1 Institutions for Implementing Constitutional Law Mark Tushnet1
This Essay considers institutions for implementing constitutional law understood as political law. After a brief discussion discussion of what it means to describe constitutional law as political law the Essay e!amines the ways in which institutions can be designed to respond to different "proportions# of politics and law in one$s understanding of constitutional law. The aim is primarily to map map understandings of constitutional law on to institutional designs in a way that illuminates the obser%able % ariations in institutions actually used to implement constitutional law. Larry &ramer has recently brought to the attention of '.(. constitutionalists an understanding of constitutional law that had been lost from %iew for a while in the 'nited (tates although it has remained close to the forefront of discussions of constitutional law elsewhere in the world.) *n that understanding constitutional constitutional law is political law. law. +oth terms on the right side of the e,uation matter. matter. Constitutional law is political political in a sense to be described shortly shortly but it is also law in the usual sense. Law is a set of normati%e rules and principles designed to guide decision in particular cases pursuant to a disinterested -"neutral# to use a term familiar from '.(. constitutional discussions application of the rules and principles to the the facts at hand. In addition the guidance law pro%ides is is reasonably well/defined0 no doubt there is a range of choices a%ailable to sincere
Carmack a aterhouse terhouse 2rofessor of Constitutional Law 3eorgetown 'ni%ersity Law
) interpreters but that range is reasonably small. small.4 Constitutional law could be understood to be ordinary albeit supreme law. Marbury v. Madison e!emplifies Madison e!emplifies that understanding. The %iew that constitutional constitutional law is political law is different5 different5 "2olitical# does not mean -merely supreme. 6ather constitutional law is political political law for se%eral other reasons. Consider first a standard distinction between law and politics. politics. 2olitical actors empowered to adopt statutes and regulations e!ercise a relati%ely unconstrained discretion.7 They are charged with with enacting statutes that ad%ance the general welfare for e!ample or that are compatible with an e!pansi%ely understood police power. power. Constitutional law is political in part because the range of interpreti%e discretion with respect to constitutional constitutional rules and principles is closer -in si8e to the range of discretionary discretionary action that legislators ha%e than to the range of discretion that e!ists with respect to ordinary law. law. In this sense as continental legal theorists emphasi8e a constitutional interpreter is a special type of legislator.9 In addition constitutional law is political law because it deals with what :rank 4
I do not ha%e a metric for determining the si8e of the range of choice though.
This is not the only distinction between law and politics but it is the one that matters
I belie%e that there is something to the idea that constitutional interpreters are legislators
because they ha%e a relati%ely wide range of discretion in interpretation whereas interpreters of ordinary -non/political -non/political law ha%e a smaller smaller range of discretion. discretion. I also belie%e though that this idea should not carry a great deal of weight because the range a%ailable to interpreters of ordinary law is wider than co ntinental legal theorists belie%e it to be.
4 Michelman calls the laws of law/making.; These are the rules that go%ern the processes by which ordinary laws are made. It is law ;
*ne standard distinction between law and politics is howe%er ruled out by treating
constitutional law as political law. That distinction treats politics as a forum for -mere preference aggregation and law as a location of deliberations that include some elements other than preference aggregation -and that on what is probably the pre%ailing %iew e!cludes preference aggregation altogether. Treating constitutional law as political law strongly suggests that the political component of constitutional law is different from preference aggregation although perhaps the political component could be understood to be that portion of law if there be one that does implicate mere preference aggregation. Combining the latter %iew with standard notions that ba r ?udges from implementing their preferences suggests that constitutional law as political law should not be implemented
7 though because it is not nearly as discretionary as other law/making processes. Clearly constitutional law can ha%e its political and legal components in any number of combinations. :or con%enience in the this Essay$s mapping e!ercise I di%ide the possibilities into two types and then the latter type into two groups. The two types are institutions of constitutional change through formal amendment and similar action and institutions for the consideration and application of e!isting fundamental law. ithin the latter type which includes importantly systems of ?udicial re%iew for constitutionality the political component of constitutional law has the larger role in the first group I consider the legal one the larger role in the [email protected] ritten constitutions can be easy to amend or hard to amend. Consider first a constitution that is ,uite easy to amend = in the limit of course by a simple ma?ority %ote of the usual legislati%e ,uorum. (uch a constitution is simultaneously entirely legal and entirely political. Its legal component is e!pressed when it is applied its political component when it is amended perhaps in response to an une!pected or disfa%ored application. As the difficulty of amendment increases so does the legal component of constitutional law. 2olitics will still play a large role in determining fundamental law when a constitution can be amended by a ,ualified ma?ority of a sitting legislature -a ma?ority of the house rather than a ma?ority of a ,uorum or a super/ma?ority of the by ?udges. @
My e!amples of how different conceptions of constitutional law map on to different
institutional forms are ,uite styli8ed. :or e!ample in discussing parliamentary supremacy I ignore the complications introduced b y the effects in 3reat +ritain of the uman 6ights Act [email protected] and 3reat +ritain$s agreement to be bound by the decisions of the European Court of uman 6ights.
9 house and e%en when it can be amended by ma?ority action in successi%e parliamentary sessions. +eyond those forms though the legal component begins to play the predominant role. There are I think two reasons for making constitutions difficult to amend. The first is a desire for stability in fundamental law. And yet it is not immediately apparent why stability in fundamental law needs to be secured by pro%isions making the constitution difficult to amend. After all legislators able to amend the constitution easily could nonetheless recogni8e on their own the importance of stability in fundamental law. (uch legislators would amend the constitution only when they thought the benefits of change e!ceeded the costs of instability.B The second reason for making constitutions difficult to amend is precisely that a difficult amendment process is a way of demonstrating that the constitution is more legal than political. In the 'nited (tates for e!ample there is a rhetoric that opposes constitutional amendment in principle but resting instead on the ground that particular proposed amendments are bad ideas that rests on the assertion that getting into the habit of amending the Constitution would lead the people to think of the Constitution less as law and more as a mere instrument of politics. 1 Constitutional amendments change the constitution permanently. There are B
2erhaps the idea is that legislators will percei%e short/run benefits from amending the
constitution and under%alue the long/run costs in instability from doing so. Det the idea of instability as a long/run phenomenon sits uneasily with the ob%ious fact that society and its laws change without anyone being concerned that such change e%idences instability. 1
Cite to (ulli%an
; de%ices for changing it pro tanto with respect to a particular proposal at hand. These de%ices too respond to different conceptions of the role of the political in constitutional law and can be organi8ed by identifying the triggering e%ent that leads to displacing the e!isting constitutional rule. The most modest e%ent is a sense arising from the culture of constitutionalism that the proposal is inconsistent with the constitution. Legislators can respond to that sense by denying the e!istence of an inconsistency by modifying the proposal or by acknowledging the inconsistency and concluding that present e!igencies ?ustify a one/time departure from the constitution.11 In one of its uses (ection 44 of the Canadian Charter of 6ights ser%es this function. (ection 44 authori8es legislatures to make legislation effecti%e -for no longer than fi%e years notwithstanding its inconsistency with some of the rights guaranteed by the Charter. The Canadian (upreme Court has held that (ection 44 can be used prospecti%ely that is before any court has actually found such an inconsistency. 11
The +ritish anti/terrorism legislation adopted in )1 illustrates this possibility with
the peculiar twist that the constitutional norms applied are e!ternal to the +ritish legal system. The uman 6ights Act [email protected] directs +ritish courts to construe legislation to be consistent with the European Con%ention on uman 6ights where such a construction is fairly possible and to make a declaration that legislation is incompatible with the Con%ention when it cannot be so construed -and is in the court$s ?udgment incompatible with the Con%ention. *ne pro%ision of the proposed anti/terrorism legislation was clearly incompatible with the Con%ention. 2ursuant to Con%ention pro%isions the +ritish go%ernment issued a declaration derogating from the rele%ant Con%ention pro%ision with respect to the legislati%e proposal -that is declaring that the Con%ention pro%ision was inapplicable. citesF
> 2rospecti%e uses of (ection 44 are rare and in the Canadian constitutional culture seem likely to occur only when the legislation$s proponents belie%e that the courts would find the legislation in enacted to %iolate Charter rights.1) (ection 44 used prospecti%ely displaces the otherwise applicable constitutional law in the ser%ice of politics. There are also more formal institutions to trigger that sort of reconsideration. A legislature might ha%e a committee on constitutional matters charged with %etting legislati%e proposals for constitutionality. a%ing concluded that a proposal is inconsistent with the constitution the committee will notify the legislature which again has the opportunity to respond in the ways I ha%e mentioned. The +ritish 2arliamentary Goint Committee on uman 6ights is an e!ample of a legislati%e committee on constitutional matters. The Committee consists of members from the ouse of Commons and the ouse of Lords.14 It is charged with "e!amining matters relating to human rights in the 'nited &ingdom# and has taken a fairly acti%e role in %etting proposed legislation. Although it is a relati%ely recent inno%ation it appears to ha%e worked reasonably well in bringing constitutional problems with proposed legislation to the attention of members of 2arliament. Legislati%e committees on constitutional matters are of course composed of members of a go%ernment$s political branches and so are likely to gi%e the political 1)
:or e!ample (ection 44 was in%oked prospecti%ely by the go%ernment of Alberta to
insulate its statutes limiting marriage to heterose!uals from a Charter challenge that Canadian constitutionalists fully e!pect to succeed within a few years with respect to some other pro%ince$s marriage statute. citeF 14
At present si! Committee members are from the Labour 2arty two are Liberal
@ component of constitutional law a large role in their deliberations. Det the members chosen to ser%e on such committees and willing to do so are likely to think that the constitution has a substantial legal component as well. And a legislature that sets up such a committee is = by that %ery fact = indicating its %iew that the legal components of constitutional law deser%e more attention than the legislature would gi%e them on its own.17 *ne can increase the legal component e%en more by charging a court with the responsibility to determine whether legislation is consistent with the constitution and then gi%ing the legislature an opportunity to respond to the court$s action.19 The +ritish uman 6ights Act [email protected] is an e!ample. 'nder the Act courts ha%e the power to declare statutes incompatible with the European Con%ention on uman 6ights. (uch declarations -of which only a handful ha%e been made so far ha%e no immediate legal effect on any ordinary person$s rights. 6ather the declaration is a trigger for reconsideration of the legislation. The minister responsible for the legislation can do nothing in response of course or may introduce amendatory legislation in the ordinary course. More important for present purposes though the minister has the power to introduce such legislation and automatically place it on a fast track for adoption and e%en to amend the legislation himself or herself if doing so is urgently necessary sub?ect to parliamentary ratification.1; 17
Cite to political science literature on committee composition
*b%iously the legal component is high here because of the general %iew that court
decisions ha%e as large a legal component as one can ensure in institutional design. 1;
:inally it is worth noting that we can make a legislature$s power to o%erride e!isting
constitutional pro%isions pro tanto less or more difficult to e!ercise. As noted (ection 44
B I turn now to the second type of institution concerned with constitutional law as political law. (ystems of parliamentary supremacy ob%iously gi%e the political component of constitutional law a %ery large role. Indeed one might wonder whether such systems treat constitutional law as law at all. The answer though is that they do = or at least that they can. :irst a culture of constitutionalism places some limits on what legislators in such a system belie%e they can properly do. The difficulty is that such a culture is a form of long/term politics and legislators may succumb to the pressures of short/term politics and treat constitutional fundamentals in the same way they treat ordinary issues of public policy. *r put another way a culture of constitutionalism may not count as an institution for implementing constitutional law. (till culture may matter at least sometimes. The political ,uestions doctrine of '.(. constitutional law when understood in a particular way pro%ides an e!ample of how a constitutionalist culture operating through politics can constrain action and thereby enhance the effecti%e role of the legal component of constitutional law in a system that treats legislati%eHe!ecuti%e action as final. 2olitical ,uestions are ,uestions of constitutional interpretation fairly open to disagreement where the political branches$ interpretation is final. The problem for the political ,uestions doctrine has always been to identify the criteria for determining when a constitutional ,uestion is a political ,uestion. The best answer to that problem I belie%e is that political ,uestions are those where there is good reason to belie%e that the political branches ha%e strong incenti%es to has an automatic sunset pro%ision unlike all other legislation. *ne could re,uire that legislati%e o%errides of constitutional pro%isions be adopted by superma?orities or by ordinary ma?orities in successi%e parliamentary sessions or by any of the other de%ices that are used to make permanent amendments difficult to adopt. Cite to +ork$s proposal.
1 interpret the Constitution in a reasonably disinterested way.1> (ome hints of this answer come in the opinion of Gustice Gohn 2aul (te%ens in Walter Nixon v. United [email protected] That case in%ol%ed an impeachment of a federal ?udge. After the ouse impeached the ?udge the (enate con%ened a committee to hear li%e testimony. That committee prepared a report which pro%ided the basis for a %ote on con%iction or ac,uittal by the entire (enate. Gudge i!on argued that this procedure did not gi%e him the "trial# to which he was entitled by the Constitution. The Court held that the ,uestion of whether Gudge i!on recei%ed a trial within the meaning of the impeachment clauses presented a political ,uestion. In doing so the Court necessarily held that Gudge i!on could not challenge in court a "trial# that consisted of a coin toss. In response to that conclusion Gustice (te%ens wrote " 6espect for a coordinate +ranch of the 3o%ernment forecloses any assumption that improbable hypotheticals . . . will e%er occur. . . .#1B Gustice (te%ens$s thought here is that (enators will be constrained by something to pro%ide basic fairness in their procedures. The best candidate is a sense of constitutional responsibility that is a sense that with respect to constitutional fundamentals the (enate should treat the Constitution as ha%ing a significant legal component. That sense is induced by electoral considerations5 (enators reasonably fear that their constituents will retaliate against them at election time unless the (enators take their constitutional responsibilities seriously.) 1>
:or discussions of actual practice see Mark Tushnet
Gudicial 6e%iew ar%. G. on Legislation.
11 (econd and perhaps more important in the present conte!t "parliamentary supremacy# should not be taken to refer e!clusi%ely to systems in which e!ecuti%e and legislati%e power is combined as they are in 3reat +ritain. 6ather it refers to a system in which decisions about constitutional matters taken by ordinary political actors = not by ?udges = are final. (o a separation of powers system can gi%e a %ery large role to the political element of constitutional law. *ne aspect of The Federalist $s argument for a separation of powers system deser%es note here. (eparation of powers worked to secure constitutionalism according to The Federalist because it set ambition against ambition and linked the political interests of particular power/holders to the interests of the institution in which they held power. The libertarian/leaning interpretation of this argument is that separation of powers makes it hard to get the go%ernment to do anything which itself preser%es liberty. There is another interpretation though more resonant with the interests of this Essay. The purely self/ or constituency/oriented interests of particular power/holders cancel each other out when ambition is set against ambition. Det ambition means that power/holders want to get something done. In the absence of self/interest or constituency/interest all that power/holders can do is enact laws that ad%ance the public interest = or in the terms I ha%e been using promote fundamental constitutional %alues. A constitutionalist culture and separation of powers then are two institutions that fit reasonably well the conception of constitutional law as law that is primarily political. *ther possibilities deser%e mention as well. The pre%iously mentioned legislati%e standing committee on constitutional matters is one. Another is an office in the e!ecuti%e branch whose charge is to %et e!ecuti%e proposals and about/to/be/enacted legislation for constitutionality.)1 And finally a "court# understood as largely political might be gi%en )1
The *ffice of Legal Counsel in the '.(.
1) power to determine = either pro%isionally or finally = that legislation is unconstitutional. These institutions gi%e the legal component of political law a slightly larger role than pure parliamentary supremacy does. These agencies speciali8e in constitutional matters and therefore are likely to think of their "mission# as taking constitutional fundamentals more seriously than do the generalist politicians elsewhere in the go%ernment. Their staffs are likely to ha%e a large number of lawyers and may de%elop an institutional culture in which responsibility to the legal aspects of constitutional law plays a significant role. *f course they are members of the go%ernment and so are not likely to treat constitutional law as law only. :urther they report to ordinary politicians and their reports will be acted on or not with an eye to the political dimension of constitutional law.)) This is only to say that these institutions still fall in the category of institutions in which the political component of constitutional law plays a larger role than the legal one. Also in this category but mo%ing toward the "more law/like# end of the spectrum are constitutional courts on the &elsenian model. According to &elsen the political component of constitutional law meant that the ordinary courts could not properly be in%ol%ed in "?udicial# re%iew. Instead a court that speciali8es in constitutional law should e!ercise the power of re%iew. (peciali8ation and the concomitant remo%al from the administration of ordinary law would make such a court sensiti%e to the political e!ecuti%e branch agency with similar responsibilities. ))
2olitical circumstances might be such that the go %ernment would not be embarrassed
by a report from one of these agencies that a legislati%e proposal %iolates constitutional principles or might be willing to acknowledge the %iolation but enact the law that e!cepts the statute from ordinarily applicable constitutional principles.
14 component of constitutional law. :urther &elsen thought the constitutional court should be remo%ed from case/specific ad?udication where indi%idual claimants would assert that they had constitutional rights to be determined according to law. Indi%idual claims that is would induce re%iewing courts to reduce the political component of constitutional law and fa%or the legal component more than they should. :inally the decision/makers who compose a &elsenian constitutional court should be selected with an e!plicit concern for their sensiti%ity to the political component of constitutional law. *f the modern systems of constitutional re%iew probably only :rance$s fits the &elsenian model reasonably well.)4 :rench legal theorists conceptuali8e the Conseil Constitutionel as an e!tension of the legislati%e process thereby emphasi8ing the political o%er the legal. It e!ercises re%iew before legislation goes into effect entirely di%orced from case/specific ad?udication. 2olitics of the ordinary sort plays an e!plicit role in selecting members of the Conseil Constitutionel as is symboli8ed by the entitlement of former presidents of the 6epublic to sit on the Conseil .)7 Contemporary international courts pro%ide an e!ample of a nother institutional arrangement that responds to the dual components of constitutional law although of )4
*ther post/orld ar II constitutional courts depart in %arying degrees from the
&elsenian model. In particular most such courts ha%e some mechanism for consideration of indi%idual complaints either by reference from the ordinary courts when a constitutional ,uestion arises in connection with a pending case or by direct consideration of applications from aggrie%ed indi%iduals. In such courts the balance between the political and the legal components of constitutional law begins to tip in fa%or of the legal. )7
The point is symbolic because the entitlement has gone largely une!ercised.
17 course they do not enforce constitutional law in the usual sense. The most ob%ious recognition of the political in international tribunals is their composition5 6epresentati%es of the states whose actions are in ,uestion are entitled to sit on the tribunal considering the complaint. At present for e!ample each member/state of the European 'nion appoints a ?udge on the European Court of Gustice.)9 In other tribunals a temporary ?udge will be appointed when the court does not ha%e a ?udge from a state whose action is at issue. In addition the European Court of uman 6ights uses a legal doctrine that recogni8es the political component of the law it administers. This is the "margin of appreciation# doctrine which seems likely to b e increasingly emulated in international tribunals. According to that doctrine each nation is to be gi%en some leeway in its application and interpretation of fundamental human rights because nations face different arrays of social economic and political problems such that what is acceptable as an interpretation of fundamental law in one nation -perhaps one e!periencing high and persistent le%els of crime would not be acceptable in another. At the same time the actual ad?udications of international tribunals tend to be highly legali8ed. ith the introduction of -real courts into the institutional mi! I can turn to the second group of institutional arrangements I mentioned earlier. These are arrangements that emphasi8e the legal component of constitutional law o%er the political component essentially by gi%ing courts a large role in implementing the constitution.); Earlier I )9
That may change as the 'nion$s membership e!pands although my understanding is
that those responsible for institutional design in the 'nion e!pect the ECG to grow. );
These courts could be speciali8ed constitutional courts or generalist courts. :or reasons
suggested earlier generalist courts are likely to gi%e the legal component of constitutional
19 discussed prospecti%e uses of Canada$s (ection 44 procedure but more commonly (ection 44 is seen as pro%iding opportunities for legislators to respond to ?udicial rulings.)> The idea is that the Canadian courts can approach constitutional interpretation in a relati%ely legali8ed way knowing that if they under%alue the political component of constitutional law the political branches can read?ust the balance by o%erriding the courts$ interpretation. Constitutional law as a whole then consists of ?udicial interpretations which treat constitutional law as mainly ordinary law and legislati%e decisions to o%erride or not which treat constitutional law as mainly po litical. That way of understanding the power to o%erride allocates the different components of constitutional law to different institutions. (ection 44 says that legislatures can declare legislation effecti%e notwithstanding Charter pro%isions -rather than notwithstanding ?udicial interpretations of the Charter. This suggests that the legislature is insisting on implementing the statute despite its inconsistency with the Charter because it regards the statute$s policy as more important than the Charter$s law. (ection 44 treats constitutional law as articulated by the courts as almost entirely legal but constitutional law as articulated by the Canadian go%ernment taken as a whole inserts a political element through the possibility of the legislati%e o%erride for policy or political reasons. *ne who belie%es that constitutional law is political law might be concerned with the incenti%es (ection 44 pro%ides courts and legislatures on this conceptuali8ation. There is howe%er an alternati%e conceptuali8ation that uses (ection 44 to show that constitutional law is political law through and through and that its different components need not be allocated to different institutions. *n the alternati%e law a larger role than they gi%e the political component. )>
(ee "dialogue# literature.
1; conceptuali8ation a legislature$s o%erride is not a -mere policy ?udgment inserting political considerations into an otherwise legali8ed constitution. 6ather it is an e!pression of disagreement with the courts$ interpretation of the constitution. That is a legislature using (ection 44 is not necessarily making a statute effecti%e notwithstanding the Charter but is making it effecti%e notwithstanding an erroneous Charter interpretation pro%ided by the courts. The legislature$s action is ?ust as legali8ed as the courts$. This alternati%e conceptuali8ation is a useful correcti%e to the final mapping of constitutional law as political law on to institutional forms. That mapping is of course the '.(. system of ?udicial re%iew which I call strong/form re%iew.)@ In strong/form re%iew courts$ interpretations of the constitution are final and binding = in terms of political morality if not in terms of enforceable law = on all political actors. The '.(. (upreme Court$s recent decision in%alidating the 6eligious :reedom 6estoration Act e!emplifies strong/form re%iew.)B There the Court insisted that Congress lacked the power to specify the content of constitutional norms differently from the way the Court itself did e%en where Congress$ specification could not be dismissed as entirely unreasonable.4 The tradition of strong/form re%iew goes back a long way though. *ne can find it in what is perhaps the easiest reading of the passage in Marbury v. Madison that has been taken to articulate the political ,uestions doctrine. Chief Gustice Marshall wrote5 The pro%ince of the court is solely to decide on the rights of indi%iduals not to )@
:or additional discussion see Mark Tushnet Alternati%e :orms of Gudicial 6e%iew
Mich. L. 6e%. -forthcoming. )B
City of +oerne %. :lores cite.
The most noted e!pression of strong/form re%iew is Cooper %. Aaron cite.
1> in,uire how the e!ecuti%e or e!ecuti%e officers perform duties in which they ha%e a discretion. Juestions in their nature political or which are by the constitution and laws submitted to the e!ecuti%e can ne%er be made in this court.41 The most straight/forward reading of this passage has Marshall distinguishing between ,uestions of law which implicate indi%idual rights and political ,uestions which do not.4) The Constitution that is is a legal document remitted to the ordinary courts for ordinary interpretation when indi%iduals raise claims that their rights ha%e been %iolated. Constitutional law has no political component at all0 politics is the realm of the discretion conferred by the Constitution on the political branches. If howe%er constitutional law is -by definition political law strong/form re%iew might cause some problems unless it is tempered with some other institutional de%ices. (trong/form re%iew allocates the political and legal components of constitutional law to different institutions. As noted in connection with the first understanding of the (ection 44 o%erride procedure described abo%e doing so may create incenti%es that lead constitutional law to be o%erlegali8ed and underpolitici8ed. The 'nited (tates lacks a (ection 44 procedure that tempers strong/form re%iew but it contains some institutional de%ices aimed at ensuring that the courts ha%e some incenti%es to treat constitutional law as at least in part political. These are the mechanisms usually enumerated as ones of political ontrol of the courts5 Congress$s 41
*n this reading the reference to ,uestions that are submitted to the e!ecuti%e b y the
constitution is e,ui%alent to a statement that the e!e cuti%e$s action in resol%ing such a ,uestion implicates no legal rights.
[email protected] power to regulate the ?urisdiction of the federal courts the impeachment power and the power of nomination and confirmation. The first two may seem primarily as methods of retrospecti%e control or responses to decisions by the courts and in that they seem to resemble the retrospecti%e uses of the (ection 44 power. +ut there is an important difference. An o%erride used to correct an erroneous ?udicial decision need not affect the ?ustices who made that decision in the slightest0 they need not e%en be embarrassed by the o%erride to the e!tent that they take it to e!press only disagreement on a matter about which reasonable people can disagree. The e!istence of a power to o%erride ?udicial decisions retrospecti%ely that is need ha%e no incenti%e effects on ?udges as they consider what to do ne!t. The impeachment power is ob%iously different because it gi%es the legislature a power to place a ?udge$s continuation in office in ,uestion. The power to control ?urisdiction has some incenti%e effects as well albeit more modest ones5 The threat of losing work may be significant to a ?udge who has taken the position to do something.44 The retrospecti%e powers do ha%e some forward/looking effects. +ut in the 'nited (tates those powers ha%e fallen into disuse to the point that a rational ?udge looking forward should not gi%e any more weight to them than he or she does to the possibility that a legislator dismayed at a ruling would hire an assassin to kill the ?udge. The problem with the powers o%er nomination and confirmation is somewhat different. These powers operate as screening de%ices at the point of entry to the ?udicial role. They can be used to ensure the selection of ?udges who understand constitutional law to ha%e a political component as well as a legal one. They cannot influence the beha%ior of ?udges once seated. :or ?udges to treat the political component of 44
6ecall the interpretation of The Federalist offered earlier.
1B constitutional law with appropriate seriousness then the president and Congress must insist on choosing ?udges who already ha%e and are likely to continue to ha%e the appropriate understanding of constitutional law. ith respect to the latter one might think that something like a minimum age ,ualification would be desirable. The theory would be two/fold5 An older nominee has a longer track record on which the president and Congress can base their ?udgment about the candidate$s understanding of constitutional law and probably more important an older nominee$s %iews are less likely to change because of the conser%atism associated with age. e might return to &elsen$s %iews in addressing the concern that nominees ha%e the right understanding of constitutional law. orking within a ci%il law tradition in which ?udges were members of a certain kind of bureaucracy &elsen thought that ?udges on a constitutional court ought to be drawn from outside the ?udiciary precisely to ensure that such ?udges take seriously constitutional law$s political component. &elsen$s position in its strongest form seems inappropriate for a system of strong/form re%iew like the 'nited (tates$ particularly where as in the 'nited (tates the constitutional court is also a generalist court. +oth the generalist character of the court and perhaps more important the system$s commitment to a strong/form system of ?udicial re%iew suggest that the ?udges on such a court should treat the legal component of constitutional law as the predominant one. hat can be said though is that these commitments must be tempered a bit to ensure that the ?udges also understand that constitutional law is political law. 2robably the best that can be done is to de%elop a norm for nominations and confirmations that ?udicial e!perience is not a prere,uisite for ser%ice on a court with substantial responsibility for constitutional law and that the desirability of ?udicial
) e!perience in a nominee will %ary depending on the composition of the court already in place. Det this set of institutions seems incomplete. It gi%es courts some incenti%es to treat constitutional law as political. It does not howe%er contain arrangements that gi%e the legislature incenti%es to treat constitutional law as law. 2erhaps though we might return to points made earlier about systems of parliamentary supremacy. A culture of constitutionalism and the institutions of separated powers may gi%e political actors the appropriate incenti%es. It would be con%enient were I able to conclude by arguing that some particular mi! of the legal and the political in constitutional law was normati%ely desirable but I cannot. Instead in the spirit of the descripti%e effort I ha%e made I end by obser%ing that different constitutional systems appear to reflect co mmitments to different mi!es at any one time. A few weak normati%e conclusions may flow from that obser%ation. :irst institution designers probably should attempt to ensure that the institutions of constitutional re%iew correspond to the mi! that their po litical/legal culture deems correct. (econd it seems likely that within any constitutional system the mi! that seems desirable may change from time to time. Institution designers can respond to the possibility of change in se%eral ways. They might make it relati%ely easy to change the institutions of constitutional re%iew.47 Alternati%ely they might be attenti%e to the ways in which the political/legal culture affects the way in which the institutions they put in 47
That course seems acceptable only if their political/legal culture accepts the %iew that
whate%er is true about other matters the institutions of constitutional re%iew are almost entirely political in nature.
)1 place are likely to operate in fact. +ut as noted earlier such cultural factors may be particularly difficult to build into relati%ely stable institutional designs. In the end then we may face a situation in which the institutions of constitutional re%iew recurrently come into tension with the political/legal culture within which those institutions operate.