KEEPING SCORE

The use of legislative scorecards dates back to the early twentieth century, though there is some uncertainty as to their first appearance. Clemens (1997) points to California in 1913, when an enterprising lobbyist published the voting records of state lawmakers on labor legislation.Footnote ⁷ By 1916, she says, labor unions in several states had followed suit, adopting the “checking up system,” as it became known—keeping tabs on what lawmakers did, not what they said they did. At the national level, Rozell, Wilcox, and Franz attribute the first scorecard to the Farmers Union in 1919.Footnote ⁸ Congressional testimony, however, suggests that the American Federation of Labor (AFL) was already publicizing roll calls on individual measures as early as 1910 and providing the voting record of any lawmaker (on “measures of vital importance”) to members on request.Footnote ⁹

Whenever a multivote scorecard specifically emerged, two points are important to consider in regard to its use. First, this was a lobbying tactic with an electoral edge, intended to produce legislative results by threatening hostile lawmakers with the prospect of electoral retribution. Scorecards were meant to wean group members off party labels as their sole guide in the voting booth, under the theory that greater influence could be secured through nonpartisanship. Instead, members should support a group's true “friends”—lawmakers who reliably backed favored measures, as shown by their record, not their rhetoric—and punish those “enemies” who did not. Second, scorecards were controversial. Though they quickly became commonplace, Clemens observes, Californians in 1913 “found the legitimacy of this technique extremely questionable.”Footnote ¹⁰ Legislators did not take kindly to this scrutiny and its coercive flavor. They “looked upon it as black-listing,” she explains.Footnote ¹¹ Some rank-and-file members were also troubled by its imperative quality regarding their own vote.Footnote ¹² During congressional hearings in 1913, AFL President Samuel Gompers denied it was “improper” to publish voting records, but claimed that the information was presented neutrally, so as to help members “form their own conclusions,” not instruct them how to vote.Footnote ¹³ Presentation mattered, in sum, to overcoming widespread hostility toward this new and potentially powerful tactic.

Such hostility was not immediately neutralized, but the practice steadily expanded among interest groups and beyond, since the information these groups compiled proved useful to journalists and political operatives. Newspapers had occasionally published individual roll call votes, but they did not produce convenient reference tables. Nor did the official Congressional Record, which could run to many thousands of pages for a given congressional session and was available in only a few depository libraries.Footnote ¹⁴ The average journalist had little time for such intense research activity, while few interest groups had the resources. Recognizing these limitations, a new subscription service appeared in 1924: Editorial Research Reports (ERR), which tracked congressional activity on a wide range of issues. By the early 1930s, ERR was tabulating roll calls on “leading measures” and “outstanding votes”—selected according to perceived importance rather than issue area—and providing some basic analysis.Footnote ¹⁵

Thus scorecards were steadily creeping into the evaluation of Congress, and into elections, too. A 1932 ERR report explained that these records “provide a direct check upon the claims of senators and representatives standing for reelection.”Footnote ¹⁶ Despite providing the broadest range of information, however, ERR would offer no guidance in making that check—no perspective from which to assess legislators’ votes in normative terms: as good or bad, right or wrong, as deserving of reelection or not. Interest groups, in contrast, had a clearer perspective, but those engaged in voting analysis were still careful not to appear too prescriptive in their scorecards—presenting them as objective and informational, while holding back from explicit judgments as to the rectitude of a lawmaker's position.

In 1938, for example, Labor's Non-Partisan League (LNPL)—an electoral group formed two years earlier and associated with the Congress of Industrial Organizations (CIO)—produced a “voting record” that reviewed recent votes on “major labor, farm, and other social legislation.”Footnote ¹⁷ In detailed charts, they examined the votes of each state's lawmakers across twelve different issues in the House and Senate. “The League stand on how Congressmen should have voted” was indicated atop each table with a “Yes” or “No”—a relatively blunt description, but one that stopped short of right or wrong. The tables themselves, moreover, required intensive study. There was no way to glean a congressman's overall record at a glance—the sea of Y's and N's could be made comprehensible only one row and column at a time. In this sense, LNPL's scorecard appears primarily designed for use by staffers in its state affiliates, as they geared up for the midterm elections. Yet LNPL also intended to distribute it more widely, “to voters in every Congressional district throughout the country,” the CIO News proudly declared.Footnote ¹⁸ Perhaps recognizing the interpretive challenges involved, however, LNPL soon announced to the press a list of selected incumbents it had graded “A” or “D” on labor issues—to be supported or opposed accordingly.Footnote ¹⁹ Such grades were far simpler to understand, but were never directly linked back to the scorecard, nor did they perfectly correspond with it.Footnote ²⁰ Objective voting analysis and more overt electoral recommendations, it seemed, were kept at a distance in public.

The scorecards of the 1930s, then, were tabular and technical; often justified as “nonpartisan” educational tools; and geared primarily to a narrow market of journalists, policy wonks, and politicos. But if the late 1930s had shown hints of a more popular approach, developments in the 1940s would soon begin to transform voting records into very different political instruments—bringing them to a much wider audience and infusing them with a new partisan purpose.

THE UNION FOR DEMOCRATIC ACTION

The story of modern index scores begins not with the ADA, but with its organizational predecessor, the Union for Democratic Action (UDA), founded in the spring of 1941.Footnote ²¹ And it owes much to the influence of one man, former Wisconsin Congressman Thomas R. Amlie. A sometime socialist, then Republican-turned-progressive, Amlie had been defeated for reelection in 1938 and subsequently languished in underemployment before landing at the UDA in early 1942, hired to head up its newly opened Washington, DC office.Footnote ²²

The UDA had been formed by leading American liberals with an eye on the international scene, dismayed by the fascist and communist governments that had emerged in Europe and pushed the world toward war. It was a vehicle for the expression of liberal hostility to totalitarian systems, whether left or right, designed to vigorously articulate “the intimate relationship between an all-out anti-totalitarian foreign policy and a strongly pro-democratic domestic policy.”Footnote ²³ Its slogan, “A Two-Front Fight for Democracy—at Home and Abroad,” encapsulated that linkage. Abroad, it advocated U.S. entry into World War II in support of the Allies and urged forceful prosecution once war was declared. At home, it urged social policies that would protect the average American, framing “pro-democratic” in essentially liberal terms. Such policies also made for a strengthened homefront more capable of waging war, the UDA claimed.

In 1942, the UDA took its message into the congressional midterms, seeking the election or reelection of lawmakers committed to their vision of domestic and international liberalism. And under Amlie's tutelage, they turned to roll call analysis as an electoral weapon. Through much of the 1930s, Amlie had toured the country as a spokesman for liberal causes and learned a powerful lesson about appearance and reality in politics. Even in crowds sympathetic to his most radical ideas, he would find enthusiasm for local members of Congress whom he considered “reactionary.” Constituents simply lacked any real sense of their representative's record, Amlie concluded.Footnote ²⁴ Roll call data could thus help to correct their false impressions, he believed, and thereby aid in the election or reelection of candidates more sympathetic to liberal goals. Now the UDA put Amlie's insight to use in the service of their wartime aims—to defeat the “obstructionists” who had opposed American entry into WWII.

To fund the venture, they partnered with The New Republic, the flagship journal of mainstream American liberalism.Footnote ²⁵ Together, they would produce and publish a special supplement to the magazine, entitled “A Congress to Win the War.” It would identify the worst “obstructionists” in Congress through hard data—roll call votes—and urge their defeat. Amlie worked feverishly preparing the supplement. Relevant measures had to be selected. Huge numbers of votes had to be checked. Tensions between the UDA and The New Republic—over credit and attribution—had to be ironed out.Footnote ²⁶ “I am putting in long hours in an effort to get this job finished,” he wrote UDA Executive Secretary James Loeb Jr., “but about once a week I attend a cocktail party and fail to use judgment.”Footnote ²⁷ Finally, as summer approached, it was ready.

“A Congress to Win the War” appeared in May 1942—five months after the attack on Pearl Harbor that had led the United States to declare war. An accompanying article explained the key issues in the election and offered interpretive analysis of the voting record provided within. So great were the stakes in wartime, it emphasized, that Americans “cannot trust in public office any man whose record indicates that he lacks either the intelligence or the moral courage to wage this fight relentlessly to a successful conclusion.”Footnote ²⁸ Such deficiencies would be evident in a lawmaker's failure to support the UDA's favored position on various measures, both foreign and domestic, which they deemed important for military preparedness and effectiveness.

“The reader will quickly see that the tests applied make no distinctions between Democrats and Republicans in terms of their party affiliations,” the supplement explained. There was but one question applied: “Does this candidate's record show that he has consistently understood the implications of the battle for freedom at home and abroad, and does it indicate that he has the character to work until full victory for the democratic cause?”Footnote ²⁹ Foresight, understanding, moral courage, even character—these were complex attributes to distill from a legislator's voting record. But the UDA was undaunted. The lawmakers of the 77^th Congress were to be put “on trial,” judged by the people according to their roll call votes, and in many cases, according to the supplement, must be found “guilty.”Footnote ³⁰ A list of particularly egregious “obstructionists” was also included, whose “guilt” was discussed in depth.Footnote ³¹

But at the heart of “A Congress to Win the War” were the tables. Two charts, one for each chamber, showed how lawmakers had voted on twenty key measures: ten directly tied to foreign policy or military preparedness and ten on domestic issues the UDA claimed were indirectly important to the war effort. There was the Lend-Lease Bill, the “destroyers for bases” deal, and the amendment of the Neutrality Act, for example, along with a major Works Progress Administration appropriation, the “court-packing” plan, and various housing, labor, and public power measures.Footnote ³² Not all measures were identical across both chambers, with entirely different issues substituted in some cases.Footnote ³³ The reader was to understand that such variation was irrelevant, so careful was the UDA's selection and so compelling the picture it painted.

Members of the House were arrayed by party, then by state delegation, and senators were listed alphabetically under party label. Only those senators up for reelection were included.Footnote ³⁴ How a lawmaker had voted on each measure was indicated by a simple “+” or “−.” These symbols did not indicate whether the legislator had voted for or against the bill in question, but whether they had voted with or against “the government,” by which the UDA meant the administration of Democratic President Franklin Roosevelt—for the UDA's favored position matched Roosevelt's in every case.Footnote ³⁵ The accompanying article explained this accord in essentially patriotic terms, casting the administration as the upholder of “freedom” and “democracy” in a “war for liberty,” against the forces of “obstruction” and “reaction” in Congress, though ideological terms like “liberal” and “progressive” did occasionally appear.Footnote ³⁶

Unsurprisingly, the supplement quickly came under fire from the so-called obstructionist lawmakers, who denounced it as a “blacklist,” a “purge,” a “smear campaign,” and even a “campaign of hate,” and who saw subversive “red” forces at work beneath UDA's liberal veneer.Footnote ³⁷ The House Un-American Activities Committee (HUAC), chaired by Rep. Martin Dies (D-TX) issued a report in June implying that the UDA was a communist front organization—despite its declared opposition to both communism and fascism.Footnote ³⁸ But more subtle criticisms appeared, too, of the methods UDA had employed and even the legitimacy of its enterprise.

Despite claiming that the “two-year record” of legislators would be examined, the UDA's analysis drew on votes from as early as 1937, long before the current Congress and the outbreak of war.Footnote ³⁹ The international context had changed dramatically since then, critics argued, and crucially, other elections had intervened. The isolationist positions so distasteful to the UDA had been largely popular at the time, it was pointed out, and many espousing them had been handily reelected in 1938 and 1940.Footnote ⁴⁰ Why should they be penalized retroactively for representing the people? “The purge of non interventionists is one of the most extraordinary movements in our political history,” Arthur Sears Henning wrote in the Chicago Tribune, criticizing the attempt to repudiate lawmakers whose constituents “at the time gave ample evidence of overwhelming support of the attitude of their public servants in Congress.”Footnote ⁴¹

Another major concern was the consistency of the analysis across the tables and the article, and across the “two fronts” of importance to the UDA. Whether domestic matters such as the court-packing proposal could really be considered military preparedness measures was one thing.Footnote ⁴² That several lawmakers identified in the article as obstructionists had actually supported the administration on most international matters was a thornier issue. These were primarily Southern Democrats in the House, such as Howard Smith of Virginia, Eugene Cox of Georgia, and HUAC members Martin Dies and Alabama's Joe Starnes. Republican Leland Ford of California also proved problematic, having voted with the administration on all of the international measures, but against it on all the domestic ones. In opposing these men, one critic observed, “The New Republic and the Union for Democratic Action repudiate their own chart.”Footnote ⁴³

This disjuncture was a key piece of evidence for those who saw communist infiltration at work in the supplement. Supposed noninterventionism was just a pretext for attacking conservative members who hunted out subversive activity, Representatives Clare Hoffman (R-MI) and Harold Knutson (R-MN) charged on the House floor.Footnote ⁴⁴ Even fellow liberals were concerned with the “scrambling” of domestic and international issues.Footnote ⁴⁵ As Morris Rubin explained in The Progressive, equating interventionism with “liberalism” yielded confusing results. If readers counted up the “right” and “wrong” votes (as he deemed them) and tried to calculate a lawmaker's “batting average,” they would find individuals like Ford—“one of the most intolerant critics of everything progressive”—coming out as a relative moderate, with a .500 average or 50%.Footnote ⁴⁶ Southern Democrats would often look even better, having voted for at least some domestic New Deal measures. Thus Carl Vinson of Georgia had an average of .777, he calculated, while Dies emerged with .666 (an apt statistic given the contempt in which he was held by progressives).Footnote ⁴⁷ “In short,” Rubin concluded, “the roll call is a mess.”Footnote ⁴⁸

But the scrambling was intentional, designed to protect domestic liberals who might otherwise do poorly in the analysis. The New Republic’s editors had initially hoped to focus exclusively on foreign policy, but Amlie had pushed back. “If we were to get out a supplement dealing only with foreign affairs,” he had told them, “we should play directly into the hands of the reactionists.”Footnote ⁴⁹ Including domestic issues would “serve to give our own friends some protection,” he added.Footnote ⁵⁰ That the records of some conservatives were enhanced by the foreign policy measures would be explained away in the article. They were accordingly excoriated for their records of “black reaction,” while several domestic liberals with poor international records were exonerated for “a change of heart” since Pearl Harbor.Footnote ⁵¹ An internal UDA memo admitted that “certain congressmen with interventionist records were criticized because their records on domestic issues seemed to indicate that their first enemy was the New Deal and social progress at home, rather than Hitler, Mussolini, and Hirohito.”Footnote ⁵²

In short, the numbers alone did not make the UDA's case, which may partly explain why they offered no summary percentage scores of the kind Rubin calculated. Only once did the supplement even hint at them, describing Hoffman's record as “almost 100-percent bad on both foreign and domestic issues.”Footnote ⁵³ But it may also reflect a concern to appear informational rather than instructive, at least in the purportedly “objective” tabular part of the supplement, in an effort to dispel criticism that it was telling readers how to vote. In some ways, though, the UDA's record was less reserved and more accessible than that of prior scorecards, such as LNPL's in 1938. Its use of +/− symbols, for example, made it easier to interpret, dispensing with the need to cross-check the group and the individual's stance, and allowing for some visual assessment. These symbols, moreover, conveyed approval of a lawmaker's vote, even without the use of normative language (while in the accompanying articles, certain votes were explicitly described as “wrong”).Footnote ⁵⁴ Most importantly, it was published by a national magazine—ultimately selling 300,000 copies—expanding the audience beyond political professionals to educated elites, if not a mass public.Footnote ⁵⁵

Indeed, UDA leaders intentionally pursued a strategy of elite influence, distributing the supplement to “politically conscious” opinion leaders across the country, which they felt had been highly successful.Footnote ⁵⁶ As Amlie wrote just before the election, “It would be no exaggeration to state that as far as this campaign for control of Congress is based upon issues, it is based upon the UDA New Republic Supplement.”Footnote ⁵⁷ The average voter was expected to absorb the supplement's roll call analysis second or third hand, through the mediation of these local opinion leaders. “This supplement is not intended for mass distribution,” Amlie confirmed to then-Congressman Lyndon Johnson in the summer of 1942.Footnote ⁵⁸ But it could be utilized in campaigns through a targeted strategy, he explained. “Each Dem. Candidate ought to get the names of all people listed in the telephone Directories in his District who can read and understand material as difficult as this,” Amlie advised, send them the supplement, and encourage them to spread word of the obstructionists’ failings.Footnote ⁵⁹

Amlie's letter to Johnson was revealing in another way, for he was clearly touting the supplement as an aid to Democratic candidates.Footnote ⁶⁰ Yet the UDA was meant to be nonpartisan—indeed, its list of “obstructionists” included names from both parties. But its support went almost exclusively to pro-administration Democrats, and it appeared far more concerned with the liberalization of the Democratic Party than the liberalization of Congress as a whole.

In this respect, the UDA appeared to be building on the strategic orientation of the industrial labor movement, which had emerged during the 1930s as an important ally of President Roosevelt, and increasingly of New Deal Democrats at large. Labor's Non-Partisan League (LNPL), for example, had backed Roosevelt in 1936 and expressed support for pro-labor (“A” grade) candidates in the 1938 midterms—most of whom were liberal Democrats. It had opposed a number of conservative Democrats, too, even taking a stand in some primary elections, and in so doing had complemented a controversial administration effort to “purge” the Democratic Party of Roosevelt's most recalcitrant critics.Footnote ⁶¹ The overlap was only partial, though—where the administration emphasized Senate targets, primarily opponents of the court-packing bill and executive reorganization plan, LNPL focused more on the House, where its key concern was the vote on recommitting the Wages and Hours Bill (this was the main consideration, in fact, behind the grades LNPL awarded, separately from its scorecard).Footnote ⁶² Both LNPL and the administration, especially, were denounced for blacklisting and interfering in local affairs, and both campaigns proved largely unsuccessful.Footnote ⁶³ Still, in reinforcing the administration's efforts, however slightly, LNPL had backed a nascent attempt to remake the Democratic Party in Roosevelt's image.

Yet LNPL's CIO leaders were not fully committed to such a goal in 1938—one that moved away from both a nonpartisan “friends and enemies” strategy of interest group influence within the existing party system and the radical dream of a new labor party. LNPL flirted with third-party politics, but also emphasized its willingness to support pro-labor congressional candidates from both major parties, and in 1940, it even turned against the Democratic standard bearer altogether.Footnote ⁶⁴ Despite internal disagreements, LNPL endorsed the Republican nominee for president—its leaders claimed Roosevelt had neglected key labor concerns in his second term and that defection might produce greater dedication to their cause. The CIO's rank-and-file membership, however, proved unwilling to punish Roosevelt at the ballot box, and LNPL was itself repudiated, with the organization crumbling soon afterward.Footnote ⁶⁵ Drawing from this experience (and the persistent failure of third parties), the CIO would thenceforth adopt a more partisan approach, seeing the Democratic Party as the best vehicle for achieving its aims, provided that the party could—as Roosevelt's purge had intended—be made more liberal.Footnote ⁶⁶ The UDA, with its sympathetic views on labor policy and broader liberal ideology, adopted a similar strategy and now sought to manipulate its scorecard to that end.

For a start, the supplement emphasized existing differences between the parties. As characterized in the article, the Republican Party's record on both the domestic and foreign policy fronts was “negative and bleak,” exhibiting almost “complete obstruction” (save for a handful of liberal Republicans from the Northeast and California), and the party would likely continue in an obstructionist direction.Footnote ⁶⁷ The Democratic record, in contrast, showed “an extremely high average” across the UDA's favored positions, the article claimed, and a progressive “underlying trend” (claims for which no statistical confirmation was provided).Footnote ⁶⁸ In this context, supporting individual liberal Republicans could actually bolster the “obstructionist” Republican picture as a whole, the supplement warned. If the GOP regained a House majority in 1942, hardcore isolationists would take the Speaker's gavel and a string of committee chairmanships. “Even the best Republican candidates will carry this stigma upon them.”Footnote ⁶⁹ The lesson was clear: Electing a good Republican could pave the way for a bad party.

The supplement also made clear that there could be bad Democrats, too. Indeed, some conservative columnists saw the administration's hand behind the UDA's pronouncements—another “New Deal blacklist” without Roosevelt having to get his hands dirty.Footnote ⁷⁰ The failed intervention in 1938 had been personally damaging to the president, after all, and it had also shown how few tools national party leaders had to punish lawmakers who strayed from the administration line. The Democratic Party was decentralized, and the national campaign committees were weak; their main role was raising money to aid all candidates running under their label, which was itself an identifier they did not control. The president had in any case recused himself from campaigning in 1942, telling the press that “politics is out” due to the war.Footnote ⁷¹ “The Democrat High Command” had “missed a glorious opportunity,” Amlie complained, by failing to mount a vigorous campaign based on the supplement.Footnote ⁷² “Just because some of their own members had bad records they have not felt like going all out against the Republicans with bad records,” he fumed.Footnote ⁷³ But the UDA had done so, going up against the Democrats with “bad records,” too, and in this respect it had waged a kind of purge campaign of its own—from outside the party apparatus. If scorecards had once been a means of weaning voters off party labels in elections, the UDA was now urging their use for different purposes: to shape the meaning of those labels and who would get to use them.

The UDA sought to build on this effort two years later, with another supplement in The New Republic and a new target: the “unholy alliance” of conservative Republicans and Southern Democrats who had banded together to block “many essential war measures.”Footnote ⁷⁴ This 1944 edition, creatively titled “A Congress to Win the War and the Peace,” would address some earlier criticisms with refinements of presentation and analysis. It was more forthright in its ideological language—a “+” vote was now a “progressive” rather than a pro-administration vote “in the opinion of the editors” at least.Footnote ⁷⁵ And it was more ideologically consistent in its analysis, too. Though the UDA's conception of war measures remained broad, the new supplement offered a clearer linkage between liberal domestic policies and the war effort and, ultimately, international peace, because “poverty is indirectly a fertile cause of war.”Footnote ⁷⁶ On this basis, more domestic issues were included and less emphasis placed on foreign affairs, making the records of most Southern Democrats appear more conservative.Footnote ⁷⁷ Had readers chosen to calculate batting averages for those members, they would find the results far lower in 1944 than in 1942.Footnote ⁷⁸ But they would have had to take this cumbersome step themselves. Once again, the supplement did not provide any scores, even though UDA staff had clearly calculated averages for internal use by this point—noting in correspondence that most House Democrats had scores of “66⅔% or better,” for example, while most Republicans fell below this level.Footnote ⁷⁹ The Democrats, then, were still the superior party and might be made more so in 1944.

This second supplement did not sell as well as its predecessor—shifting around 100,000 copies—and UDA explained the drop in terms of competition, as “many other organizations were doing similar organizational work” that year.Footnote ⁸⁰ The concept of widely distributed congressional scorecards was spreading. And Thomas Amlie was helping to spread it. He had moved on from the UDA to the newly formed Political Action Committee (PAC), created by the CIO in 1943 as a revamped successor to LNPL.Footnote ⁸¹ Designed to marshal the money and manpower of labor unions in elections, Amlie helped set up the PAC's research arm for the 1944 campaign, and a congressional scorecard duly appeared.Footnote ⁸²

The PAC's sister organization, the National Citizens Political Action Committee (NCPAC) also honed in on roll call analysis. It even featured in NCPAC radio ads during the fall campaign (one airing in Ohio stressed the need to assess Republican Senator Robert Taft “objectively” and “scientifically” and proceeded to “tune in Senator Taft's record” direct from the National Archives).Footnote ⁸³ The NCPAC also took steps to make its analysis more accessible to the average voter, temporarily at least. The August 1944 edition of its newsletter provided a “box score” showing overall party breakdowns on selected votes, with a black skull-and-crossbones flag indicating the “reactionary” position, contrasted with an American flag to show “democratic” votes.Footnote ⁸⁴ It was perhaps an unfortunate ink choice that had the American flag depicted entirely in red, for both the NCPAC and PAC's efforts were attacked as “Communistic” by HUAC members and others.Footnote ⁸⁵ (Subsequent issues toned down the imagery.Footnote ⁸⁶)

In neither case, however, did the PAC or NCPAC publish individual summary scores—though once again, internal correspondence shows that batting averages were being calculated on the basis of PAC scorecards.Footnote ⁸⁷ Why were they not published? The intended audience and use of these records may offer a partial explanation. As national PAC staff explained, scorecards were primarily designed to aid state and local PAC or NCPAC affiliates in making endorsement decisions and planning campaign activities.Footnote ⁸⁸ They were “master records” rather than propaganda tools, packed with information to be utilized by trained workers, who would distill their message for the general public.Footnote ⁸⁹ As such, summary scores were not really necessary and perhaps not even useful in comparison to the raw votes and explanations.

Yet a lingering concern among interest group leaders, that appearing too instructive could be dangerous when venturing into the electoral sphere, may have continued to play a constraining role. Providing master records to local PAC and NCPAC workers helped avoid the charge of instructing voters directly, while also preserving the appearance of local autonomy in electoral contexts—another source of anxiety as national groups became more involved in congressional campaigns. And there were other steps they took to package their analysis in less threatening ways. Like the UDA's scorecards, the PAC's versions initially refrained from describing votes as right or wrong, and they avoided explicit judgments of lawmakers, too.Footnote ⁹⁰ As PAC director Jack Kroll explained before a congressional committee investigating campaign expenditures in 1946, “We do not single out an individual and say, ‘Here is Bill Smith; Bill Smith is altogether bad. …’ We say, ‘We are for this issue. Here is how all the Congressmen voted on it.’”Footnote ⁹¹ The point was deniability, in essence, that the group producing the scorecard did not command or even explicitly urge its members to vote for or against a particular candidate.

By the mid-1940s, however, more “neutral” organizations were beginning to experiment with shorthand summary scores. Congressional Quarterly (CQ), which emerged out of Editorial Research Reports (ERR), began to publish percentage measures of “party unity” in 1946—how often a member voted with his or her party when majorities in each divided on an issue. The 1947 CQ Almanac also included “bipartisan support” scores, showing how often lawmakers voted with the congressional majority as a whole, when the bulk of both parties favored a bill.Footnote ⁹² An array of other scores appeared over the next decade and beyond, showing how often a lawmaker backed the president's agenda, for example, or supported the “conservative coalition”—where Southern Democrats joined with conservative Republicans to vote against Northern Democrats—all available for use by paying clients.Footnote ⁹³ (Their audience, like that of ERR, was still journalists and politicos.)

Interest groups with pressing political concerns, however, were still hanging back from summary scores, though they were becoming bolder in other respects. By 1948, the AFL had joined the scorecard fray, with the Taft-Hartley Act of 1947 rousing its leadership into creating a political action committee: Labor's League for Political Education (LLPE). Though Taft-Hartley served as its “acid test” for supporting candidates in the campaign, LLPE scorecards included other votes and explicitly described lawmakers’ stances as right or wrong.Footnote ⁹⁴

The UDA had also introduced occasional “box scores” into its newsletter in the mid-1940s, thus making a small move toward summaries.Footnote ⁹⁵ And it had been active in the 1946 midterms, producing a third roll call supplement and continuing to push for its liberal vision.Footnote ⁹⁶ But it would not do so in 1948. International developments had placed strains on the American left: The emerging Cold War had divided liberal opinion, with former Vice President Henry Wallace emerging as a voice of reconciliation with the USSR and ultimately mounting a third-party presidential bid to take that message to the electorate. The UDA's anti-communist liberals saw a need to band together with like-minded groups to offset the Wallace threat.Footnote ⁹⁷ In January 1947, therefore, they terminated their old organization and reconstituted themselves as a new, larger group: the Americans for Democratic Action (ADA).Footnote ⁹⁸

But the UDA would not be forgotten. “A Congress to Win the War” marked a major shift in the use of lawmakers’ records within modern election campaigns. As UDA staffers themselves reflected in the mid-1940s, it had “set the pattern for the use of roll call votes in political campaigning.”Footnote ⁹⁹ The UDA supplement had broken new ground in extending roll call analysis beyond single issues or a single Congress, in expanding the audience for such analysis and harnessing it to the cause of party change. But “A Congress to Win the War” was complicated in presentation and ideologically confusing. ADA would pick up where the UDA left off.

THE AMERICANS FOR DEMOCRATIC ACTION

Founded by Eleanor Roosevelt, Arthur Schlesinger Jr., Reinhold Niebuhr, and Walter Reuther, among others, the Americans for Democratic Action (ADA) would become the foremost liberal interest group in the United States, advocating for liberal causes and supporting liberal candidates in elections.Footnote ¹⁰⁰ It publicly endorsed candidates, provided some financial resources to them during primary and general campaigns, and sought to mobilize liberal voters.Footnote ¹⁰¹ And congressional roll call analysis was the thread that bound these political activities together.

As a 1955 Republican Party report explained: “The ADA revived the idea of the Congressional ‘Score Card’ which the antecedent Union for Democratic Action had first employed in 1942.” The scorecard appeared in its member newsletter—ADA World—in August 1947, the report noted, “and has been carried as a ‘Congressional Supplement’ every year since then.”Footnote ¹⁰² Though distributed to regular members, not just political professionals, these early ADA scorecards still lacked summary scores, and much like their predecessors (except the AFL's scorecards), they avoided describing votes as right or wrong in the tables themselves.Footnote ¹⁰³ Much like those who had previously ventured into scorecard territory, moreover, the ADA was subjected to heavy criticism for doing so. Amid rising Cold War tensions, in fact, this criticism would have a distinctly red glaze, despite the ADA's avowed anti-communism.

Thus when an ADA-backed candidate challenged Louisiana Congressman F. Edward Hébert in his 1948 primary, Hébert took to the airwaves to denounce the group and its methods.Footnote ¹⁰⁴ A Southern Democrat who supported Strom Thurmond's Dixiecrat presidential candidacy and a HUAC member to boot, Hébert had little sympathy for the forces of liberalism. The ADA was a communist group running a smear campaign against him, Hébert told his constituents in a radio address.Footnote ¹⁰⁵ All the talk about his legislative record was mere “poppy-cock” designed to obscure the ADA's real purpose in the South: the destruction of states’ rights and the dismantling of segregation.Footnote ¹⁰⁶ The ADA had, after all, engineered passage of the civil rights plank at the Democratic convention, leading to the Dixiecrat bolt.Footnote ¹⁰⁷ But Hébert also stressed an objection made earlier against the UDA, that it was democratically illegitimate to use votes from previous congresses. His record had been placed before his constituents every two years, Hébert said, and they had approved it three times since his initial election in 1940, with their “ballots of confidence.”Footnote ¹⁰⁸ In effect, Hébert argued that his record was forged anew with each new Congress, and thus there was a statute of limitations on raising issues from the past.

“There is a widespread attitude that it violates the cannons of good taste to explore too far into a politician's past record,” Thomas Amlie acknowledged in 1950. Amlie had moved on from the PAC by this point, but remained in the research game under his own steam. He even self-published a book that year, Let's Look at the Record, aiming to spread his mantra that roll call analysis could rectify the ills of American politics.Footnote ¹⁰⁹ Too many people seemed to believe “that each election serves to wipe out past political sins,” Amlie complained, but only over a longer timeframe of analysis could a lawmaker's political posture “stand out with clarity.”Footnote ¹¹⁰

Let's Look at the Record was Amlie's magnum opus. “A Political Kinsey Report on Congressmen,” he called it.Footnote ¹¹¹ Perhaps unsurprisingly, given the subject matter, it proved far less popular than Kinsey's exploration of American sexual behavior. But the presentation was less than accessible, too, being highly detailed, technical, and oriented toward elites. Part philosophical treatise, part practical manual, Amlie's book set out to illuminate the records of all legislators by first guiding readers through the votes of just one, examined across eleven policy areas. Thus enlightened, the reader could look up other members of Congress in the extensive voting tables that made up most of the book. This meant wading through 300 or more roll calls, going as far back as twenty years for some lawmakers. There were no shortcuts or summary scores in the tables, though Amlie did summarize the bigger picture in percentage terms, claiming the average record of Republican members was “close to 80 per cent conservative,” but that of Northern Democrats was “close to 80 per cent liberal.”Footnote ¹¹² He even identified a handful of individuals and characterized their records in similar terms.Footnote ¹¹³ But he held back from providing scores for all.

The rise of CQ and the popularity of its “party unity” or “presidential support” scores among journalists, however, was beginning to show the publicity value of summary measures.Footnote ¹¹⁴ And the lesson was not lost on the ADA. Throughout the 1950s, the ADA began to flirt with percentages in its public pronouncements, sporadically and tentatively at first, but with increasing confidence. ADA World’s 1953 “Congressional Supplement,” for example, featured percentage language for the first time—describing how often the “average Senator” or “average Congressman” voted “right,” as the ADA now defined it, with a further party breakdown.Footnote ¹¹⁵ The supplement also provided averages for some individual party leaders for the first time. House minority whip John McCormack (D-MA), for example, voted the “right” way—the ADA way—100% of the time, it announced.Footnote ¹¹⁶ These examples, offered in the article introducing the voting tables, might invite members to make similar calculations. But for journalists the ADA was willing to do more of the legwork. A press release accompanying the 1953 supplement listed lawmakers who had “supported the liberal position over 90% of the time.”Footnote ¹¹⁷

The following year, the ADA's press release included percentage scores for all individual senators and a party breakdown for the House, including a list of 100% liberal and “never voting liberal” members.Footnote ¹¹⁸ But these percentages did not feature in the ADA World’s Congressional Supplement itself. Indeed, mention of percentages disappeared from the supplement entirely.Footnote ¹¹⁹ The ADA was under fierce attack in 1954—the high tide of McCarthyism—for its supposed communistic connections, perhaps leading to reticence in the official scorecard sent to all members. Though Senator Joe McCarthy (R-WI) himself would soon be censured, numerous candidates in the 1954 elections disavowed any relationship with the ADA, even rejecting its endorsements, which suggests that ADA support could be viewed as a “kiss of death.”Footnote ¹²⁰

Similarly, percentage analysis does not appear in the ADA's Guide to Politics 1954, a book-length organizing manual for liberals, co-edited by Schlesinger and largely prepared by future Senator Daniel Patrick Moynihan (then a young graduate student).Footnote ¹²¹ It combined issue analysis with practical guidance on political campaigns and included some roll call analysis in its “Political Scoreboard” section. But there were no scores on the board: The tables offered no summary measures, simply indicating with a “+” those votes “which ADA believes to be in harmony with liberal policies” and a “–” otherwise.Footnote ¹²²

Throughout the rest of the 1950s, the ADA's use of percentage scores was haphazard. With the exception of 1957, percentages received no mention in the annual “Congressional Supplement” carried in ADA World. Roll calls even disappeared entirely from the ADA's Voting Guide 1956. Footnote ¹²³ Yet from 1956 onward, the press release accompanying the annual supplement consistently mentioned percentages and even began to include tables listing individual lawmaker scores, at least for the Senate.Footnote ¹²⁴ Moreover, in 1956, the press release introduced a new label for the scores—the liberal quotient— and proceeded to identify eleven senators, all Democrats, who “scored 100% on the liberal side on every vote.”Footnote ¹²⁵ From then on, liberal quotient, or LQ, became a regular shorthand in the ADA's correspondence and press releases and even made it into the New York Times in 1957.Footnote ¹²⁶ But the term did not appear in the published supplement, suggesting some controversy and fear of backlash still lingered around the overt use of scores.Footnote ¹²⁷

As late as 1958, the Massachusetts ADA—an especially active and important ADA chapter—warned against summary scores in its own statewide “Legislative Supplement.”Footnote ¹²⁸ “We trust the form of presentation will not lure the hasty reader to count +’s and –’s and rate his legislators in percentage fashion,” they said.Footnote ¹²⁹ Just two years earlier, another interest group—the American Farm Bureau Federation—had been hauled over the congressional coals for publishing a “batting average,” summarizing lawmaker stances on ten amendments to the recent Farm Bill.Footnote ¹³⁰ Testifying before a congressional committee investigating lobbying and campaign activities, chaired by Senator John McClellan (D-AR), Farm Bureau leaders sought to downplay any hint of electoral instruction or punitive intent in presenting such scores. “You will notice that we did not recommend to our members that any of the members of this team be retired,” Farm Bureau president Charles B. Shuman told the committee. “We simply reported on their performance in this game. We hope in the next game they will have a better batting average.”Footnote ¹³¹

Committee member Barry Goldwater, Arizona's conservative Republican Senator, was less than impressed, disparaging scorecards like the Farm Bureau's as “so-called voting records of Congressmen,” or even “so-called right or wrong voting record[s].”Footnote ¹³² “I think, gentlemen, that we are getting on some rather thin ice when we try to evaluate the voting of Congress by our own feelings,” Goldwater warned.Footnote ¹³³ Shuman clarified that “we indicate the position taken on each of these issues by the American Farm Bureau, but we do not say that the Congressman was right or wrong. We say that he disagreed with the Farm Bureau position or agreed with it.”Footnote ¹³⁴ “It is written that way on purpose,” another official pitched in.Footnote ¹³⁵ Thus the Farm Bureau tried to submerge its judgment and grope toward plausible deniability of instructing voters.

But other organizations had become more assertive about their own feelings, boldly proclaiming what they considered to be right or wrong. Indeed, Goldwater's ire was primarily directed at the Committee on Political Education (COPE)—the political action committee created after the AFL-CIO merger in 1955—and the intense dislike was mutual.Footnote ¹³⁶ COPE had expanded the range of issues included in its scorecard beyond strictly labor votes and offered a lengthy timeframe for their analysis (drawing on PAC and LLPE records back to 1947). Though it refrained from percentage analysis, COPE began offering “box scores” summarizing a lawmaker's individual performance, describing in words how often he or she had voted “right” or “wrong.”Footnote ¹³⁷ These box scores embraced entire careers, in effect, and conservatives like Goldwater did not come out well. If not quite a blacklist, COPE's 1958 voting record was a “political blackball,” according to the Republican Congressional Committee—containing just “2% of the Record!” a print ad exclaimed, “Old Stuff” and “Plenty of Distortion.”Footnote ¹³⁸

The ADA was operating in a changing context, in part due to its own innovations and those of other groups now producing scorecards, and in part due to political circumstance. The 1950s had already seen the greater use of box scores, “right” or “wrong” characterizations, and the ADA's own tentative promotion of percentages; now the 1960 election played an important role in furthering the advance of scores. For the first time since systematic roll call analysis began, one of the presidential candidates was an incumbent lawmaker—not a governor or general or vice president, but a senator.Footnote ¹³⁹ Democrat John F. Kennedy of Massachusetts had a legislative record that could be analyzed, therefore, from his years of service in both the House and the Senate, and it was one the ADA was keen to promote.Footnote ¹⁴⁰ Thus in September 1960, the ADA put out a campaign bulletin highlighting analysis from its most recent congressional scorecard: Senator Kennedy's record had been “100% liberal” during the 86th Congress, it announced.Footnote ¹⁴¹ Moreover, as the ADA had been “keeping its scorecard’ since 1947 (“the very year that Kennedy entered Congress”), it could calculate his “lifetime score” of 90%.Footnote ¹⁴²

Despite this newfound prominence in the presidential campaign, however, percentages remained curiously absent from ADA World’s “Congressional Supplement” in 1960 and likewise in 1961 and 1962.Footnote ¹⁴³ Finally, in the 1963 supplement—published in early 1964—the dam broke.Footnote ¹⁴⁴ ADA World itself had been redesigned, and the “Voting Record Supplement” given a facelift. At last, percentage analysis was placed center stage. The front page signaled the change, featuring prominent photographs of “Four U.S. Senators with 100% Liberal Voting Record.”Footnote ¹⁴⁵ The liberal quotient label appeared in the text, and for the first time, its meaning was explained: “a measure of the liberalism of a member of Congress determined by the percentage of his votes in harmony with liberal policies.”Footnote ¹⁴⁶ The ADA was not couching what it “believed” to be liberal, but pronouncing what was liberal, plain and simple. And this explanation prefaced the voting tables where, also for the first time, the individual liberal quotient scores of each lawmaker appeared in a new column. This summary measure was now instantly available to the reader, as a single convenient score.

Thenceforth, ADA World’s “Congressional Supplements” always included individual percentage scores, and the liberal quotient had truly arrived. It appeared in ADA promotional materials, which were themselves increasingly oriented to the average member or voter. “Find out how your Congressman and Senators voted last year on a dozen or more key issues. Discover their ‘Liberal Quotient,’” a 1964 mailing advised.Footnote ¹⁴⁷ They could discover the 1964 presidential candidates’ LQs, too, for once again, a sitting senator topped the ticket—this time Barry Goldwater on the Republican side.Footnote ¹⁴⁸ ADA World reported Goldwater's dismal LQs with barely concealed glee: just 7% in 1953, 15% in 1954, 7% in 1956, and 8% in 1958. “For every other year his ADA percentage was zero.”Footnote ¹⁴⁹ Goldwater's average across all eleven years in the Senate was a lowly 3%, summing up an ignominious career as far as the ADA was concerned.Footnote ¹⁵⁰

But the ADA did not simply report liberal quotients in increasingly public ways, it acted on them too. The ADA was much more than an elite advocacy group based in Washington, DC; it had a mass membership with supporters across the country, and it participated actively in elections.Footnote ¹⁵¹ Scores helped guide endorsement decisions made by state and local chapters, for example (in the ADA's federated structure, only presidential endorsements were made at the national level).Footnote ¹⁵² And scores helped direct practical campaign support at all levels. Thus each election year from at least the early 1960s, the national ADA raised a general campaign fund to aid liberals, with deserving candidates identified by scores.Footnote ¹⁵³ In August 1960, for example, a staffer compiled a list of fourteen incumbent legislators “deserving of ADA campaign support,” almost all of whom had a “100% voting record.”Footnote ¹⁵⁴ Almost all those who were endorsed or aided, moreover, were Democrats. And while the ADA could claim a “nonpartisan” rationale behind this apparently partisan outcome—that there was simply a larger number of liberals in the Democratic Party, hence their support would naturally be skewed—its leaders also hoped to make that party even more liberal, by deploying scores in the service of party change and promoting an ideological realignment.

There had, of course, been a partisan impetus behind the UDA's “objective” analysis, and the ADA continued in this vein—aiming to show that one party was already better and how it might be made even better, and urging that the other be cast aside. “Our aim is non-partisan,” the first ADA World “Congressional Supplement” had proclaimed; the ADA sought “to elect liberal Congressmen and Senators regardless of party affiliation.” But as it quickly informed readers, “the voting record makes it fairly clear in which party such Congressmen are most likely to be found.” In 1947, “liberal opinion” had “failed to express itself in a politically significant way through the Republican party,” it noted.Footnote ¹⁵⁵ Yet far from inspiring an effort to improve liberal standing within the GOP, it was the Democratic Party, the ADA said, which could and should be improved.Footnote ¹⁵⁶ “The Democratic Party presents the most likely medium for the progressives in this country,” Arthur Schlesinger Jr. had told an early ADA meeting. “The surest way to make the Democratic Party a liberal party is to go into the Democratic Party,” Franklin Roosevelt Jr. had added.Footnote ¹⁵⁷

Thomas Amlie's Let's Look at the Record illuminated the thinking behind this approach. It was no longer possible “to work through the Republican Party,” as “sincere liberals” such as he himself had once hoped, Amlie said.Footnote ¹⁵⁸ The continued election of liberal Republicans, moreover, served “only to confuse the issue” and bolstered an overall reactionary picture in Congress since they would uphold procedural control for their conservative colleagues.Footnote ¹⁵⁹ But if the Democrats were to be the party of liberalism, the Southern wing proved a thorn in its side, diluting the party divisions that Amlie's analysis revealed.Footnote ¹⁶⁰ Roll call analysis served a twofold purpose, therefore, to show that the GOP was a lost cause for liberals and to identify those conservatives who had to be driven from the Democratic Party. As Amlie later wrote, it would help impress upon the public “that a Republican is someone who votes like a Republican.”Footnote ¹⁶¹

In this objective, Amlie and the ADA reflected the scholarly zeitgeist—the idea of “responsible parties,” which political scientists had pushed since the late 1940s.Footnote ¹⁶² The ideal, scholars and liberal activists agreed, was to have cohesive liberal and conservative parties competing against each other, with uniformity and discipline imposed by their national committees. The national committees, however, still lacked the power and enforcement mechanisms to do this, though they were growing in prominence through the 1950s. But ADA scores, along with targeted campaign support in primary and general elections, could provide something of a substitute. Indeed, the ADA criticized CQ’s “party unity” scores as “misleading,” since they inferred “party policy” from the tangled votes of congressional Democrats. In contrast, ADA provided an external ideological standard of evaluation—more in line with the Democratic National Committee's (DNC’s) liberal views. Its publicity director even pitched the Guide to Politics 1954 to the DNC's research director as “good liberal Democratic (capital D) propaganda,” noting that the authors were “all good Democrats—except Wayne Morse.”Footnote ¹⁶³

The ADA's commitment to a strategy of liberalization through party change was clear in their reaction to an apparent setback in 1957. When ADA chairman Robert Nathan reviewed the “crummy” 85th Congress, he found that the average Democratic LQ had actually dropped to its lowest recorded point, just 57% compared to a 73% average in 1947.Footnote ¹⁶⁴ In contrast, the Republican trend was in a liberal direction, achieving “a liberal high” of 43% in 1957 compared to just 17% ten years before.Footnote ¹⁶⁵ Rather than try to promote this GOP trend further, however, Nathan sought to reinforce the northeastern liberal Democrats—ADA's chosen foot soldiers—who were being hamstrung by a resurgent Southern bloc.Footnote ¹⁶⁶ The goal remained defeating those conservatives, replacing them with liberal Democrats, or urging their defection across the aisle.

Even if ADA leaders had concerns as to how far liberalism was progressing in their chosen party, Republicans and conservatives still cast a wary eye on their intent. In 1955, the Senate Republican Policy Committee prepared a report charting what they saw as the ADA's steady infiltration of Democratic Party ranks.Footnote ¹⁶⁷ “The ADA is the first political organization of its kind in our history and it would be idle to dismiss it as a freak,” the report warned.Footnote ¹⁶⁸ In June 1958, conservative journalist M. Stanton Evans, writing in Human Events, highlighted the ADA's “scoresheets” as an important tactic in its mission to impose “piecemeal socialism” in the United States.Footnote ¹⁶⁹ “The picture is clear,” Stanton said, “although some of its ‘liberal’ friends are shedding crocodile tears over its prospects, ADA, outside the South, is taking over the Democratic party.”Footnote ¹⁷⁰ By 1964, Republican ads would confirm this assessment, calling out the ADA's voting record as less an “Honor Roll” than “a Roll of Shame.”Footnote ¹⁷¹

This ad also suggested the potential to invert ADA scores for alternative ends. Indeed, for all their criticism of scores (intertwined with criticism of the groups that made them), at least some conservatives had already caught on to this possibility. The American Retail Federation (a trade association with a conservative economic orientation), for example, reprinted the ADA's 1958 voting charts in its member newsletter—describing ADA as “a leftist group which carries the banner for extreme liberals.”Footnote ¹⁷² “If your Congressman voted a straight ADA party line,” it explained, “you can be certain that his voting record does not adequately reflect the attitude of businessmen who want a sound Government and sound laws.” There followed detailed guidance on interpretation, all the while avoiding outright instruction, for which conservatives had long denounced liberal groups when dabbling in electoral politics. “Measure your own beliefs against your Congressman's voting record and decide whether or not you think he voted right,” it advised. “Above all,” it urged, “DON'T let the complicated looking manner in which the ADA compilation was prepared scare you away.”Footnote ¹⁷³

But would conservatives go a step further and create their own score? Back in 1956, Goldwater had warned against evaluating votes “by our own feelings” and attacked the whole practice of scorekeeping.Footnote ¹⁷⁴ “We ought to start a little organization of our own up here with our own secretary to get up our own voting records just the way we want them for our own States, and then I think we would be all right,” he had joked.Footnote ¹⁷⁵ “In other words,” said Senator McClellan, warming to the theme, “every time we go to bat let's give ourselves a hit.”Footnote ¹⁷⁶ If not quite so self-congratulatory, a conservative group would soon emerge to do just that.

THE AMERICANS FOR CONSTITUTIONAL ACTION

Having spent more than a decade attacking congressional scorecards as inaccurate, misleading, and even subversive, conservatives changed course as the 1950s drew to a close. Scores were generating publicity, and conservatives were increasingly convinced that they were helping labor and liberal groups to make gains in Congress, to the detriment of opposing interests. COPE was now distributing millions of scorecards in election years, while the ADA circulated tens of thousands of its own—to members and beyond.Footnote ¹⁷⁷ And conservatives had nothing to counter with.

They had not entirely ignored roll call analysis to that point. Since the mid-1940s, conservative-oriented business interest groups like the National Association of Manufacturers and U.S. Chamber of Commerce had compiled lawmaker voting records for lobbying purposes, and occasionally published details of individual votes in their member newsletters.Footnote ¹⁷⁸ But as labor and liberal groups turned their roll call analysis outward, business scorekeeping had remained largely behind closed doors.Footnote ¹⁷⁹ In the late 1950s, conservative commentators increasingly pushed for more—arguing that to stem the liberal tide, conservatives must embrace the methods that liberals had pioneered.Footnote ¹⁸⁰ And so it was that conservatives entered the world of congressional ratings once so distasteful to them, with a new ideological interest group founded in 1958, the Americans for Constitutional Action (ACA), leading the way.Footnote ¹⁸¹

The ACA was officially formed in Pittsburgh on August 4, 1958. It was headed by retired U.S. Navy Admiral Ben Moreell, former chairman of the Jones and Laughlin Steel Corporation. Former President Herbert Hoover was a trustee, along with Edgar Eisenhower and an assortment of ex-military leaders and businessmen.Footnote ¹⁸² The new organization was “to be a conservative version of Americans for Democratic Action,” the Christian Science Monitor informed its readers.Footnote ¹⁸³ It would try to “force back together the conservative coalition” that the ADA was doing so much to destroy, by aiding the election of conservative candidates regardless of party.Footnote ¹⁸⁴ And since scorecards played a major role in the ADA's approach, so the ACA would “fight fire with fire, or, more exactly, roll call with roll call,” as the Milwaukee Journal reported in 1960.Footnote ¹⁸⁵ The battle lines of ideological scores were about to be drawn.

More than a year in the making, the ACA released its ACA-Index in 1960.Footnote ¹⁸⁶ Over 122 pages, it offered “an analysis of the voting record of every Senator and Representative now in office on questions vital to the survival of our free society” and showed lawmakers’ positions “on issues which in ACA's judgment tend to strengthen or weaken American Constitutional Government.”Footnote ¹⁸⁷ “Constitutional conservatism” was the ACA's underlying philosophy, it explained, looking to the enduring principles of the Founding Fathers as a bulwark to liberty, now under threat from ADA- and COPE-inspired “statism” and “piecemeal socialism.”Footnote ¹⁸⁸

From the outset, ACA analysis drew from votes across several Congresses, going back as far as 1955 in the case of the Senate.Footnote ¹⁸⁹ And from the outset, too, it provided summary scores for each lawmaker, called the “ACA Consistency Index,” which showed the percentage of a legislator's votes in accordance with the ACA's preferred positions. There was also a series of “Subsidiary Indices” in 1960, using votes from within particular policy areas: for example, the “FOR Sound Money and AGAINST Inflation Index” and the “FOR Individual Liberty and AGAINST Coercion Index.”Footnote ¹⁹⁰ This labeling, in fact, suggests an important reason why ACA leaders chose to create their own scoring system rather than relying upon and reversing those of liberal groups. It allowed the ACA to seize the initiative and argue for something.Footnote ¹⁹¹ As the Reverend I. E. Howard explained in Christian Economics, the ACA-Index was “a help to the conservative senator or representative, because he is no longer in the position of merely having voted against welfare state measures. … The A.C.A. Index gives him a positive platform.”Footnote ¹⁹² Substantively, too, it allowed ACA to define the legislative contours of its conservative agenda, rather than accept the ground on which liberal groups chose to fight. It put lawmakers on notice that their actions in specific policy areas would be monitored and publicized.Footnote ¹⁹³

Indeed, lawmakers who had long suffered at the hands of ADA and COPE ratings, and even critiqued their practices, were pleased to find themselves rated positively by the ACA. “I'm sure glad to learn that I am 100 per cent on somebody's list” wrote Minnesota Republican Congressman Ancher Nelsen to the ACA's Executive Director, “thanks a million.”Footnote ¹⁹⁴ Even if a low ADA score might serve as a badge of honor in some quarters, the psychological and publicity value of doing “well” should not be underrated. As the Milwaukee Journal reported, the ACA “has reversed the roles of hero and villain” that the ADA presented.Footnote ¹⁹⁵ And these newly minted conservative “heroes” tended to view the ACA and its methods much more positively than they had regarded the ADA or COPE. As Rep. Jack Westland (R-WA) wrote, “a group such as yours has long been needed to present a true picture of voting records and to act as an antidote to some of the so-called ‘vote records’ of organizations such as ADA and COPE.”Footnote ¹⁹⁶

For all he had once denounced “so-called voting records” and warned against evaluating Congress “by our feelings,” even Senator Goldwater was willing to publicly endorse the ACA project and attest to its “objectivity.” “Your ACA-Index is the finest work of its kind ever put between covers,” Goldwater wrote Moreell, “for it tells the truth about each Member of Congress without bias of any kind.”Footnote ¹⁹⁷ As ACA Assistant Director John Synon wrote in Human Events, the indices were constructed from a conservative perspective but without “favoritism”: “The same criteria that took apart Lyndon Johnson are applied to every other member of Congress and just as thoroughly.”Footnote ¹⁹⁸ “Objectivity” was being reinterpreted not as a disinterested standard of assessment, but as a consistent method of application. For conservatives, scores were no longer bad in themselves, but they could be badly done.Footnote ¹⁹⁹

For their part, liberals accepted the ACA's right to engage in scorekeeping, focusing their attacks on the organization itself. Much as conservatives had once denounced the ADA and other liberal groups as communistic, liberals now portrayed the ACA as a part of an extremist “radical right” with links to controversial groups like the John Birch Society (JBS).Footnote ²⁰⁰ (ACA leaders claimed they were part of the “responsible” right and denied any official connection to the JBS, though the unofficial links were somewhat murkier.Footnote ²⁰¹) But liberals were not the only ones to level criticism at the ACA.

Self-described conservatives who fared poorly by the ACA's metrics were less enamored of its project. Perhaps the strongest complaint came from those representing farming regions, who objected to the negative scoring of agricultural subsidies and price supports.Footnote ²⁰² “We of the Midwest, have tried to be conservative in all matters,” Kansas Republican Congressman Wint Smith wrote to an ACA staffer, but the group was failing to recognize the realities of the post–New Deal agricultural system.Footnote ²⁰³ His objections were relevant for Southern Democrats, too, whose scores were often lower than they expected. Mississippi Congressman William Colmer noted “much criticism” of the 1960 ACA-Index among Southern conservatives, for example, and expressed concern over his 66% score in the updated ACA ratings released in 1961.Footnote ²⁰⁴ “Since I regard myself as a conservative and am frequently referred to as such and since according to my own scoring, I should be nearer 100%, I find it difficult to understand your methods,” he wrote.

Methodologically, this might be viewed as a concern about weighting or the lack thereof. Where specific issues might play a larger role in a lawmaker's self-conception and reputation as conservative (such as opposition to federal civil rights legislation in the case of most Southern Democrats), each vote counted equally in calculating the ACA's scores. In response, ACA actually took steps to downplay agricultural issues in subsequent years—fudging some of its supposedly “objective” standards so as to aid lawmakers from farm areas in the South and West. “At one time there was sharp criticism because of too much emphasis on agriculture,” Executive Director Charles McManus explained to ACA's Trustees in early 1963. “There were three or four votes given on agricultural bills. We have corrected that.”Footnote ²⁰⁵ The scores were still not weighted, but agricultural votes were quietly de-emphasized in constructing them.

Beyond the agriculture case, however, the ACA's scores were envisaged as instruments of change, not changeable instruments. And in part, they were psychological instruments, intended to encourage improved legislative performance. Numerous legislators with high scores expressed a touch of jovial regret that they had not done better or asked for more serious feedback on how they might improve their scores.Footnote ²⁰⁶ As a 1964 ACA newsletter acknowledged, most congressional inquiries boiled down to a “plaintive query”: “What did I vote wrong on?”Footnote ²⁰⁷ The ACA was tapping into the powerful human urge to do well against standards by which we are judged, whatever those standards may be. And the lure of a perfect score was not lost on the competitive personalities in the Congress. ACA even distributed special pins to legislators with 100% scores in the 1961 Index.Footnote ²⁰⁸ They also introduced a “Distinguished Service Award” for consistent high scorers, a “public relations ‘gimmick’” staff admitted, intended “to provide favorable publicity to conservative members of Congress in off-election years.”Footnote ²⁰⁹ A total of 136 lawmakers received the award in 1961, 154 in 1963, and 128 in 1965.Footnote ²¹⁰

To be eligible for a Distinguished Service Award in 1961, a lawmaker must have obtained an ACA score of 65% or better.Footnote ²¹¹ Where a .400 rating in baseball would afford entrance to the Hall of Fame, expectations for political “batting averages” were thus far higher. Yet 65% was originally viewed as a relatively low bar, Moreell acknowledged in a confidential memo, reflecting concerns over the “somewhat debatable validity” of certain votes included in the original scores (perhaps referencing the agriculture issues) and the possibility that lawmakers had not fully appreciated “the probable effects of pending legislation on the integrity of the Constitution,” since the ACA had not yet had chance to inform them properly.Footnote ²¹² As the ACA “sharpened up” its methodology and expanded its educational efforts, Moreell believed “we will be able to raise the qualifying percentage to an appreciably higher figure than 65%.”Footnote ²¹³

Despite such aims, the cutoff appears to have hovered around the 65% mark thereafter (applied to cumulative career scores, rather than performance in a single congressional term), while Moreell found himself fighting with those who sought a higher standard of conservative “purity.”Footnote ²¹⁴ “One of the most difficult tasks I have is to convince dedicated constitutional conservatives that they should not expect a candidate to agree 100% with their ideology,” Moreell wrote in September 1962.Footnote ²¹⁵ McManus sounded a similar theme in 1965, critiquing “so-called ‘conservatives’” who worked to defeat candidates “who did not agree with them on each issue to the exact same degree,” and warning of “even greater problems from this element” in future.Footnote ²¹⁶ Yet scores like the ACA's index helped to propagate that kind of perfectionism.

Like their counterparts in the ADA, the ACA's leaders had seen the 1960 presidential election as an opportunity to promote their scores, rating Kennedy as just 11% “conservative” by their measure.Footnote ²¹⁷ Numerous other groups had joined the scoring bandwagon, too. Kennedy was a “100%” candidate for both COPE and the Farmers Union, while the Farm Bureau was less enamored, rating him a mere 8% on the agricultural issues they cared about.Footnote ²¹⁸ Perhaps due to the attention so generated, CQ began to gather all the major ratings together for the first time in its October 1960 Weekly Report, listing scores for individual lawmakers side by side, in convenient tables ripe for dramatic analysis by journalists.Footnote ²¹⁹ Indeed, CQ went so far as to generate percentage ratings for groups such as COPE, which still produced only right/wrong box scores in their congressional analysis.Footnote ²²⁰

So normalized and widespread had ratings become by 1966 that a Washington Post journalist could speak of annual “report card time on Capitol Hill” with rankings “from every political angle.”Footnote ²²¹ Journalists typically highlighted the extreme scores—casting “100 percenters” for different groups against each other, as leading warriors for their respective political philosophies.Footnote ²²² “Tower, Price Rated 100, Yarborough 0 by ACA,” the Dallas Morning News informed its readers in 1967, emphasizing the wide ideological gulf between the two Texan Senators: one a Republican and the other a Democrat.Footnote ²²³ Scores also lent a horse-race quality to comparisons of state delegations, much as polls do in election coverage. The Wichita Eagle fully embraced the metaphor in 1967, describing how freshman GOP representative Larry Winn Jr. had “nosed out” his colleague Bob Dole “for first place in the Kansas congressional delegation” (in “a photo finish,” of course).Footnote ²²⁴ Dole's 92% ACA score was still high (just behind Winn's 95%), but such relative assessment could further fuel the drive toward ideological perfection.

Beyond the media coverage and psychological pressure that might alter congressional performance from within, the ACA's and other scores were also designed to shape Congress from without, through elections. But if roll call analysis had once been oriented to elites and the presentation of scorecards designed to downplay any direct influence on voter choice, the introduction of percentages signaled an extended audience and a desire for greater impact.

Senator Harry F. Byrd of Virginia had commended the 1960 ACA-Index for its accessibility—something that could “be understood by the man on the street.”Footnote ²²⁵ But the ACA had initially stuck to an elite strategy of influence, distributing free copies of the full 122-page 1960 Index to conservative politicians and sending out nearly 8,000 “ACA Performance Fact Sheets” to newspaper editors across the country, providing information on local incumbents.Footnote ²²⁶ Further reproduction of ACA scores was tightly regulated, though, in hopes of generating funds through sales.Footnote ²²⁷ This was the last vestige of scorekeeping as a kind of service for which political professionals would pay. Thus ACA sold or distributed only about 3,000 copies of the full Index in 1960.Footnote ²²⁸ At $15 for the Index and $5 for a fact sheet, they were too expensive for the average citizen.”Footnote ²²⁹ The key was quality, not quantity, in the ACA's original vision.

But feedback from conservative candidates following the 1960 campaign caused them to reconsider. “Recognition was given to the fact that the INDEX, in its present form, was directed toward thought leaders,” McManus summarized, but candidates hoped moving forward “that it would be directed to the thought level of the average voter.”Footnote ²³⁰ Thus in 1961, the ACA produced an abbreviated Index for use in campaigns, listing only recent votes, providing cumulative “consistency index” scores, and eliminating some of the subindices.Footnote ²³¹ It also quietly recalculated the 1960 “batting averages” in line with these new categories.Footnote ²³² This time, both the full and abbreviated publications were distributed free to all ACA members and beyond: 12,000 copies of the full 1961 ACA-Index went out, and 15,000 in 1962.Footnote ²³³ The print runs expanded to 20,000 in 1964, ’65, and ’66.Footnote ²³⁴ In 1962, moreover, candidates were permitted to reproduce parts of the Index in their campaign ads, and news outlets were allowed to reprint the scores in their entirety.Footnote ²³⁵ And in the mid-1960s, the ACA began to produce interim ratings before the end of a Congress, too, made possible by the introduction of computer processing.Footnote ²³⁶ The goal was to keep the ACA and its scores continuously in the public eye.

The goal was also to reach voters more directly. “ACA ratings should be in the hands of every voter if possible,” one trustee urged at a 1965 meeting, “just to show whether the fellow votes the way he talks.”Footnote ²³⁷ Reflecting this aim, the presentation became still simpler, ever more focused on the ultimate consistency score and less on what went into making it. The remaining subindices were gradually eliminated from ACA publications aimed at a broad audience, and by 1967, all references to specific bills had disappeared from at least one leaflet. It simply listed lawmakers’ names and their scores—current and cumulative—a percentage to sum up an entire career.Footnote ²³⁸ As distribution of these simplified publications expanded (e.g., 150,000 copies of a similar 1968 leaflet), legislative lives were increasingly reduced to a contest of numbers.Footnote ²³⁹

Stripped bare of their moorings, those numbers took on new meaning in the electoral context—becoming the functional equivalent of endorsements. The ACA did make formal endorsements, too, as did the ADA (with decisions made at the national level in the ACA's case, reflecting its top-heavy organizational structure). But for most voters, there was little distinction between hearing that a candidate had been endorsed by a group or that he or she had rated highly according to a group's criteria.Footnote ²⁴⁰ And scores helped determine those endorsements in any case.

Beyond endorsements, scores served other functions in the campaign context, where both the ACA and ADA were expanding their activities in the 1960s, coming to operate almost as PACs—even experimenting with providing financial support directly to candidates, which would be a defining feature of PACs as codified in the campaign finance reforms of the 1970s.Footnote ²⁴¹ The ACA's broader purpose, after all, was “to help elect any conservative who will work toward our Salvation,” and they offered tangible support to favored candidates.Footnote ²⁴² In 1960, the ACA prioritized general election assistance for “incumbents rating relatively high in the ACA-INDEX” and “candidates who are attempting to replace some of those rating low in the ACA-INDEX,” thus providing some level of support in ninety-four House and Senate contests.Footnote ²⁴³ More complex guidelines were formulated in subsequent years, taking into account the competitiveness of districts and the strength of the opposing candidate, but scores still played a key role.Footnote ²⁴⁴ Indeed, the ACA even provided behind-the-scenes support to high-scoring candidates who, sensitive to local conditions, had refused their public endorsement.Footnote ²⁴⁵ The ACA's backing could be a kiss of death, too, it appeared. In 1966, for example, it provided campaign assistance to 225 candidates, but only publicly endorsed 135.Footnote ²⁴⁶ It also offered primary campaign support in some cases.Footnote ²⁴⁷

The ACA's brand of assistance was initially in-kind. It provided “manpower and other material aid to candidates rather than money contributions.”Footnote ²⁴⁸ On the “manpower” side, the ACA provided staff support—hiring campaign managers, press secretaries, and pollsters to work with candidates for the course of the election.Footnote ²⁴⁹ In 1964, for example, it hired seventy-five temporary employees for the duration of the campaign.Footnote ²⁵⁰ In 1966, it “arranged for the services of eighty-three campaign specialists, firms or agencies, to work in behalf of constitutional conservative candidates.”Footnote ²⁵¹ In terms of “material aid,” ACA staff wrote personalized press releases for candidates, provided “speech kits,” and purchased direct mail lists on their behalf.Footnote ²⁵² The ACA expected to need between $250,000 and $300,000 to support all these activities in 1966.Footnote ²⁵³

That year, however, the ACA's leadership began to think more about direct financial support to candidates, though funneled in indirect ways. They “inaugurated a program of special assistance to large [ACA] contributors,” which effectively recommended “worthy candidates,” to whose campaigns they should donate.Footnote ²⁵⁴ In some cases, the donors sent checks “made payable to the candidate committee for conveyance by ACA to the candidate.”Footnote ²⁵⁵ The ACA was essentially operating as a clearinghouse for campaign contributions, but a confidential one.Footnote ²⁵⁶ It expanded this initiative substantially in 1968—now called Programmed Financial Support—raising nearly $116,000 for candidates (while spending a further $237,200 on in-kind campaign assistance).Footnote ²⁵⁷ The ADA, too, was experimenting with a similar program at this time. In 1966, for example, ADA staffers sent a confidential memo to their top contributors, providing details on Senate candidates (primarily the LQ) and encouraging them to “make some contribution to the candidate of your choosing.”Footnote ²⁵⁸

But where the ADA was always committed to a strategy of party realignment, the ACA was formed with more bipartisan intent—aiming to stop the progress of liberalism in the Democratic Party and “force back together the conservative coalition.”Footnote ²⁵⁹ Its leaders were not, at first, intent on fashioning the Republican Party into a more conservative counterweight, though their relationship to the Southern Democrats was sometimes strained. The ACA wanted to expose those “posing as conservatives,” even as they downplayed farm issues to placate Southern members of the conservative coalition.Footnote ²⁶⁰ But increasingly in the early 1960s, sustaining that coalition became more of a backstop, with a conservative GOP moving into sight as the key goal.

The ACA's analysis encouraged people to see nascent party divisions, that is, to recognize real differences between the Democratic and Republican parties, often for the first time.Footnote ²⁶¹ “Many of us in the pundit business have fallen into errors of political hypothesis” the editors of the Richmond News Leader admitted to their readers, but they were now, as a result of the 1960 ACA-Index, “happily subject to some revision and correction.”Footnote ²⁶² Over the course of the past decade, they had subscribed to the general view that “there really isn't much difference” between the parties.Footnote ²⁶³ But the Index had revealed that there was “a great difference,” one “capable of being objectively measured.”Footnote ²⁶⁴ As the ACA's John Synon explained in Human Events, the weight of each party's membership fell in different ideological directions. “That every member of Congress the ACA-Index rates at zero is a Democrat and every member rated 100 per cent is Republican is evidence enough of the ideological difference in parties,” Synon concluded.Footnote ²⁶⁵ Indeed, for the Richmond News Leader editors, the Index raised questions about the South's commitment to the Democratic Party—showing the mismatch between the conservative principles of Southern voters and the national thrust of their party.Footnote ²⁶⁶ “If the Republic ever comes to its political senses,” they reflected, “we will choose up sides all over again.”Footnote ²⁶⁷

Choosing sides was also becoming important inside the ACA. “The Republican Party, as a whole, is the Party of constitutional conservatism and individual freedom,” Moreell told his colleagues just before the 1964 election—a contest in which they would endorse the Republican nominee for president and offer lopsided support to GOP congressional candidates.Footnote ²⁶⁸ The Democratic Party, in contrast, stood for “democratic socialism, government intervention in the people's affairs and regimentation of the citizens.”Footnote ²⁶⁹ There were exceptions, he acknowledged, but “the preponderance of each of the Parties is as I have stated.” The Republicans had a 77.8% average ACA score, while the Democrats languished at 21.3%. That there remained some outliers compared to these averages—liberal Republicans and conservative Democrats—Moreell continued, “confirms the view of many students of political science that the time is here for a re-alignment of the two Parties into conservative constitutionalists and liberal centralists.”Footnote ²⁷⁰ Goldwater's presidential candidacy or South Carolina Senator Strom Thurmond's recent switch from Democrat to Republican may have helped to convince them.Footnote ²⁷¹ Or the fear that liberalism had progressed too far in the Democratic Party to be undone. But this statement suggests that the ACA was now committed to the same realignment that the ADA had long sought to bring about.Footnote ²⁷²

And scores were a tool for achieving that realignment. As Richard Viguerie, the direct mail guru of the 1970s New Right, has observed, the ACA-Index offered conservatives “a precise way of knowing which Republicans to support or oppose, based on their voting record, and they used these ratings to help change the face of the GOP.”Footnote ²⁷³

THE LEGACY OF SCORES

The system of scoring legislators according to ideological criteria, first championed by Thomas Amlie and the UDA, emulated and refined by both the ADA and ACA, reflected a profound shift in the methods and the mentality of politics in the mid-twentieth century. In seeking to reshape the party's composition at the candidate level, these ideological interest groups undertook something that the regular party could not do, even if it wanted to. While both national committees had established research divisions by the 1950s, conducting some roll call analysis themselves, they had a limited capacity to pronounce party positions or punish deviation from them.Footnote ²⁷⁴ Primaries denied parties control over use of their label, and national campaign funds were traditionally distributed on the basis of need, not party orthodoxy. The committees could do little, therefore, to make their parties more “responsible” in the sense that many scholars and activists desired.

Interest groups faced no such constraints, and in the 1950s and 60s, especially, the ADA and ACA sought to use their scores to change the congressional parties—offering a form of ideological “quality control,” in V. O. Key's phrase, from the outside. This was ironic, since proponents of “responsible parties,” such as E. E. Schattschneider, had viewed interest groups more as an impediment to their emergence rather than an instrument of their creation.Footnote ²⁷⁵ But these groups came to provide a kind of institutional graft onto the national committees, lending their assistance in an area of notable weakness, while operating within an informal network around the formal party apparatus, as contemporary scholars have conceptualized such relationships.Footnote ²⁷⁶

The ADA and ACA hoped to encourage incumbents to improve their scores, but more importantly, they wanted to urge party outliers to switch to the “correct” party for their ideology or to replace them with new candidates whose principles and party label were better aligned. Thus they leaned toward improving the team by poaching from the opposition or by “retiring” some of the existing members, to use the Farm Bureau's baseball analogy. More generally, these groups hoped to steadily increase the number of liberal Democrats or conservative Republicans in Congress. And they would see their hopes increasingly fulfilled over the next few decades.

The conversion or electoral replacement of party outliers has been accorded an important role in the growing polarization of Congress through the latter twentieth century, and has been traced to the 1960s and even the late 1950s in origin.Footnote ²⁷⁷ To what extent ideological scores contributed to this polarizing trend cannot be ascertained quantitatively, but it is clear that ADA and ACA leaders intended them to help bring just such a trend about.Footnote ²⁷⁸ They pioneered an important tactic that could measure and promote party purity according to broad ideologies and specific issues—one that was adopted and refined by numerous groups on the left and right as the years progressed. It is also clear that both organizations felt their actions were influential, and that they each feared the advance of their opponents.Footnote ²⁷⁹ Index scores helped direct their money and manpower along ideological and partisan lines, and served as a straightforward signal to primary and general election voters about the candidates the ADA or ACA would have them support.

Indeed, the ADA today is still providing those signals. It created an official PAC in 1979 and remains an advocate of liberal principles in elections and legislation, if a less prominent voice than in the past. The ACA, in contrast, is no longer active. It was, in some respects, a victim of its own success, finding itself edged out by rival conservative groups as the movement gathered steam in the 1960s. A new umbrella group founded in 1964, the American Conservative Union (ACU), would ultimately eclipse it.Footnote ²⁸⁰ Originally a clearinghouse intended to coordinate conservative efforts, the ACU increasingly ventured into ACA territory and began compiling its own ratings in 1971.Footnote ²⁸¹ That year proved to be one of the ACA's worst financially, and its scale of activity, already in decline, fell dramatically.Footnote ²⁸² The ACA continued to produce an Index until the mid-1980s, but then disappeared from view. In contrast, the ACU has endured and continues to produce conservative scores, which scholars have utilized as replacements for the ACA's.Footnote ²⁸³

The use of scores, of course, is not confined to scholars alone. They have been widely publicized in the media and diffused into the discourse and culture of American politics. As a rhetorical shorthand, they encourage the public to categorize politicians in distinct and divided ideological terms, and they draw particular attention to the upper and lower ranges deemed “good” or “bad.” Even in their own materials, ideological groups emphasized the extremes. A 1968 ADA World article, for example, contrasted “100 percenters” with lawmakers receiving “the dubious honor of Perfect Zero.”Footnote ²⁸⁴ Journalists readily adopted the divisive frame of “heroes” and “zeroes,” further encouraging such diametric comparisons. In this sense, scores may contribute to a polarized mentality, shaping the way people think and talk about politicians.Footnote ²⁸⁵

At face value, scrutiny of the legislative record may seem an exemplar exercise in democracy—in ensuring transparency and accountability—with index scores a manifestation of that noble vision. The pioneering liberal groups claimed that a lawmaker's record could speak for itself. But in their selection of issues to examine, rating groups often aimed to benefit their friends. There was something to the critique that records could be distorted through selection, and that tabulating a legislator's record across a few issue areas, let alone offering a single comprehensive “score” of their performance, did a disservice to the complexities of the legislative process and to the nuanced motivations of legislators. There was something to the critique that a lawmaker's votes from long ago should not come back to haunt him or her. And there was something to the critique that voters, too, should not be told exactly how to vote.

Indeed, ideological interest groups were sensitive to these criticisms, even as they embraced roll call analysis as a way to promote their political aims. Conservative attacks were undoubtedly motivated as much by who was doing the scoring as the practice of scoring itself, but their objections tapped into a broader cultural discomfort with interest group involvement in elections in the early twentieth century. Elections had long been seen as an arena for party combat, and the entrance of new players, organized nationally, was viewed with suspicion in many quarters—as somehow subverting democratic norms and individual decision making, seeking undue political influence, and bringing outside actors into local elections. In response, liberal groups modulated their language, format, and presentation—seemingly minor matters that were here deployed meaningfully, aiming to minimize criticism, even alter these negative attitudes. The ADA's reluctance to publicize clear percentage scores for years after they had implicitly begun to use them, for example, is testament to these concerns. So is the careful language in which they initially packaged scores to the voting public as guidance, not instruction.

That conservatives ultimately ceased attacking the practice of scorekeeping and created their own score instead speaks to the transformation in attitudes that a decade of union and liberal group electoral activity had already achieved, and it helped eliminate the last vestiges of opposition. Ratings became acceptable tools of political conflict, waged by interest groups in pursuit of ideological and ultimately partisan goals. The transmutation of scores into informal endorsements, for example, showed that concerns over electoral instruction were no longer a constraint. ACA claims—echoed by the ADA and other rating groups—that voting records were intended to guide but not determine a voter's choice, became ever more tenuous as the underlying roll call information disappeared from its publications.Footnote ²⁸⁶ With no other information than a lawmaker's score—high or low—what room was there for independent judgment? As columnist Charles McDowell noted in a last-ditch critique of the ACA's 1967 interim scores, it was making citizens into mere “meter-readers.”Footnote ²⁸⁷

Groups signaled in the small print that scores might not be everything—that there were “inherent limitations” to roll call analysis and aspects of legislator performance that could not be assessed “in statistics or percentages,” as the ADA World’s “Congressional Supplement” acknowledged in 1959.Footnote ²⁸⁸ Yet by the mid-1960s, ideological scores were serving to guide not only endorsement decisions but also more fundamental evaluations of a lawmaker's worth. Entire careers, across years or even decades, were being compressed and contorted into a single number. As Ed Foreman, a former Republican congressman and ACA trustee, told his colleagues in 1965: “When I looked at a Member of Congress, I got to where I looked at him as a number. I got to where I could look at a guy and tell you what his ACA-Index rating was. And, in a way, that helped form my opinion as to what kind of a person he was.”Footnote ²⁸⁹

The interest group progenitors of ideological scores encouraged this reductive outlook. They sought to boil the complexities of congressional performance down to a simple metric, to categorize legislators on that basis as liberal or conservative, and to shape how those terms were understood. They deployed their efforts in the service of party realignment, believing that ideological party purity would better secure their policy goals than a traditional nonpartisan approach. Their key innovation, the index score presented as a percentage, might be viewed as a small tactical move and as an obvious way to summarize and report legislative records. But its path to prominence was more complex, and the strategic purposes behind it were more ambitious and partisan than previously acknowledged by scholars. By detaching such scores from their context, in sum, we have missed much of their meaning. The numbers themselves have a story to tell.