"From the census of 1870 to the census of 1910 the Federal reapportionment acts require the State legislatures to devise districts of equal population and contiguous territory, and after the census of 1900 the requirement of compactness was added. And yet, during this period, even though it preceded the big wave of urbanism, there were important disparities of population between districts inside many of the States. That the disparities were not greater than they were is to be attributed not so much to an attempt to abide by the equality requirement as to the fact that urbanism had not reached full flower.

"Examples of population disparities between districts under the 1910 census as given in Schmeckebier's 'Congressional Apportionment' (Washington, D.C., Brookings Institution, 1941) are the following: Colorado, largest district, 228,444 and smallest district, 134,469; Connecticut, 250,182 and 190,403; Illinois, 349,883 and 167,634; Louisiana, 234,382 and 165,563; New York, 300,000 and 194,708; Ohio, 264,297 and 164,474. A list covering all the States for the year 1897 is found in appendix I to Justice Frankfurter's opinion in Colegrove v. Green (328 U.S. 549, 557-559 (1946)).

"One method of compelling compliance with congressional regulations in this field would be for the House of Representatives to refuse to seat Congressmen from improperly devised districts. The House has never done this, although as Schmeckebier notes, the matter at least was discussed on 3 occasions: 1843, when 4 States did not use districts at all; 1901, a Kentucky district, and 1910, a Virginia district. In view of the interminable wrangles and interstate reprisals which would flow from a serious attempt of the House of Representatives to police a districting formula, this sanction appears inadvisable. The vital task of legislating might suffer.

"Can the Federal courts be looked to for help in enforcing statutory or constitutional formulas for districting? Under present precedents the answer appears to be in the negative. In 1932 two Federal district courts invalidated State districting laws in Mississippi and Kentucky for noncompliance with the standards fixed in the congressional statute of 1911. But the Supreme Court reversed. The majority placed the decision on the ground that the 1911 law had expired, thus leaving open the question of whether the Federal courts would enforce congressional standards for districting. But four members of the Court, Justices Brandeis, Stone, Roberts, and Cardoza, would have dismissed the suit for want of equity (Wood v. Broom (287, U.S. 1 (1932)); Mahan v. Hume (287, U.S. 8 (1932))). The question left open by Wood v. Broom now appears to be settled by the later cases of Colegrove v. Green (328 U.S. 549 (1946)) and South v. Peters (339 U.S. 276 (1950)).

"For discussions of Colegrove v. Green see Burdette, "The Illinois Congressional Redistricting Case,' the American Political Science Review, volume 40 (October 1946), pages 958-962; note, 'Constitutional Right to Congressional Districts of Equal Population,' Yale Law Journal, volume 56 (1946), pages 127139.

"Colegrove v. Green was a suit in Federal district court for a declaratory judgment that Illinois congressional districts were in violation of the Federal Reapportionment Act of 1911 and the Federal Constitution, and for an order restraining the holding of the 1946 congressional election on the basis of these districts.

"The district court dismissed on authority of Wood v. Broom. The Supreme Court affirmed in a 4 to 3 decision. Three members of the majority, speaking through Justice Frankfurter, placed the affirmance on the broader ground of nonjusticiability; that is, resorted to the doctrine of political questions. Justice Frankfurter said that this issue was of a peculiarly political nature and therefore not mete for judicial determination (p. 552).

"Further, he said:

66

'Nothing is clear 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. It is hostile to a democratic system to involve the judiciary in the politics of the people. And it is not less pernicious if such judicial intervention in an essentially political contest be dressed up in the abstract phrases of the law' (pp. 553-554).

"South v. Peters involved a challenge of the Georgia county unit system under the 14th and 17th amendments. In a per curiam rejection of the challenge the Supreme Court said:

"Federal courts consistently refuse to exercise their equity powers in cases posing political issues arising from a State's geographic distribution of electoral strength among its political subdivision' (p. 277).

70784-61-pt. 4- 4

"It is also axiomatic that a court will not issue a mandamus or an injunction compelling a legislature to act or to desist. Fergus v. Marks (321 Ill. 510, 152 N. E. 557 (1926)); McChord v. Louisville & N. R. Co. (183 U. S. 483 (1902)). On the general doctrine of political questions in the Federal courts see Post, The Supreme Court and Political Questions' (Baltimore: Johns Hopkins Press, 1936). Also in point are Harris, "The Judicial Power of the United States' (Louisiana State University Press, 1940) and my own "The Doctrine of Separation of Powers and Federal Immunity Statutes,' to appear this spring in the April and June issues of the George Washington Law Review. * * * I feel that the district system should be postponed-perhaps forever. The problem of gerrymandering, and by this I include of course the so-called silent gerrymander, i. e., that slow strangulation of metropolital political strength by simple failure to redistrict, is a virtually insoluble problem. The Federal judiciary will not, and probably should not, interfere even to enforce constitutional mandates. "The legislatures have the power, indeed, have always had the power. the power of the rural and small-city legislative majorities naturally has not been modified by a desire for political immolation. It would be quixotic to expect otherwise, now or in the future. Likewise, it is unlikely that the problem of the gerrymander will be solved by mass population movement within the foreseeable future."

But

Mr. DOUGLAS. In summary, Mr. President, as Professor Dixon has said, Congress has never enforced effectively, and probably never will enforce penalties in an effective way. Moreover, we cannot look to the courts for relief, because the Supreme Court has ruled in Colegrove v. Green that the gerrymander question is political and, therefore, non justiciable. This alone is a sufficient reason for rejecting the Mundt-Coudert plan.

IT WOULD HANDICAP REPUBLICANS

An eighth evil which would arise from operation of the Mundt-Coudert plan is that it creates all sorts of anomalies. First of all, the plan would tend to favor the party which failed to carry the State in the popular election. In 1952 for example, Eisenhower carried Tennessee, but he would have received only 5 electoral votes to Stevenson's 6. In 1948 Mr. Truman carried California and Ohio; but Mr. Dewey would have captured a majority of electoral votes in each of these States. Application of the Mundt-Coudert formula to the statistics for the last 10 elections produces dozens of anomalies similar to these examples. The Mundt-Coudert plan, like the Lodge-Gossett plan, would make presidential elections closer contests by enlarging the electoral vote of the candidate who was defeated in the popular election.

Mr. President, I ask unanimous consent to have printed at this point in the Record a table of statistics which I have prepared on the electoral vote margin. There being no objection, the table was ordered to be printed in the Record, as follows:

NOTE. Because the popular vote for President has not been tabulated in a number of congressional districts, the electoral vote margin under the Mundt plan can be given only in terms of a mimimum and maximum. In 1944, for example, Roosevelt would have had at least 37 and perhaps as many as 205 more electoral votes than Dewey would have had.

Source: Hearings, op. cit., p. 341.

Mr. DOUGLAS. Mr. President, at best, this situation would merely enlarge the electoral vote consolation prize of the defeated candidate. At worst, it would enlarge the electoral vote of the candidate defeated in the popular election to the point that he could snatch electoral victory from the jaws of popular defeat. Perhaps, equally as bad, it would enlarge the electoral vote of the popularly defeated candidate sufficiently to throw the election into the Congress. In speaking about the eighth evil of the Mundt-Coudert plan, let me say a word about how the plan might operate to the advantage of one party and the disadvantage of the other. In general, I think the plan would benefit almost nobody except the South. In Democratic years, we would trade large blocs of electoral votes for nothing. In Republican years, the Republican candidate would trade large blocs of northern electoral votes for a mere handful in the South. In 1952, for example, Eisenhower would have lost 58 northern and 9 southern electoral votes. In 1924, Mr. Coolidge would have traded somewhere between 95 and 177 northern electoral votes for 3 or 4 southern electoral votes.

This all brings me to a point about which I have heard much, namely, that the amendment before us would greatly handicap the Republican Party in years when the GOP polls a national plurality in the popular election. I am not here to promote the interests of the Republican Party. But I am here to defend our two-party system, and the Republican Party is a vital part of that system. I think there is a very real possibility that this plan would deprive the Republican Party of the presidency in years of Republican popular victory, because they would trade such large blocs of electoral votes in the North for so few in the South. The Lodge-Gossett plan is likely to be more unfair than the Mundt plan in this respect. I think, however, that the Mundt plan would also handicap the Republican candidate in this respect even though a majority of northern States have Republican legislatures which could further gerrymander the northern State in favor of the Republican Party.

THIRD-PARTY COMPLICATIONS

Let us look at this formula if a third-party complicated the situation. If a third-party candidate drew popular votes mainly from the Democratic candidate, such as Wallace did in 1948, the Republican could win more electoral votes under the present system than under either the Mundt-Coudert or Lodge-Gossett plans. For the present system allows the Republican to garner all of a State's electoral votes by polling a mere plurality of the popular vote. In 1948, for example, Dewey's 46.3 percent of the popular vote in New York gave him all 47 of that State's electoral votes. Under the Mundt-Coudert plan, however, Dewey would have received only 24 to Truman's 23 electoral votes. Under the Lodge-Gossett plan, he would have had a margin of only four-tenths of an electoral vote-21.6 to Truman's 21.2 and Wallace's 3.9.

Conversely, if there were a third-party candidate who drew a sufficient number of popular votes from the Republican candidate in normally Republican areas to reduce the Republican to second or third place, as La Follette did in some areas in 1924, the Republican would likely gain more electoral votes under the proposed amendment than under the present system. The Lodge-Gossett plan would benefit the Republican most if the third-party candidate's strength were spread throughout the State so that the Republican could carry only a few if any congressional districts. In Wisconsin, for example, in 1924, Coolidge was able to win a popular plurality in probably no more than one and certainly no more than three congressional districts. Consequently, he would have received 1, 2, or 3 electoral votes under the Mundt plan but would have received 4.9 electoral votes under the Lodge-Gossett plan and actually received none under the present system.

If a third-party candidate's strength were concentrated in only certain congressional districts in a normally Republican State, however, the Mundt plan would give the Republican more electoral votes than would the Lodge-Gossett plan. The result in Minnesota in 1924 gives an example. La Follette carried at least 2 and no more than 4 congressional districts and, consequently, would have received 2-4 electoral votes to Coolidge's 8-10. But Coolidge's 8-10 electoral votes under the Mundt plan would have been cut to 6.2 votes, and La Follette would have received 5.0 electoral votes under the Lodge-Gossett plan.

In considering the Mundt-Coudert plan, we are actually operating in the dark. The presidential vote has never been completely tabulated by congressional districts except for the two most recent elections.

On this point, let me quote from the testimony of Dr. Ruth Silva, professor of political science at the Pennsylvania State University. Professor Silva is the most skilled expert on the whole subject. I am quoting from the hearings, starting on page 342:

"It is actually impossible to tell exactly how the Mundt plan might work, how it might affect party alinement in general, or how it might affect the fate of either major party until a more thorough statistical analysis has been made. This, in turn, will require a tabulation of the popular vote cast for each presidential candidate in each congressional district. It will also require making a statistical analysis of other elections. We need to know how the plan would operate if the South were not solid. Therefore, we should apply the MundtCoudert formula to the elections of 1864, 1868, 1872, and 1876. We should know exactly what would happen under the Mundt-Coudert plan if there developed a third-party which drew on Republican strength. Therefore, we should apply the Mundt formula to the elections of 1912 and 1924. Certainly, we ought to know how the Mundt formula would operate if there were a third-party candidate who drew votes from the Democratic candidate in nonmetropolitan areas. Therefore, the formula should be applied to the election of 1892. Perhaps we should also know how the Mundt formula would have operated in the election of 1824.

"If the Mundt-Coudert plan should be adopted without first applying the formula to a sufficient number of past elections to get some idea how the plan would work under various conditions, we are likely to find that we have opened a political Pandora's box. We have been told, however, that it is not necessary to tabulate the popular vote cast for each presidential candidate in each congressional district in order to apply the Mundt formula to past elections. For, one witness has said, we may safely assume that each political party would have elected one presidential elector for each House seat won. As a matter of fact, elaborate tabulations of electoral votes under the Mundt plan have been made on the basis of this assumption. In the following table is reprinted a part of such a tabulation:

"NOTE.-Electors assumed elected where parties' representatives were elected and States were carried by presidential candidates. (Congressional Directories and World Almanacs, 1935 and 1945 editions, the New York Times on 1948 election. Hearings on S.J. Res. 8, Senate Committee on the Judiciary, 83d Cong., 1st sess., p. 217.

"The figures for 1952 and those in parentheses were supplied by Ruth C. Silva. The figures in parentheses are electoral votes won under Mundt-Coudert plan on the basis of tabulations of the actual popular vote for President in congressional districts.

"It is submitted that tabulations such as the ones presented in the above table are not a satisfactory substitute for tabulating the actual vote cast for each presidential candidate in each congressional district. The weakness of assigning electorial votes on the basis of House seats won is shown in the assignment of electoral votes to third-party presidential candidates. In 1948, Thurmond would have received only eight electoral votes for carrying four States according to the method used in the table. Yet 39 electoral votes were assigned to Thurmond in the table for no reason consistent with the method. This method of assigning electoral votes on the basis of House seats won means the assigning of electoral votes to third parties which had no presidential candidates. It means the assigning of electoral votes to third parties which actually endorsed a major party's presidential candidate. Such electoral votes, it seems, should be assigned to the major party endorsed.

"The fallacy of assigning electoral votes on the basis of congressional seats won is clear in 1952, when the presidential and congressional candidates of different parties carried 84 congressional districts. Thus, there would have been

Some

84 errors in assigning electoral votes on the basis of House seats won. of these errors cancelled each other out, however, so that only 76 errors remained in the 1952 total given in the table. In 1948, the presidential and congressional candidates of different parties carried 100 congressional districts.

"In defense of tabulating the electoral vote under the Mundt formula for past elections on the basis of House seats won, it has been said: "The assumption regarding district presidential electors is based on the fact that in New York and some other States the name of the party's candidate for district elector would appear on the voting machines right next to that of the party's candidate for Representative in Congress, the offices being arranged on the party's row on the voting machine in descending order from the statewide to the smallest constituency. ***

"At first blush it may seem that the *** (tabulations) should be based on the popular vote for President in congressional districts. However, reflection from common knowledge will support the view that such a basis would be a distortion due to the mere mechanics of ballot and voting-machine arrangements of candidates.'

"The mere mechanics of ballot and voting-machine arrangements of candidates will not insure that a party's candidate for Representative in Congress and its candidate for district elector will run together like siamese twins if the Mundt plan were adopted for the following six reasons:

"1. Actually, 27 States controlling 363 electoral votes use the so-called presidential short ballot. In these 27 States the only change the Mundt plan would make would be that a popular vote for a presidential candidate would be tabulated as a vote for his party's candidate for district elector and its two or more candidates for electors at large instead of being tabulated as a vote for each of his party's candidates for elector in the entire State. If there is anything in the Mundt-Coudert plan which would require abolition of the presidential short ballot, adoption of the plan would reduce rather than increase the democratic character of presidential elections.

"2. In 18 States controlling 252 electoral votes, candidates are arranged on the ballot according to the offices they seek rather than according to party affiliation. In all of these States except Pennsylvania, the voter would have to select his party's candidate for Congress and its candidate for district presidential elector from two separate groupings on the ballot. In Pennsylvania, the voter may vote a straight-party ticket by marking a single X in the party square or circle. Therefore, we cannot safety tabulate the electoral vote for past elections under the Mundt plan on the basis of House seats won.

"3. All ballots are not arranged to encourage straight-party voting. In 17 States using the office-block type of ballot plus 5 States using the party-column or party-row ballot, there is no party square or circle on the ballot or party lever on the voting machine to encourage or facilitate straight-party voting. These 22 States control 256 electoral votes. This fact too shows the fallacy in tabulating electoral votes on the basis of congressional seats won.

"4. At least one State (Michigan) facilitates splitting the ticket between the presidential candidates and the congressional-State ticket by providing the voter with a separate presidential ballot or a seperate presidential lever on voting machines.

"5. In New York, the name of the party's candidate for district elector probably would not appear on the voting machine right next to the party's candidate for Representative in Congress. The State legislature would probably continue to consider the presidential candidate as the leader of the statewide ticket and not merely as a candidate in the congressional district. As a matter of fact, many State officials doubtless hope to slide into office on the coattails of a popular presidential candidate. Consequently, the State law would probably continue to place the presidential candidate at the top of the ticket before the other statewide candidates and would not move his name down on the ballot to the district level.

"6. Under the Mundt-Coudert plan, the voter would be voting for at least three candidates for presidential elector-one in his congressional district and at least two in the State at large. Therefore, even under the New York State law mentioned above, it is likely that the courts would decide that the statewide electors were controlling and thus that the presidential candidate should be placed at the head of the ticket rather than next to the district congressional candidate.