rather than determining what is momentarily expedient, we may easily say that the judge is respecting a right-based claim. Of course, in practice, a judge may dress a Type I decision in the garb of a Type II decision. Recognizing this possibility, however, does not advance the analysis beyond the writings of the early realists. The oft-debated question of whether requiring Type II justifications actually constrains decision making and advances the cause of individual liberties is still unanswered. (32) [-----609-----]

     In sum, Spann has simply proven that if all decisions are labeled political, a puzzle arises over whether decisions to adhere to principle are actually based on policy. Because we can easily avoid the ambiguity, and hence the semblance of self-reference, the dilemma is unconvincing. As a practical matter, the liberal rhetoric of rights may be merely emotive, but this argument against liberalism reveals no logical dilemma.

C. THE NEGATIVE RIGHTS DILEMMA

     Spann defines negative rights theories as those that 'seek to promote individual liberty by defining a private sphere of autonomy and self- determination.' (33) He finds this contradictory because ' whenever the government protects one individual's autonomy from intrusion by another, the government is interfering with the autonomy of the second individual.' (34)

     This complaint confuses a private sphere of autonomy with unbounded autonomy. Few people would claim that individuals have a right to do whatever they want. Total autonomy on the part of selfish individuals would produce irreconcilable conflicts. Thus, any plausible theory must distinguish the private, protected sphere from the public, unprotected sphere. The only analytical dilemma related to negative rights theories (35) comes from speaking of autonomy without differentiating between the private and public spheres.

     Nonetheless, Spann rejects the possibility of distinguishing between permissible and impermissible failures of government to preserve autonomy. (36) Essentially repeating the arguments already considered in the previous two sections, Spann reasons that a political institution would have to interpret any principle defining a private sphere of autonomy that is immune from governmental interference. Having already concluded that all decisions of political organs are political rather than principled, Spann sees the same dilemma in negative rights theories that he advanced against liberal theories.

     Spann's argument is equally weak in this context. Indeed, Spann seems to attribute to negative rights theorists the view [-----610-----] that every interference with total autonomy is equally serious. Spann suggests that negative rights theories cannot differentiate between a claim that one has the right to burn children alive for sadistic pleasure and the claim that one has the right to read Joyce's Ulysses in one's own home.

     Yet negative rights theories can and do make such distinctions without becoming internally inconsistent. Consider the classic negative rights principle that denies to the government any legitimate power to prevent an individual from acting as long as the action does not engender harm to others. Adherence to this principle can produce countermajoritarian results in specific instances. It also establishes a private sphere of autonomy--I can do what I want as long as my actions do not harm any one else--which interferes with no one else's parallel sphere. Everyone else, likewise, may do what he or she wants as long as it does not harm another person.

     Of course the harm-to-others principle has its shortcomings. Difficulties become apparent as one moves from the canonical case of physical harms (such as burning children) to merely offensive behaviors (such as public fornication), to risks of harms, and so on. Contemporary political philosophers have striven mightily to clarify or supplement the harm-to-others principle in order that it might produce reasonable results. (37) Their efforts may not have been crowned with complete success, (38) but there is nothing logically wrong with their attempts or the underlying principle.

     Inasmuch as Spann fails to mention the modern harm-to-others literature, which purports to delineate a defensible scope of inviolable interests (and could thus answer his queries about speed limits, heroin laws, and Sunday closing laws), (39) his claim that all negative rights theories have failed seems premature or rash. Still less substantiated is his sweeping claim that 'any effort to rehabilitate a theory of negative rights is doomed to failure because no analytically acceptable distinction between public and private spheres of interest exists.' (40) The sole reason for this nonexistence thesis is that 'the private sphere simply has no meaning independent of that given it by the public [-----611-----] sphere.' (41) This may be solid, deconstructionist rhetoric, but it ignores the analytically acceptable distinction, advanced in the previous two sections, between Type I and Type II social preferences. Institutions that enforce Type II preferences may keep the spheres separate.

D. THE COMMUNITY RIGHTS DILEMMA

     Societal rights theories give greater priority to community well-being than to individual interests. Because only individuals can interpret and apply community rights, Spann finds that 'some version of each dilemma that arose in the context of individual rights . . . reappear[s] in the context of communitarian rights.' (42) In particular, he contends that a variant of the antimajoritarian dilemma arises because 'no communitarian right could secure recognition unless an individual were willing to honor it.' (43) He maintains that the public-private dilemma appears because 'the sphere of collective interests comprising a communitarian right cannot be defined in any way that an individual decision maker does not ultimately determine.' (44)

     I have said perhaps more than necessary to indicate why I find that these arguments fall short of demonstrating 'a distinction between the individual and society . . . destined to lack analytical coherence.' (45) Spann's treatment of terms like majoritarian, political, and public is redolent of the gravy served at Woolworth's--it covers everything. While no rights theory has universal support, the claim that all of them are logically incoherent rests on putative errors of self-reference or circularity (46) that one can avoid within the context of customary rational analysis. [-----612-----]

II. IS RATIONAL ANALYSIS WRONG?

     Spann concludes that, because his treatment of rights theories leads to counterintuitive results, the culprit is rational analysis. (47) He insists that 'the rules of the governing paradigm' permit his analysis. (48)

     If each dilemma arises from the use of an ambiguous term that suffocates crucial distinctions, Spann errs in presuming that he has provided impeccable logical arguments. (49) Rather than unearth fundamental flaws in logical theory, he steps into pitfalls that logicians normally avoid. (50)

     It is unclear, however, whether Spann actually is claiming that he properly proved that a logical flaw inheres in rights theories. At one point he characterizes his argument about rights quite differently. He states: '[I]t is always possible to generate counterintuitive conclusions without violating any of the rules of rational analysis . . . . [T]he only thing wrong with what I have done is that it leads to conclusions that are intuitively [-----613-----] unacceptable.' (51)

     Because formal logic permits us to be certain of the validity of arguments, including counterintuitive ones, this modification of what appeared to be the original thesis is hardly an indictment of logic or a justification for some other kind of reasoning. Spann defines rational analysis as 'the application of logical rules to valid initial premises.' (52) I conclude that Spann's complaint lies not with logic itself but with premises of a logical argument that are subtly false and, when unrecognized, produce false conclusions. As he states, 'Our intuitions cannot properly be relied upon to help distinguish between the manipulated conclusions and the arguably genuine ones.' (53) Thus, his modified thesis has two parts: garbage in produces garbage out, and the incoming garbage is nearly impossible to spot in arguments about rights. Because we cannot smell the garbage with language and logic, Spann suggests that we use some other mode of analysis.

     This modified thesis requires elaboration and defense. Are we unable, even in theory, to root out false premises? Are coherence theories that ask us to reconcile our intuitions about specific results with the general principles to which we purport to subscribe theoretically inadequate? (54) Or is the argument more pragmatic: because people's opinions differ on these matters, no practical way exists to measure the better argument concerning rights? Will the alternative mode of discourse that Spann advocates produce more consensus or more correct results? And, regardless of one's responses to these questions, would not logic retain a role in revealing the consequences of accepting a statement as true or rejecting it as false?

     By and large Spann does not answer these questions. Instead, he launches a more elaborate assault on deductive logic as a component in the analysis of rights. In the course of this discussion he rejects coherence theories as illicit 'peeking ahead,' (55) stating that a new paradigm would work better. (56) [-----614-----]

     Spann limits his criticism to deductive, first order logic. (57) He initially complains that this logic not work with premises whose truth values can never be determined (58) and that 'demonstrating the metaphysical truth of a premise describing a social phenomenon would be difficult.' (59) While noticeably not criticizing deductive logic itself, Spann realizes that the rules of deductive logic do not assign truth values to premises.

     If taken at his word, Spann's skepticism concerning the value of deductive argument about rights would be unwarranted. I interpret his assertion about 'difficulty' to be tantamount to the stronger claim that establishing the truth of premises describing social phenomena is impossible because Spann's most skeptical conclusions presuppose impossibility rather than difficulty. Furthermore, I doubt that Spann intends to characterize descriptive propositions about social phenomena as indeterminate. Consider a sentence S, which reads, 'In the United States, exactly 4281 people live in traditional families.' S describes a social phenomenon, and S is false. Admittedly, the words people, living, and traditional families contain some ambiguity, but I am confident that I could clarify such terminology sufficiently that most would agree that S has an exceedingly high probability of falsity. Now, to make inferences involving S as a premise, I need an inductive logic. (60) [-----615-----] While Spann contends, without any basis or apparent reason, that inductive logic is less useful than deductive logic for establishing the truth of propositions like S about society, (61) his real target must be normative propositions like the claim that all people are entitled to equal protection of the laws. (62) Such propositions are not empirical, making it unclear what it means to say that such a proposition has a given probability of being true.

     As far as I can tell, Spann only questions the use of deductive logic with normative, as opposed to descriptive, propositions. If so, in expressing his concern for metaphysical truth he merely restates the classic problem of objectivity in ethics. Even assuming, as Spann seems to, that no objective procedure for discerning the truthfulness of ethical premises exists, it does not follow that deductive logic is useless. Spann recognizes, after all, that deduction permits us to establish the truth of conditional propositions. For example, let R stand for the right not to incriminate oneself, let x stand for any person, and let n stand for the person named Oliver North. The sentence C, which may be written as (x)Rx --> Rn, or less formally as 'If all people have the right not to incriminate themselves, then Oliver North has a right not to incriminate himself,' is true. As Spann explains, 'As long as the analysis complies with the logical rules, any conclusion generated by the analysis should be acceptable at least to those who agree with the premises.' (63)

     Spann implies that knowing whether conditional sentences like C are true does not aid philosophical or legal discourse about rights. He complains that '[i]n the context of most social problems, the act of ascribing a truth value to a premise--deciding whether or not the premise is acceptable--replaces rather than facilitates logical analysis.' (64) Perhaps this is an empirical claim, but why it should be true is obscure. In the end I think that it restates his concern with the ambiguity of nature language. (65) Because we do not really know what R is, Spann suggests, we cannot use statements like (x)Rx to derive C. If we do not share a core understanding of the property R, [-----616-----] sentences like C, while formally correct, tell us nothing about rights. (66) Yet, the view that no canonical case of compulsory self- incrimination can be constructed strikes me as extreme and untenable skepticism about the language of rights.

     Even more troubling is the observation that '[p]eeking ahead to see how a premise will be used before taking a position on its acceptability . . . overrides the logical analysis.' (67) This claim ignores the usefulness of indirect proof in mathematics. For instance, Euclid established that no largest prime number exists by providing that the assumption of a largest prime number leads to a contradiction. We are forced to reject the premise as false by examining its consequences. Such reasoning, which involves peeking ahead, does not defeat 'the whole purpose of conducting a logical analysis' even though 'the conclusion has generated the premise.' (68) More generally, logic serves a purpose even if one starts with the conclusion. By determining initially what premises establish the conclusion, one can turn to moral, legal, or other extralogical discussion of less complex ideas about which persons are more likely to agree or, at least, to recognize the core of their disagreement. (69)

     In partial reply Spann might concede that logic requires us to brand as false a premise that generates a contradiction, while asserting that we would be cheating to reject a premise that leads to merely counterintuitive conclusions. It takes considerable ingenuity, however, to argue that one may peek ahead when making mathematical deductions but not legal or philosophical deductions. The problem with peeking ahead cannot be self-evident, considering that several respected philosophers have introduced coherence theories as devices for adjusting premises and conclusions. Examining unexpected conclusions that logically flow from attractive premises seems like a reasonable way to test our intuitions about both the premises and the conclusions. Perhaps we will accept what first appeared to be a counterintuitive conclusion if the premises remain attractive. In other words, we will modify our intuitions about the conclusions. Alternatively, our intuitions about the conclusions may compel us to modify or reject the premises. This process of adjusting [-----617-----] and accommodating our intuitions about both premises and conclusions will not resolve all controversies, but neither is it illogical.

     Spann may, of course, believe that the acceptance of tentative and modifiable intuitions is a sham. Indeed, he claims that '[t]he system has been closed before the analysis even begins, thereby ensuring the desired result.' (70) But this is a complaint about human psychology rather than an indictment of deductive logic. Spann's argument will not convince anyone who does not already subscribe to the same belief. (71)

     In addition to his claim that deductive logic fails to improve the discussion of rights, Spann asserts that it may be affirmatively [-----618-----] counterproductive. (72) This thesis rests on three weak legs. To the extent that Secret Rights convinces us that fallacious deductive arguments often seduce people, it does not give us any reason for a paradigm shift. It only gives us a reason to look at legal rhetoric with care. (73) To the extent that Secret Rights stresses that logic induces us to think in dichotomous terms and to effect a binary reduction that conveys 'an over-simplified version of the social world in which we live,' (74) it errs. (75) To the extent that Secret Rights contends that logic 'dilutes the incentive . . . to imagine new, more satisfying forms of conceptualization,' (76) it must identify a superior mode of thought. Let us turn, then, to the new paradigm that Spann espouses.

III. THE NEW PARADIGM

A. SCIENTIFIC REVOLUTIONS

     Spann finds in the increasing number of books and essays like Secret Rights (77) an indication that the legal world may be in the midst of a paradigm shift in rational analysis. (78) The notion of a paradigm emerged from Thomas Kuhn's studies of the history of science. (79) By and large the paradigms that Kuhn identified [-----619-----] in physics and related sciences rely heavily on deductive logic and mathematics. To this extent 'the rules of the analytical game' (80) have never changed.

     Viewing this conservatism as blindness to new ways of looking at the universe, Spann treats all paradigms as equally effective in explaining how we reach conclusions. 'If we had another organizing system,' he writes, 'it would work just as well as our rational organizing system does because it would have to.' (81) Reducing all theory to convention seems bizarre when applied to scientific paradigms. The Copernican view of the universe is superior to the Ptolemaic even though both can predict the same observations of planetary motions. Newton's laws of motion work better than Aristotle's. A scientific realist or pragmatist would surely disagree with Spann's observation that 'our logic-based system of rationality is neither natural nor compelled by any need to correspond to objective reality.' (82) If we are witnessing a revolution in legal theory that will overthrow the present paradigm of rational analysis, (83) the paradigm shift is far more revolutionary than previous shifts, drawing only metaphorical support from those that have occurred in the scientific world. (84) [-----620-----]

B. MODERNISM IN GENERAL

     Fortunately, these allusions to scientific revolutions are not central to the analysis in Secret Rights. The more important issue is whether the nonrational system which Spann favors for political discourse has greater efficacy than the established mode of political and legal argument. His source of inspiration is '[p]ost-structuralist, modernist schools of thought that have influenced theology and the humanities.' (85) 'Modernism,' he reports, 'is a term that can be used to convey the belief that fundamental, structural assumptions are proper subjects for scrutiny and reexamination.' (86) This belief is not itself a new paradigm. 'Skepticism about religious authority, biblical doctrine and about the historical accuracy of theological events' (87) does not substitute nonrational conceptualization for deductive and inductive logic. In the humanities, modernism which refers to 'such works as nonrepresentational painting, atonal music, and stream of consciousness literature' (88) may seem less compatible with logical reasoning. Nevertheless, this catalog does not explain how 'modernism offers the promise of a qualitative advancement in analytical thought' (89) about rights.

     The only modernist insights about rights that Spann makes explicit are the arguments about equivocal terms, circular definitions, and self-reference. (90) Subject to the criticisms outlined in Part I, these arguments do nothing to establish the existence or reveal the contents of rights. (91)

C. LEGAL MODERNISM

     Like his treatment of scientific revolutions, Spann's description of modernist tenets is but a prologue to a more concrete description of the new paradigm. As opposed to expository writing, literary language should be privileged because [-----621] of its 'greater capacity to capture and convey ineffable concepts.' (92) Literary language, according to Spann, provides a 'direct evocation of potentially complex concepts.' (93) Although a well-written short story or poem can evoke an emotional response or capture the nuances of a particular situation, it is hard to see how this quality of literature helps us analyze 'social phenomena that are now being constructed to correspond to our current perceptions about social reality.' (94) Descriptions of social reality are one thing; normative claims about social arrangements are another. A novel may depict a social phenomenon like corruption with great insight, it may spur reform, and so on, but is it the type of writing to which attorneys and legal academics should aspire? Would such writers know good from bad, right from wrong, without recourse to deductive, normative reasoning? Could we more confidently rely on this new wave of literary briefs than on the tired, old analytical ones?

     A concrete example of literary language's use in the new paradigm would help to dissipate the mystery. Spann identifies the Supreme Court's opinion in American Textile Manufacturers' Institute v. Donovan (95) as illustrative of our present tendency to 'carry our penchant for precise expository definition to ridiculous extremes.' (96) In Spann's eyes Justice Brennan 'pretended that a profoundly difficult social issue, the proper manner of regulating exposure to toxic substances when the effects of given exposure levels is uncertain, could be resolved by reliance on the dictionary definition of a statutory term.' (97) Justice Brennan, of course, pretended no such thing, (98) but how [-----622-----] would a modernist construe the statute? By devising a Dickensian novel about laborers in the industrial revolution? What ineffable concept should the Court have addressed? The pain and anguish of victims of byssinosis or the pleasures of cotton dungarees?

     Spann offers the tale of Robin Hood as an example of the superiority of modernist legal analysis. In the new paradigm, 'we would just tell the jury the story and allow Robin's claim of right to resonate on the same intuitive level at which the operative rule of decision resides.' (99) If this is all that is required, the new paradigm looks like unbridled intuitionism, and story-telling to the jury seems similar to the technique that good litigators use now. Perhaps Spann's proposal differs because it contains no rules of relevance (because explicit legal liability rules would only impede intuitive decision making) so that Robin could tell his story better. More significantly, Spann may be arguing for a new logic that modulates the resonances of intuitions, producing correct outcomes.

     In addition to the reliance on literary language then, the new paradigm involves a reconceived concept of logic that has a more formal role for intuitions. (100) Presumably, this new logic will supply the literary lawyers with more valid intuitions than are presently available. This leads us again to the suggestion that we create more formal or well-defined theories to treat intuitions about premises and conclusions. Spann only can say in explanation of his suggestion that 'the precise model of how the two will interact is presently unclear.' (101) The mystery continues.

     Spann, nevertheless, asserts that the reconceived system of logic will improve upon deductive logic with its presently excluded intuitions. (102) Yet in existing normative and empirical argument we may, and usually do, inquire into the plausibility of an argument's premises. Because the rules of a particular formal logic only tell us how to transform one set of symbols into another, they do not help us in this inquiry. But neither do they preclude us from consulting our intuitions about premises, transformation rules, or conclusions. The choice is not between a new way of reasoning, of 'which we are as yet unable to conceive,' and an 'obsolete' system 'that tries to exclude intuition [-----623-----] altogether.' (103) A more rigorous procedure for dealing with intuitions would be welcome, but in this regard Spann proselytizes for an improved epistemology, not a better propositional calculus. In calling for new logic rather than improved moral epistemology, Secret Rights may point in the wrong direction.

     Spann also wants the new logic to contain 'the ability to account for a broader range of interactions between conceptual ideas.' (104) He reiterates the view that 'binary reductionism entails too great a loss of descriptive accuracy.' 105) This objection indicates that Spann misunderstands the power of deductive logic to handle multivalued functions. At various points Spann asserts that social phenomena are beyond the reach of binary logic because they are multidimensional. (106) Spann seems to define multidimensional as multivalued; he alludes to properties or quantities that cannot be described with yes or no propositions. Consider the measurement of an object's temperature. If binary logic restricted us to yes-no measurements of temperature, major branches of chemistry and physics would not have progressed very far. Fortunately, two-valued truth functions do not constrain us to the assertion that an object is hot or cold, and binary logic does not impose upon us such a poor approximation of physical reality (or our perceptions of that reality, to use modernist language). To say how hot an object is, we must refine our language to allow sentences T subj of the form 'The object has temperature j.' Each of these sentences is either true or false but not both. I can detect no advantage over the conventional and fruitful rule that a sentence cannot be simultaneously true and false in a new logic that would assign a true value to the compound proposition T1 and not-T1.

     Spann may not disagree with this. At one point he suggests that the new logic may reserve a domain for the old logic, (107) and he could say that I have uncovered a part of this domain. I merely point out, however, that criticism of binary logic as incapable of handling multivalued functions and multidimensional, multivalued properties does not constitute a substantial argument for the new logic. Rather, it reflects mesmerization with the word binary. [-----624-----]

     Furthermore, binary logic can accommodate the private-public distinction that troubles Spann. We do not need a new logic 'to allow something to be both public and private at the same time.' (108) We could define the property P of privacy on a scale ranging from zero (completely public) to one hundred (completely private) and use a series of numerically indexed sentences to describe the degree of privacy. As in the temperature example, each sentence remains either true or false but not both.

     Legal rules reflect a binary feature arising from the binary nature of many legal proceedings. Either the government allows the individual to undertake an activity without penalty or it does not. A new logic, allowing statements like 'the action is private and the action is not private' to be true (partly true, or whatever) still would have to confront this fact. Perhaps we could devise a new system of outcomes to legal disputes, but Spann does not relate this idea to his call for a new logic. (109)

IV. THE END OF THE ROAD

     Perhaps these criticisms would not matter to Spann. He concludes that 'when we ultimately do effect our escape to a post-binary logical paradigm, we cannot expect the route to be paved with traditional logic appeal.' (110) I have argued, however, that even the pragmatic justifications that he adduces have little appeal. Conceding that technical arguments about rules and terms frequently entangle lawyers and courts, and recognizing that many rules are open-textured and difficult to apply, I find no obvious reason to think that the intuitionist, storytelling approach would be superior to a sincere effort to induce and apply general rules to individual cases.

     I fear that my reluctance to experience the new perceptions and the leap into conceptual hyperspace (111) makes me part [-----625-----] of a regressive force. (112) But my intransigence should not distress a believer in Secret Rights. After all, is it not written that

When the best student hears about the way
He practices it assiduously;
When the average student hears about the way
It seems to him one moment there and gone the next;
When the worst student hears about the way
He laughs out loud.
If he did not laugh
It would be unworthy of being the way. (113)

Like Taoism, Secret Rights has mystical, rather than logical, allure.


NOTES

* Professor of Law, Arizona State University. I am grateful to Jeffrie Murphy, Kevin Saunders, James Weinstein, and Girardeau Spann for their comments on a draft of this Article.

1. Girardeau Spann, Book Review, 84 Mich. L. Rev. 628 (1986).

2. Girardeau Spann, Secret Rights, 71 Minn. L. Rev. 669 (1987).

3. Spann observes that '[s]ince the seventeenth century, when the concept of rights first came into vogue, philosophers and social theorists have struggled to articulate an acceptable theory of individual rights, but their efforts remain largely unsatisfactory. . . . [T]he persistent failure of rights theory suggests that we have done just about all that can be done from within the analytical paradigm that has governed rational discourse to date.' Id. at 669.

4. Id. at 670.

5. See, e.g., C. Winquist, Epiphanies of Darkness: Deconstruction in Theology (1986); Balkin, Deconstructive Practice and Legal Theory, 96 YALE L.J. 743 (1987).

6. Spann, supra note 2, at 671.

7. While Spann contends that a right is 'beyond the reach of juristic modification,' he does not mean that a right is absolute. Spann only requires 'some sort of immunization from societal abrogation.' Id. at 671 n.2.

8. Id. at 672. Spann does not limit his analysis to rights in democratic societies. The antimajoritarian dilemma applies, mutatis mutandis, to any sovereign. Id. at 672 n.3.

9. Id. at 672.

10. Id.

11. Id.

12. Cf. R. Sartorius, Individual Conduct and Social Norms 154-55 (1975) (defending an absolute prohibition against governmental interference even though such interference may periodically have advantageous consequences); Rawls, Two Concepts of Rules, 64 Phil. Rev. 3 (1955) (demonstrating the importance of distinguishing between 'justifying a practice' and 'justifying a particular action falling under it').

13. Spann, supra note 2, at 673.

14. Cf. Wollheim, A Paradox in the Theory of Democracy, in Philosophy, Politics & Society 71, 72 (1962) (identifying concepts of democracy, both ancient and modern, as 'the rule of the people').

15. Cf. Sunstein, Legal Interference with Private Preferences, 53 U. Chi. L. Rev. 1129, 1142 (1986) ('Some rights represent second-order preferences reflected in constitutional text.').

16. Spann, supra note 2, at 675.

17. Id. Spann anticipates that some readers may detect in his foregoing argument 'a concept not fully articulated.' Id. at 676. Indeed, Spann mentions the distinction among preferences upon which I have elaborated. Id. at 676 n.11. Yet Spann does not identify any inconsistency that arises from defining a right as a Type II preference prevailing over a Type I preference. He merely asserts that one could attack the Type I-Type II conception of a right as not resting on 'a meaningful difference between first- and second-order preferences,' as precluding natural rights, or as resting on an unjustified elevation of Type II rights. He expresses his faith that if one were to pursue such an argument, one would always generate a dilemma. Id. Inasmuch as Spann's argument does not lead to logical inconsistency, Spann's treatment of the antimajoritarian quality of rights falls far short of establishing the inevitability of such a dilemma.

     Spann's second tack for addressing the Type I-Type II solution is even less satisfying. Spann implicitly redefines his thesis. Instead of demonstrating the unavailability of a coherent theory of rights, he emphasizes 'that analytical manipulations can always be performed once the initial premises have been specified and locked in.' Id. If this means that scholars may constantly refine terms to permit further argument, he is probably correct. To a great extent, the history of philosophy is a history of refining the terms and rethinking the premises of arguments.

18. Id. at 672-73.

19. Not all instances of self-reference are troublesome. Many are simply amusing. See D. Hofstadter, Metamagical Themes 5-48 (1985).

20. For a discussion of self-reference in legal argument, see Fletcher, Paradoxes in Legal Thought, 85 Colum. L. Rev. 1263, 1266-67 (1985). As Professor Kevin Saunders has pointed out to me, the self-reference (if there is any) in Spann's antimajoritarian dilemma does not lead to the semantic and mathematical paradoxes that have been important or intriguing in philosophical logic. The self-reference in those paradoxes comes from making a statement about the statement itself (for example, 'This statement is false.') or asking of a term whether it has the property denoted by the term (for example, 'Is heterological heterological?'). See, generally, Recent Essays on Truth and the Liar Paradox (R. Martin ed. 1984). If we ask, 'Is it antimajoritarian for the majority to respect a(n) (antimajoritian) right?,' the answer merely depends on how we use antimajoritarian in each part of the question.

21. The point can also be made without refining phraseology. One could compare the antimajoritarian dilemma to the argument that if God is omnipotent, it can limit its own powers, in which case, it is not omnipotent. The fallacy is obvious--a previously omnipotent being can change its status. Likewise, an executive or administrative officer can bind himself by a regulation that he does not have the power to rescind. See, e.g., United States ex. rel Accardi v. Shaughnessy, 347 U.S. 260, 267 (1954). ('[A]s long as the regulations remain operative, the Attorney General denies himself the right to sidestep the Board or dictate its decision in any manner.'). Similarly, the majority can agree to institutional arrangements that result in outcomes that would not be acceptable in a plebiscite but retain the power to change these outcomes by constitutional amendment or revolution. So too, a majority can adopt a two- thirds voting rule that produces individual outcomes distasteful to a majority but that nevertheless commands the support of the majority. To view individual rights as antimajoritarian in this sense does not present a logical dilemma. Cf. Wollheim, supra note 14, at 85 (distinguishing between 'direct' and 'oblique' moral principles and arguing that they are not incompatible even though they cannot be simultaneously realized).

22. Spann, supra note 2, at 676.

23. Id.

24. Id. at 677-80.

25. Id. at 678.

26. Id. at 679.

27. Id. at 680.

28. Id.

29. Id.

30. Id. at 680 n.15

31. Spann asserts that congressional decision making may be more protective of individual liberty than Supreme Court decision making. Id. at 681 n.15.

32. See, generally, Kennedy, Freedom and Constraint in Adjudication: A Critical Phenomenology, 36 J. Legal Educ. 518 (1986) (discussing a reasoning process through which a hypothetical case involving a conflict between law and a judge's personal beliefs may be resolved).

33. Spann, supra note 2, at 681.

34. Id. at 682.

35. Id.

36. Id. at 682-83. I would restate this distinction as the distinction between the private, protected sphere and the public, unprotected sphere. Then we may say that negative rights theory must demarcate these spheres. In those terms I understand Spann's argument to be that one cannot identify the private sphere in a principled way.

37. See, e.g., 1 J. Feinberg, Harm to Others: The Moral Limits of the Criminal Law (1984); 2 J. Feinberg, Harm to Others: The Moral Limits of the Criminal Law (1985).

38. See, e.g., Dalton, Book Review, 96 Yale L.J. 881 (1987).

39. Spann, supra note 2, at 683.

40. Id.

41. Id. at 684. This aspect of Spann's argument does not amount to an effort to undermine a theory of rights with a paradox of self- reference. His argument is simply that negative rights theories are circular. Private is defined as the absence of public, and public as the absence of private. I am indebted again to Professor Kevin Saunders for reminding me that there is nothing illogical about defining a term as the negation of its negation. Arguments involving such circular definitions may not take us out of the starting gate, but they are neither contradictory nor paradoxical, and they do not impeach deductive logic itself.

42. Id. at 685.

43. Id. at 687.

44. Id. at 686.

45. Id. at 688.

46. Id.

47. Id.

48. Id.

49. Spann believes that his analytical dilemmas result from self- reference, as in the Liar paradox, in which Epimenides the Cretan asserts that all Cretans are liars. Id. at 692. This reliance on classic paradoxes of self-reference seems misplaced. See supra note 20. Spann confuses the vice of circularity with the paradox of self-reference. Compare Spann, supra note 2, at 692 ('Good can be defined as the absence of evil . . . . The terms . . . become not only mutually supporting but self-referential . . . .') with supra note 41 (distinguishing circularity from self-reference).

     Furthermore, even if Spann has identified a true paradox of self-reference in political argument, his discovery would not impeach formal logic. Scholars have used the self-referential argument that Spann considers impeccable to show that set theory is incoherent. The Russell Paradox about the class of all classes that are not members of themselves made this point. E. Nagel & J. Newman, Godel's Proof 24 (1958); B. Russel, Introduction To Mathematical Philosophy 181-93 (1967). With his theory of types, however, Russell successfully avoided the paradoxes of set theory. The lesson here is not that formal logic is wrong, but that a certain class of arguments is inadminissible in any valid analysis that assigns definite truth values to all the propositions in a formal language.

50. As indicated, supra note 41, circularity is not a defect in logic. We even can explain the self-reference in the Liar Paradox without rejecting logic. See, e.g., C. Lewis & C. Langford, Symbolic Logic 438-85 (2d ed. 1959). I do not suggest that there is presently a fully adequate theory of truth that enables us to identify the pernicious instances of self-reference. We have informal methods, however, for identifying the troublesome cases and once they are identified we can remove their sting. Cf. G. Harmon, Change In View: Principles of Reasoning 87 (1986) (discussing the value of self- referential beliefs).

51. Spann, supra note 2, at 688.

52. Id.

53. Id. at 688-89.

54. The use of reflective equilibrium, for example, is prominent in moral and political philosophy. See, e.g., J. RAWLS, A THEORY OF JUSTICE 48-51 (1971). Logicians also may rely on reflective equilibrium analysis to justify the very rules of logic. See G. HARMAN, supra note 50, at 9; Resnik, Logic: Normative or Descriptive? The Ethics of Belief or a Branch of Psychology, 52 PHIL. SCI. 221 (1985).

55. Spann, supra note 2, at 697.

56. Spann also discusses the limitations of natural language. Id. at 689-94. Because I am particularly interested in his criticism of logic, I shall offer only a few comments on his discussion of language. Spann points out the ambiguity of natural language, stressing that the linguistic map is not itself the territory. Nevertheless, Spann does not rest here. He further criticizes natural language as a vehicle for analyzing rights because self- referential terms present a 'potential problem.' Id. at 693. He also suggests that the concept of rights is ineffable, stating that it 'simply refuses to be confined by any linguistic formulation.' Id. Consequently, a conclusion expressed in expository language may be too artificial, and the meaning of words may change in subtle ways in the course of an argument. Id. at 693-94.

     What conclusions should be drawn from these observations is less than clear. When language is ambiguous, we should reject arguments that depend on equivocation. Similarly, the solution to any occasional problem of self-reference is not a paradigm shift but the use of better terminology if a paradox surfaces. If we encounter ineffable concepts, we should say as much as we can while admitting that we cannot tell the whole story. Expository prose that describes color to a congenitally blind person or pregnancy to a man cannot depict the experience accurately or completely, but it is not necessarily worthless.

57. Id. at 695 n.48.

58. Id. at 695.

59. Id.

60. An argument is deductively valid if and only if it is impossible that its conclusion is false given that its premises are true. B. SKYRMS, CHOICE AND CHANCE: AN INTRODUCTION TO INDUCTIVE LOGIC 7 (2d ed. 1975). An argument is inductively strong if and only if it is improbable that the conclusion is false while the premises are true. Id. A deductive logic suited to probabilistic claims also can be devised. See Saunders, A Logic for the Analysis of Collateral Estoppel, 12 RUTGERS COMPUTER & TECH. L. REV. 99 (1986).

61. Spann, supra note 2, at 695 n.48.

62. Id. at 696.

63. Id. at 695.

64. Id. at 696.

65. See supra note 56.

66. But see Kress, The Interpretive Turn, 97 ETHICS 834, 855-59 (1987) (describing theories of meaning).

67. Spann, supra note 2, at 697.

68. Id.

69. Professor Kevin Saunders called my attention to this fact, in language quite close to that which I have used.

70. Spann, supra note 2, at 697.

71. He merely asserts that 'we have not developed an epistemological model--a paradigm--in which such a form of argument is explicitly cognizable.' Id. at 698 n.57. Apparently, the substantial body of writing in moral philosophy that explicitly indulges in discussion about the priority of various intuitions does not present cognizable argument.

     Neither is Spann's reference to Godel's theorem very revealing. While Spann wisely recognizes that 'it is probably not useful to attempt a direct application of Godel's theorem to legal or philosophical analysis,' id. at 699 n.58, he thinks that '[o]ne of the implications of Godel's theorem is that logic' does not constitute 'a closed system in which things happen in a predictable way, in accordance with orderly rules that are understandable and reliable.' Id. This is not what the theorem states or implies. The orderly rules of an axiomatized system do not cease to work. A theorem that is correctly proven within such a system is not subject to doubt by reason of Godel's discovery. See, e.g., R. JEFFREY, FORMAL LOGIC: ITS SCOPE AND LIMITS 185 (2d ed. 1981). As for predictability, even an extremely simple system, to which Godel's theorem does not pertain, can produce unexpected results. See W. POUNDSTONE, THE RECURSIVE UNIVERSE: COSMIC COMPLEXITY AND THE LIMITS OF SCIENTIFIC KNOWLEDGE 13-32 (1985).

     Godel's proof does demonstrate that there are arithmetic truths that cannot be proved within a strictly formal system. Yet, even these truths can become known to mathematicians:

     Godel's proof should not be construed as an invitation to despair or as an excuse for mystery-mongering. The discovery . . . does not mean that there are truths which are forever incapable of becoming known, or that a 'mystic' intuition (radically different in kind and authority from what is generally operative in intellectual advances) must replace cogent proof. It does not mean . . . that there are 'ineluctable limits to human reason.' It does mean that the resources of the human intellect have not been, and cannot be, fully formalized, and that new principles of demonstration forever await invention and discovery. We have seen that mathematical propositions which cannot be established by formal deduction from a given set of axioms may, nevertheless, be established by 'informal' meta-mathematical reasoning. It would be irresponsible to claim that these formally indemonstrable truths established by meta-mathematical arguments are based on nothing better than bare appeals to intuition.

E. NAGEL & J. NEWMAN, supra note 49, at 101.

72. Spann, supra note 2, at 699.

73. Theories about rights do not purport to offer causal or psychological explanations for why we fine or imprison people. These theories explain why we are justified in our actions. Reliance on legal rules or principles by way of justification 'perpetuates a distorted account of why we do the things that we do to each other,' id. at 699, only if one fails to distinguish between the internal and external aspects of a legal system.

74. Id. at 700.

75. See infra text accompanying notes 104-07.

76. Spann, supra note 2, at 700.

77. Id. at 703.

78. Id.

79. T. KUHN, THE STRUCTURE OF SCIENTIFIC REVOLUTIONS (1962). In his writings Kuhn uses the term paradigm in two senses. A paradigm-as-achievement is the accepted way of solving a problem, serving as a model for future researchers. A paradigm-as-a-set-of-shared-values encompasses the methods, standards, and generalizations shared by those trained to carry on the work. SCIENTIFIC REVOLUTIONS 2-3 (I. Hacking ed. 1981). As one might expect, philosophers of science have challenged Kuhn's use of paradigms on which Spann builds. One of the more caustic critics, Paul Feyerabend, writes:

     Kuhn's ideas are interesting but, alas, they are much too vague to give rise to anything but lots of hot air. If you don't believe me, look at the literature. Never before has the literature on the philosophy of science been invaded by so many creeps and incompetents. . . . We do not get interesting false ideas, we get boring ideas or words connected with no ideas at all.

Feyerabend, How to Defend Society Against Science, in SCIENTIFIC REVOLUTIONS, supra, at 160.

80. Spann supra note 2, at 701.

81. Id. at 702.

82. Id. at 701. The same point may be made in regard to everyday, practical reasoning. Although imperfect, the correspondence between formal logic and pragmatic reasoning schemas is more than coincidence or convention. J. HOLLAND, K. HOLYOAK, R. NISBETT & P. THAGARD, INDUCTION: PROCESSES OF INFERENCE, LEARNING AND DISCOVERY 44-45, 255-86 (1986).

83. I fully agree with Spann that alternatives to inductive and deductive logic such as Zen Buddhism exist. To produce a credible argument for this kind of paradigm shift in science, however, Spann must establish that these alternatives have greater explanatory power than contemporary scientific reliance on logic.

84. Spann uses quantum theory as an example of how our culture relies heavily on nonlogical organizing systems. Spann, supra note 2, at 702. Quantum mechanics is not nonlogical. Quantum mechanics is full of deductions from assumedly true premises. Rather than eschew deductive logic, quantum mechanics employs it. Rather than undermining mathematical logic, the Heisenberg uncertainty principle reflects a mathematical property of the position and momentum operators. See, e.g., E. MERZBACHER, QUANTUM MECHANICS 156 (1961). In this, as in other instances, quantum mechanics uses different mathematical representations of physical systems than does classical mechanics, but it does not reject the logico-deductive schema in favor of some 'nonrational conceptualization.' Spann, supra note 2, at 702. To understand the behavior of elementary particles like photons, physicists no longer believes, in the spirit of Democritus, that particles are tiny, immutable billiard balls. In quantum field theory, the fields are the fundamental units and the particles are merely manifestations of these fields. While this conception of matter is far removed from everyday experience and produces profound difficulties in understanding causation in traditional terms, it is not, as Spann seems to think, 'replete with . . . logical impossibilities.' Id.

85. Spann, supra note 2, at 703.

86. Id. at 704.

87. Id.

88. Id.

89. Id.

90. Id. at 705-07

91. Spann views this as an advantage. He states that '[b]oth theoretical and practical discussions of rights completely depend on perceptions concerning those rights rather than any actual content of the rights.' Id. at 706.

92. Spann, supra note 2, at 707.

93. Id. at 710.

94. Id.

95. 452 U.S. 490 (1981).

96. Spann, supra note 2, at 708.

97. Id. at 708 n.81.

98. The textile manufacturers challenged OSHA's cotton dust standard. American Textile, 452 U.S. at 494. OSHA contended that the pertinent statute required the most stringent standard that was technologically and economically feasible. Id. at 495. The Court stated the issue to be 'whether the Occupational Safety and Health Act requires the Secretary in promulgating a standard . . . to determine that costs of the standard bear a reasonable relation to its benefits.' Id. at 506. After examining the language, structure, and legislative history of the Act, the Court held that the statutory phrase 'to the extent feasible' did not require OSHA to equate marginal benefits to marginal costs when setting standards. Id. at 512-13. Justice Stewart dissented on the ground that, without any evidence of the total cost of the standard, there was not substantial evidence to support the agency's determination that its standard was economically feasible. Id. at 543.

99. Spann, supra note 2, at 716.

100. Id. at 712.

101. Id.

102. Id.

103. Id.

104. Id. at 713.

105. Id.

106. Id. at 695, 700, 710, 713.

107. Id. at 712.

108. Id. at 713 n.95.

109. I also should register disagreement with Spann's related claim that 'it is possible to uncover internal contradictions and logical inconsistencies in almost any formulation of a binary premise.' Id. at 713. To establish this claim, he must exhibit a method or procedure for uncovering such contradictions. From the standpoint of logic, the deconstructionist style is not such a method. Indeed, for the reasons stated in the early parts of this Article, the 'logical dilemmas generated in Part I,' id., are resolvable within standard logic. Spann's assertions notwithstanding, he has not shown that the method of generating the dilemmas vitiates reformulations that avoid or dispose of the dilemmas.

110. Id. at 714.

111. Id. at 722.

112. Spann, supra note 2, at 720.

113. LAO TZU, TAO TE CHING 102 (E. Rieu trans. 1963). In quoting from Lao Tzu, I am calling attention to the link between Secret Rights and the fundamental Taoist premise that:

The way that can be told
Is not the constant way;
The name that can be named
Is not the constant name.

Id. at 57.