REALIGNMENT IN THE DEEP SOUTH STATE OF MISSISSIPPI

Stephen D. Shaffer and Monica Johnson

Mississippi State University

Since V.O. Key's (1955) seminal article, "A Theory of Critical Elections," and Walter Dean Burnham's classic book, Critical Elections and the Mainsprings of American Politics, political scientists have been fascinated with the extent to which an election or series of elections can usher in new political eras dominated by a political party that captures the support of a majority of voters. Everett C. Ladd and Charles D. Hadley's (1978) Transformations of the American Party System meticulously details the rise and evolution of the most recent New Deal party era dominated by the Democratic party, as well as the long-term move of the South away from the Democratic party, a process that Key (1959) labelled "secular realignment." The changing nature of party politics in the modern South has sparked considerable debate with Swansbrough and Brodsky (1988) concluding that both realignment and dealignment may characterize this key region, and Lamis (1990) pointing out that GOP gains have produced a true Two-Party South.

Recent political events have rekindled the scholarly fascination with the study of partisan realignment, as the GOP capture in 1994 of both houses of Congress for the first time in forty years prompted Burnham (1996) to characterize it as a "seismic event" in an article entitled, "Critical Realignment Lives: The 1994 Earthquake." Particularly interesting was that for the first time since Reconstruction, Republicans now held a majority of the U.S. House and Senate seats from the eleven states of the old Confederacy. Underscoring the importance of The Vital South (Black and Black 1992) to both parties, five of the top six GOP congressional leadership positions in both chambers by 1995 were held by southerners, and all three Democratic presidents since 1963 hailed from the South. Yet only eighteen months after the 1994 elections, polls showed the GOP Congress and Speaker Newt Gingrich in public disrepute and Democratic president Bill Clinton with a 20 point lead over his GOP opponent, clearly illustrating one of Sundquist's (1973) themes that the final outcome of a realignment process can be shaped by the strategically wise or unwise actions of political leaders.

To gain insight into the nature of the current realignment process taking place in Dixie and its likely outcome, we focus on the state of Mississippi, which has been a leader historically of the southern Democrats and today's modern Republicanism. Mississippi governor Fielding Wright was the Vice Presidential candidate of the States' Rights Party in 1948, and Democratic senators James Eastland and John Stennis held such powerful positions as chairs of the Judiciary, Armed Services, and Appropriations committees and President Pro Tempore. By 1996 Mississippians Haley Barbour and Evelyn McPhail were the chairman and co-chairman of the Republican National Committee, while Majority Whip Trent Lott and Senate GOP Conference Chairman Thad Cochran faced off for the Senate Majority Leadership position vacated by Bob Dole, and Mississippi GOP Congressman Roger Wicker served as chairman of the House Republican Freshman class. Yet state Democrats remained a force to be reckoned with, winning all statewide races in 1995 except governor and retaining two-thirds of the state legislative seats in each chamber. Especially interesting was the changed nature of the Mississippi Democratic party, which in the 1980s had provided the nation with the first African-Americans to serve as a state party chairman (Ed Cole) and as a Congressman from a rural district (Mike Espy), and in the 1990s continued to provide such progressive leaders as Attorney General Mike Moore, a leader of the state fight against the tobacco companies (to recover state medicaid payments for smoking-related diseases). With such successful political leaders battling for the souls of a state's voters, important insights may be gained into the nature of political conflict in the "New South."

During the Reagan 1980s the Democratic New Deal majority was clearly in danger as the Democratic advantage over the Republicans in the public's partisan identifications narrowed significantly, primarily because of the decay of Democratic loyalties among white Southerners (Petrocik 1987). By November 1994 the Democratic advantage nationally had fallen to a mere 3 point edge (Stanley 1996). Though migrants into the South were more Republican than native southerners, most scholars concluded that GOP gains were primarily a product of conversion by native white southerners (Campbell 1977). Debate raged over the policy correlates of partisan change among whites with Carmines and Stimson (1981) stressing the importance of race, and Beck (1982) and Abramowitz (1994) pointing to the importance of ideological issues in general, such as social welfare and national defense. Downplaying conversion, other realignment scholars studied the mobilization of previous non-voters to a new majority party, a possible source of the 1930s realignment (Campbell 1985). More recently, while younger white southerners are more Republican in identification than older whites, the GOP has made gains among all age cohorts (Petrocik 1987).

A less researched area is the important role that political leaders can play in promoting or hindering partisan realignment. As Franklin Roosevelt did for the Democrats in the 1930s, Ronald Reagan may have boosted Republican hopes, particularly in the South (Swansbrough and Brodsky 1988). Yet excessive GOP conservatism in the 1980s may have cost them a more enduring status as the majority party (Hurley 1989). By the 1990s southern Democratic party officials had moved to the left ideologically, while Republicans were clearly conservative, causing considerable speculation about which (if either) party might be out-of-touch with the average southerner (Moreland 1990; Hadley and Bowman 1995). And in the final analysis, the decreased saliency of political parties to voters today and their tendency to vote for the candidate rather than for the party gives candidates of both parties an opportunity to moderate their ideological message in order to win office, a strategy increasingly employed by today's southern Democrats (Stanley 1988).

Our information on the party identification ties of Mississippians is drawn from a unique series of statewide telephone polls conducted by the Social Science Research Center at Mississippi State University from 1981 to 1996. These surveys, usually conducted in a two-week period in April and having a sample error in the three to five percent range, utilize a random digit dialing procedure and are weighted by demographic characteristics (education, race, sex). Independent leaners are classified as partisans, as our analyses confirm those of other studies in finding that leaners behave in as partisan a manner as do weak partisans (Petrocik 1987; Stanley 1988; Abramowitz 1994). While other studies of the public's partisan ties have concentrated heavily on all adults (Swansbrough and Brodsky 1988), we also examine the partisanship of an even more critical group--"likely voters." Our likely voter indicator is a simple additive index of three highly correlated items reflecting campaign interest, political knowledge, and expressed likelihood of voting. Comparisons of actual state election results over the 1980s with likely voter intentions were consistently within sample error.

Though most studies of in-migration have defined migrants based on the region of birth or the region where one grew up, some have pointed out that political socialization is a life-long process (Dawson, Prewitt, Dawson 1977; Jennings 1987). Mississippi has historically had a very traditionalistic culture, characterized by a "closed society" reflected in segregation and minimal public participation in government, so any exposure to a different political environment could significantly affect one's political orientations (Elazar 1984). Therefore, we define pure natives as those who have lived in Mississippi all of their lives, and examine the influx of "newer ideas" by singling out those who have lived outside of the Magnolia state for one or more years.

Reflecting party identification patterns in the South more generally, Mississippi has been transformed from a Democratic bastion in the early 1980s to a competitive two party state today. From a greater than 20 point Democratic advantage among all adults in the 1980s, by the mid-1990s Democrats held only a six point edge over Republicans (table 1). While Democrats retained the support of over 80 percent of African-Americans (data not shown), their support plummeted among whites. While Democrats retained a slight plurality in party identification among whites in the early 1980s, the partisan gap essentially disappeared in the late 1980s. Particularly dramatic were Republican gains among whites after 1990. By 1996, the party that had previously symbolized the old Confederacy retained the support of only 31 percent of white Mississippians, while the rising GOP was favored by 58 percent. The "white flight" from the Democratic party left an increasingly "black" Democratic party. The African-American presence among Democratic identifiers rose from 40 percent in 1981 to a majority of 56 percent in 1996. No longer threatened by the possibility of exclusion by white racists, Democrats at the 1996 state party convention abandoned their racial quotas, and promptly elected African-Americans to both national committeemember slots.

The party realignment taking place in Mississippi is even more striking when focusing on likely voters. In the 1980s during the terms of Democratic governors William Winter, Bill Allain, and Ray Mabus, and Senator John Stennis, a majority of likely voters described themselves as Democrats. This partisan gap disappeared in 1990, the year before the election of the first Republican governor and lieutenant governor since Reconstruction, and since then a majority of likely voters now identify with the Republican party. In 1992 Mississippi gave Republican President George Bush his highest vote total of any state, and in 1994 Republican Roger Wicker easily captured (with a 63% vote total) a U.S. House seat held by the retiring Democratic chairman of the House Appropriations Committee, Jamie Whitten, who was regarded as the "last of a breed--the conservative Southern Democrats" (Duncan 1993).

The in-migration of new ideas has played an important role in transforming the partisan ties of white Mississippians. While whites who had lived their entire lives in the Magnolia state were over 10 percent more Democratic than Republican throughout the 1980s, those who had lived part of their lives in another state were in the vanguard of partisan change (table 2). Even as early as 1981 such "in-migrants" were pretty evenly split between the two parties. In the critical period between 1984 and 1990 when GOP gains eliminated the partisan gap among whites, these GOP gains occurred only among whites who were not life-long residents of Mississippi. After 1990 these Republican values began to diffuse across the white population more generally. Such diffusion was so widespread that by 1994 for the first time in state polling history, a majority of life-long residents were now calling themselves Republicans, a development that had occurred among other residents four years earlier.

A key element in the dramatic GOP gains after 1990 has been the exodus of white conservatives from the old Democratic party. Conservatives had always found a comfortable home in the old segregated Democratic party, and even as late as 1990 a sizable 41 percent of self-identified white conservatives regarded themselves as Democrats (table 3). Conservative support for the Democratic party sank after the election of conservative Republican Kirk Fordice as governor to only 20 percent in 1992, and remained low at 18-21 percent in 1994 and 1996. A historically staggering 68-73 percent of white conservatives now regarded themselves as Republicans.

The realignment of the state parties along race and ideological lines has clearly transformed the state Democratic party into a more progressive organization. In the early 1980s, moderate and conservative whites comprised about half of Democratic party identifiers, while African-Americans and white liberals were equally numerous (table 4). By the mid-1990s, moderate and conservative whites now made-up only about one-third of Democratic identifiers, with African-Americans and white liberals dominating two-thirds of the party. And while African-American Democrats in the population are nearly as split on the ideological self-identification question as are white Democrats, African-American Democratic party activists are nearly always more liberal than their white party colleagues on a host of specific policy issues (Hadley and Bowman 1995, 171). Given the moderate to conservative values of the great majority of Mississippians, the increasingly liberal orientation of state Democrats could pose some serious electoral problems.

Whites' partisan ties may also be shifting in reaction to national and state political leaders, as suggested earlier. After the segregationist Democratic governors of the 1960s, the Magnolia state elected "New South" Democrats, whose moderate philosophies prevented Mississippians from viewing them in ideological or partisan terms. The election of a staunch Reagan conservative in 1991, Republican Kirk Fordice, has clearly changed that situation. A majority of whites rating Fordice's performance as excellent or good in the 1990s called themselves Republicans, and less than 30 percent were Democrats. Those rating him as only fair or poor were clearly less Republican in orientation (table 5). Furthermore, Republicans made significant gains between 1992 and 1996 among whites rating Fordice favorably, but not among those rating him unfavorably. And while Fordice's popularity appeared to be helping GOP hopes, President Clinton was a clear drag on Democratic prospects. A sizable 75-78 percent of white Mississippians rated Clinton's job performance as only fair or poor in 1994 and 1996, and Republicans maintained a three-to-one advantage over Democrats among these Clinton detractors. Clinton's clear negative impact on his party's hopes can be contrasted with Bush's positive, though modest, help to the GOP. Bush backers were more Republican than Bush detractors, though Republicans made relatively equal gains between 1990 and 1992 among those rating Bush favorably and unfavorably.

Some insight into the future of party politics in Mississippi may be gained by examining the attitudes and orientations of young whites, as generational replacement proceeds to further transform the political landscape. Young people have been in the vanguard of partisan change in Mississippi. Even in the early 1980s, a plurality of young whites preferred the Republican to the Democratic party, while a plurality of other age groups generally backed the Democrats (table 6). Middle aged whites were the next to shift towards the GOP, and by the late 1980s about 50 percent of both the young and middle aged now regarded themselves as Republicans. Older whites, who most reflected historic Democratic strength grounded in Reconstruction and the Great Depression, were the most resistant to change. A majority of older whites in the 1980s backed the Democrats, and the GOP sometimes found itself outnumbered in this age group by a two-to-one margin. The older white generation finally moved more towards the GOP in the 1990s, thereby bringing their partisan views more into line with young and middle aged whites, and explaining much of the Republican gains among whites after 1990.

On balance, partisan change has been particularly dramatic among young whites, as Republicans gained 20 percentage points and Democrats lost 17 points between 1981 and 1996. By 1996 Republicans outnumbered Democrats among young adults by well over a two-to-one margin, as 65 percent of whites under 35 called themselves Republicans and only 27 percent identified with the Democratic label. Republican gains among whites has had a profound impact on the racial composition of Mississippi's Democratic party, as African-Americans comprise a growing percentage of each age group of Democrats. Comprising only 37 percent of middle aged and 28 percent of older Democrats in 1981, by 1996 African-Americans constituted 54 percent and 50 percent of the transformed Democratic party. Particularly striking were the effects of racial shifts among young Democrats under 35. A true biracial party in 1981, when blacks made up 47 percent and whites 53 percent of young Democrats, was by the mid-1990s in danger of becoming viewed as "the black" party, as African-Americans began to comprise over 60 percent of younger generation Democrats.

In addition to declining numbers, Mississippi Democrats face the growing problem of turning out their supporters to vote. As expected given their generally lower socioeconomic status levels compared to Republicans, Democrats have consistently been less likely to vote than Republicans (table 7). Most alarming for Democrats, though, has been the widening of this gap in the 1990s. Whereas in the 1980s Democratic identifiers were generally only 10 percent less likely to vote than the Republicans, this Democratic disadvantage has widened to a 20 percent or greater margin in the 1990s.

Some political observers have spotted low turnout in black majority counties in the Mississippi Delta and concluded that depressed turnout among only one racial group has hurt Democratic candidates, yet we find that turnout decreased in the 1990s among Democrats of both races. Before 1990, turnout exceeded 40 percent for whites in four of five years and for blacks in three of the years. Beginning in 1990, turnout fell below 40 percent for both whites and blacks in three of the four years. Indeed, declining turnout hurt white Democrats at least as much as African-Americans to such an extent that the African-American proportion of likely Democratic voters was generally as high in the 1990s as it was among all Democratic party identifiers. An alternative explanation for declining turnout-- a conservative Democratic revolt against the increasingly liberal nature of the state party-- also receives little evidential support. Turnout dropped fairly equally among Democrats of all ideological groups from 1981 to 1994--from 51 percent to 33 percent among liberals and moderates, and from 52 percent to 36 percent among conservatives (Shaffer and Johnson 1995).

Instead, Democrats seem to be having an increasingly difficult time at enticing young, uneducated males to vote. Consistently more likely than women to vote up until 1992, by 1994 only 26 percent of Democratic men were likely voters compared to 38 percent of women. A turnout gap between high school dropouts and those with at least some college education of less than 20 percent before 1986 swelled to the 30s starting in 1986. By 1994 only 23 percent of Democratic high school dropouts were likely voters, compared to 50 percent of the college educated. Regarding age, a turnout gap between young and old Democrats in the early 1980s of less than 10 percent magnified to about 15 percent by the late 1980s. In 1994 only 24 percent of Democrats under 35 were likely voters, compared to 40 percent of those over 55. Hence, Democrats appear to face rising disillusionment among young, uneducated men, whose fathers were able to provide satisfactory living conditions for their families, but who envision little hope in today's changed economy. Disillusioned with political leaders of all stripes but lacking the political awareness to devise a consistent ideological philosophy, many of these "embattled" workers have neither the time nor the enthusiasm to even exercise the most basic responsibility of citizenship (Shaffer and Johnson 1995).

IMPLICATIONS FOR PUBLIC POLICY AND ELECTORAL POLITICS

Republican gains in the public's partisan identification ties have been reflected in electoral victories. The GOP has carried Mississippi in all four of the last presidential elections, and the last time that a Democrat has won a U.S. Senate election was in 1982. With the upset of two Democratic incumbents in the 1991 state elections by GOP candidates, Kirk Fordice for governor and Eddie Briggs for lieutenant governor, and the steady rise of Republican representation in the state legislature from less than one percent in 1970 to 27 percent in 1994, state Republicans eagerly awaited the 1995 state elections when all statewide offices and legislative seats would be up for grabs. We now turn to an examination of public policy controversies in the first Fordice administration, which illustrate the Sundquist (1973) theme about the importance of the actions of public officials in shaping the outcomes of realigning eras.

Despite Republican gains, state polls consistently showed strong public support for an activist government that would improve the quality of lives for Mississippi families through increased state and local government spending on such important domestic programs as education, health care, and highways. After a tax cut in 1979, the legislature raised taxes in 1982 for a landmark Education Reform Act and in 1987 for a major Highway Bill. The 1990-1991 recession necessitated unpopular spending cuts, however, prompting public repudiation of the Democratic governor (Ray Mabus) and lieutenant governor in the 1991 elections. The threat of a four-day school week in 1992 generated considerable public support for a one cent increase in the sales tax, which was enacted over the veto of the first-year Republican governor who had campaigned for a tax cut. The legislature's earmarking of over 90 percent of the new revenue for the popular program of education at all levels (elementary, secondary, and higher) during an economically distressed period prevented this from becoming a realigning issue. While over 80 percent of Democratic legislators voted to override the governor's veto of the tax increase, so too did many Republican lawmakers (table 8). Fordice proceeded to muddy the water by publicly attacking the "pseudo-Republicans" who had failed to support him and threatening to campaign against them, rather than concentrating on the larger numbers of pro-tax Democrats.

As economic times improved with a revived national economy and the advent of casino gambling, Fordice urged the legislature to pass an income tax cut. Early in the election year of 1995, the tax cut easily passed the senate with 74 percent of white Democrats joining with all of the Republicans, thereby innoculating themselves against the charge that they were "tax-and-spend" liberals. In the house, many Democrats backed a "fairer" tax cut, reducing the sales tax on groceries, rather than cutting a tax that only benefitted those wealthy enough to pay an income tax. Sharp ideological differences were quite evident between Republicans and African-American Democrats, as over 90 percent of GOP lawmakers voted for the income tax cut and against the sales tax cut (which was also being championed by the progressive Democratic nominee for governor, Secretary of State Dick Molpus), while 89 percent of black lawmakers backed the sales tax cut and all opposed the income tax cut. The dominant house faction, white Democrats, who comprised about half of the lawmaking body, innoculated themselves against the "tax-and-spend" label, as had their senate counterparts. Nearly 90 percent voted for at least one of these tax cuts, with one-third of these backing both tax cuts, and the remaining two-thirds split equally between those favoring only the sales tax cut and those favoring only the income tax cut. Both tax measures received a majority vote but failed to achieve the 60 percent margin required under the constitution for tax measures, thereby killing any tax cut.

State polls had also shown tremendous public concern over crime, and support for tougher measures such as longer prison sentences and retaining the death penalty. In 1994 and 1995 the legislature enacted a massive prison construction program, required inmates to work in order to receive amenities such as television, and passed Megan's Law providing for community notification when sex offenders were released. Republicans, as the more conservative party, took the lead in being "tough on crime." Over 90 percent, for example, backed a truth-in-sentencing bill requiring inmates to serve at least 85 percent of their sentences (table 9). White Democrats also responded to public sentiment and innoculated themselves against the campaign charge of being "soft on crime." About 90 percent backed truth-in-sentencing, for instance. Indeed, in the house, 82 percent of white Democrats and 91 percent of Republicans voted to permit judges to impose caning, a vote symbolizing public resentment over crime (the bill then died in a senate committee). The most liberal faction of the legislature, African-American Democrats (elected from majority black districts) generally opposed such anti-crime measures that they viewed as punitive rather than preventive or rehabilitative.

Issues less salient to average Mississippians were privatization and affirmative action, though both also exhibited tremendous ideological polarization between Republicans and black Democrats with white Democrats holding the balance of power. As Republicans overwhelmingly backed continuing Fordice's pilot program to privatize child support collections in two counties and black Democrats overwhelmingly opposed it, white Democrats narrowly split in favor of continuing it as proponents and opponents explained their votes to constituents by selectively presenting the pros or cons of the issue (high startup cost, loss of public jobs versus higher collection rates and improved computerization with privatization). Particularly interesting was Fordice's veto of a telecommunications conference and training center bond bill that included the usual 20 percent set-aside for African-Americans and women. Proponents of affirmative action quickly shifted the focus of public debate by stressing that this was an "economic development, pro-business" bill, and that Fordice's veto was election-year politics since he had signed similar bills in previous years. Maintaining party unity on an issue of vital interest to African-Americans, over three-fourths of white Democrats joined with all black lawmakers to successfully override Fordice's veto. Indeed, even one-fourth of Republicans ended up backing the override.

While adroit white Democrats maneuvered to take the popular positions of supporting education and being tough on crime, Republicans proceeded to violate Ronald Reagan's 11th commandment and started acting like Democrats had historically acted by attacking each other. While Fordice was on a hunting safari in Africa, acting governor Eddie Briggs expanded Fordice's call for a special legislative session to include streamlining death penalty appeals, a move attacked by Fordice's staff as a power grab. In early 1995 Fordice threatened to back a more conservative primary challenge to Briggs' renomination, prompting the moderate conservative lieutenant governor to threaten to run for governor as an Independent. Fearing a party split that would elect Democrat Molpus governor, both Republicans "publicly" made up and were renominated to their respective offices.

Stunned in the face of GOP gains (symbolized by the 1994 tidal wave), Democrats braced for a bitter fight in 1995. Lawmakers moved the candidate filing deadlines for the August primaries up to March 1 when the legislative session was only half over, thereby reducing the number of challengers likely to oppose them. White Democrats, whose ideologically eclectic roll call records illustrated their essentially moderate orientations, faced the greatest electoral pressure. Thirty-five percent of white Democrats in the house and 50 percent in the senate faced a primary challenge, compared to only 10-20 percent of black Democrats and 22-31 percent of Republicans (senate/house respectively), as such moderates faced possible attack from either end of the ideological spectrum. If surviving the fight for their own party's renomination, the typical white Democrat then faced a Republican opponent. Fifty-two percent of white Democrats in the house and 70 percent in the senate faced a general election foe, compared to only 8-40 percent of black Democrats and 28-61 percent of Republicans (house/senate respectively). Battered from the left and right, a few white Democrats sought retirement. "Only" 93 percent of white Democrats in the house and 87 percent in the senate sought re-election, compared to 95-100 percent of other lawmakers.

Meanwhile, Fordice backers focused their attention on re-electing the first GOP governor since Reconstruction, airing non-ideological "feel good" commercials about the state's economic recovery and building up a massive campaign warchest that could have otherwise gone to helping other GOP candidates. A prominent conservative Republican columnist blasted Lieutenant Governor Eddie Briggs for failing to "completely" back the pro-life movement and the conservative Fordice agenda, while challenger Ronnie Musgrove, a moderate Democrat who projected a conservative image (despite education association backing as chairman of the senate education committee), won the right-to-life endorsement. Other progressive-to-moderate Democratic candidates for statewide offices included: a liberal PhD and college professor, Eric Clark, running for Secretary of State; incumbent Attorney General Mike Moore, a leader of the states' lawsuits against the tobacco companies, who had fought against the abuse of elderly in nursing homes, and worked with at-risk juveniles; and Treasurer Marshall Bennett, who had sought (unsuccessfully) to unseat Democratic Congressman Mike Parker (before his switch to the GOP) due to Parker's moderate conservative record that he deemed too conservative.

Democrats stressed such issues as their support for "locking up criminals" and fiscal responsibility, showed their families as an illustration of how they were living "family values," and greatly outspent their GOP opponents who were basically invisible. One prominent progressive columnist, for instance, stressed how state representative Eric Clark had driven long distances home every night (instead of partying in the state capital) to be with his wife and child, a victim of Downs' syndrome. Such Democratic positioning to take the "conservative" position on crime and family values issues was particularly wise given a post-election poll illustrating that majorities of voters felt that the GOP would do a better job in "reducing the crime rate" and "preserving traditional values (table 10)."

Legislative Democrats avoided self-destruction, as 92 percent of white Democrats in the house and 100 percent in the senate won renomination, a success rate comparable to the 94-100 percent rates of black Democrats and Republicans. On election night Republicans found little to cheer about. Fordice was re-elected with 56 percent, but Musgrove upset Briggs with 52 percent, and Democrats won all other statewide offices with at least 60 percent of the vote. A pre-election pledge to gain control of the state senate fell far short, as Republicans actually lost two seats to pro-education Democrats, while gaining only two seats in the house. Realignment had been stifled by the adroit actions of Democrats and the counterproductive campaign tactics of Republicans.

Democratic staying power was particularly impressive in the face of a major effort by the state Republican Party, with considerable financial support from the national party, to exploit public distrust of politicians. State party mailouts sent to voters in many legislative districts portrayed Democratic incumbents as pigs feeding at the public trough, and as criminals who should be jailed for raising taxes so much. But efforts to exploit public cynicism and a perceived "conservative" mood in the state fizzled, as the power of incumbency continued to sway voters.

The race, partisanship, and ideological orientations of incumbent lawmakers generally exerted no significant effect on election outcomes. For instance, on the house side a respectable 94 percent of white Democrats running in November were victorious, compared to 97 percent of Republicans and 100 percent of black Democrats. On the senate side, 95 percent of white Democrats were re-elected in November, compared to 82 percent of Republicans and 100 percent of black Democrats. Renomination and re-election success in the house was unrelated to lawmakers' liberal or conservative voting orientations on social issues and taxes, suggesting that incumbents had been effective in representing the basic values of their constituents. Indeed, fiscal and social conservatives in the senate were somewhat less successful in winning the general election than were liberals and moderates (in the range of 85% probabilities of victory versus 98%), consistent with the upsets of key Republican senators by pro-education Democrats.

The persisting power of incumbency in Mississippi elections helps to put the one prominent Republican victory in perspective--Governor Fordice's reelection defeat of Dick Molpus. Presiding over a booming economy, Fordice began 1995 with a 56-44 percent edge over the popular pro-education Democratic Secretary of State. Instead of building up his name recognition with positive ads that introduced himself to more uninformed voters, Molpus went negative and repeatedly attacked Fordice's "extreme" conservatism and even characterized his differences with Fordice as "irreconcilable differences," a perceived jab at the governor's reported marital problems of two years before. In the face of the governor's "Feel Good" television ads, Molpus' negatives in polls went up, and Fordice's lead edged up to a 60-40 margin. Only a Fordice campaign proposal to amend the constitution through the initiative process to permit parents to control the local school systems and to institute merit pay gave the Molpus forces a renewed opportunity to brand the governor as an opponent of public education. With television ads featuring education leaders and community leaders, rather than Dick Molpus, attacking Fordice's education stance, Molpus regained the ground he had lost, and ended the campaign with 44 percent of the vote. Yet some political observers gave the Fordice education initiative credit for emphasizing an issue area usually stressed by Democrats (though with very ideologically different proposals). Combined with the GOP state legislative support for education spending cited earlier, perhaps the Fordice education plan helped to minimize the party gap in public perception of which party could do a better job "supporting education" (table 10).

Another challenge facing state Democrats is the growing racial polarization of the two parties, and the concern that whites will view them as "the black party." Among the many Fordice initiatives that were turned back by the Democratic-controlled state legislature in 1996 was his appointment of four highly qualified white men to the four vacancies on the state College Board. Democratic Lieutenant Governor Ronnie Musgrove had previously increased the racial diversity of committee chairs through his appointments, which included African-American Democrat Hillman Frazier as chair of the Universities and Colleges committee, a committee entrusted with confirming Board nominees. To consider Fordice's nominees, Frazier appointed a five-member subcommittee that included three African-American Democrats, one of whom was the state Democratic party chairman, Johnnie Walls. The subcommittee, seeking more racial and gender diversity among the nominees who would replace a retiring woman and black Board member, permitted all four Fordice nominees to die in committee. Yet while these events seemed to justify the public perception of the Democratic party as the better party in "protecting the interests of African-Americans," they appeared to generate little "white backlash." Instead, the state press vigorously attacked Fordice for being "insensitive" to the "diverse" interests of and people in the state, and most average citizens showed little interest in the political skirmish (though some letters to the editor did react against the liberal advocacy of the press).

Democratic retention of the great majority of state and local offices in Mississippi contrasted with GOP gains in Congress across the South, especially in 1994. Gaining a majority of congressional seats in the South, Republicans displaced white Democrats as the largest of the three factions, and black Democrats remained the smallest faction. Despite these differing outcomes, an examination of ideological pressure group ratings of Congress members illustrates similar policy splits regionwide as exists in the Mississippi state legislature. Dividing the ideology scale into five categories with 20 points in each one illustrates the ideological polarization of Republicans and African-American Democrats. Over 90 percent of Republicans fall into the most conservative category, while nearly 90 percent of African-American Democrats fall into the most liberal category (table 11). White Democrats occupy a middle ground with a mean score of 44, as fully 70 percent in 1994 fell into the "moderate" or "moderate liberal" category.

Perhaps some Republican gains at the expense of white Democrats have been because of the latter moving too far to the left. White Democrats in Congress ideologically are somewhat closer to African-American Democrats than to Republicans (33 versus 46 points), though their districts appear philosophically closer to Republican-held districts (in 1992 presidential vote and in black population presence). The slightly left-of-center 44 mean score of white Democrats is a far cry from this faction's behavior as recently as 1972. In 1972 such southern Democratic senators as John Stennis, Sam Ervin, and Herman Talmadge helped produce a mean southern Democratic senate score of 74 percent conservative. Not only have many conservative white Democrats left the political scene, but within the Democratic party they have been replaced by moderate whites or liberal blacks. Indeed, the leftward drift of the Democratic congressional party in the South may even be accelerating, as white Democrats elected most recently have more liberal voting records compared to those elected before 1990.

CONCLUSIONS

Partisan change in the Deep South state of Mississippi appears to be very similar to changes transforming the South as a whole. The Vital South (Black and Black 1992) has indeed been transformed from a Democratic bastion to a Republican electoral majority, as reflected in the rising GOP identification of white Mississippians. Historic home of southern whites fighting to preserve a segregated society, today's Democratic party is experiencing a growing African-American presence which is displacing moderate and conservative whites, who are switching to the GOP. Yet race is not the only motivating force for "white flight," as a conservative self-identification more generally, positive reactions to GOP presidents and governors and negative reactions to some prominent Democratic officeholders also generate party switching.

Sundquist's (1973) Dynamics of the Party System reminds us of the critical roles that political leaders play in ushering in a new party era or preventing its emergence. Both parties nationally responded to the Progressive movement, thereby preventing this issue from supplanting the party system of 1896. So too have many Mississippi Democratic officeholders effectively responded to the public mood by being conservative on many social issues while progressive on key domestic issues like education, thereby blunting further GOP electoral gains. Such Mississippi Democrats have successfully coped with the public perception of Republicans generally as being better able to fight crime and preserve traditional values by basically arguing that they are conservative "Mississippi Democrats," not "liberal national Democrats."

Democratic domination of state and local offices in the face of growing parity in the public's partisan identification is an illustration of Harold Stanley's (1988) theme that both processes of realignment and dealignment characterize the contemporary South. In short, Mississippi is a long way from helping to usher in a Solid "Republican" South. Moderate to conservative Democrats pragmatically pursuing ideologically eclectic policies and completely divorcing themselves from the liberal national party can still defeat "purist" Republicans content with waging their own ideological "holy war" within their own ranks.

Abramowitz, Alan I. 1994. "Issue Evolution Reconsidered: Racial Attitudes and Partisanship in the U.S. Electorate." American Journal of Political Science 38:1-24.

Beck, Paul Allen. 1982. "Realignment Begins? The Republican Surge in Florida." American Politics Quarterly 10:421-438.

Black, Earl and Merle Black. 1992. The Vital South. Cambridge: Harvard University Press.

Burnham, Walter Dean. 1970. Critical Elections and the Mainsprings of American Politics. New York: W.W. Norton and Company.

Burnham, Walter Dean. 1996. "Critical Realignment Lives: The 1994 Earthquake." In The Clinton Presidency: First Appraisals, ed. Colin Campbell and Bert A. Rockman. Chatham, NJ:Chatham House.

Campbell, Bruce A. 1977. "Patterns of Change in the Partisan Loyalties of Native Southerners: 1952-1972." Journal of Politics 39: 730-763.

Campbell, James E. 1985. "Sources of the New Deal Realignment: The Contributions of Conversion and Mobilization to Partisan Change." Western Political Quarterly 38: 357-376.

Carmines, Edward G. and James A. Stimson. 1981. "Issue Evolution, Population Replacement, and Normal Partisan Change." American Political Science Review 75:107-118.

Cotter, Patrick R. and James G. Stovall. 1996. "Party identification and political change in Alabama: a mid-1990s update." Presented at the Citadel Symposium on Southern Politics, Charleston, S.C.

Dawson, Richard E., Kenneth Prewitt and Karen S. Dawson. 1977. Political Socialization, 2nd ed. Boston: Little, Brown.

Duncan, Phil. 1993. Politics in America, 1994. Washington D.C.: CQ Press.

Elazar, Daniel J. 1984. American Federalism: A View from the States. New York: Harper and Row.

Hadley, Charles D. and Lewis Bowman. 1995. Southern State Party Organizations and Activists, ed. Westport, Conn: Praeger.

Hurley, Patricia A. 1989. "Partisan Representation and the Failure of Realignment in the 1980s." American Journal of Political Science 33: 240-261.

Jennings, M. Kent. 1987. "Residues of a Movement: The Aging of the American Protest Generation." American Political Science Review 81: 367-382.

Key, V.O., Jr. 1955. "A Theory of Critical Elections." Journal of Politics 17: 3-18.

Key. V.O., Jr. 1959. "Secular Realignment and the Party System." Journal of Politics 21: 198-210.

Ladd, Everett C. and Charles D. Hadley. 1978. Transformations of the American Party System. New York: W.W. Norton and Company.

Lamis, Alexander P. 1990. The Two-Party South. 2nd expanded ed. New York: Oxford University Press.

Moreland, Laurence W. 1990. "The Ideological and Issue Bases of Southern Parties." In Political Parties in the Southern States, ed. Tod A. Baker, Charles D.Hadley, Robert P.Steed, and Laurence W. Moreland. New York: Praeger.

Petrocik, John R. 1987. "Realignment: New Party Coalitions and the Nationalization of the South." Journal of Politics 49: 347-375.

Shaffer, Stephen D.and Monica Johnson. 1995. "Rejection of Positive Liberalism? Exploring Partisan Realignment in the Deep South." Presented at the Annual Meeting of the American Political Science Association, Chicago.

Stanley, Harold W. 1988. "Southern Partisan Changes: Dealignment, Realignment, or Both?" Journal of Politics 50: 64-88.

Stanley, Harold W. 1996. "The Parties, the President, and the 1994 Midterm Elections." In The Clinton Presidency: First Appraisals, ed. Colin Campbell and Bert A. Rockman. Chatham, NJ: Chatham House.

Sundquist, James L. 1973. Dynamics of the Party System. Washington D.C.: Brookings Institute.

Swansbrough, Robert H. and David M. Brodsky. 1988. The South's New Politics: Realignment and Dealignment. Columbia: University of South Carolina Press.

ABSTRACT

Modern Republican gains in the South and the 1994 GOP Congressional landslide have sparked renewed interest in the possibility that we are in a realigning era. Statewide opinion polls from Mississippi over a sixteen year time frame illustrate significant GOP gains in the public's partisan identifications, especially among white conservatives. Republican gains also appear to be prompted by positive public reactions to GOP presidents and a governor and negative reactions to their Democratic counterparts. Yet adroit actions by pragmatic white Democratic officeholders and emerging Republican factionalism have hindered further GOP electoral gains. The ability of state Democrats to successfully combine conservative social values with progressive domestic policies illustrates the relevance of dealignment as well as realignment in contemporary southern politics.

TABLE 1

The Party Identification Ties of Mississippians (%)

Source: Mississippi State University SSRC telephone polls.

Note: Percentages total 100% down each column.

Source: Mississippi State University SSRC telephone polls.

Note: Percentages total 100% down each column.

The residential length question was not asked in 1986 and 1988.

TABLE 3

The Role of Ideology in Shaping Party Realignment among Whites (%)

Source: Mississippi State University SSRC telephone polls.

Note: Percentages total 100% down each column.

Question wording was "What about your political beliefs? Do you consider yourself very liberal, somewhat liberal, moderate or middle of the road, somewhat conservative, or very conservative."

TABLE 4

Democratic Factionalism and Mississippians' Ideology (%)

Source: Mississippi State University SSRC telephone polls.

Note: Percentages total 100% down each column.

TABLE 5

Executive Leadership Sources of Realignment among Whites (%)

RATING PRESIDENT EXCELLENT/GOOD RATING PRESDIENT FAIR/POOR

GEORGE BUSH BILL CLINTON GEORGE BUSH BILL CLINTON

RATING GOVERNOR EXCELLENT/GOOD RATING GOVERNOR FAIR/POOR

MABUS (D) KIRK FORDICE (R) MABUS (D) KIRK FORDICE(R)

Source: Mississippi State University SSRC telephone polls.

Note: Percentages total 100% down each column.

TABLE 6

Generational Differences in Partisanship and Race (%)

Source: Mississippi State University SSRC telephone polls.

TABLE 7

The Changing Turnout of Mississippians (%)

Source: Mississippi State University SSRC telephone polls.

Note: Predicted turnout is based on likely voter scale.

TABLE 8

Policy Orientations of State Lawmakers on Fiscal Issues (%)

STATE HOUSE OF REPRESENTATIVES MISSISSIPPI STATE SENATE

Source: Shaffer and Johnson (1995).

Note: NV indicates that no vote was held in that chamber.

A special legislative election in 1992 was required due to court-ordered redistricting, resulting in the election of more African-American lawmakers.

TABLE 9

Policy Orientations of State Lawmakers on Social Issues (%)

STATE HOUSE OF REPRESENTATIVES MISSISSIPPI STATE SENATE

Source: Shaffer and Johnson (1995).

Note: NV indicates no vote held in that chamber.

TABLE 10

Public Images of the Two Parties Among Whites and Blacks (%)

Source: The April 1996 Mississippi State University SSRC telephone poll.

Questions were adapted from Cotter and Stovall (1996).

Note: Percentages total 100% across each row.

Question Wording is "Which of the two parties-Democratic or Republican-do you think would do the best job in the following areas?"

TABLE 11

Orientation of U.S. House Factions in the South, 1993-94 (%)

Source: Shaffer and Johnson (1995).

Note: Mean ideology = (ACU score + (100-ADA score)) / 2, averaged for the three most recent years. The ACU is the American Conservative Union, and the ADA is the Americans for Democratic Action.

NA indicates not applicable.