Indiana Law Review Karcher v. Daggett: The Supreme Court Draws the Line on Malapportionment and Gerrymandering in Congressional Redistricting I. Introduction The framers of the United States Constitution were very explicit as to how the seats in the House of Representatives were to be apportioned among the several states. 1 The framers omitted, however, the standards that the states should uphold when drawing the congressional districts once the House seats had been apportioned. That task fell upon the United States Supreme Court, which has read into article I, section 2 of the Constitution certain guidelines with respect to redistricting. This Note will review the apportionment process and the Supreme Court's involvement in redistricting. 2 The recent case of Karcher v. Daggett , 3 in which the Court held that New Jersey's congressional district plan was unconstitutional because its .6943% interdistrict population variance4 was unjustified, will then be discussed and analyzed at length. Finally, the Karcher decision will be used as a standard to assess the con- stitutionality of the Indiana congressional district plan enacted after the 1980 census. •U.S. Const, art I, § 2, cl. 3: Representatives and direct Taxes shall be apportioned among the several States which may be included within this Union, according to their respective Numbers .... The actual Enumeration shall be made within three Years after the first Meeting of the Congress of the United States, and within every subsequent Term of ten Years, in such Manner as they shall by Law direct. The Number of Repre- sentatives shall not exceed one for every thirty Thousand, but each State shall have at Least one Representative .... , 2Every ten years, state legislatures redistrict both federal and state legislative districts. Thus, the legislators draw the districts of the United States House of Representatives as well as the districts of the state's House and Senate. This Note will focus on the constitu- tional requirements state legislatures must observe in drawing federal congressional districts. The drawing of federal congressional districts must comport with the United States Con- stitution, but the drawing of state legislative districts has been treated differently by the Supreme Court. See infra note 71. 3 103 S. Ct. 2653 (1983). 4A state's total interdistrict population variance is the percentage difference between the smallest district's population and the average district population plus the percentage difference between the largest district's population and the average district population. Im- agine a state with two congressional districts with populations of 10 and 14. Because the state's total population is 24, and it has two districts, the average size district for this state is 12. The smallest district's population, 10, is 16.7% lower than 12. The largest district's population, 14, is 16.7% higher than 12. Thus, this state's total interdistrict population variance is 16.7% + 16.7% = 33.4%. Notice that this method could exaggerate the variance in a congressional district plan 651 652 INDIANA LAW REVIEW [Vol. 17:651 II. Reapportionment and Redistricting A. Reapportionment Reapportionment refers to the process of assigning each state the number of congressional representatives to which it is entitled. 5 With the bicameral legislature compromise of 17876 came the troubling question of how many House seats there should be and how those seats should be distributed among the various states. At the Constitutional Conven- tion, the framers formed a committee which settled on a House seat distribution plan and incorporated it into the Constitution. 7 However, ar- ticle I, section 2 did not specify any guidelines for future apportionments. Therefore, Congress passed the first apportionment bill after the 1790 census. 8 Washington felt that this bill was unconstitutional because the apportionment scheme was not based on the population of the states, and because it allotted eight states more than one representative for every 30,000 persons, contrary to article I, section 2, clause 3; therefore, Washington exercised the first presidential veto on this bill. 9 The reap- portionment bill which was finally approved based the distribution of House seats on the population of the states, allotting one house seat for every 33,000 persons. 10 Various refinements in the reapportionment process occurred in the nineteenth century, particularly with respect to the structure of the districts themselves. The 1842 Reapportionment Act required that House members be elected from districts composed of contiguous territory equal in number to the number of representatives to which that state was entitled, with since only extremes, the least and most populous districts, are used. For example, if a state had seven districts, five of which had identical populations, the total interdistrict popula- tion variance would only take into account the two districts above and below the average sized district. However, a state must justify any variance, no matter how small. Kirkpatrick v. Preisler, 394 U.S. 526 (1969). Therefore, this exaggeration would not necessarily place a higher burden of justification on the state. The courts have also referred to the interdistrict population variance as the maximum population deviance. 5 U.S. Const, art. I, § 2, cl. 3. See supra note 1. This compromise was between large and small states, resulting in the creation of two legislative houses. The upper house would be composed of two delegates from every state, regardless of its size; and the lower house would be composed of delegates assigned in number to the states on the basis of population. B. Mitchell, A Biography of the Constitution of the United States 70 (1964). 'Id. at 69-72. See U.S. Const, art. I, § 2, cl. 3. The precise apportionment of represent- atives in the Constitution was premised on little more than an estimate of each state's popula- tion, since reliable population figures were unavailable. L. Schmeckebter, Congressional Apportionment 107 (1941) [hereinafter cited as L. Schmeckebter] . 'L. Schmeckebier, supra note 7, at 107. 9 Id. at 108. '"Id. Act of April 14, 1792, 1 Stat. 253. Reapportionment of House seats was done after each decennial census, the custom being to give from one to three seats to any state entering the Union between censuses. L. Schmeckebier, supra note 7, at 117-22. 1984] CONGRESSIONAL REDISTRICTING 653 only one representative per district allowed." The Reapportionment Act of 1872 added the requirement that districts contain, as nearly as prac- ticable, an equal number of inhabitants. 12 Finally, the Act of 1901 added the requirement of compactness. 13 These requirements threatened the hold which rural state legislators had on the redistricting process. 14 Rural areas were often over-represented in state legislatures, 15 and because population and compactness re- quirements were previously not included in congressional reapportionment statutes, rural areas were often over-represented at the congressional level as well. The requirements codified by Congress made it likely that urban areas would receive greater representation. In an attempt to stop such a shift of power, rural congressional legislators blocked passage of a new reapportionment bill following the 1920 census. 16 A reapportionment bill was finally passed in 1929, 17 but the requirements of contiguity, popula- tion equality, and compactness were not included in the legislation. 18 This exclusion led to the first major Supreme Court case dealing with the struc- tural requirements of congressional districts, Wood v. Broom. 19 Broom, a resident of New Jersey, asserted that it was the right of every voter to reside in fairly drawn congressional districts. Consequently, he challenged a Mississippi statute which redrew congressional district lines after Mississippi's congressional delegation was decreased from eight to seven following the 1930 census. 20 The dispute arose because the statute "Act of June 25, 1842, ch. 47, 5 Stat. 491. See L. Schmeckebier, supra note 7, at 113. 12Act of Feb. 2, 1872, ch. 11, 17 Stat. 28. See L. Schmeckebier, supra note 7, at 118. 13Act of Jan. 16, 1901, ch. 93, 31 Stat. 733. Compact districts are those which con- tain the requisite population in as circular an area of the state as possible. Compactness can be measured in several ways. One method is the ratio of the perimeter of the district to the circumference of a circle with the same area as that district; another method is the ratio of the area of the smallest circle that could be drawn around the district. B. Morrill, Political Redistricting and Geographic Theory 22 (1981). See also Karcher v. Daggett, 103 S. Ct. 2653, 2673 n.19 (Stevens, J., concurring). '"Congressional Quarterly, Congressional Districts in the 1970's 221 (2d ed. 1974) [hereinafter cited as Congressional Districts in the 1970's], l5 This was due to state legislative districts being based primarily on geographical boun- daries rather than population. 1 Congressional Districts in the 1970's, supra note 14, at 221. "Act of June 18, 1929, ch. 28, 46 Stat. 26 (codified as amended at 2 U.S.C. § 2a (1982)). While the size of the House of Representatives remains a constant 435 members, the populations of the states with respect to one another change. Thus, the distribution of the 435 congressional seats changes. Pursuant to 2 U.S.C. § 2a(a), each state receives one seat automatically, and the remaining 385 seats are apportioned using the method of equal proportions. For a description of this complex formula, its effect on the reapportion- ment process, and an assessment of alternative methods of reapportionment, see L. Schmeckebier, supra note 7, at 125. "L. Schmeckebier, supra note 7, at 1-107. "287 U.S. 1 (1932). 20 Id. See also B. McKay, Reapportionment: The Law and Politics of Equal Representation 357 (1965). 654 INDIANA LAW REVIEW [Vol. 17:651 created congressional districts which were not compact and contained disparities in population. The Supreme Court held that since the 1929 Reapportionment Act did not incorporate the requirements of population equality, compactness, and contiguity included in earlier reapportionment bills, those requirements had expired. 21 Federal legislation presently in ef- fect calls for the Secretary of Commerce to evaluate redistricting plans to assure the implementation of unspecified neutral objectives. 22 B. Redistricting After the reapportionment process ends, the redistricting process begins. 23 Redistricting is the process a state legislature undertakes to divide the state into the number of districts Congress has apportioned to it. Redistricting has often been characterized by two practices which give the political party in power in a state legislature a higher probability of win- ning congressional seats. These practices are gerrymandering and malapportionment. 24 1. Gerrymandering.—Gerrymandering refers to the excessive manipula- tion of geographic boundaries of legislative districts to benefit a certain incumbent party. 25 This perversion of the redistricting process may take one of three forms: the majority party draws district lines to perpetuate the status quo and its position of power; bipartisan gerrymandering oc- curs when both parties act to protect the seats of their incumbent con- gressmen; or, the majority party's power over redistricting is traded for support of legislative proposals or wielded to punish political opponents. 26 Gerrymandering has at least four adverse effects on voters and the goal of fair and effective representation for all citizens. First, because many districts are virtually guaranteed to one party, the value of the voter's political participation is diluted. 27 Second, incumbents in these districts may be less responsive to the interests of all constituents since the prob- ability of defeat in an election is small. 28 Third, gerrymandering allows political parties to field weak candidates in districts where they will have little chance of losing thereby weakening the parties. 29 Finally, the political 21 287 U.S. at 8. 22Act of Aug. 31, 1954, ch. 1158, 68 Stat. 1019 (codified as amended at 13 U.S.C. § 141 (1982)). This evaluation is ultimately carried out by the judicial branch when redistrict- ing plans are the subject of litigation. 25 See supra note 5. 24Congressional Districts in the 1970's, supra note 14, at 228. 2 The term was coined in 1812 when the Massachusetts legislature drew a bizarrely shaped district which critics thought looked like a salamander. One critic dubbed the district the "gerrymander" after Elbridge Gerry, then Governor of Massachusetts. Id. at 225. 26Adams, A Model State Reapportionment Process: The Continuing Quest for "Fair and Effective Representation," 14 Harv. J. on Legis. 825, 839-41 (1977). 11 Id. at 843. "Id. 29 Id. at 844. 1984] CONGRESSIONAL REDISTRICTING 655 strength of racial, ethnic, and other minorities may be diluted by lump- ing them into as few districts as possible or by putting pockets of minorities into many districts. 30 In spite of these ill effects, gerrymandered plans seem to go unchallenged most of the time. This is probably due to recognition by the courts that the redistricting process is a political animal, and partisan motives are often behind the choices legislatures make in drawing district lines. 31 Eventually, however, the Supreme Court focused on one characteristic 32 of gerrymandering, malapportionment, to provide some guidance to state legislatures in the redistricting process. 2. Malapportionment.—Malapportionment refers to gross disparities in the populations of a state's congressional districts, 33 and is commonly measured by a state's total interdistrict population. 34 Originally, many states did not base district lines on population, 35 and population dispar- ities have frequently arisen from the failure of state legislatures to redis- trict over long periods of time. 36 The adverse effects of malapportion- ment are quite similar to the adverse effects of gerrymandering, 37 but malapportionment is particularly damaging to the goal of fair and effec- tive representation for all. Depending on whether a congressional district's population is larger or smaller than the state's average-sized dis- trict, the voting power of individuals in that district will be decreased or increased proportionally. Unlike gerrymandering, the opportunity to use malapportionment for political purposes has decreased in recent years be- cause United States Supreme Court decisions have placed severe restric- tions on interdistrict population variances. 38 The Supreme Court's involvement in assessing the constitutionality of malapportionment has undergone a major evolution. At first, the Court was reluctant to become involved in the redistricting process, which it viewed as purely political. 39 Later, however, the court became more active in this area, and began to elucidate a standard for congres- sional redistricting. 40 30 id. 3i See infra note 131 and accompanying text. "Gerrymandering has many components. The United States Supreme Court has iden- tified several. See infra note 131. "Congressional Districts in the 1970's, supra note 14, at 228. l *See supra note 4. "Congressional Districts in the 1970's, supra note 14, at 228. ib Id. inSee supra text accompanying notes 27-30. "See, e.g., Karcher v. Daggett, 103 S. Ct. 2653 (1983) (New Jersey's plan with a .7% variance struck down); White v. Weiser, 412 U.S. 783 (1973) (Texas' plan with a 4.13% variance struck down); Kirkpatrick v. Preisler, 394 U.S. 526 (1969) (Missouri's 5.97% variance struck down). 39See Colegrove v. Green, 328 U.S. 549 (1946). '"See, e.g., Wesberry v. Sanders, 376 U.S. 1 (1964); Kirkpatrick v. Preisler, 394 U.S. 526 (1969). 656 INDIANA LAW REVIEW [Vol. 17:651 a. Justiciability: 41 The Court hesitates.— The existence of gross malapportionment in Illinois prompted a Northwestern University political science professor to take court action. In Colegrove v. Green/2 Colegrove argued that the officers of Illinois should be restrained from conducting the congressional election because interdistrict population variances violated the equal protection clause of the fourteenth amendment/ 3 At the time, congressional districts in Illinois varied in population from 112,116 to 914,053, a total interdistrict population variance of over 264%. 44 In dismissing the action as not justiciable, Justice Frankfurter, writing for the majority, stated the traditional rationale why the Court would not act: "Nothing is clearer than that this controversy concerns matters that bring courts into immediate and active relations with party contests. From the determination of such issues this Court has traditionally held aloof." 45 Justice Black replied with a vigorous dissent. 46 Citing prior case law in support of justiciability, 47 Justice Black stated that "[n]o one would deny that the equal protection clause would also prohibit a law that would expressly give certain citizens a half-vote and others a full vote." 48 Colegrove was not well-received by many legal scholars impatient with the Court's position, who wanted some judicial action to correct the ex- treme population disparities which existed in the congressional districts of the states. 49 Over time, the complexion of the Supreme Court changed to include new members more inclined toward judicial action on the redistricting problem. 50 By 1962 only three members of the Colegrove Court remained, 51 when the landmark justiciability case of Baker v. Carr52 was decided. 41A justiciable controversy is one which is appropriate for judical determination. Black's Law Dictionary 777 (5th ed. 1979). The four categories of wowjusticiability are lack of ripeness, mootness, lack of party standing, and political questions. The Colegrove case presented a political question. For a discussion of the subcategories of political questions, see Baker v. Carr, 369 U.S. 186, 217 (1962). 42 328 U.S. 549 (1946). * l Id. The plaintiffs did not base their action on art. I, § 2 of the U.S. Constitution. Since the court decided that the issue was nonjusticiable, the theory of liability was prob- ably irrelevant. "Id. 4i Id. at 553. Actually, because Justice Rutledge's concurring opinion asserted that the controversy was justiciable but should be dismissed for want of equity, a majority of the Court (Rutledge and the four dissenters) disagreed with the plurality opinion as to justiciability. 46 Id. at 566. 47 Smiley v. Holm, 285 U.S. 355 (1932) (holding that the Constitution does not exempt redistricting statutes from a governor's veto). 48 328 U.S. at 569 (Black, J., dissenting). 49Congressional Districts in the 1970's, supra XiOtt 14, at 233. "Id. 5 'Only Justices Black, Douglas, and Frankfurter were on the Court in both Colegrove and Baker v. Carr. 52369 U.S. 186 (1962). 1984] CONGRESSIONAL REDISTRICTING 657 In Baker, a group of Tennessee citizens sued to enjoin elections, claim- ing that the state legislative districts violated the Constitution. Tennessee had not redrawn its districts in over fifty years, and by 1960 the Ten- nessee House districts had populations varying from 3,454 to 36,031, and Senate districts varying from 39,727 to 108,094. 53 The plaintiffs claimed that their votes were debased because they were from overpopulated districts and that this denied them equal protection under the law. 54 The district court, relying on Colegrove, dismissed the complaint for lack of subject matter jurisdiction, 55 but the United States Supreme Court reversed the judgment and remanded the case, holding that Colegrove was dismissed for want of equity, and not because the cause of action was nonjusticiable. 56 With the barrier of justiciability set aside, Baker made it clear that the Court would no longer shy away from involvement in the political process of redistricting. b. The "As Nearly As Practicable'" Standard: The Court steps in.— Although Baker was a case dealing with state legislative districts, not con- gressional districts, it paved the way for the Court to find that controver- sies concerning congressional districts were justiciable. Because many states' congressional districts still had gross interdistrict population variances, 57 it is not suprising that a landmark congressional redistricting case was de- cided by the United States Supreme Court only two years after Baker. The case was Wesberry v. Sanders, 5 * and the state was Georgia. Georgia's congressional districts ranged in population from 272,154 to 823, 860. 59 Voters in that state's most populous district claimed that Georgia's con- gressional districts violated 42 U.S.C. sections 1983 and 1988, since their votes were worth less than the votes of other Georgians. 60 5 'Id. "Id. at 204. 55 179 F. Supp. 824 (M.D. Tenn. 1959). 56369 U.S. at 234. On remand, the Tennessee districts were invalidated. 206 F. Supp. 341 (1962). The district court held that the equal protection clause requires that at least one house of a state legislature have districts based on population. 57The interdistrict population variances during the 88th Congress in states with more than one congressional district, in increasing order, were: Maine, 9%; North Dakota, 11%; Rhode Island, 14%; New Hampshire, 19%; Iowa, 23%; Massachusetts, 24%; Minnesota, 25%; Nebraska, 27%; New York, 30%; Missouri, 30%; West Virginia, 32%; Montana, 37%; Kansas, 38%; Washington, 41%; Idaho, 46%; North Carolina, 52%; Arkansas, 54%; Virginia, 57%; Utah, 57%; Oregon, 58%; Pennsylvania, 60%; Kentucky, 60%; Illi- nois, 65%; South Carolina, 65%; Louisiana, 67%; California, 69%; Mississippi, 72%; Connecticut, 73%; Wisconsin, 74%; Oklahoma, 77%; New Jersey, 82%; South Dakota, 93%; Indiana, 96%; Tennessee, 102%; Florida, 103%; Colorado, 105%; Maryland, 106%; Arizona, 107%; Ohio, 116%; Georgia, 140%; Michigan, 144%; Texas, 169%; Alabama, Hawaii and New Mexico elected all of their congressmen at-large. The variances were com- puted from population figures in Congressional District Data Book (Districts of the 88th Congress) (1964). i9 Id. 60 Id. Because voters in the smallest district could elect a representative with one third of the votes needed in the largest district, the votes of individuals in the smallest district 658 INDIANA LAW REVIEW [Vol. 17:651 Writing on behalf of the Wesberry majority, Justice Black interpreted article I, section 2 of the Constitution to require, "as nearly as . . . practicable," 61 that one man's vote in a congressional election be worth as much as another's. 62 The Court did not provide precise guidelines as to how much interdistrict population variation the Constitution would allow, saying only that While it may not be possible to draw congressional districts with mathematical precision, that is no excuse for ignoring our Constitution's plain objective of making equal representation for equal numbers of people the fundamental goal for the House of Representatives. That is the high standard of justice and com- mon sense which the Founders set for us. 63 The Supreme Court attempted to elucidate this "as nearly as prac- ticable" standard in Kirkpatrick v. Preisler. 64 The total variance involved in Kirkpatrick, 5.97%, was much smaller than the disparities involved in earlier cases; nevertheless, the Missouri congressional district scheme was struck down by the Supreme Court on article I, section 2 grounds. 65 Justice Brennan, writing for the majority, said that the "as nearly as prac- ticable" standard "requires that the State make a good-faith effort to achieve precise mathematical equality. . . . Unless population variances among congressional districts are shown to have resulted despite such ef- fort, the State must justify each variance, no matter how small." 66 Brennan' s mathematical standard elicited a number of responses from various members of the Court. For Justice Fortas, Brennan's method of determining the constitutionality of Missouri's congressional districts was too strict. Justice Fortas concurred in the judgment, but not in the stand- ard of near perfection: "[T]he majority's pursuit of precision is a search for a will-o'-the-wisp." 67 The dissenters68 believed that a variance of five percent was permissible. Justice White said that a variance of ten to fif- teen percent was the upper limit of constitutionality, 69 but Justices Harlan and Stewart disagreed with the use of a mathematical standard at all: were three times as powerful. Also, persons in the smallest district had three times as much congressional representation. 6l /d. at 7-8. This language is much older than the Wesberry case. Congress included this requirement in the Reapportionment Act of 1872, but it did not have the judicial back- ing of the Supreme Court until Wesberry. See supra text accompanying note 12. 62 376 U.S. at 7-8. "Id. at 18. "394 U.S. 526 (1969). 65 Id. "Id. at 530-31 (citation omitted). 67 Id. at 538 (Fortas, J., concurring). 68 Harlan, Stewart, and White dissented to Kirkpatrick in a companion case decided the same day, Wells v. Rockefeller, 394 U.S. 542, 549 (1969). 69 Id. at 553 (White, J., dissenting). 1984] CONGRESSIONAL REDISTRICTING 659 "[T]he Court's exclusive concentration upon arithmetic blinds it to the realities of the political process . . . . ,,7 ° The restriction of allowable interdistrict population variance under Wesberry and Kirkpatrick prompted swift state action in the 1970's to conform with these decisions. 71 In fact, every state with more than one congressional district dramatically reduced its interdistrict population variance after the 1970 census. 72 After the 1980 census, when seventeen congressional seats were reapportioned from the northeast and midwest to the south and southwest, 73 states made even greater efforts to achieve interdistrict population equality. 74 In fact, Michigan achieved almost perfect interdistrict population equality: sixteen of that state's eighteen congres- 10 Id. at 551 (Harlan, J., dissenting). 7 'The United States Supreme Court has been more lenient as to population requirements in state legislative districts. In Chapman v. Meier, 420 U.S. 1 (1975), the Court held that minor population deviations in such districts did not establish a prima facie constitutional violation, "[a]s contrasted with congressional districting, where population equality appears now to be the preeminent, if not the sole, criterion on which to adjudge constitutionality . . . ." Id. at 23 (citations omitted). See also Brown v. Thomson, 103 S. Ct. 2690 (1983), where the Court upheld Wyoming's state legislative district plan even though it embodied an 89% interdistrict population variance. The Court found that sacrificing population equality to allow one representative for the Wyoming county in question was a legitimate state interest. 72The interdistrict population variances of the 93rd Congress, in increasing order, were South Dakota, .01%; Utah, .02%; Connecticut, .04%; Wisconsin, .07%; Montana, .14% Nebraska, .15%; Idaho, .20%; Arizona, .22%; Oregon, .22%; Indiana, .23%; Rhode Island .24%; Arkansas, .27%; Florida, .28%; Louisiana, .33%; Kentucky, .40%; Oklahoma, .43% Maine, .46%; Ohio, .50%; Michigan, .54%; Missouri, .63%; Colorado, .64%; Iowa, . 65% Virginia, .68%; Alabama, .78%; West Virginia, .79%; New Hampshire, .96%; New Jersey .98%; Georgia, 1.1%; New Mexico, 1.2%; Illinois, 1.3%; Minnesota, 1.4%, Kansas, 1.6% Massachusetts, 1.6%; Pennsylvania, 2.2%; Maryland, 2.6%; New York, 2.7%; California, 2.8%; North Carolina, 3.8%; Mississippi, 4.1%; Texas, 4.9%; South Carolina, 8.2%; Ten- nessee, 8.3%; Washington, 8.5%; and Hawaii, 11.9%. These variances were computed from population figures in Congressional District Data Book (Districts of the 93rd Con- gress) (1973). "Florida gained four seats; Texas three, California two, and Tennessee, Washington, Colorado, Arizona, Oregon, New Mexico, Utah, and Nevada each gained one. New York lost five seats; Pennsylvania, Illinois, and Ohio each lost two; and Michigan, New Jersey, Massachusetts, Indiana, South Dakota, and Missouri each lost one. Congressional Direc- tory, 98th Congress 438 (1983). 74The interdistrict population variances of the 98th Congress, in increasing order, are Michigan, .0002%; Colorado, .0025%; Minnesota, .009%; Hawaii, .01%; Illinois, .03% Idaho, .04%; Arizona, .08%; Iowa, .10%; Florida, .13%; Wisconsin, .14%; Oregon, .17% Missouri, .18%; Mississippi, .21%; Nebraska, .23%; Pennsylvania, .24%; New Hampshire .24%; Texas .28%; South Carolina, .29%; Kansas, .34%; Maryland, .35%; California, .38% Louisiana, .42%; Utah, .43%; Connecticut, .48%; Oklahoma, .57%; Ohio, .61%; Nevada .68%; New Jersey, .70%; Arkansas, .77%; New Mexico, .87%; Massachusetts, 1.1% Washington, 1.4%; Kentucky, 1.4%; New York, 1.6%; North Carolina, 1.8%; Virginia 1.8%; Georgia, 2.0%; Indiana, 2.4%; Tennessee, 2.4%; Maine, 6.6%; Rhode Island, 7.8% Montana, 8.5%; West Virginia, 12.8%; and Alabama, 48%. These variances were com- puted from district population figures in Congressional Directory, 98th Congress (1983). 660 INDIANA LAW REVIEW [Vol. 17:651 sional districts have exactly the same population, while the remaining two each have but one person fewer. 75 Although the United States Supreme Court's redistricting decisions caused state legislatures to consider interdistrict population variance when drawing new district maps, the exact constitutional parameters were not yet settled. The Supreme Court had indicated that absent a good-faith effort to achieve interdistrict population equality, even minute variances had to be justified. 76 Yet, the Court also recognized that exact interdistrict population equality would be difficult, if not impossible, to achieve. 77 Thus, while it was clear an interdistrict variance of 5.97% was too large in Kirkpatrick, what percentage the Court would deem acceptable was unknown. Ironically, Indiana's variance of about 2.4% is the greatest (along with Tennessee) of the states which lost or gained seats after the 1980 census, yet Indiana's districts have not been attacked as unconstitu- tional; while New Jersey, with a comparatively minute variance of less than .7%, was the subject of the Supreme Court's most recent attempt to express the specific requirements of the "as nearly as practicable" standard. 78 III. Karcher v. Daggett and the New Jersey Plan Judicial involvement in the drawing of legislative districts in New Jersey occurred in twelve of the sixteen years immediately prior to the 1980 census; 79 thus, it was not surprising that the congressional district plan adopted by the New Jersey legislature after the 1980 census also became the subject of litigation. The census revealed that New Jersey had grown at a slower rate than many other states; 80 so after the 1980 appor- tionment, it lost one of its congressional seats. Consequently, an entirely new congressional district map had to be drawn with fourteen, rather than fifteen, districts. The map adopted by the Democratic-controlled New Jersey legislature was the Feldman Plan, 81 signed into law by the Democratic governor one day before his Republican successor took office. 82 Under the Feldman Plan New Jersey's fourteen congressional districts had an average population of 526,059; the largest district differed from 15 Id. at 92-99. 76 Kirkpatrick, 394 U.S. at 530-31. "Id. at 527. But see Karcher v. Daggett, 103 S. Ct. 2653, 2659 (1983). "Karcher v. Daggett, 103 S. Ct. 2653 (1983). "Torricelli and Porter, Toward the 1980 Census: The Reapportionment of New Jersey's Congressional Districts, 7 Rutgers Comp. & Tech. L.J. 141 (1979). See, e.g., David v. Cahill, 342 F. Supp. 463 (D.N.J. 1972) (holding that New Jersey's congressional districts, which had a total population variance of 51.54%, were patently unconstitutional). l0New Jersey's population increased by 2.7% from 1970 to 1980, while that of the nation as a whole increased by 11.5%. World Almanac and Book of Facts 207 (1983). "Karcher v. Daggett, 103 S. Ct. 2653, 2657 (1983). Feldman was the President Pro Tem of the New Jersey Senate. ,2N.J. Stat. Ann. §§ 19:46-4, -5 (West Supp. 1983-84). 1984] CONGRESSIONAL REDISTRJCTING 661 the average by about .27%, and the smallest differed from the average by about .43%. 83 Although these variances were quite small, plans with even smaller variances had been offered to the legislature but were re- jected, and the Feldman Plan became law. 84 The Feldman Plan eliminated one Republican district, paired Republican incumbents in one district, created a new district leaning Democratic, and removed some Republican territory from the third district which had a Democratic incumbent. 85 The Feldman plan was described as a ''four-star gerrymander that boast[ed] some of the most bizarrely shaped districts to be found in the nation." 86 Rather than challenge the Feldman Plan as a gerrymander, however, the plaintiffs attempted to show that the plan failed the "good-faith effort to achieve population equality" test of Kirkpatrick. The challengers, who included New Jersey's entire Republican congressional delegation, sought a judicial declaration that the plan violated article I, section 2 of the Constitution, and an injunction against New Jersey officials to prevent them from holding primary elec- tions under the districts in the Feldman Plan. 87 In the United States District Court of New Jersey, a three-judge panel, convened pursuant to federal statute, 88 denied the defendants' motion for summary judgment, and, relying largely on the two-step Kirkpatrick analysis, 89 held the Feldman Plan unconstitutional and granted the injunction. 90 By a thin margin, the United States Supreme Court af- firmed the district court's decision. 91 Justices Brennan, Marshall, O'Con- nor, Blackmun and Stevens were in the majority; and Justices White, Rehnquist, and Powell, and Chief Justice Burger dissented. The sharp split of the Court is further illustrated by the number of opinions writ- ten; besides Brennan's majority opinion, Stevens wrote concurring opin- ion, and White and Powell each wrote dissenting opinions. "Karcher v. Daggett, 103 S. Ct. 2653, 2657 (1983). Thus, the total interdistrict popula- tion variance is .43 + .27, or about .70%. "Id. See N.J. Stat. Ann. §§ 19:46-4, -5 (West Supp. 1983-84). 8 Congressional Quarterly, State Politics and Redistricting Part II 20 (1982). * 6 Id. 87Karcher v. Daggett, 103 S. Ct. 2653, 2657 (1983). 88 28 U.S.C. § 2284(a) (1982): "A district court of three judges shall be convened . . . when an action is filed challenging the constitutionality of the apportionment of congres- sional districts . . . ." Note the broad usage of the word apportionment, which covers redistricting as well. i9See supra text accompanying note 66. 90Daggett v. Kimmelman, 535 F. Supp. 978 (D.N.J. 1982), aff'd sub nom. Karcher v. Daggett, 103 S. Ct. 2653 (1983). This order was stayed pending appeal, and pursuant to 28 U.S.C. section 1253, the case was appealed directly to the U.S. Supreme Court, which noted probable jurisdiction. "Karcher v. Daggett, 103 S. Ct. 2653 (1983). After the Supreme Court remanded Karcher, the New Jersey District Court fixed February 3, 1984, as a deadline by which the New Jersey legislature was required to enact a new plan. This deadline passed and no new plan was enacted; therefore, the district court convened to choose a plan from those 662 INDIANA LAW REVIEW [Vol. 17:651 A. Brennan's Two-Level Inquiry for Malapportionment According to Justice Brennan's opinion, there are two levels of in- quiry to be undertaken when the constitutionality of a state's congres- sional district map is challenged. 92 Initially, the burden is on the challenger to show that the state did not make a good-faith effort to achieve precise mathematical equality. 93 If the challenger carries this burden, then the burden shifts to the state to show that precise interdistrict population equality was sacrificed to achieve some other legitimate state interest. 94 If the state fails to make such a showing, the plan will be declared unconstitutional. 1. "Functional Constitutionality" . —In an effort to circumvent this two-step analysis, New Jersey argued that the population of its congres- sional districts should be regarded as functionally equivalent and therefore exempt from the scrutiny normally present in challenges to congressional district plans. 95 New Jersey attacked the legitimacy of the census popula- tion figures which showed a comparatively large interdistrict population variance in the Feldman Plan; alternatively, New Jersey asserted that even if the census figures were correct, the variance in the Feldman Plan was small enough to be ignored. First, New Jersey argued that there was a systematic undercount in the census that was not uniformly distributed. 96 In other words, although a state might achieve precise mathematical equality based on census figures, in reality the population of the districts would not be equal since the cen- sus could reflect neither the exact population nor the precise distribution of the population within the state. Brennan thoroughly countered this argument: offered by the parties involved. Plans with districts similar to the Feldman Plan's districts, but with greatly reduced population variances, were rejected by the district court in favor of a plan with more compact districts. Daggett v. Kimmelman, 580 F. Supp. 1259 (D.N.J. 1984). The court-approved plan favored the Republican Party in New Jersey, therefore, the Democratic proponents of the Feldman Plan applied to the U.S. Supreme Court for a stay of the district court's order. This application was denied, Karcher v. Daggett, 104 S. Ct. 1691 (1984), but Justice Brennan dissented. Id. Brennan wrote that the district court had abused its discretion by not accepting the alternate plan which most closely resembled the Feldman Plan. The judgment of the district court was subsequently affirmed by the Supreme Court. Karcher v. Daggett, 52 U.S.L.W. 3873 (U.S. June 4, 1984) (No. 83-1526). Brennan referred to his dissent of the denial of application for stay, in dissenting to the affirmance of the district court's decision. 92 Id. at 2658. Brennan repeated the test he articulated in Kirkpatrick. "Id. "Id. "Id. "Id. at 2660-62. For a discussion of the political ramifications of the undercount see McKay, Constitutional Implications of a Population Undercount: Making Sense of the Census Clause, 69 Geo. L.V. 1427 (1981). 1984] CONGRESSIONAL REDISTRICTING 663 To the contrary, the census data provide the only reliable—albeit less than perfect—indication of the districts' "real" relative population levels. Even if one cannot say with certainty that one district is larger than another merely because it has a higher cen- sus count, one can say with certainty that the district with a larger census count is more likely to be larger than the other district than it is to be smaller or the same size. That certainty is suffi- cient for decisionmaking. 97 Second, the state argued that because the population variances in the Feldman Plan were smaller than the margin of error in the census, 98 the districts should be treated as functionally equivalent in population. 99 However, no de minimis 1 ™ figure was acceptable to Brennan, who noted that due to the arrival of computer technology, compliance with a stand- ard of precise mathematical equality would not be burdensome. 101 By rejecting New Jersey's theories, Brennan made it clear that the two-step analysis introduced in Kirkpatrick would be used in every challenge to a congressional district plan, no matter how small the plan's variance. Thus, the burden was on the challengers to show that the Feldman Plan was not the result of a good-faith effort to achieve precise interdistrict population equality, which if carried would shift the burden to New Jersey to justify the variances of the Feldman Plan. 2. Good-Faith Effort.—The challengers argued that the Feldman Plan was not a good-faith effort to achieve interdistrict population equality because other plans with smaller interdistrict population variances had been offered to the New Jersey legislature but were rejected. 102 Brennan agreed with the district court that this action by the legislature cast serious doubt on a determination that the Feldman Plan was a good-faith effort to achieve precise mathematical equality. 103 Additionally, Brennan held that the ease with which the district lines could be moved slightly to achieve smaller interdistrict population variances made it clear that the Feldman "Karcher v. Daggett, 103 S. Ct. 2653, 2662 (1983) (citation omitted). "The margin of error in the 1980 census is between l°7o and 2°7o. Id. at 2680 n.3 (White, J., dissenting). "Id. at 2658-59. 100(1De minimis" refers to the doctrine of de minimis non curat lex, or the law does not concern itself about trifles. Black's Law Dictionary 388 (5th ed. 1979). Thus, a de minimis percentage of interdistrict population variance is the point at which the Supreme Court would ignore that a variance existed at all. l01 Karcher v. Daggett, 103 S. Ct. 2653, 2659 (1983). ,02 Id. at 2662. l03 Id. at 2664. The Reock Plan contained a total deviation of .3250%, and only .2960% after it was amended. The DiFrancesco Plan had a total deviation of .1253%. The Hard- wick Plan contained a total deviation of .4515%. The Bennett Plan had a total deviation of .1369%, and the Kavanaugh Plan had a total deviation of .0293%. Daggett v. Kim- melman, 535 F. Supp. 978, 982 (1982), aff'd sub nom. Karcher v. Daggett, 103 S. Ct. 2653 (1983). 664 INDIANA LAW REVIEW [Vol. 17:651 Plan was not a good-faith effort. 104 Thus, the challengers carried their burden, and the burden shifted to New Jersey to justify the variances in the Feldman Plan. 3. Legitimate State Interests.—Brennan recognized that some legitimate state interests could justify the enactment of a particular redistricting plan when other plans with smaller variances were available, or could have been made available. Some permissible state interests identified by Bren- nan include making districts compact, 105 respecting established political boundaries, preserving the cores of prior districts, and avoiding contests between incumbent congressmen of the same party. 106 New Jersey's only attempt to justify the deviations in the Feldman Plan was to assert that the plan preserved the voting strength of minority groups. 107 Brennan flatly rejected that argument, finding no causal link between the asserted goal and population variances in districts with little minority strength to preserve. 108 Brennan said that the showing required for a legitimate state interest was flexible, and the factors to be weighed include the size of the deviation, the importance of the state interest, the consistency with which the redistricting plan reflected the asserted interest, and the availability of alternative plans with smaller deviations. 109 Bren- nan was silent regarding the legitimacy of the goal of preserving minority voting strength, and the district court expressly stated that because that goal was not supported by the facts, it did not have to reach the legitimacy question. 110 B. Problems with the Two-Level Inquiry for Malapportionment 1. Good-faith Effort.—Brennan's two-level inquiry contains both prac- tical problems and logical inconsistencies. The burden placed on the challengers of a redistricting plan is so small as to be almost nonexistent. The challengers in Karcher carried their burden as to New Jersey's lack of a good-faith effort to achieve interdistrict population equality by simply showing that the Feldman Plan could be modified to achieve a lower population variance. In bolstering this notion, Brennan noted that other plans with smaller variances were rejected by the legislature. 111 In fact, it was possible to transfer entire political subdivisions between districts in the Feldman Plan and achieve a lower interdistrict population ,04Karcher v. Daggett, 103 S.Ct. 2653, 2665 (1983). ]0iSee supra note 13. ,06 Karcher v. Daggett, 103 S. Ct. 2653, 2663 (1983). 107 Id. at 2664. l0i Id. at 2665. 109 Id. at 2663. "°Daggett v. Kimmelman, 535 F. Supp. 978, 982, aff 'd sub nom. Karcher v. Daggett, 103 S. Ct. 2653 (1983). '"Karcher v. Daggett, 103 S. Ct. 2653, 2662 (1983). 1984] CONGRESSIONAL REDISTRICTING 665 variance. 112 Thus, the Court found these factors as determinative that good- faith was not present. Because any redistricting map which does not have exact interdistrict population equality can have its variances decreased by shifting the lines slightly, the burden placed on the challenger is really no burden at all. Thus, Brennan's analysis will have the effect of requir- ing a state to justify any variance in its redistricting plan, since the absence of absolute interdistrict population equality establishes the challenger's prima facie case of lack of good-faith. The ease with which a challenger can carry the initial Karcher burden will further the Supreme Court's standard of absolute interdistrict popula- tion equality. In terms of political realities, the majority party in a state legislature should realize the ease with which the minority party can over- come the burden of showing lack of good faith. The majority party can be assured that its redistricting plan will be easily challenged unless it is one whose interdistrict population variance could not be decreased; that is, a plan with an interdistrict population variance of zero. Justice Brennan indicated that congressional district plans must be drawn in a good-faith effort to achieve interdistrict population equality, 113 and that the population variances in the plan must be unavoidable despite such an effort. 114 The unavoidability question must be answered by the state if the challenger carries the initial burden. However, Brennan im- plied that even if the challenger cannot show a lack of good-faith effort, the challenger may assert that the variances in the plan were avoidable. If this is true the intial burden of showing that the plan is not a good faith effort is mere surplusage, since whether or not this burden is met, an inquiry into the legitimacy of the reasons for the variance will be undertaken. The result, then, of the ease with which a challenger can carry his burden, furthers the goals of the Kirkpatrick Court, that developed the "as nearly as practicable" doctrine. Additionally, however, any redistrict- ing plan challenged for its interdistrict population variance must be justified by some legitimate state interest unless it has no variance. This result is probably not what Brennan intended, for he admitted that "[p]recise mathematical equality . . . may be impossible to achieve in an imperfect world; therefore the 'equal representation' standard is enforced only to the extent of requiring that districts be apportioned to achieve population equality 'as nearly as is practicable.'" 115 Further, if exact population equali- ty is really required by the Court, the two-step Brennan analysis of shif- 112 Id. at 2663. 111 Id. at 2658. "*Id. ll5 Id. at 2658 (citation omitted). Brennan seems to contradict this idea by asserting that computer technology has made redistricting much simpler for state legislatures, so that an equal population requirement would not be overly burdensome. Id. at 2659. 666 INDIANA LAW REVIEW [Vol. 17:651 ting burdens is meaningless, for the only issue in litigation would be whether the state could justify the variances, no matter how minute, in its plan. Apparently, if the state had districts with equal populations, the plan would be upheld, but if there were any variance the only burden in the litigation would be on the state to justify it. Thus, the slight burden Brennan has placed on potential challengers is inconsistent with his asser- tion that interdistrict population equality is impossible to achieve. 2. Legitimate State Interests.—Brennan's first level of inquiry is also inconsistent with his second level of inquiry: Whether a state can justify its population deviations, shown not to be the result of a good-faith ef- fort, by demonstrating legitimate state interests. One relevant factor in assessing the causal relationship between the state interest and the specific deviations is the size of the deviation. 116 Presumably, the smaller the devia- tion, the more readily the court will accept the state's justification for it. Yet, this type of balancing test implies that there is some point at which any quasi-legitimate justification will be accepted. Because Bren- nan noted that absolute population equality is impossible to achieve, this point will be above zero variance, at some minute figure. Thus, Brennan implied that there is a de minimis population figure at which the state's justification will, as a matter of course, satisfy the requirement of prov- ing a legitimate state interest. Yet, in his discussion of the challenger's initial burden, Brennan rejected a de minimis figure at which the state could be said to have engaged in a good-faith effort to achieve inter- district population equality, implying that redistricting plans can be placed in only two categories: those which have no population variance, and those which have some population variance. Thus, Brennan's enunciation of a balancing test to assess the legitimacy of the state's asserted interest is inconsistent with his refusal to recognize a de minimis figure to raise a presumption of good-faith on the part of the state. Brennan also identified the availability of plans with lower variances as a test to determine if the state has a legitimate interest. 117 The prac- tical effect of such a test is to insure that any plan challenged by a plan with a lower interdistrict population variance will be struck down. One source of alternative plans is the minority party of a state legislature. After Karcher, these minority parties are on notice that if they offer a plan to the state legislature that embodies the basic goals of the majority party's plan, but has smaller interdistrict population variances, it will prob- ably succeed in having the majority party's plan judicially nullified. The minority plan would carry the burden of showing the state's lack of a good-faith effort simply by showing that its plan, which embodies smaller variances, was offered to the state legislature but was rejected. In addi- tion, because the majority plan would reflect goals similar to those in- See supra text accompanying note 109. Karcher v. Daggett, 103 S. Ct. 2653, 2663 (1983). 1984] CONGRESSIONAL REDISTRICTING 667 eluded in the minority plan, the state's legitimate interests justification for its variances would be unacceptable because alternatives embodying the same values were available. 118 The side effect of Brennan's analysis is that, in the real political world, a challenged plan with any variance will likely be struck down. Brennan, seemingly, did not desire such a result in light of his view that absolute population equality is impossible to achieve, as well as his enunciation of legislative policies that would justify some variance. Unfortunately, however, it is apparent that the practical results compelled by Brennan's two-level analysis are inconsistent with the components of the analysis itself. C. Justice Stevens' Concern and Gerrymandering That prior case law in the congressional redistricting area has been concerned almost exclusively with interdistrict population equality is sur- prising since it is but one of the requirements that had been included in early federal reapportionment statutes. 119 For example, compactness does not necessarily exist in districts with equal populations. Rather than be- ing a constitutional requirement, however, in Karcher compactness was treated as a legitimate state interest that might justify some population variance. 120 That Brennan relied too heavily on population equality and failed to recognize other requirements of congressional district plans is the contention of the Karcher concurring and dissenting opinions. 121 Justice Stevens, who concurred in the result in Karcher, suggested another constitutional basis upon which a congressional district plan could be challenged. While Brennan's holding was based on the Feldman Plan's violation of article I, section 2 of the Constitution, Stevens noted that the equal protection clause of the fourteenth amendment could be invoked to support a cause of action for gerrymandering. 122 Stevens accepted the Brennan approach to article I, section 2 based on stare decisis, but felt that particular provision was inadequate to guarantee equality of representation. 123 Rather, Stevens said, the equal protection clause should be used in applying the one man, one vote 11 "It is presumed that, as a practical matter, there are only a few goals that the majority could consider in drawing its redistricting plan. See supra text accompanying note 106. Thus, it would not be difficult for the minority party to create a plan which includes any legitimate state interests embodied in the majority's plan. n9See supra text accompanying notes 11-13. 120Karcher v. Daggett, 103 S. Ct. 2653, 2663 (1983). 121 See Id. at 2667 (Stevens, J., concurring); id. at 2678 (White, J., dissenting); id. at 2687 (Powell, J., dissenting). 122 Id. at 2669 (Stevens, J., concurring). 12i Id. 668 INDIANA LAW REVIEW [Vol. 17:651 standard. 124 Stevens observed that in racial bias voting cases at the state level, the Supreme Court has said that the dilution of votes of a distinct political group may be unconstitutional. 125 Extending these cases to the federal level, Stevens analogized that the equal protection clause is a guard against congressional redistricting plans which discriminate on the basis of political grouping. Stevens demonstrated that gerrymandering and malapportionment causes of action are distinct with the assertion that a gerrymander would not be immune from constitutional attack even if the districts were of equal population: It is plainly unrealistic to assume that a smaller numerical disparity will always produce a fairer districting plan. Indeed, ... a stand- ard "of absolute equality is perfectly compatible with 'ger- rymandering' of the worst sort. A computer may grind out district lines which can totally frustrate the popular will on an overwhelm- ing number of critical issues." 126 Therefore, said Stevens, the equal population requirement must be sup- plemented with inquiries into the plan's adverse effect on identifiable political groups and the state's evidence that the plan serves the neutral legitimate interests of the community. 127 1. The Cause of Action for Gerrymandering.—The cause of action for gerrymandering enunciated by Justice Stevens puts the burden on the challenger to show that he is a member of an identifiable political group and that the redistricting plan has an adverse impact on that group. 128 Additionally, the challenger must show that the redistricting plan departs from other neutral criteria. 129 Upon such a showing, according to Stevens, the burden of justification falls on the state. 130 This burden can be car- ried by showing that the plan embodies acceptable neutral objectives. 131 i2A id. l25 See, e.g., Gomillion v. Lightfoot, 364 U.S. 339 (1960) (the Court struck down the newly created city boundaries of Tuskegee, Alabama, which excluded black voters from the city). '"Karcher v. Daggett, 103 S. Ct. 2653, 2671 (1983) (Stevens, J., concurring) (quoting Wells v. Rockefeller, 394 U.S. 542 (1969) (Harlan, J., dissenting)). l27 Karcher v. Daggett, 103 S. Ct. 2653, 2670 (1983) (Stevens, J., concurring). ]2i Id. 129 Id. at 2672. no Id. n> Id. at 2670. These criteria include large interdistrict population deviations, irregularly shaped districts, substantial diversion from a mathematical standard of compactness, exten- sive deviation from established political boundaries, and discrimination in the process of formulating and adopting the plan. Id. at 2672-74. Apparently, statements by legislators which indicate that politics were the motivation behind formulation of the plan will not raise a presumption of discriminaton: "Legislators are, after all, politicians; it is unrealistic to attempt to proscribe all political considerations in the essentially political process of redistricting." Id. at 2671-72. 1984] CONGRESSIONAL REDISTRICTING 669 It is evident that Stevens saw the danger of a torrent of litigation if gerrymandering supported a cause of action, for he went to great lengths to make it clear that the burden on a challenger in a gerrymandering case is an extremely high one. First, Stevens stated that this burden will be carried in few cases. 132 Also, the components of the test by which the challenger carries his burden are difficult to meet. The challenger must first show that he belongs to a politically salient class whose geographical distribution is ascertainable and could have been taken into account by the state; second, he must show that his proportional voting influence has been adversely affected because this distribution either was not taken into account, or was taken into account with the purpose of adversely affecting the group; finally, the challenger must make a prima facie show- ing which raises a rebuttable presumption of discrimination. 133 Stevens concluded his opinion with the caveat that due to the posture of the Karcher case, a challenge based on population deviations, it could not be concluded with certainty that the Feldman Plan violated the equal protection clause. 134 The plaintiffs did not raise, and the state did not have the opportunity to offer justifications for, the characteristics of the Feldman Plan which might indicate the existence of a gerrymander. 135 Stevens did note, however, that the Feldman Plan's lack of compactness, the fact that county boundaries were ignored, and the obvious political motivation in the drafting of the plan strongly indicated the existence of a constitutional violation. 136 Thus, since four other justices were willing to strike down the plan on the basis of stare decisis, Stevens concurred. 137 2. Problems with the Cause of Action for Gerrymandering.—Due to the onerous burden Stevens has put on challengers, as well as his failure to expand on how political groups must be taken into account by state legislatures, the practical value of his cause of action for gerrymandering is questionable. Stevens has made the burden so heavy for those challenging an alleged gerrymander, 138 relief will only be available in a small number of cases where there is a blatant gerrymander. Because the Brennan ap- proach will prompt legislatures to enact plans with zero interdistrict popula- tion variances, one method articulated by Stevens for the challenger to carry his burden in a gerrymander case, evidence of interdistrict popula- tion variance, is not useful. In fact, even if the challenger shows that m Id. at 2672. lii Id. [i4 Id. at 2677. lii Id. li6 Id. at 2676. Stevens gave two examples of bizarrely shaped districts in the Feldman Plan: the "swan" (district five), and the "fish hook" (district seven). Id. See infra Appen- dix A, p. 683. 137 Karcher v. Daggett, 103 S. Ct. 2653, 2667 (1983). liS See supra text accompanying notes 132-33. 670 INDIANA LAW REVIEW [Vol. 17:651 the plan has interdistrict population variances, that alone would probably be insufficient to carry the burden of proving a gerrymander. Courts will likely invalidate such a plan only on the basis of an article I, section 2 malapportionment violation, and such invalidation would not necessarily vindicate the voting rights of salient political groups claiming an equal protection violation. Thus, the gerrymander challenger must rely on ir- regularities in the map itself and its effect on the political group involved to carry his burden. It is, however, Stevens' failure to identify how these salient political groups must be taken into account by a state legislature to ensure that it has not enacted a gerrymander that most undermines the value of the gerrymander cause of action. Stevens gave examples of salient political groups, saying that they may be based on political affiliation, race, ethnic group, national origin, religion, or economic status. 139 The geographical distribution of these groups is revealed in many cases by the decennial census, and thus they may be taken into account by the state legislature when it draws new congressional districts. The ability of a challenger to show discrimination if these groups are not taken into account by the state in redistricting seems to create a duty on the part of the state to consider all of these groups in the process of drawing congressional districts. While such a duty may be desirable, though extremely burdensome, the problem facing the state is how to take these groups into account during the redistricting process. For example, if disgruntled Republicans challenge an alleged gerrymander by the Democrats, and if the challenge is successful, the guidelines the state legislature should use in drawing a new map are unknown. It would be unwise to require that the number of districts under the control of the state's majority party be limited to the percentage of state voters in that party. For example, this would re- quire a state which is sixty percent Republican to have a redistricting plan which would assure that party of winning no more than sixty percent of the districts. Such a requirement neither takes into account indepen- dent voters, nor the fact that people do not always vote for the candidate of their party. Not only would such a requirement fail to guarantee the desired split in the congressional delegation, it would also thwart the idea of a representative government even more than gerrymandering. Because the minority party would be guaranteed a certain percentage of the districts, the court in essence would determine the make-up of the House of Representatives, thereby engaging in judicial gerrymandering. 140 l3,Karcher v. Daggett, 103 S. Ct. at 2672 n. 12 (1983) (Stevens, J., concurring). l40Though it is true the majority party in state legislatures have, theoretically, the power to determine the political makeup of the House, this power is contingent on the electorate voting as the majority party projected when they drew the redistricting map. This safeguard is not present under a duty to take political groups into account, since the state must then guarantee the minority party a certain percentage of seats under its plan. Such a duty is patently unworkable. 1984] CONGRESSIONAL REDISTRICTING 671 Another difficulty with Stevens' approach is that he did not limit the definition of "salient political groups" to race, religion, or political party, but said that other characteristics may become politically significant in a particular context. 141 Thus, any significant special-interest group whose geographic distribution is ascertainable must be considered by the legislature in order to ensure that the plan is not a gerrymander. Again, it is unclear how these groups are to be taken into account. The interest group example illustrates the two principal problems with gerrymander- ing as a cause of action. First, it is naive to expect the majority party to pass a congressional district plan not based on the assumption that it would favor that party, as even Stevens recognized. 142 However, Stevens went no further in concretely identifying conduct by the legislature which would give rise to a prima facie showing of gerrymandering. Indeed, it is the map itself upon which Stevens relied in formulating the characteristics of a gerrymander. 143 Since, in most states, it is impossible to make every district competitive between Democrats and Republicans due to the uneven statewide distribution, a plan favoring one party will almost inevitably disfavor the other. Stevens did not identify the degree of disfavor that would be tolerated by the Constitution. Second, requiring the state legislature to consider the geographic distribution of salient political groups is a vague and unworkable require- ment. If the state recognizes the geographical distribution of a political interest group by including its members in a restricted number of districts, thus giving the group a better chance to win representation in the House, the group could claim that its voting power had been diluted in the other districts. Stevens said that "in case after case arising under the Equal Protection Clause the Court has suggested that 'dilution' of the voting strength of cognizable political as well as racial groups may be unconstitutional . " 1 44 Alternatively, if the state legislature assigns to several districts a percentage of persons representing the political group, to reflect the overall state percentage of that group, the group could claim uniform vote dilu- tion, and a gerrymander cause of action would again arise. As a practical matter, the only course left open to the legislature is to ignore the distribu- tion of the group, but this action squarely contravenes the duty of the state, as implied by Stevens, to take into account those salient political groups whose existence and geographic distribution are ascertainable by the legislature. Another problem with the gerrymander cause of action is the required level of review of challenged congressional disctrict plans. Stevens made it clear that only the most blatant gerrymanders will be struck down, 145 mKarcher v. Daggett, 103 S. Ct. 2653, 2672 n.12 (1983) (Stevens, J., concurring). 142 /rf. at 2671-72. it3 Id. at 2612-15. ,44 Id. at 2669. lA, Id. at 2672 (Stevens, J., concurring). 672 INDIANA LAW REVIEW [Vol. 17:651 implying that a challenged gerrymander will carry a strong presumption of constitutionality. Given that the equal protection clause is the basis for the gerrymander cause of action, this blanket deference is inconsis- tent with the traditional method of inquiry the Supreme Court has developed to review such challenges. This method consists of three levels of inquiry in equal protection clause cases: strict scrutiny, middle level scrutiny, and lower level scrutiny. 146 The review which accords the state the least deference is strict scrutiny, which occurs in cases involving fundamental rights 147 or suspect classifications. 148 In Reynolds v. Sims, 149 the Court identified the right not to have one's vote for a state legislator diluted as a fundamental right. The alleged abridgement of that right, then, would demand strict scrutiny to determine the challenged law's constitutionality. Unless the Court finds that this fundamental right does not exist with respect to the vote for a congressman, strict scrutiny should be invoked where there is an equal protection challenge to a congressional district plan. The deference Stevens is willing to give the state in gerrymander cases does not comport with strict scrutiny. Similarly, where the gerrymander cause of action is brought by a racial minority, courts must apply strict scrutiny to comply with earlier decisions. 150 A congressional district plan which does not treat minorities equally, then, should only be upheld if the plan was necessary to achieve some compelling state interest. 151 Again, a presumption of constitutionality should not arise, as it is inconsistent to defer to the state by invalidating only the most blatant gerrymanders. Similarly, the error of Stevens' use of one standard of review is demonstrated by the Court's use of two other standards, besides strict scrutiny, when a law is challenged as violative of the equal protection clause. For example, if a gerrymander is challenged on the theory that it did not treat women equally, it would probably be subject to middle level scrutiny, 152 which is more rigorous than lower level scrutiny, but more deferential than strict scrutiny. 153 The review called for by Stevens 146 J. Nowak, R. Rotunda, & J. N. Young, Constitutional Law 591-93 (1983) [hereinafter cited as J. Nowak]. ]A1 Id. See, e.g., Kramer v. Union Free School District, 395 U.S. 621 (1969) (right to vote); Shapiro v. Thompson, 394 U.S. 618 (1969) (right to travel); Griswold v. Connecticut, 381 U.S. 479 (1965) (right to privacy). ,48Suspect classifications include race, Brown v. Board of Education, 347 U.S. 483 (1954); and national origin Yick Wo v. Hopkins, 118 U.S. 356 (1886). I49 377 U.S. 533 (1964). ]$0See supra note 148. ,5, See Korematsu v. United States, 323 U.S. 214 (1944). 1 "Middle level scrutiny is usually applied in sex-discrimination cases. See Orr v. Orr, 440 U.S. 268 (1979); Craig v. Boren, 429 U.S. 190 (1976); Reed v. Reed, 404 U.S. 71 (1971). '"Middle level scrutiny involves evaluating the law's substantive relationship to a govern- mental interest. J. Nowak, supra note 146 at 592-93. 1984] CONGRESSIONAL REDISTRICTING 613 in gerrymander cases seems to be lower level scrutiny, which demands only that the means used by the legislature be reasonably related to its purpose. 154 Thus, in order to be consistent with equal protection clause analysis, the deference given to congressional district plans challenged as ger- rymanders should depend largely on two circumstances: first, whether the Court extends the fundamental right of Reynolds to congressional district plans, and second, whether different challengers must be afforded dif- ferent levels of scrutiny. It may be that these levels of scrutiny will be applied when evaluating the neutral criteria of the plan, but it is inconsis- tent with traditional equal protection analysis to assert that after these criteria are evaluated, only the most blatant gerrymanders will be struck down. These criticisms of the gerrymander cause of action are made with the realization that Stevens, in proposing it, was navigating in unchartered waters. It may be that future litigation will refine this cause of action to the point where it will be a workable one. However, as articulated by Stevens, it is not. D. The Kracher Dissenters The Brennan approach in redistricting cases is more inclined to prompt states to meet the standard of zero population variance than the Stevens approach. The four dissenters in Karcher, however, felt that exact popula- tion equality was too strict a requirement. 155 The contention that exact interdistrict population equality is impossible to achieve is untenable in light of advances made in computer technology. 156 It is likely that the true concern of the dissenters was that traditional boundary lines such as those surrounding cities and counties would have to be sacrificed in order to achieve such precision: The more likely result of today's extension of Kirkpathck is to move closer to fulfilling Justice Fortas' prophecy that "a legislature might have to ignore the boundaries of common sense, running the congressional district line down the middle of the corridor of an apartment house or even dividing the residents of a single- family house between two districts." 157 ,5i Id. at 591. 155Karcher v. Daggett, 103 S. Ct. 2653, 2678 (1983) (White, J., dissenting) and id. at 2687 (Powell, J., dissenting). 156For a discussion of how computer technology can be used to prevent gerrymanders see Torricelli and Porter, supra note 79 (Computers can be used to draw compact districts with no interdistrict population variance, but the prevention of gerrymanders also requires the removal of the redistricting process from partisan legislators and creating an apportio- nment commission). 157 103 S. Ct. at 2682 (White, J., dissenting) (quoting Fortas, J., in Kirkpatrick v. Preisler, 394 U.S. at 538). 674 INDIANA LAW REVIEW [Vol. 17:651 The more interesting aspect of the Karcher dissents is, however, that they agreed with Justice Stevens that gerrymandering is as important a problem, if not more so, than interdistrict population variance. Justice White said that "[o]ne must suspend credulity to believe that the Court's draconian response to a trifling 0.6984% maximum deviation promotes 'fair and effective representation' for the people of New Jersey.'' 158 White added that it would be a different matter if the plan discriminated against a racial or political group because such discrimination is a legitimate reason to hold that a redistricting plan is unconstitutional. 159 Justice Powell's dissent recognized the extraordinary shape of New Jersey's congressional districts, 160 and opined that injuries in voter representation that result from gerrymandering "may rise to constitutional dimensions." 161 The significance of the dissenters' agreement with Stevens as to the recognition of the gerrymander cause of action is that there are at least five members of the Court, 162 a majority, willing to recognize that cause of action. Thus, although Karcher v. Daggett is a population equality case, it also stands for the proposition that gerrymandering may give rise to a separate cause of action. Those challenging a state's congressional district map, then, can do so on two theories: that the population variances in the plan violate article I, section 2, and that the gerrymandering characteristics violate the equal protection clause. IV. Karcher: The Indiana Congressional District Map The results of the 1980 census revealed that the State of Indiana had a population of 5,490,224. 163 Although the state's population had increased by 5.7% since 1979, 164 it had increased at a slower rate than other sec- tions of the country. 165 After the 435 congressional seats were reappor- tioned, Indiana lost one seat, placing its congressional delegation at ten. Therefore, when the Indiana legislature drew new districts, an entirely new map was necessary. A. The Sutherlin Plan The Republican-controlled Indiana General Assembly began work on a new map in January, 1981. 166 The congressional district map adopted "'Karcher v. Daggett, 103 S. Ct. at 2653, (1983) (White J., dissenting). n9 Id. at 2686 (White, J., dissenting). ' e0See Appendix A p. 683. 161 Karcher v. Daggett, 103 S. Ct. 2653, 2689 (1983) (Powell, J., dissenting). I62 C.J. Burger, JJ. Stevens, White, Powell, and Rehnquist. '"United States Census Bureau, Number of Inhabitants—Indiana (1981). 1 "World Almanac and Book of Facts 207 (1983). l65The eight states with the greatest percentage of increased population from 1970 to 1980 are: Nevada (63.5%), Arizona (53.1%), Florida (43.4%), Wyoming (41.6%), Utah (37.9%), Alaska and Idaho (32.4%), and Colorado (30.7%). Id. '"Congressional Quarterly, State Politics and Redistricting Part I 113 (1982) [hereinafter cited as State Politics and Redistricting Part I]. 1984] CONGRESSIONAL REDISTRICTING 675 by the legislature 167 was drawn by Allan Sutherlin, the former Secretary of the Indiana Republican State Committee, and was passed on the last day of the legislative session after a plan introduced by the Democrats was rejected. 168 The latter plan had smaller interdistrict population variances than the Sutherlin Plan, and split only one county. 169 The Sutherlin Plan contains ten districts whose average population is 549,022. 170 The most populous district, the third, has a population of 558. 100. 171 The least populous district, the sixth, has a population of 540.939. 172 Thus, the interdistrict population variance is 17,161 people, or about 2.4%. 173 Additionally, the Sutherlin Plan splits thirteen of Indiana's ninety-two counties, 174 as well as the city of Bloomington. At the time of the creation of the Sutherlin Plan, there were seven Democratic and four Republican Indiana congressmen. 175 The plan divided Democrat Floyd Fithian's district among four new districts, effectively splitting his old constituency and leaving him without a district in which to run. 176 The same result was achieved with Democrat Dave Evans' district, and he subsequently ran against another incumbent Democrat, Andy Jacobs, Jr., in the primary. 177 Of the ten redrawn districts, only the districts of Jacobs, Benjamin, and Hamilton were considered safely Democratic, thus, the Republican Party stood a fair chance of capturing seven of the ten seats; 178 they succeeded in winning only five, however, in the 1982 election. 179 After the Sutherlin Plan was signed into law, the President Pro Tern of the Indiana Senate and the Indiana House Speaker, both Republicans, filed suit in state court in an effort to establish the constitutionality of the plan. 180 The plaintiffs' counsel asserted that the lawsuit was designed 167Act of May 5, 1981, 1981 Ind. Acts, Pub. L. No. 18, § 1 to -15 (1982). 168State Politics and Redistricting Part I, supra note 166, at 113 (1982). 165 Id. 170 See Appendix C, p. 685. Since the population of Indiana was 5,490,224, and it had ten congressional districts, Indiana's average sized congressional district is actually 549,022.4. 17 'Congressional Quarterly, Congressional Directory of the 98th Congress 67. 172 Id. '"Indiana's variance is the seventh greatest of the 44 states with more than one con- gressional district. See supra note 74. 174The split counties are Lake, Porter, Laporte, Kosciusko, Delaware, Henry, Rush, Monroe, Marion, Fayette, Washington, and Crawford. See Appendix C, p. 685. 175The Democrats were Adam Benjamin, Phil Sharp, Floyd Fithian, Lee Hamilton, Andrew Jacobs, Jr., and Dave Evans. The Republicans were John Hiler, Dan Coats, Elwood Hillis, John Meyers, and Joel Deckard. State Politics and Redistricting Part I, supra note 166, at 115. ll6 Id. at 112. ll7 Id. at 112-13. 111 Id. '"Republicans won the third, fourth, fifth, sixth, and seventh districts. '"Indianapolis News, Aug. 25, 1981, at 21, col. 4. The state legislative districts con- currently enacted by the Indiana legislature are the subject of litigation that is still pending in federal court. Bandemeer v. Davis, IP 82-56-C; NAACP v. Orr, IP 82-1669-C, (con- 676 INDIANA LAW REVIEW [Vol. 17:651 to avoid the confusion which frequently accompanied judicial changes in redistricting plans made close to election time. 181 The suit was removed to federal district court, and was later dismissed after the Democratic defendants failed to raise any issues. 182 While this preemptive court ac- tion by the Republicans was unusual, it reflects the great uncertainty under which state legislatures enact redistricting plans since the courts' increased involvement in the process following Baker v. Carr. The impetus of this uncertainty is that, nationwide, between twenty-five and thirty-five per- cent of current House district lines were drawn by courts. 183 B. The Sutherlin Plan under Karcher Thus, the constitutionality of the Indiana congressional district map was not determined. An analysis of the background and passage of the plan, as well as the characteristics of the map itself, indicate how the Sutherlin Plan would fare under the doctrines presented in Karcher v. Daggett. 1. Malapportionment Analysis.—Using the article I, section 2 theory of constitutional violation based on interdistrict population variances, the challenger has the burden of showing that the plan was not the result of a good-faith effort to achieve interdistrict population equality. 184 Because this burden can be carried by showing that plans with smaller population variances were proposed to, but rejected by, the state legislature, 185 such evidence fulfills the plaintiff's prima facie requirements. Therefore, if Indiana's congressional district map were the subject of litigation, the challenger's burden could be carried, since before the Sutherlin Plan was adopted, the Indiana General Assembly rejected an alternative plan with smaller population variances. 186 That such a plan was rejected would be viewed as strong evidence of a lack of a good-faith effort to achieve population equality. 187 The other factor identified by Brennan in Karcher with respect to good faith was the ease with which the interdistrict population variances in the plan could have been reduced. 188 In an effort to assess the difficulty solidated challenges based on the equal protection clause). Bandemeer alleges vote dilution of Democrats, and NAACP v. Orr alleges vote dilution of blacks. '"Indianapolis News, Aug. 25, 1981, at 21, col. 6. i>2 Id. at col. 3. It also appears that the defendants did not have enough money to pay their lawyer. Id. mKarcher v. Daggett, 103 S. Ct. 2653, 2684 (1983) (White, J., dissenting) (quoting American Bar Association, Congressional Districting 20 (1981)). n *See supra text accompanying note 93. n5See supra note 103 and accompanying text. n6See supra text accompanying note 169. ltlSee supra note 99 and accompanying text. n *See supra text accompanying note 104. 1984] CONGRESSIONAL REDISTRICTING 611 of achieving interdistrict population equality in Indiana, this author redrew the Indiana congressional districts using population figures which were only reduced to the township level. 189 The result of this effort 190 was a decrease in the total interdistrict population variance from 2.4% to .28%, a ninety percent reduction. Given that legislatures have access to com- puters, while the author redrew the district manually, it would clearly have been quite easy for the Indiana legislature to greatly reduce the inter- district population variances contained in the Sutherlin Plan. Thus, based on the analysis of the Karcher plurality, the Sutherlin Plan is not a good- faith effort to achieve interdistrict population equality. As the Karcher Court pointed out, however, the showing of a lack of good-faith effort to achieve interdistrict population equality does not mean that the plan is unconstitutional; it merely shifts the burden to the state to justify the deviations. 191 Karcher makes clear that every deviation in every district must be justified. Four examples of legitimate state in- terests which may justify a deviation were supplied by the Court. 192 '""United States Bureau of the Census, Number of Inhabitants—Indiana (1981). 190The author's redrawn districts are: One: Lake County (except Eagle Creek, Cedar Creek, and Winfield townships), and Portage Township of Porter County. Population: 548,944. Two: Lake county (remainder), Porter County (remainder), Newton, Benton, White (except Round Grove and Prairie townships), Pulaski, Laporte, Jasper, Warren (Pine and Prairie townships only), Starke (except Washington and North Bend townships), and St. Joseph counties. Population: 548,911. Three: Elkhart (except Locke township), Steuben, Lagrange, De Kalb, Noble, Allen, and Adams (Union township only) counties. Population: 548,834. Four. Marshall, Fulton, Miami, Grant, Huntington, Blackford, Jay (except Jeffer- son, Madison, and Pike Townships), Elkhart (Locke township only), Kosciusko, Whitley, Cass (except Clinton, Washington, Tipton, Deer Creek, and Jackson townships), Wabash, Wells, Adams (except Union township), Howard (except Ervine and Monroe townships), and Starke (Washington and North Bend townships only) counties. Population: 548,362. Five: White (Round Grove and Prairie townships), Carroll, Cass (remainder), Howard (Er- vine and Monroe townships only), Tippecanoe, Clinton, Tipton, Boone (except Sugar Creek, Jefferson, and Jackson townships), Hamilton, and Marion (Pike, Washington, and Lawrence townships only). Population: 548,877. Six: Posey, Gibson, Vanderburgh, Warrick, Pike, Spencer, Knox, Daviess, Martin, Dubois, Perry, Lawrence, Orange, Crawford, Washington, and Sullivan counties. Population: 549,652. Seven: Vigo, Clay, Owen, Greene, Monroe, Morgan, Brown, Vermillion, Parke, Putnam, Hendricks, Fountain, Montgomery, Warren (except Pine and Prairie townships), and Boone (Sugar Creek, Jefferson, and Jackson townships only) counties. Population: 548,908. Eight: Harrison, Floyd, Clark, Scott, Jackson, Bartholomew, Jennings, Jefferson, Ripley, Dearborn, Ohio, Switzerland, Decatur, Franklin, Johnson, and Shelby (Sugar Creek, Hendricks, Jackson, and Washington townships only) counties. Population: 549,018. Nine: Marion (Wayne Center, Warren, Decatur, Perry, and Franklin townships only), Hancock (Sugar Creek township only), and Shelby (Moral and Brandywine townships only) counties. Population: 549,875. Ten: Madison, Delaware, Ran- dolph, Henry, Wayne, Jay (Jefferson, Madison, and Pike townships only), Fayette, Union, Rush, Hancock (except Sugar Creek), and Shelby (remainder) counties. Population: 548,843. Average deviation from the average, .049%. Total deviation: .28%. See infra Appendix D, p. 686. mKarcher v. Daggett, 103 S. Ct. 2653, 2663 (1983). 192See supra text accompanying note 106. 678 INDIANA LAW REVIEW [Vol. 17:651 The first of these legitimate state interests is the effort to make districts compact. 193 That the population variances in the Sutherlin Plan are not due to an effort to make the districts compact can be shown in two ways. First, an examination of the map reveals that some of the districts in the Sutherlin Plan are not at all compact. 194 The second, eighth, and ninth districts are especially irregular. Second, the plan drawn by the author, 195 which incorporates much smaller interdistrict population variances, con- tains districts as compact, if not more so, than those in the Sutherlin Plan. The desire for compactness, then, does not justify the Sutherlin Plan's population variances. The second legitimate state interest identified in Karcher is that of respecting established political boundaries. The pursuit of this interest also does not justify the variances in the Sutherlin Plan. First, if the Indiana legislature were truly concerned with respecting municipal and county boun- daries, it could have accepted a plan like the Democratic one, which split only one county—the Sutherlin Plan splits thirteen. 196 In addition, the author's plan illustrates that much smaller variances could have been achieved by splitting only one more county than in the Sutherlin Plan. 197 Finally, the Sutherlin Plan splits the city of Bloomington in half, a result the author's map shows to be unnecessary. Clearly, the justification for the variances in the Sutherlin Plan cannot be claimed in respecting established political boundaries. The third legitimate state interest identified by the Karcher Court is preserving the cores of prior districts. An examination of the congres- sional district map in effect before the Sutherlin Plan was adopted 198 reveals that the Sutherlin Plan did preserve the cores of the first, third, fourth, seventh, eighth, and ninth districts. While it might be argued that because Indiana lost one district in 1980, it would be difficult to preserve the cores of all of the old districts; the fact that the districts whose cores were not preserved (the second, sixth, fifth, and tenth) had significant popula- tion variances from the average 199 illustrates that the preservation of the cores of the old districts was not the reason for the variances in the Sutherlin Plan. Additionally, because Brennan said in Karcher that the state interest offered as a justification must be consistently applied, 200 the fact that some districts whose cores were not preserved still had substan- tial variances eclipses the legitimacy of this justification. t9lSee supra note 12. 194 See Appendix C, p. 685. l9iSee supra note 179 and Appendix D, p. 686. 196 See supra note 174. l97 Lake, Porter, White, Warren, Starke, Elkhart, Adams, Jay, Cass, Howard, Boone, Marion, Shelby, and Hancock counties are split in the author's map. ,9iSee Appendix B, p. 684. '"The second district has a population which is 4000 people above the average, while the sixth and tenth districts are about 10,000 people below the average. 200Karcher v. Daggett, 103 S. Ct. 2653, 2663 (1983). 1984] CONGRESSIONAL REDISTRICTING 679 The fourth goal mentioned by the Karcher Court is avoiding contests between two incumbents. While the decrease in Indiana's apportioned con- gressional delegation meant that one incumbent would have to lose, the Sutherlin Plan effectively caused the defeat of two Democratic congressmen—Evans and Fithian. 201 Thus, the Sutherlin Plan created con- tests between incumbents, and the justification of avoiding such contests could not be used to vindicate the variances in that plan. The justification offered unsuccessfully by New Jersey, preserving the voting strength of minorities, would also fail as a justification for variances in the Sutherlin Plan. The two districts in the latter plan with the smallest precentage of blacks, the seventh, and ninth, have district populations that vary by over 10,000 people. 202 Again, because any justification of- fered must be consistently applied throughout the map, preserving the voting strength of minorities, if offered as the sole justification would fail. Finally, an attempt to offer the justification of preserving the voting strength of minorities in tandem with another justification would fail, because the other justifications themselves would fail. The legitimate state interests identified by the Court in Karcher, then, would not justify the interdistrict population variances which exist in the Sutherlin Plan. Though it is true that the Court did not limit the possible justifications to the examples given, 203 it must be remembered that because an alternative to the Sutherlin Plan was available, and because the variances in the plan are high compared with those of other states, 204 the showing of the legitimate state interest must be especially strong. 205 Indiana's lack of justification for the Sutherlin Plan's variance is fur- ther bolstered by the boldness of the mapmakers in identifying their over- riding concerns in the redistricting process. Some Republican legislators admitted during the redistricting process that they would do all that was possible to undermine the Democrats. 206 Such assertions diminish the prob- ability that Indiana could justify the variances in the Sutherlin Plan for two reasons. First, the statements indicate that none of the legitimate state interests identified by the Karcher Court are embodied in the plan. Sec- ond, an assertion that the state interest served by the plan was to allow the majority party to serve its own best interests would not succeed in justifying the variances in the Sutherlin Plan, since Brennan required that any justification offered must be nondiscriminatory. 207 Thus, it is highly probable that the Sutherlin Plan would not survive an article I, section 2 constitutional attack as formulated in Karcher. 201 See supra note 176. 202See supra note 188. 203Karcher v. Daggett, 103 S. Ct. 2653, 2663 (1983). 20ASee supra note 74. 20iSee supra note 109 and accompanying text. 206 Indianapolis Star, March 22, 1981, § II, at 3, col. 1 207Karcher v. Daggett, 103 S. Ct. 2653, 2663 (1983). 680 INDIANA LAW REVIEW [Vol. 17:651 2. Gerrymander Analysis.—Not only would the assertions of Indiana Republican legislators regarding the motives behind the Sutherlin Plan fail to serve as a justification for the plan's interdistrict population variance, they might also prompt a constitutional attack based on the cause of ac- tion discussed in Karcher's concurring and dissenting opinions, gerrymandering. In his formulation of the cause of action for gerrymandering, Justice Stevens indicated that the initial burden for the challenger is difficult to overcome. 208 For the purposes of analyzing the success or failure a ger- rymander claim would have against the Sutherlin Plan, it will be assumed that the challenger would be the Indiana Democratic Party. 209 Also, because this cause of action had not been recognized before Karcher, it is unclear how the nine justices would formulate the burdens and tests to be used: the five members of the Court210 who recognized gerrymander- ing as a cause of action did so in three distinct opinions. Since the only indication of these factors was in Justice Stevens' opinion, his enuncia- tion of the cause of action for gerrymandering will be used for analysis. 211 First, the Democratic Party is an identifiable political group. Cer- tainly, the Indiana legislature was aware that there were such persons as Indiana Democrats because at the time the congressional map was adopted there were eighteen Democrats in the Indiana Senate and thirty-seven in the House. 212 Statements made by Republican legislators indicated that the geographic distribution of Democrats was known by the legislature, and was taken into account in an effort to weaken their political effectiveness. 213 Therefore, these distribution figures could have been used to prevent a gerrymander from occurring. Making the necessary showing that its voting strength had been diluted would be a more difficult task for the Democratic Party than showing that it is an identifiable political group, for it is unclear what kind of showing is required. The statements of Republican legislators are some evidence of intent to dilute Democratic voting strength, but the dilution itself must be shown. If the Democratic Party could show that the Sutherlin Plan makes it impossible for Democrats to elect any members of their party to Congress, that showing would be sufficient to demonstrate vote 20iSee supra note 132 and accompanying text. 209This assumption is made due to the clear intent of Republican legislators to deplete Democratic strength as much as possible. That is, it is probable that the Indiana Democratic Party is the political group most adversely affected by the Sutherlin Plan. Whether a challenge by a racial minority might influence the Court to invoke a stricter standard of review is unclear given Stevens' assertion that only blatant gerrymanders will be struck down. See supra text accompanying notes 145-54. 2 '"Stevens, White, Powell, Rehnquist, Burger. 2 "See supra text accompanying notes 127-32. 2i2 Index to Indiana Senate And House Journals 1, 33 (1982). 2nSee supra text accompanying note 206. 1984] CONGRESSIONAL REDISTRICTING 681 dilution. However, the minimum seats "guaranteed" to the Democrats under the plan before a dilution claim would fail is unknown. Because the Democrats are virtually guaranteed three congressional seats in the Sutherlin Plan, and because five Democrats, or half of Indiana's delega- tion, were elected to Congress under the plan in 1982, the Democratic Party would probabaly fail in an attempt to show vote dilution under the Sutherlin Plan. Furthermore, the challengers must also make a prima facie showing that raises a rebuttable presumption of discrimination. While the statements of Republican legislators regarding their intent to undermine the Democrats, as well as the fragmentation of two formerly Democratic districts, comprise extrinsic evidence of discrimination by the Republican legislative majority, Stevens relied almost exclusively on the structure of the map itself as the means by which such a presumption could be raised. One indication of gerrymandering mentioned by Stevens, the existence of interdistrict population variance, 214 would be insufficient by itself to raise a presumption of discrimination since the Court would probably invalidate the plan on the basis of malapportionment, rather than gerrymandering. Such an invalidation would not necessarily remove the harm of which the Democrats complain, since exact interdistrict population equality and gerrymandering are compatible. Further, although some of the districts in the Sutherlin Plan lack a high degree of compactness, 215 they are not nearly as irregular as the shapes in the Feldman Plan. 216 Indeed, the districts that are somewhat irregular in Indiana's congressional district map seem to have been drawn in an effort to minimize the number of counties which were split. Thus, ir- regularities in the structure of the districts in the Sutherlin Plan would not be considered extraordinary enough to raise a rebuttable presumption of discrimination, particularly in light of Stevens' desire that only blatant gerrymanders be struck down. 217 Because Indiana Democrats could not demonstrate either vote dilu- tion or a sufficient deviation from neutral criteria218 to raise a rebuttable presumption of discrimination, an attack of the Sutherlin Plan on the basis of its being an unconstitutional gerrymander would fail. This con- clusion is based, however, on the burdens and test enunciated by Justice Stevens in Karcher in formulating this new cause of action. Further refine- ment of the gerrymander cause of action could result in a new set of burdens and tests which might compel a different conclusion. The effect of Karcher v. Daggett on the Indiana congressional map, if it were challenged, would depend upon the theory of invalidation chosen 2i4Karcher v. Daggett, 103 S. Ct. 2653, 2670 (1983). 2,5 See Appendix C, p. 685, note particularly the second, fifth, eighth, and ninth districts. 216See Appendix A, p. 683. 217 See supra text accompanying note 136. 2nSee supra note 129. 682 INDIANA LAW REVIEW [Vol. 17:651 by the challenger. Under the "as nearly as practicable' ' standard as set out in Brennan's two-level inquiry, Indiana's Sutherlin Plan would be struck down: the population variance involved, nearly four times as large as the variance involved in Karcher, is unjustified, and could easily have been avoided. If attacked as an unconstitutional gerrymander, however, the Sutherlin Plan would not be struck down by the Supreme Court. Although there is some extrinsic evidence that gerrymandering was on the minds of some Hoosier Republican legislators, the results of the 1982 con- gressional election under the Sutherlin Plan, and the structure of the map itself, undermine the possibility of a challenger showing the existence of vote dilution or of raising a rebuttable presumption of discrimination. Both were required of the challenger by Stevens' formulation of the ger- rymander cause of action. V. Conclusion Karcher v. Daggett is a landmark congressional redistricting case for two reasons. First, the requirement that interdistrict population variances be mimimized "as nearly as practicable" was enforced against a state whose congressional district plan embodied smaller deviations than the Court had ever considered unconstitutional before, paving the way for precise mathematical equality in all of the districts of the House of Representatives, which furthers the goal of equal representation. Second, a majority of the Court recognized that a claim of gerrymandering would support an alternative cause of action against a state and its congressional district plan. Thus, those who use the Karcher decision to challenge a state's con- gressional district plan have two constitutional theories under which to bring a claim: article I, section 2 for malapportionment, and the equal protection clause of the fourteenth amendment for gerrymandering. Indiana's congressional district plan would not survive a malapportion- ment attack, but would survive a gerrymandering attack, where the burdens on the challenger are much greater. The Karcher decision should please those who have demanded exact interdistrict population equality from congressional district plans, as well as those who assert that there are other interests that the Supreme Court should recognize. In either case, Karcher sends a powerful message to state legislatures: the Court will not hesitate to enter into the redistricting process if malapportionment or gerrymandering occurs. This check on the redistricting process is a healthy one in light of the conflict of interest which abounds between the ability of state legislators to create the districts and the desire of state legislators to maintain or increase their political power. William B. Powers 1984] CONGRESSIONAL RED/STR/CTING 683 APPENDIX A: The Feldman Plan 684 INDIANA LA W REVIEW [Vol. 17:651 APPENDIX B: Congressional Districts of the 1970's 1984] CONGRESSIONAL REDISTRICTING APPENDIX C: The Sutherlin Plan 685 686 INDIANA LA W REVIEW APPENDIX D: The Author's Plan [Vol. 17:651