WEBGALLEYS.DOC 05/05/99 5:44 PM


    Dan M. Kahan*

“Democracy” is often played as the trump card in the an-tidelegation hand. Delegation, the critics charge, relieves our elected representatives of their responsibility to make law and binds us to the will of unaccountable bureaucrats. However sensi-ble delegation might be on policy grounds, its antidemocratic na-ture constrains us to reject it as a matter of constitutional princi-ple.1

My goal in this essay is to expose the vacuity of this argument. Democracy, I’ll argue, is an empty standard for judging the desir-ability or constitutionality of the delegation schemes familiar to American law. This is so because democracy is an essentially con-tested concept: there is not just one, but rather a plurality of com-peting conceptions of democracy, each of which emphasizes a dif-ferent good commonly associated with democratic political regimes. Delegated lawmaking schemes never disregard these goods entirely but rather give more or less prominence to one or another relative to some nondelegation alternative. Thus, before someone can persuasively criticize a delegation scheme on grounds of democracy, she must normatively justify the particular concep-tion of democracy that informs her critique. That means that any democracy-grounded critique of a particular delegation scheme will turn out to be derivative of some independent normative cri-tique. It also means that if we feel we do have good normative grounds to delegate, democracy will never be a persuasive reason against delegating.

I’ll present this argument in the form of five claims: (1) De-mocracy is a contested concept. (2) Because democracy is a

con-* Professor of Law, University of Chicago Law School. I am grateful to Einer El-hauge, Elizabeth Garrett, Paul Kahn, Eric Miller, Mark Rosen, Richard Pildes, David Strauss, Cass Sunstein, and Barry Friedman for comments, and to the Russell J. Parsons and Jerome S. Weiss Faculty Research Funds at the University of Chicago Law School for generous financial support.



 WEBGALLEYS.DOC 05/05/99 5:44 PM

796 CARDOZO LAW REVIEW [Vol. 20:795

tested concept, it doesn’t uniquely determine institutional struc-ture. (3) Delegation schemes invariably advance some conception of democracy, and impede others, relative to the nondelegation baseline. (4) The ends promoted by delegation are therefore nor-mative for democracy, not vice versa. (5) Adding the Constitution adds nothing to the vacuous democracy critique of delegation.


The idea of “essentially contested concepts” is central to con-temporary political theory.2 An evaluative concept can be deemed essentially contested when it is sufficiently complex and open-textured to sustain multiple reasonable interpretations of what the concept values.3 Such concepts give rise to “endless disputes about their proper uses,” in which all sides are able to support their posi-tions with “perfectly respectable arguments and evidence.”4 “Lib-erty,” “power,” and “rule of law” are all essentially contested con-cepts in this sense.5

So is democracy.6 Very generally speaking, it is possible to identify two competing conceptions of democracy in standard legal and political discourse. Each of these, moreover, can be divided into two competing subconceptions. The pluralist conception views government as more or less democratic depending on the ex-tent to which official decisions conform to the aggregated prefer-ences of the electorate. The subconceptions diverge on how pref-erences should be aggregated: the populist variant assigns equal weight to each voter’s policy preferences and then simply adds them; the market variant takes the intensity of voters’ preferences into account as well.7

The civic republican conception treats government as


3 See id. at 10.
4 W.B. Gallie, Essentially Contested Concepts, 56 PROC. ARISTOTELIAN SOC’Y 167, 169 (1956).

     5 See CONNOLLY, supra note 2, chs. 3 & 4 (discussing power and freedom, respec-tively); Richard H. Fallon Jr., The Rule of Law as a Concept in Constitutional Discourse, 97 COLUM. L. REV. 1 (1997) (discussing rule of law).

6 See Gallie, supra note 4, at 184-85.
7 The populist view is more often assumed than explicitly defended in writings on de-mocracy. For accounts that self-consciously develop the market variant, see generally DONALD A. WITTMAN, THE MYTH OF DEMOCRATIC FAILURE: WHY POLITICAL INSTITUTIONS ARE EFFICIENT (1995); ROBERT DAHL, A PREFACE TO DEMOCRATIC THEORY (1956); and JOSEPH A. SCHUMPETER, CAPITALISM, SOCIALISM, AND DEMOCRACY (2d ed. 1947).

 WEBGALLEYS.DOC 05/05/99 5:44 PM


cratic only to the extent that official decisions are reached through a process of reflective deliberation on the “common good.” Within civic republicanism, it’s possible to distinguish between a dialogic approach, which defines the “common good” as the view that emerges, or that would emerge, from a full and open process of deliberation,8 and a communitarian approach, which views the good as consisting of the shared values that make up a political community’s distinctive culture or way of life.9

These conceptions and subconceptions of democracy generate conflicting results across a myriad of practical issues. As is so with disputes over the application of other contested concepts, these disagreements can’t be resolved by showing that one conception of democracy or another is most in keeping with standard usage—no such usage exists—but only by demonstrating that a particular conception is, either in general or for a specific purpose, norma-tively superior to its competitors.10


My second claim is that the concept of democracy, by itself, doesn’t uniquely determine the structure of government institu-tions.
11 This conclusion follows from the status of democracy as an essentially contested concept. For no matter how well a particular institutional arrangement implements one conception of democ-racy, there will always be another arrangement that better imple-ments some other conception or variant thereof.

Consider the choice between “direct democracy” and “repre-sentative democracy.” A plausible case can be made for direct democracy under the pluralist conception: What better way to de-termine whether a particular policy or law accords with collective preferences than to have all of the community’s citizens vote on

8 See, e.g., AMY GUTTMANN & DENNIS THOMPSON, DEMOCRACY AND DISAGREEMENT 52-164 (1996) (developing and defending “deliberative” conception of democracy that realizes common good in midst of moral pluralism); CAROLE PATEMAN, PARTICIPATION AND DEMOCRATIC THEORY 22-27, 103-11 (1970) (building on Rous-seau).


10 See CONNOLLY, supra note 2, at 29.
11 See generally SAMUEL ISSACHAROFF, PAMELA S. KARLAN & RICHARD H. PILDES, THE LAW OF DEMOCRACY: LEGAL STRUCTURE OF THE POLITICAL PROCESS 1 (1998) (“There are many possible forms democracy can take, many different institutional em-bodiments of democratic politics.”).

 WEBGALLEYS.DOC 05/05/99 5:44 PM

798 CARDOZO LAW REVIEW [Vol. 20:795

it?12 But this is pluralism in its populist form, which conceives of aggregation as the simple addition of equally weighted prefer-ences. Under the market variant, which takes intensity of prefer-ences into account, legislation might be “more democratic” after all, since the electoral process affords certain advantages to well organized and intensely interested minorities.13

Civic republicanism generates prescriptions that are no more determinate. The dialogic position might see representative insti-tutions as the locus of democracy. Genuine deliberation is more feasible among a small number of elite legislators than it is among the citizenry at large. Moreover, if allowed to serve terms of suffi-cient duration and if elected to represent constituents sufficiently large in number on issues sufficiently diverse in nature, representa-tives will enjoy a modest degree of independence from their con-stituents. As a result, they will be more likely to reflect dispas-sionately on the common good than would self-interested citizens making law directly on their own behalf. This, at least, was Madi-son’s claim in his influential defense of representative democracy in Federalist No. 10.14

But, by design, such representatives will never enjoy more than modest independence from the demands of their constituents. As a result, there can be no guarantee that they will be perfectly responsive to the shared values that construct the community’s dis-tinctive political culture, particularly when those values purport to temper the immediate interests of society in the name of “higher” ends, such as liberty and equality. We can well understand why an individual, in the interest of “self-determination,” might submit to the authority of another (e.g., a fitness trainer) who is charged with assuring that that individual follows through on his preference to subordinate certain immediate desires (to eat rich foods) to his higher goals (to be physically fit). Likewise, a community that wants to be governed by its higher values and not just its immedi-ate interests might submit to the authority of an institution, say, a constitutional court, that is specifically designed to invalidate

leg-12 Cf. Matthew L. Spitzer, Evaluating Direct Democracy: A Response, 4 U. CHI. L. SCH. ROUNDTABLE 37, 43-44 (1997) (examining evidence that referenda move law in the direction of the median voter’s preference).

13 See Einer R. Elhauge, Does Interest Group Theory Justify More Intrusive Judicial Review?, 101 YALE L.J. 31, 65-67 (1991); David A. Strauss, Corruption, Equality, and Campaign Finance Reform, 94 COLUM. L. REV. 1369, 1378-79 (1994).

14 THE FEDERALIST NO. 10 (James Madison); see also Marci A. Hamilton, The People: The Least Accountable Branch, 4 U. CHI. L. SCH. ROUNDTABLE 1, 6-7 (1997) (developing this view).

 WEBGALLEYS.DOC 05/05/99 5:44 PM


islation in the name of those higher values. This is a communi-tarian rationalization of judicial review that sees it not as “counter-majoritarian” but as essential to perfecting democracy.15

Matters get even more complicated when it is recognized that the fit between particular conceptions and particular lawmaking apparatuses is empirically contingent. It’s not demonstrably clear, for example, that representative democracy is superior to direct democracy under the market variant of pluralism, since referenda and legislation can be made more or less sensitive to intensity of preferences depending on how referenda are conducted and elec-tion campaigns financed.16 Civic republican generalizations are similarly fragile: legislators can plausibly be made more attentive to the common good through devices like term limits;17 courts can be expected to be more or less responsive to the political commu-nity’s fundamental values depending on the philosophy of consti-tutional interpretation that they adopt.18

None of this is to say that it’s pointless to argue over the de-sirability of referenda, the need for campaign finance reform, the scope of judicial review, and the like. But it is to say that the point of having such arguments can’t be to figure out which positions on these issues are compelled by democracy; all the relevant positions make government simultaneously more and less democratic de-pending on which conception of democracy one has in mind.

We might reasonably conclude that the best institutional ar-rangement is one that doesn’t pursue a single variant of any par-ticular conception of democracy but that instead partakes of all the conceptions and subconceptions to some degree. But this ecu-menical conception of democracy doesn’t uniquely determine in-stitutional structure either. To say we want populist pluralism and market pluralism and dialogic civic republicanism and

communi-15 See RONALD DWORKIN, FREEDOM’S LAW: THE MORAL READING OF THE AMERICAN CONSTITUTION 30-31, 344-46 (1996). For accounts that seek to dissolve the “countermajoritarian” dilemma in this fashion, see Barry Friedman, Dialogue and Judicial Review, 91 MICH. L. REV. 577 (1993); Paul W. Kahn, Community in Contemporary Con-stitutional Theory, 99 YALE L.J. 1 (1989); Frank I. Michelman, The Supreme Court, 1985 Term—Forward: Traces of Self-Government, 100 HARV. L. REV. 4 (1986).

16 See, e.g., Elizabeth Garrett, Who Directs Direct Democracy?, 4 U. CHI. L. SCH. ROUNDTABLE 17 (1997).

17 See GEORGE F. WILL, RESTORATION: CONGRESS, TERM LIMITS AND THE RECOVERY OF DELIBERATIVE DEMOCRACY (1992) (defending term limits as device for blunting effect of self-interest and accentuating civic virtue in democratic politics). But see Elizabeth Garrett, Term Limitations and the Myth of the Citizen-Legislator, 81 CORNELL L. REV. 623 (1996) (challenging this view).

18 See, e.g., RONALD DWORKIN, TAKING RIGHTS SERIOUSLY 141 (1977) (arguing that different conceptions of democracy support different theories of interpretation).

 WEBGALLEYS.DOC 05/05/99 5:44 PM

800 CARDOZO LAW REVIEW [Vol. 20:795

tarian civic republicanism doesn’t tell us how much of each we want. Moreover, even if we could figure out the proportions we wanted, our political science lacks empirical and conceptual in-struments precise enough to measure how much a particular insti-tution or practice—say, term limits or even a particular philosophy of judicial interpretation—alters the overall balance. We can still make arguments about whether such an institution or practice makes sense, but for those arguments to make sense they have to be grounded in normative considerations outside the concept of democracy.


What goes for institutional structures in general goes for dele-gation in particular. The concept of democracy, by itself, doesn’t uniquely determine a position on the desirability of delegation be-cause all of the delegation schemes familiar to American adminis-trative law can plausibly be seen as both “less” and “more” demo-cratic than some nondelegation alternative depending on what conception of democracy one has in mind.

The nondelegation alternative that delegation critics have in mind is congressional legislation. Congressional lawmaking quali-fies as democratic under both the pluralist and civic republican conceptions of democracy: Congress is responsive to citizens’ preferences by virtue of the popular election of its members, and it deliberates before it enacts legislation. But no one can plausibly say that Congress is perfectly responsive or perfectly deliberative. Depending on how it is structured, a delegation scheme can always be defended as making government more democratic along one of these dimensions.

Start with the civic republican conception of democracy. Civic republicans tend to see Congress as excessively responsive to interest groups and insufficiently attentive to the common good; the role of money in contemporary electoral politics has defeated Madison’s deliberative aspirations. Delegation to agencies, how-ever, may be one way to restore it. Administrative rulemakings af-ford all interested citizens an opportunity to voice their views, and the Administrative Procedure Act’s “arbitrary and capricious” standard19 obliges agencies to address the substance of these views, instead of merely aggregating the preferences of interest groups, when it makes law. This deliberative conception of delegated

19 See 5 U.S.C. 706(2)(A) (1994).

 WEBGALLEYS.DOC 05/05/99 5:44 PM


lawmaking furnishes a plausible description of what at least some agencies do.20

Indeed, it’s possible to see at least some delegations as making Congress itself more deliberative. Congress is in its least delibera-tive cast of mind in the appropriations process, where members routinely auction off government largesse to the interest groups that are best positioned to support members’ reelections. The ill-fated Line Item Veto Act (“LIVA”),21 which delegated authority to the President to cancel appropriations, would have disrupted this appropriations market. Because LIVA allowed the President to impound any appropriated funds, the deals members could offer to interest groups would have been worth substantially less; be-cause they would have been worth less, interest groups would have been unwilling to pay as much for them; because interest groups would have been unwilling to pay as much, members of Congress would have had less incentive to pander to them; and because they would have had less incentive to pander, members would have been more likely to deliberate on the common good. It’s even possible that over time LIVA would have made Congress more deliberative by making federal office less attractive to individuals who are willing to exchange largesse for electoral support, or by making it more difficult for those persons to entrench themselves in office.22

Certain delegations can also be seen as perfecting the com-munitarian variant of civic republicanism. The occasions on which Congress summons the political courage to subordinate constitu-ents’ immediate interests and desires to higher national values are not nonexistent, but they are definitely rare. Giving courts the power to review the consistency of legislation with such values is one way for a community to assure that it will be governed by its principals and not just by its interests; another is to delegate some lawmaking power to an agency that is likely to give more weight to those principles than is the legislature. Because its members needn’t fear being depicted as “soft on crime,” the Federal Sen-tencing Commission, for example, is much more likely to be re-spectful of liberty when it promulgates sentencing guidelines than

20 See Mark Seidenfeld, A Civic Republican Justification for the Bureaucratic State, 105 Harv. L. Rev. 1511, 1541-43 (1992); Cass R. Sunstein, Factions, Self-Interest, and the APA: Four Lessons Since 1946, 72 VA. L. REV. 271, 282 (1986).

21 2 U.S.C. 691-692 (Supp. II. 1996). The Supreme Court held LIVA unconstitu-tional in Clinton v. City of New York, 118 S. Ct. 2091 (1998).

22 See generally Elizabeth Garrett, Accountability and Restraint: The Federal Budget Process and the Line Item Veto Act, 20 CARDOZO L. REV. 871 (1999).

 WEBGALLEYS.DOC 05/05/99 5:44 PM

802 CARDOZO LAW REVIEW [Vol. 20:795

is Congress when it enacts sentencing statutes.23 Likewise, were the Equal Employment Opportunity Commission to be delegated the authority to issue legally binding rules, it would likely be more responsive than Congress to the democratic value of equality.

These arguments undermine any claim that delegation is anti-democratic merely because it diminishes “congressional account-ability.” Too much legislative accountability, the civic republican view suggests, can disserve certain democratic goods and experi-ences; when that is so, it perfects democracy for elected represen-tatives to delegate some of their power to agencies structured to respect those goods and experiences. To be sure, this is not an un-contestable account of what democracy requires. But precisely because the concept of democracy is contestable, the preference for congressional accountability over delegated lawmaking can’t be justified as compelled by democracy.

Some versions of the democracy critique attack the civic re-publican account of delegation on empirical grounds. On this view, Congress doesn’t genuinely give up power when it delegates. Instead, it continues to call the shots from behind the scenes by threatening to cut agencies’ budgets or discipline them in various ways if they take positions contrary to the expectations of the rele-vant congressional oversight committees. Members of those com-mittees predictably use this control to pander to the parties whom the agencies regulate. Thus, delegation is actually a means of so-lidifying the influence of interest groups; the appearance of inde-pendent deliberation or guardianship of higher values is an illu-sion.24

But even assuming this “agency capture” story is well-founded empirically—an issue that many dispute25—it does not

23 The Sentencing Commission has taken more liberal positions than Congress, for ex-ample, on the method for calculating sentencing of offenders convicted of distributing LSD and on the disparities in the sentences of offenders convicted of distributing powder cocaine and those convicted of distributing crack cocaine. See Act of Oct. 30, 1995, Pub. L. No. 104-38, 109 Stat. 334 (congressional rejection of Sentencing Commission recom-mendation to reduce severity of sentences of crack cocaine relative to powder); Neal v. United States, 516 U.S. 284 (1996) (overturning attempt by Sentencing Commission to dispel statutory anomaly in LSD sentencing).

24 See, e.g., Morris P. Fiorina, Legislative Choice of Regulatory Forms: Legal Process or Administrative Process, 39 PUB. CHOICE 33, 54-55 (1982); Morris Fiorina, Legislator Uncertainty, Legislative Control and the Delegation of Legislative Power, 2 J.L. & ECON. 33, 49 (1986).

25 See JERRY L. MASHAW, GREED, CHAOS, AND GOVERNANCE: USING PUBLIC CHOICE TO IMPROVE PUBLIC LAW 188 (1997). See generally Terry M. Moe, An Assess-ment of the Positive Theory of “Congressional Dominance,” 12 LEGIS. STUD. Q. 475 (1987).

 WEBGALLEYS.DOC 05/05/99 5:44 PM


unambiguously demonstrate that delegation is undemocratic. To say that delegation makes Congress more responsive to interest groups than to citizens generally is akin to saying that delegation connects the law more reliably to the preferences of the parties who care about the law the most, since they are the ones who will have the incentive to form an effective interest group. Thus, rela-tive to a nondelegation baseline, the “agency capture” story sug-gests that agency lawmaking actually enhances democracy accord-ing to the market variant of pluralism.26 Indeed, some means of structuring agency proceedings, including “negotiated rulemak-ings,” adopt this rationale explicitly.27

Obviously, it isn’t the case that every conceivable form of delegation can be justified as making government more demo-cratic according to some plausible conception of democracy. Just as we wouldn’t accept the claim that an individual could, in the name of self-determination, agree to become another person’s slave, so we wouldn’t accept the proposition that a community could, in the name of self-government, vote to relinquish its sover-eignty to a benevolent dictator. But it would be hyperbole bor-dering on hysteria to describe the types of delegations that are fa-miliar to American law as equivalent to that. Indeed, my guess is that no democratically organized community would ever enact a delegation scheme that couldn’t be seen as making its government more democratic under some plausible conception of that term. But in any case, the critics of delegation have yet to identify an existing delegation scheme in American law that can’t be seen this way.


So far I’ve argued that democracy is a contested concept, that this contested concept does not uniquely determine institutional structure, and that all conventional delegation schemes are consis-tent with one conception of democracy or another. If I’m right, then it follows inescapably that the interests and values served or disserved by delegation are normative for democracy, and not vice versa. Any argument that critiques delegation based on one

con-26 Cf. Richard B. Stewart, The Reformation of American Administrative Law, 88 HARV. L. REV. 1667, 1711-1813 (1975) (developing pluralist model of administrative deci-sionmaking and critiquing proposed means of muting dominance of highly organized in-terest groups).

27 See generally Lawrence Susskind & Gerard McMahon, The Theory and Practice of Negotiated Rulemaking, 3 YALE J. ON REG. 133 (1985).

 WEBGALLEYS.DOC 05/05/99 5:44 PM

804 CARDOZO LAW REVIEW [Vol. 20:795

ception of democracy will be answerable by an argument that de-fends delegation based on some other conception of democracy. To choose between these arguments, we must first decide which conception of democracy is more compelling. And to avoid circu-larity in that decision, we must appeal to some interest or value ex-ternal to the concept of democracy itself.

There are different strategies for normatively justifying con-ceptions of democracy, and they have different implications for the status of delegation generally. A top down strategy would try to solve the problem of democracy globally by showing that one con-ception or another advances some master value like efficiency or autonomy. It would then take a position on delegation, pro or con, depending on whether delegation implements or impedes this normatively justified conception of democracy. Some public choice critiques of delegation,28 and certain civic republican de-fenses of it,29 have this character. But I’m skeptical of the top down approach, given the evident intractability of arguments over conceptions of democracy and the limited power of political sci-ence to predict how institutional structures will operate across di-verse settings.

A bottom up approach strikes me as more sensible. This strategy asks not which conception of democracy and correspond-ing position on delegation are “best” in the abstract, but which make the most sense in a particular regulatory setting, given the values and interests at stake there. On this account, whether dele-gation is desirable is decided locally, not globally. Giving a private citizen authority to sue under the Endangered Species Act of 1973,30 for example, might be great in light of the interest group distortions that afflict enforcement policy within the Executive Branch; giving authority to a private citizen (a so-called “Inde-pendent Counsel”) to enforce federal criminal law against mem-bers of the Executive Branch, in contrast, might be horrible in light of the coarsening of political discourse that that institution engen-ders.31 By contracting the relevant field of inquiry, the bottom up

28 See, e.g., Peter H. Aranson et al., A Theory of Legislative Delegation, 68 CORNELL L. REV. 1 (1983).

29 See, e.g., Seidenfeld, supra note 20.

30 16 U.S.C. 1531-1544 (1994).
31 Compare, e.g., Cass R. Sunstein, What’s Standing After Lujan? Of Citizen Suits, “Injuries,” and Article III, 91 MICH. L. REV. 163, 221-22 (1992) (giving qualified endorse-ment to citizen suits to overcome political inertia in enforcement of environmental laws), with Cass R. Sunstein, Bad Incentives and Bad Institutions, 86 GEO. L.J. 2267 (1998) (sav-aging the Independent Counsel law).

 WEBGALLEYS.DOC 05/05/99 5:44 PM


approach makes it much easier to get a handle on the relevant normative and institutional issues.

But whatever strategy one uses to justify democracy, the need for a strategy of some kind clearly undermines the democracy cri-tique of delegation. That critique sees democracy as normative for the interests and values served by delegation: no matter how use-ful we might otherwise believe delegation to be, we are obliged to reject it because it’s antidemocratic. But this argument, it should now be clear, has things completely backwards: it’s policy that’s normative for democracy, because without some independent normative standard we can’t make a justified choice between the competing conceptions of democracy served and disserved by delegation. An argument against delegation based on democracy will thus never be genuinely independent of some other normative argument about the desirability of delegation, and if we conclude that delegation is normatively justified, democracy will never be a grounds not to delegate.


Democracy critics usually see their arguments against delega-tion as constitutionally grounded. However, adding the Constitu-tion doesn’t make the democracy critique any more persuasive.

This is especially obvious when democracy is invoked as a trump card in constitutional arguments. The constitutional argu-ments for and against delegation are evenly matched: critics point to the Articles I and II Vesting Clauses32 and to Montesquieu’s classic theory of separation of powers;33 proponents to the Neces-sary and Proper Clause34 and to the Framers’ institutional pragma-tism as evidenced in the text, the Federalist Papers, and contempo-raneous state practice.35 At this point, critics tend to argue that any uncertainty should be resolved against delegation on grounds of democracy, since the Constitution is clearly a democratic

char-32 U.S. CONST. arts. I, 1; II, 1.
33 MONTESQUIEU, THE SPIRIT OF THE LAWS, bk. XI (Anne M. Cohler trans., 1989).
34 U.S. CONST. art. I, 8, cl. 18.
35 These arguments trace their lineage to the debate between Chief Justice Taft and Justice Brandeis over the constitutional propriety of restrictions on the President’s power to remove executive officers. Compare Meyers v. United States, 272 U.S. 52, 116-18 (1926) (Taft, C.J.) (relying, inter alia, on Article I Vesting Clause and Montesquieu to in-validate restriction on the power of the President to remove executive officers), with id. at 246-49, 293-94 (Brandeis, J., dissenting) (relying, inter alia, on the Necessary and Proper Clause, contemporary state practice, and asserted pragmatism reflected in the Federalist Papers to uphold such restriction on the removal power).

 WEBGALLEYS.DOC 05/05/99 5:44 PM

806 CARDOZO LAW REVIEW [Vol. 20:795

ter. But this argument is tractionless once it’s seen that democracy is a contested concept and that delegation simultaneously pro-motes some conceptions of democracy and impedes others.

For the Constitution to save the democracy critique, the crit-ics need to show that the Constitution compels a particular con-ception of democracy, and that that conception rules out delega-tion. But they can’t, for two reasons. First, the structural provisions of the Constitution can themselves plausibly be read as embodying competing conceptions of democracy. The popular election of representatives and (now) senators, for example, attests to the centrality of pluralist preference aggregation; bicameralism and the President’s participation in legislation through present-ment demonstrate commitment to civic republican deliberation; and, depending on how we conceive of constitutional interpreta-tion, the power of the Judiciary to engage in judicial review pre-serves communitarianism. Second, as I’ve shown, each conception of democracy is itself contestable and, in one version or another and in one circumstance or another, compatible with delegation. Consequently, to justify selecting from among all of these concep-tions of conceptions of democracy one that rules out delegation in all circumstances requires resorting to some normative considera-tion outside of democracy. If that normative consideration exists, it is that normative consideration, and not any constitutional prin-ciple of democracy that is condemning delegation. If that norma-tive consideration doesn’t exist—if delegation makes sense, nor-matively, all other things considered—then no constitutional principle of democracy can justify not delegating.

Democracy, I’ve argued, can’t help us to decide whether to be for or against delegation. The reason is that the concept of democ-racy is itself in need of too much help from independent normative considerations to generate dispositive answers about institutional structure. There are multiple conceptions of democracy, some of which will be promoted and others of which will be frustrated by any particular delegation scheme. Choosing between these con-ceptions necessarily requires appeal to normative considerations outside the concept of democracy.

The critics of delegation may be right to see the appropriate-ness of delegation as one of the most vital issues confronting our democracy. But for better or worse, the concept of democracy can’t help us to resolve it.